In vivo PET imaging of mutant huntingtin using [11C]CHDI-180R as candidate marker in a mouse model of Huntington’s Disease (#4) 21
Data-driven, blood-free quantification of positron emission tomography radiotracers with irreversible kinetics using the publicly-available Source-to-Target Automatic Rotating Estimation (STARE): Validation with [18F]FDG data and simulations (#10) 22
Novel data-driven method captures spatio-temporal patterns of neurodegeneration in Parkinson’s disease (#12) 23
A high-resolution in vivo atlas of the human brain’s benzodiazepine binding site of GABAA receptors (#13) 24
An algorithm for classifying neurotransmitter signals from future ultra-high sensitivity dynamic PET (#22) 25
In vivo neuroimaging of histone deacetylases in multiple sclerosis using [11C]Martinostat (#24) 26
Optimized synthesis of [18F]FPEB and preliminary neuroimaging of mGlur5 in transgenic mouse models (#25) 27
Estimating PET partial volume full-width-half-maximum directly from human data (#30) 28
Repurposing novel PET neuroimaging radioligands in xenograft mouse models of cancer (#32) 29
[18F]MNI-1054, a novel PET ligand for lysine-specific histone demethylase 1A (LSD1): First-in-human validation including radiation dosimetry, kinetic modeling and test-retest variability (#39) 30
Towards the development of PET tracers for the imaging of melatonin receptors (#43) 31
Distinct spatio-temporal patterns of putuminal dopamine processing in Parkinson’s disease: A multi-tracer positron emission tomography study (#47) 33
Ketanserin can block but not displace [11C]Cimbi-36 binding in the pig brain (#68) 35
Alpha-synuclein preformed fibril (PFF)-triggered synucleinopathy recapitulates neurochemical features of human PD: a PET study in rats (#71) 36
Evaluation of MAO-B and TSPO PET radiotracers in a lipopolysaccharide rat model of neuroinflammation (#72) 37
Clinical measures of upper motor neuron burden strongly associate with neuroimaging in amyotrophic lateral sclerosis (#73) 38
Development of novel PET ligands to image the receptor interacting protein kinase 1 (#77) 39
[18F]NOS PET measurement of in vivo neuroinflammation in Parkinson’s disease and healthy humans (#93) 40
Functional changes in Alzheimer’s disease evaluated by multimodality PET/MRI imaging (#96) 41
PET imaging of [11C]NCGG401 for colony stimulating factor 1 receptor (#111) 42
Development of 18F-labeled radioligands for the GluN2B subunits of NMDA receptors: Synthesis and evaluation in non-human primates (#123) 43
Validation of PET-compatible chemogenetic tools in squirrel monkeys (#125) 44
Increased cerebral glucose consumption during hypoglycemia in obese patients measured using dynamic bolus-injection 18F-FDG PET/MR during hyperinsulinemic euglycemic and hypoglycemic clamp (#138) 44
Dopamine release after fear conditioning as measured using bolus-infusion11C-raclopride PET-MRI (#139) 46
Analysis of the expression of norepinephrine transporter and neuronal plasticity-related proteins in social isolation model rats (#141) 46
Tau pathology is associated with synaptic loss and altered synaptic function: a combined [18F]flortaucipir, [11C]UCB-J and magnetoencephalography study (#145) 47
Effect of 5-day regular coffee consumption and subsequent abstention on A1 adenosine receptor occupancy and availability (#156) 49
The influence of scanning time window on 18F-FP-DTBZ PET in PD (#158) 49
Sex-related differences in cerebral A1 adenosine receptor availability in the human brain (#160) 50
[18F]FDOPA PET imaging for prediction of treatment response in psychosis (#169) 51
Tetrazine-functionalised clearing agent to increase contrast in antibody imaging (#182) 52
Measure of metabolic and electric rat brain changes induced by kainic acid, an EEG and 18FDG-PET studio (#184) 53
Simultaneous PET/fMRI reveals differential D1 and D2 receptor trafficking induced by amphetamine in NHP (#185) 54
In vivo evidence for disrupted association between synaptic and glutamatergic markers in depression – A combined [11C]UCB-J and [18F]FPEB study (#188) 55
Translocator protein in occupational post-traumatic stress disorder: Preliminary findings using the [18F]FEPPA PET radioligand (#189) 56
The PET radioligands ([11C]NR2B-X; X = Me, R or S; X = SMe, S) are selective for binding to the NR2B subunit of NMDA over the s1 receptor in rat in vivo (#191) 57
Generation of a normative dopamine neuroreceptor template from [11C]PHNO PET imaging modelling population variability (#195) 58
mGluR5 as a biomarker for suicidal behavior in trauma related disorders: Evidence from in vivoPET imaging studies (#200) 59
Neuroinflammation in epilepsy: A systematic review of microglial activation imaging studies (#203) 60
Towards an opioid receptor atlas of the human brain (#206) 61
DaTscan-based progression subtypes for Parkinson’s disease (#213) 62
A single session of theta burst stimulation alters metabolic activity in the core depression network (#230) 63
MRI quantification of brain oxygenation and relationship with cerebrovascular reactivity in Moyamoya disease using simultaneous [15O]-water PET/MRI (#242) 64
In vivo assessment of astroglial activation in cognitively impaired subjects using 11C-BU99008 PET and its relationship with amyloid load (#247) 66
Relations of EEG and α2-adrenoceptor availability in patients with Parkinson’s disease (#252) 66
Electroconvulsive stimulation differentially affects [11C]MDL100907 binding to cortical and subcortical 5HT2A receptors in porcine brain (#253) 67
Network integration derived non-invasive biomarkers for early prediction of Alzheimer’s disease (#254) 69
Sucrose lowers µ-opioid and D2/3 dopamine receptor availability of porcine brain (#256) 70
Nicotine attenuates age-related memory and learning impairment in D-galactose-induced senescence in mice (#258) 71
Anesthetics differentially influence [11C]MDL100907 binding to 5HT2A receptors in porcine brain (#259) 72
Transcranial photoacoustic imaging of NMDA-evoked focal circuit dynamics in rat hippocampus (#260) 74
Intrinsic connectivity of the human brain provides scaffold for tau aggregation in Alzheimer’s disease (#262) 75
Correspondence between gene expression and neurotransmitter receptor and transporter density in the human cortex (#275) 76
A possible link between Fragile X mental retardation protein and metabotropic glutatmate receptor subtype 5 in men with fragile X syndrome (#277) 79
A translational investigation of morphine-induced neuroimmune signaling: Implications for opioid use disorder (#278) 81
[124I]IBETA PET/CT studies in Alzheimer’s disease 5XFAD mouse model of β-amyloid (Aβ) Plaques (#279) 82
Examining kinetic spectrum of extracerebral signal and its contributions to reference regions of 18F-MK6240 PET (#280) 83
The relationship between glutamate, dopamine receptors, dopamine release and cortical grey matter: A simultaneous PET-MR study (#281) 84
Peripheral red blood cell (RBC) docosahexaenoic acid (DHA) and serum triglyceride levels influence [11C]PBR28 binding to TSPO in the brain (#282) 85
Choroid plexus enlargement is associated with neuroinflammation and reduction of blood brain barrier permeability in depression (#283) 86
Evidence of blood-to-cerebrospinal fluid alterations in traumatic brain injury (#284) 87
Comparison of Lewy body distribution and monoamine oxidase A in anterior cingulate of postmortem Parkinson’s disease brains (#286) 88
A multiple regression modelling approach to investigate the coupling between [18F]fluorodeoxyglucose positron emission tomography and resting-state functional MRI (#287) 89
Comparative sensitivity of PET radioligands to partial inhibition of P-glycoprotein at the blood-brain barrier (#288) 90
Multimodal investigation of the synaptic and metabolic basis of low-frequency oscillations in the human brain: A [11C]UCB-J, [18F]BCPP-EF and resting state fMRI study (#289) 92
Receptor distribution associations to mRNA gene expression patterns in the human cerebral cortex (#290) 92
Altered neuroepigenetics in autism spectrum disorder: a [11C]Martinostat PET brain imaging study (#292) 93
Preclinical imaging of [11C]colchicine, [11C]verubulin, and [11C]HD-800 for imaging microtubules (#293) 94
Task-specific dynamics of dopamine synthesis during monetary gain and loss (#294) 95
Interdependent adaptations of glucose metabolism and functional connectivity elicited through learning (#295) 96
Harmonization of neuroimages acquired on two brain PET scanners (#296) 97
PET simulations investigating the effect of off-target binding in [F-18]MK-6240 PET scans for the detection of early-stage Alzheimer’s disease (#298) 98
Visual memory scores are associated with lateralization of tau in the medial temporal lobe (#299) 99
Investigating tauopathy in military occupational blast: a [18F]flortaucipir positron emission tomography study in Canadian armed forces members (#302) 101
In vivo imaging of microglial activation by positron emission tomography with 11C-ER176 in a mouse model for post-infectious autoimmune encephalitis (#304) 102
Development of a novel carbon-11 radiotracer for brain PET imaging of sulfonylurea receptors 1 (#305) 103
Imaging the endogenous opioid response to acute cannabis smoking in humans (#307) 103
Comparative analysis of immunohistochemical and autoradiographic images of Tau in postmortem human Alzheimer’s disease brain (#308) 104
Evaluation in rat, monkey, and humans of [18F]PF-06445974, a PET radioligand for phosphodiesterase 4B (#309) 105
Perfusion heterogeneity in extra-tumoural brain in grades III and IV glioma measured using positron emission tomography with radiolabelled water (#311) 106
Expression of transduced estrogen receptor-α fragment fusion protein in rhesus monkey brain using [18F]FES as a PET reporter probe (#312) 108
Quantification of cerebral nicotinic α7 acetylcholine receptors (α7 nAChRs) under gastric stimulation of the vagus nerve in piglets (#313) 109
Regional analysis demonstrates asymmetric binding with [11C]-(R)-PK11195 positron emission tomography in normal brain (#315) 110
Impact of cerebral blood flow and amyloid load on SUVR bias (#316) 111
[18F]Flotaza and [124I]IBETA, two new β-amyloid PET imaging agents in post-mortem human Alzheimer’s disease brain (#317) 112
Multi-tracer PET joint correlation analysis reveals disease-specific patterns in both Parkinson’s disease and asymptomatic LRRK2 mutation carriers compared to healthy controls (#318) 114
Quantification of cerebral blood flow by a non-invasive hybrid PET/MR method for extracting the 15O-water image-derived input function free of partial volume errors (#319) 115
Synaptic density measured by [18F]SynVesT-1 PET in ageing mice (#321) 116
The positron emission tomography brain imaging data structure (PET-BIDS) extension: A new standard for sharing PET data (#322) 117
Development of a novel PET ligand for receptor-interacting protein kinase 1 in brain (#323) 119
Positron emission tomography imaging of alpha-synuclein?In vitro and in vivo evaluation of MODAG-005 (#324) 120
Simultaneous Multifactor Bayesian Analysis (SiMBA) of PET Time Activity Curve Data (#325) 121
Investigating fatty acid amide hydrolase in mild traumatic brain injury; A PET study of [11C]CURB in occupational posttraumatic stress disorder (#327) 122
5HT1A receptor co-expression based polygenic score associates with 5HT1A predicted binding potential in frontal regions (#328) 123
[124I]IPPI: Molecular modeling and autoradiographic study for binding to Taupathies (#329) 124
Increased Tau, Aβ amyloid, and monoamine oxidase-A in post-mortem human Alzheimer’s disease anterior cingulate (#330) 125
A three-factor model of common early-onset psychiatric disorders: Temperament, adversity, and dopamine (#331) 127
[18F]2-fluoro-2-deoxy-sorbitol PET imaging for quantitative estimation of blood-brain barrier permeability in vivo (#332) 127
Relative cerebral blood flow, amyloid burden and cognition in individuals with subjetive cognitive decline are closely associated (#333) 129
Genetically identical twins show comparable tau PET load and spatial distribution (#334) 130
Assessing neurobiological correlates of [11C]WAY-100635 two-tissue compartment model parameters using SiMBA (#335) 131
[18F]FLUDA – A novel radiotracer for PET imaging of the adenosine A2A receptor (A2AR) (#337) 132
Development of a non-invasive PET/MRI method for quantifying cerebral glucose kinetics (#338) 133
Positron emission tomography with [18F]ROStrace reveals increased oxidative stress in a mouse model of alpha synuclein aggregation (#339) 134
Visual memory test equal to commonly used verbal memory test in predicting Tau in the medial temporal lobe (#340) 135
3D reconstruction of 20 neurotransmitter receptor atlases from 2D autoradiographs (#341) 136
Neuroimaging of M4 muscarinic acetylcholine receptors using [11C]MK-6884 in rhesus macaques (#342) 138
Population-based input function for [11C]PBR28 quantification in non-human primates (#343) 139
[11C]MC1 has adequate sensitivity to measure low density cyclooxygenase 2 (COX-2) in healthy human brain (#344) 140
Combining plasma p-Tau181 and p-Tau231 enhances Alzheimer’s disease in vivo classification (#346) 141
Disrupted association between Mu-opioid receptor levels and resting state activity in patients with schizophrenia: Multimodal imaging study with [11C]-carfentanil PET and resting state fMRI (#347) 142
Individual-level molecular connectivity of GABAA receptors: assessing the similarity of [11C]Ro15-4513 kinetics across brain regions (#348) 143
The pandemic brain: neuroinflammation in healthy, non-infected individuals during the COVID-19 pandemic (#349) 144
Imaging nociceptive opioid peptide (NOP) receptors in alcohol use disorders (AUD) with [11C]NOP-1A and PET: Findings from a second cohort (#350) 145
In response to social acceptance, nucleus accumbens mu opioid receptor activation is associated with increased self-esteem and positive mood (#351) 146
[18F]Nifene binding to α4β2* nicotinic acetylcholinergic receptors is reduced in human hippocampus of postmortem Alzheimer’s disease brains (#352) 146
Imaging synaptic density in aging and frontotemporal dementia (#353) 148
Associations between markers of synaptic dysfunction and tau accumulation, glial activation, and neurodegeneration in Alzheimer’s disease (#354) 149
Detecting two-task dopamine release via residual analysis (#355) 149
Using personalized longitudinal brain models to identify neurotransmitter receptor-mediated alterations in Alzheimer’s disease (#357) 151
Dosimetry and biodistribution of the novel PET radioligand (R)-[11C]Me-NB1 specific to the GluN2B subunit of the N-methyl-D-aspartate receptor (#358) 152
Longitudinal in vivo PET imaging of human ESC-derived dopamine neurons in minipig (#359) 154
Differential associations between neocortical tau pathology and blood flow with cognitive deficits in early-onset vs late-onset Alzheimer’s disease (#360) 154
12 days of sucrose consumption reduces 3H-UCB-J binding in the caudate and prefrontal cortex of healthy Göttingen minipigs (#361) 155
Longitudinal tau PET using [18F]flortaucipir: Comparison of (semi)quantitative parameters (#362) 156
From genes to networks: In-vivo CRISPR/Cas9-induced VMAT2 knockdown exerts functional connectome changes in the rat brain (#363) 157
Development of voxel-level EC50 Images for use in CNS Drug Development (#364) 158
Agonist-induced µ-opioid receptor desensitization increases radiotracer binding: A positron emission tomography study in the mice brain (#365) 159
Age and BMI associations with brain NET availability using [11C]MRB (#366) 160
Measuring synaptic density and the dopamine transporter in Parkinson’s disease: a PET imaging study with11C-UCB-J and18F-FE-PE2I (#367) 161
Longitudinal test-retest reproducibility of 11C-UCB-J, a PET tracer for synaptic density imaging (#368) 162
Simultaneous optogenetic [18F]FDG-fPET/fMRI to study brain circuits in rats (#369) 163
Manifold component analysis and its use in evaluating stage-wise cortical tau propagation measured with 18F-MK-6240 PET (#370) 164
Evaluation of brain structure and function in currently depressed adults with a history of childhood trauma (#371) 165
Impact of direct-4D PET image reconstruction on within- and between-subject variance of [11C]UCB-J in Parkinson’s disease (#373) 167
Initial validation of reconstruction parameters in [18F]FDG PET brain images aiming scan time reduction (#374) 168
Investigating autism spectrum disorder with synaptic density PET imaging (#376) 169
Centrum semiovale as reference region – An evaluation of methods to quantify [18F]FPEB PET binding data in man (#377) 169
Comparison of a simultaneous estimation method for quantifying [18F]FEPPA uptake to standard compartmental analysis (#378) 170
PET imaging of demyelination in traumatic brain injury with [18F]3F4AP in mice (#379) 171
Risk and resilience in a rodent model of posttraumatic stress disorder: an in vivo [18F]FPEB and positron emission tomography imaging study examining the role of metabotropic glutamate receptor 5 (#380) 172
Progression of Tau and neuroinflammation PET are independently associated with structural network reorganization in Alzheimer’s disease (#381) 173
A reference tissue forward model for improved PET accuracy using within-scan displacement studies (#382) 174
First-in-human use of [11C]CPPC with PET for imaging the macrophage colony stimulating factor 1 receptor in healthy brain (#383) 176
Interaction between vascular risk and Alzheimer’s disease pathology boosts neurodegeneration and cognitive decline in cognitively unimpaired individuals (#384) 176
APOE isoforms differentially modulate the associations between regional tau deposition and neuroinflammation in Alzheimer’s disease (#385) 177
Employing simultaneous (EEG-)PET-MRI to map arousal-induced hemodynamic and metabolic dynamics (#387) 178
Biomarker modeling of Alzheimer’s disease using PET-based in vivo Braak staging (#388) 180
Neuro-immune signatures in chronic low back pain subtypes (#389) 181
Tau accumulation using [18F]MK6240 PET is associated with increase in executive dysfunction in prodromal AD (#390) 182
Neuroimaging VMAT2 in Parkinson’s disease with rapid eye movement sleep behaviour disorder (#391) 183
Olfactory impairment is related to tau pathology and neuroinflammation in Alzheimer’s disease (#392) 184
Associations between neutrophils and amyloid deposition in the Alzheimer’s disease spectrum (#393) 185
Verbal Fluency associated with tau accumulation and not amyloid deposition in the Alzheimer’s disease spectrum (#395) 186
Reducing Model Bias in Measurement of Dopamine Response to Behavioral Challenge (#396) 187
Single dose of cocaine alters synaptic vesicle glycoprotein 2A density in adolescent rats (#397) 188
Protective role of β-amyloid revealed by PET of [11C]PiB and [18F]FDG in Alzheimer’s disease (#398) 188
Using a support vector machine to identify signatures of different p-tau CSF species in incipient Alzheimer’s disease (#399) 189
Ex-vivo analysis of metabotropic glutamate receptor type 5 hippocampal abnormalities in epileptogenic foci (#400) 190
Covid-19 pandemic: Quantifying the effects of the first lockdown on behavioral and cognitive measures using TASIC (#401) 192
Amyloid-PET and free-water diffusion MRI of the white matter: a multi-center mixed cohort of small vessel disease and Alzheimer’s disease pathology (#402) 193
Evaluation of [18F]FR – a potential PET tracer for the diagnosis of cerebral amyloid angiopathy (#403) 194
Cognitive reserve: Evaluating the relationship between WASI-II matrix reasoning and tau accumulation using [18F]MK6240 in monolingual and bilingual individuals (#404) 195
Amyloid beta deposition and cognitive decline in Parkinson’s disease: a study of the PPMI cohort (#405) 196
[124I]IAZA: Development of a new radioiodinated PET imaging agent for human Alzheimer’s disease brain (#406) 197
Mapping the multivariate effects of amyloid, tau, and neuroinflammation on cortical thickness in AD (#408) 198
Anti-correlations between 18F-FDG PET and resting state dynamic functional connectivity: Insights into brain network variability (#3) 199
Does ebselen: A potential new lithium mimetic, decrease synaptic glutamate availability? (#9) 200
Examining the underpinnings of loudness dependence of auditory evoked potentials with positron emission tomography (#17) 201
Physiological brain activation by visual stimulation does not alter binding of the synaptic density tracer [11C]UCB-J (#26) 202
Reliability assessment of neuronal activation using fPET, BOLD and ASL obtained with simultaneous PET/MR imaging (#38) 203
Assessment of model bias upon detection of dopamine response to challenge (#50) 204
Non-displacable binding is a potential confounding factor in [11C]PBR28 PET studies (#53) 205
Transcriptome-based human cerebral cortex parcellation (#55) 206
Neuroinflammation in autism spectrum disorder: A [18F]FEPPA PET Study. (#57) 207
Development of a carbon-11 positron emission tomography pro-radiotracer for imaging the astrocyte glutamate transporter 1 (#64) 208
Amyloid (A), tau (T) and voxel-based morphometry (N) correlates of visual memory performance (#86) 210
Neuroinflammation as a predictor of total knee arthroplasty recovery (#88) 211
Reduced metabotropic glutamate receptor subtype 5 in fragile X syndrome (#92) 212
Clinical validation of [18F]EKZ-001: A novel positron emission tomography ligand for quantifying HDAC6 in the human brain (#104) 213
Fetal imaging of synaptic vesicle glycoprotein 2A using 18F-SynVesT-1 PET on the primate miniEXPLORER (#105) 215
Simultaneous multi-parameter multi-tracer estimation with dynamic neuro-PET data (#106) 216
Thalamic distributions of α4β2* subtype nicotinic acetylcholine receptors in two independently collected positron emission tomography datasets (#107) 217
Quantification and kinetic analysis of [11C]deschloroclozapine positron emission tomography imaging for designer receptors exclusively activated by designer drugs in monkey brain (#108) 218
Pretreatment brain metabolism imaging for prediction of major depressive disorder outcome (#109) 219
Parametric mapping of [11C]PBR28 brain PET imaging using spectral analysis (#112) 220
Test-retest and α2-receptor challenge with [11C]yohimbine with search for a suitable reference region (#113) 221
Design of a clinically applicable bolus-plus-constant-infusion PET imaging scheme for gold standard quantification of amyloid-beta in Alzheimer’s disease (#133) 222
Validation of longitudinal [18F]FEPPA-PET in rat subcortical stroke revealed that TSPO misses chronic remote white matter inflammation (#134) 223
Strength in numbers: Multilevel modelling of time activity curves (#140) 224
Decreased binding of [C-11]UCB-J PET in cognitively impaired (#142) 225
Evaluation of [18F]MNI-1188: A reversible monoacylglycerol lipase (MAGL) PET radiotracer in non-human primates (#143) 226
Early stopping in clinical PET studies: Save money and mSv! (#147) 227
Meta-analysis of the glial marker TSPO in psychosis revisited: Reconciling inconclusive findings of patient-control differences (#148) 228
Bayesian partial volume correction for image derived input function (#153) 229
In vivo relationship between brain arachidonic acid incorporation and cerebral metabolic rate of glucose in the healthy brain and in bipolar disorder: A pilot study (#164) 230
Quantification of vesicular monoamine transporter type 2 (VMAT2) occupancy with [18F]AV‑133 in non-human primate (#170) 231
Full quantification of brain glucose metabolism using a portable positron emission tomography (PET) camera: A preliminary report (#171) 232
Simultaneous assessment of α4β2 nicotinic acetylcholine receptor (nAChR) availability and neuronal response to rewarding food-cues in human obesity using PET-MRI (#172) 233
Determination of the 5-HT2C receptor fraction in the human hippocampus in vivo: A [11C]Cimbi-36 PET study (#173) 234
Magnitude of translocator protein binding and its association with C-reactive protein in depression: An 11C-PK11195 PET study (#174) 236
Novel software for computer-aided differential diagnosis of Parkinsonism using positron emission tomography (#175) 237
Comparison of centiloids and amyloid load for evaluation of amyloid change in Down syndrome (#176) 238
First-in-human evaluations of [11C]PS13 for imaging cyclooxygenase-1 and [11C]MC1 for imaging cyclooxygenase-2 (#180) 239
Using hybrid PET/MRI to determine if perfusion MRI has comparable sensitivity to 15O-water PET for detecting dementia-related hypoperfusion (#181) 240
Olfactory impairment is related to tau pathology and neuroinflammation in Alzheimer’s disease (#187) 242
Automated data quality control in [18F]FDOPA brain PET imaging using deep learning (#190) 243
Human blocking study to assess selectivity of [18F]FTP PET for dopamine D3 receptors (#193) 244
ApoE4 packs a punch in women: Sex-specific vulnerability for tau (#194) 245
Measurement of HDAC6 target occupancy in macaque using [18F]EKZ-001 positron emission tomography (#201) 246
Extra-striatal D2/3 receptor availability in youth at risk for addiction (#202) 247
Evaluation of [18F]APN-1607 to image tau protein in patients with Alzheimer’s disease and progressive supranuclear palsy: Test-retest and cross-sectional analysis (#204) 248
Validation of a non-invasive hybrid PET/MRI method for imaging the cerebral metabolic rate of oxygen (#209) 249
Non-invasive quantification of [11C]PBR28 binding in non-human primates (#211) 250
Type 5 metabotropic glutamate receptor availability in youth at risk for addictions: Effects of vulnerability traits and cannabis use (#217) 251
Morphine administration increases translocator protein availability in humans (#218) 252
Evaluation of PET quantitation methods using a 3D-printed anatomically accurate brain phantom (#234) 253
Alterations in cerebral blood flow in patients with secondary progressive and stem cells treated multiple sclerosis – A perfusion study with 15O-water-PET (#237) 254
Development of novel PET imaging agents for neurodegeneration (#239) 255
Evaluation of neuroinflammation in HIV patients and “elite controllers” using [11C]-PBR28 PET (#241) 256
Extension of pseudo-CT method for attenuation correction of simultaneous PET/MR brain imaging in rhesus macaques (#244) 257
Validation of kinfitr: An open-source tool for reproducible PET kinetic modelling (#246) 258
Lower thalamic dopamine D2-receptor binding and connectivity in drug-naive patients with psychosis – A combined [11C]FLB 457 PET and DTI study (#248) 260
Tau deposition assessed by [18F]MK6240 PET is associated with longitudinal decrease in grey matter density across the spectrum of Alzheimer’s disease (#250) 261
Pre-clinical evaluation of novel COX-2 PET radiotracers for imaging neuroimmune dysregulation (#255) 262
Author Index 264
Keyword Index 295
2021-01
In vivo PET imaging of mutant huntingtin using [11C]CHDI-180R as candidate marker in a mouse model of Huntington’s Disease (#4)
Daniele Bertoglio1, Jeroen Verhaeghe1, Klaudia Cybulska1, 2, Špela Korat1, 2, Alan Miranda1, Leonie Wyffels1, 2, Sigrid Stroobants1, 2, Vinod Khetarpal3, Ladislav Mrzljak3, Celia Dominguez3, Jonathan Bard3, Mette Skinbjerg3, Longbin Liu3, Ignacio Munoz-Sanjuan3 and Steven Staelens1
1Molecular Imaging Center Antwerp (MICA), University of Antwerp, Wilrijk, Belgium
2Department of Nuclear Medicine, Antwerp University Hospital, Edegem, Belgium
3CHDI Management/CHDI Foundation, Los Angeles, CA, USA
Abstract
Introduction: Huntington’s Disease (HD) is a progressive autosomal dominant neurodegenerative disorder caused by mutant huntingtin (mHTT). Although several promising therapeutic approaches aimed at decreasing cerebral mHTT levels are currently being developed, there is a major (pre)clinical need for a noninvasive marker to monitor mHTT changes. The aim of this study was to investigate the first in class mHTT PET radioligand, namely [11C]CHDI-180R, for imaging of mHTT aggregates in the Q175DN mouse model of HD.
Methods: Dynamic microPET/CT imaging was performed longitudinally in heterozygous (HET) Q175DN mice (n = 23) and wild-type (WT) littermates (n = 20) at 3, 6, 9, and 13 months of age (M). The total volume of distribution (VT) (Logan) was calculated noninvasively using an image-derived input function (IDIF) (VT (IDIF)) for brain regional and voxel-wise analyses. Post-mortem autoradiography with [3H]CHDI-180 and mHTT immunohistochemistry (IHC) were performed at each time point (n = 10/genotype at 3, 6, 9, and 13M) to confirm the in vivo findings. Receiver operating characteristic (ROC) curves for assessment of diagnostic ability and sample size calculations at desired therapeutic effects were performed.
Results: Longitudinal PET imaging could discriminate HET from WT littermates as early as 3M (striatum: HET = 0.53 ± 0.03 mL/cm3, WT = 0.49 ± 0.03 mL/cm3; +6.5%, p < 0.001). Binding in HET mice increased successively over time, resulting in the largest difference between genotypes at 13M (striatum: HET = 0.87 ± 0.05 mL/cm3, WT = 0.48 ± 0.03 mL/cm3; +82.5%, p < 0.0001) (Figure 1). Post-mortem analyses supported the in vivo PET quantification demonstrating increased binding with age in HET mice, which was associated with disease progression (r > 0.89, p < 0.0001) as measured by in vitro [3H]CHDI-180 autoradiography and mHTT IHC. No age-related or disease progressive increase in PET signal was detectable in WT littermates. ROC curves revealed good accuracy already at 3M (AUC = 0.81). Sample size calculations showed that less than 10 animals per treatment group were required to detect a 30% therapeutic effect as of 6 (Figure 2).
Conclusion: We described the first radioligand to image mHTT in the living brain. These findings indicate [11C]CHDI-180R PET imaging is a non-invasive candidate PET tracer suitable for monitoring HD progression and potentially for evaluating the efficacy of mHTT lowering therapies with future clinical application.
2021-02
Data-driven, blood-free quantification of positron emission tomography radiotracers with irreversible kinetics using the publicly-available Source-to-Target Automatic Rotating Estimation (STARE): Validation with [18F]FDG data and simulations (#10)
Elizabeth A. Bartlett1, 2, R. Todd Ogden1, 3, John Mann1, 4 and Francesca Zanderigo1, 2
1Molecular Imaging and Neuropathology Area, New York State Psychiatric Institute, New York, NY, USA
2Department of Psychiatry, Columbia University Medical Center, New York, NY, USA
3Department of Biostatistics, Mailman School of Public Health, Columbia University, New York, NY, USA
4Department of Radiology, Columbia University Medical Center, New York, NY, USA
Abstract
Introduction: Full quantification of positron emission tomography (PET) data requires arterial blood sampling. We present STARE, Source-to-Target Automatic Rotating Estimation, which quantifies the absolute net influx rate (Ki) of PET radiotracers with irreversible kinetics. STARE operates on individual-level PET data in the absence of blood measurements and without relying on a reference region. STARE is validated here in N = 69 human [18F]FDG scans using arterial-based quantification, and its performance is assessed with simulations and across clinical populations (healthy control, Alzheimer’s Disease (AD) and mild cognitive impairment (MCI)).
Methods: STARE is a data-driven approach that uses a source-to-target tissue model, based on a two-tissue irreversible compartment model, where time activity curves of multiple “target” regions are expressed as a function of a common “source” region, and separates target regions’ Ki from source Ki by fitting this model across all target regions simultaneously. Data-driven, subject-specific anchoring ensures identifiability of the model, using the PET signal in an automatically-determined vasculature cluster in the field of view. The estimation repeats, allowing each region to act in turn as the source, and final Ki in each region is estimated by averaging across source rotations.
Results: STARE Ki estimates were in good agreement with arterial blood-based estimates (regression slope = 0.88, Pearson’s r = 0.80), and were precisely estimated (comparing STARE Ki estimates across runs of the nondeterministic algorithm; coefficient of variation = 6.74 ± 2.48%). STARE performance, relative to blood-based Ki estimation, did not differ between diagnostic groups (p = 0.17). In simulations, STARE Ki estimates were largely robust to factors that influence the subject-specific anchoring (e.g., kinetics and noise within the automatically-determined vasculature cluster).
Conclusion: Feasibility is demonstrated for STARE blood-free quantification of Ki in this sample of [18F]FDG scans. Preliminary analyses suggest that STARE may be robust across some clinical populations (AD, MCI). Further validation is required, including application to data from other scanners, populations, and radiotracers.
Acknowledgements
The National Institute of Biomedical Imaging and Bioengineering provided funding for this study (R01EB026481, PI: Francesca Zanderigo, PhD).
2021-03
Novel data-driven method captures spatio-temporal patterns of neurodegeneration in Parkinson’s disease (#12)
Jessie Fanglu Fu1, 2, Ivan S. Klyuzhin3, Tilman Wegener1, 4, Martin McKeown3, 5, A. Jon Stoessl3, 5 and Vesna Sossi1, 5
1Department of Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada
2Department of Radiology, Massachusetts General Hospital/Harvard Medical School, Boston, MA, USA
3Division of Neurology, Department of Medicine, University of British Columbia, Vancouver, BC, Canada
4Department of Medical Engineering, University of Luebeck, Luebeck, Schleswig-Holstein, Germany
5Djavad Mowafaghian Centre for Brain Health, Pacific Parkinson’s Research Centre, University of British Columbia, Vancouver, BC, Canada
Abstract
Introduction: Most neurodegenerative disorders are characterized by progressive neurodegeneration following specific spatio-temporal patterns. To quantify these patterns, traditionally one often fits pre-defined models with spatial and temporal parameters.1,2 We recently proposed a novel data-driven approach, dynamic mode decomposition (DMD),3 to decompose the overall putaminal dopaminergic changes of Parkinson’s disease (PD) patients into orthogonal spatio-temporal patterns.4 In this work, we extended the previous work by applying DMD to a distinct PD cohort scanned on a lower resolution scanner to examine the robustness of the DMD results and the effect of scanner resolution.
Methods: We applied DMD to [11C]( ± )dihydrotetrabenazine (DTBZ) PET data in 70 PD patients (disease duration: 1 to 22 years) scanned on a CTI ECAT 953B scanner5 (resolution ∼8 mm3) and compared with previous results obtained in a cohort comprising 41 PD patients (disease duration: 1 to 16 years) scanned on the Siemens High Resolution Research Tomograph (HRRT)4 (resolution ∼2.5 mm3). DMD was applied to the parametric binding potential images (occipital cortex as the reference region) of the less and more affected putamen in the common space. Robustness was tested with leave-one-out cross-validation.
Results: DMD decomposed the ECAT data into two orthogonal spatio-temporal patterns (Figure 1): 1) an anterior-posterior gradient, with higher expression in the less affected putamen; the expression of which decreased gradually with disease progression. This pattern replicated the first DMD spatio-temporal pattern obtained from the HRRT data (Figure 2). 2) a dorsal-ventral gradient, with higher expression in the less affected putamen; the expression of which only existed in the early disease. This pattern was similar to the second DMD spatio-temporal pattern previously obtained with the HRRT data.
Conclusion: We showed the applicability and robustness of DMD for extracting spatio-temporal patterns of neurotransmitter changes using two PD cohorts on different scanners. The orthogonal temporal curves may reflect different mechanisms underlying disease progression and initiation. The slight differences in the second DMD modes may be related to scanner resolution and image registration quality differences. In principle, DMD can be easily extended to other PET studies and may help to elucidate disease mechanisms in more detail compared to traditional approaches.
Acknowledgements
The authors thank the UBC PET scanning and the TRIUMF radio-chemistry production staff. The volunteer subjects who generously donated their time to this research are also most gratefully acknowledged. The study was financially supported by the Natural Sciences and Engineering Research Council of Canada (NSERC), Brain Canada, and Canadian Institutes of Health Research (CIHR) and Michael Smith Foundation of Health Research. AJS is supported by the Canada Research Chairs program. JF receives scholarship funding from the Isotopes for Science and Medicine program (NSERC-CREATE). TW is supported by ENEN+ project that has received funding from the EURATOM research and training Work Programme 2016–2017 (1#755576) and the PROMOS Study or Internship Abroad Scholarships which has received funding from the German Academic Exchange Service (DAAD).
References
NandhagopalRKuramotoLSchulzerM, et al.Longitudinal progression of sporadic Parkinson’s disease: a multi-tracer positron emission tomography study. Brain2009;
132: 2970–2979.LeeCSSchulzerMDe La Fuente-FernandezR, et al.Lack of regional selectivity during the progression of Parkinson disease: implications for pathogenesis. Arch Neurol2004;
61: 1920–1925.SchmidP.Dynamic mode decomposition of numerical and experimental data. J Fluid Mech2010;
656: 5–28.FuJFKlyuzhinISMcKeownMJ, et al.Novel data-driven, equation-free method captures spatio-temporal patterns of neurodegeneration in Parkinson’s disease: Application of dynamic mode decomposition to PET. NeuroImage Clin2020;
25: 102150.SpinksTJJonesTBaileyDL, et al.Physical performance of a positron tomograph for brain imaging with retractable septa. Phys Med Biol1992;
37: 1637–1655.
2021-04
A high-resolution in vivo atlas of the human brain’s benzodiazepine binding site of GABAA receptors (#13)
Martin Nørgaard1, 2, Vincent Beliveau3, Melanie Ganz1, 4, Claus Svarer1, Lars H. Pinborg1, 2, Sune H. Keller5, Peter S. Jensen1, Douglas N. Greve6 and Gitte M. Knudsen1, 2
1Neurobiology Research Unit & CIMBI, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark
2Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
3Medical University of Innsbruck, Innsbruck, Austria
4Department of Computer Science, University of Copenhagen, Copenhagen, Denmark
5Department of Clinical Physiology, Nuclear Medicine and PET, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark
6Athinoula A. Martinos Center for Biomedical Imaging, MGH/HMS, Boston, MA, USA
Abstract
Introduction: Gamma-aminobutyric acid (GABA) is the main inhibitory neurotransmitter in the brain, and plays a critical role for brain function and in neuropsychiatric disorders.1 Benzodiazepines act as agonists on the benzodiazepine receptor binding site (BZR) which is located between the α1,2,3 or 5 and γ subunits in the pentameric constellation of the ionotropic GABAA receptor (GABAAR). Here, we present a quantitative high-resolution in vivo atlas of the human brain’s BZRs; this atlas can serve as a reference in future studies investigating disorders or pharmacological interventions on the BZR site.
Methods: Twenty-six unique Positron Emission Tomography (PET) scans were obtained using a Siemens HRRT scanner with the radioligand [11C]flumazenil. The radioligand was given either as a bolus or as a bolus-infusion, and the metabolite-corrected arterial input function was measured. The PET data was quantified to estimate total distribution volumes (VT) for each brain region; steady-state analysis was done for the bolus-infusion experiments and Logan analysis was done for the bolus experiments.2,3 BZR density (pmol per gram protein in brain tissue) was obtained by normalizing VT with the corresponding postmortem human brain [3H]diazepam autoradiography data.4 The non-displaceable distribution volume (VND) was estimated as the intercept (Figure 2(b)). The atlas was transformed to represent protein densities in pmol/ml (Figure 2(a)). Finally, the association between protein density and mRNA expression for the 19 GABAAR subunits was assessed using the Allen Human Brain atlas.5
Results: The regional VT’s and the postmortem human brain autoradiography were highly correlated (Pearson’s R = 0.96) and the intercept, VND, was low, in average 0.51 across subjects (range 0.04–1.31). The association between BZR densities and mRNA expression showed a high correlation for each of the subunits included in the commonly expressed pentameric GABAAR subtype α1β2γ2 (>50% expression in the brain) with Pearson’s R > 0.62 (range: 0.62–0.88, P < 0.0001) for all three subunits.
Conclusion: This high-resolution atlas of the spatial distribution of BZR densities in the healthy human brain provides insights into the association between mRNA expression for individual subunits in the GABAAR and the BZR density at each location in the brain; in addition, it may be used to evaluate efficacy of pharmacological targets acting on the GABAAR.C
The regression is shown as the black line, and the intercept is the non-displaceable distribution volume (VND). The shaded area is the 95% confidence interval.
References
SequeiraAShenKGottliebA, et al.Human brain transcriptome analysis finds region- and subject-specific expression signatures of GABAAR subunits. Commun Biol2019;
2: 1–14.KnudsenGMJensenPSErritzoeD, et al.The Center for Integrated Molecular Brain Imaging (Cimbi) database. NeuroImage2016;
124: 1213–1219.FengLSvarerCMadsenK, et al.Design of infusion schemes for neuroreceptor imaging: application to [11C]Flumazenil-PET steady-state study. BioMed Res Int2016;
2016: 9132840.BraestrupCAlbrechtsenR, andSquiresR.
High densities of benzodiazepine receptors in human cortical areas. Nature1977;
269: 702–704.HawrylyczMJLeinESGuillozet-BongaartsAL, et al.An anatomically comprehensive atlas of the adult human brain transcriptome. Nature2012;
489: 391–399.
2021-05
An algorithm for classifying neurotransmitter signals from future ultra-high sensitivity dynamic PET (#22)
Heather Liu1, 2 and Evan D. Morris1, 2
1Department of Biomedical Engineering, Yale University, New Haven, CT, USA
2Department of Radiology and Biomedical Imaging, Yale University, New Haven, CT, USA
3Department of Psychiatry, Yale University, New Haven, CT, USA
Abstract
Introduction: Innovations in solid-state detector technology will soon be leveraged to build the next-generation NeuroEXPLORER (NX) brain scanner, offering an order-of-magnitude improvement in detection sensitivity compared to the current state-of-the-art, Siemen’s HRRT. Our goal was to explore how increased sensitivity could be used with the linear-parametric neurotransmitter PET (lp-ntPET) model1–4 to better detect transient dopamine (DA) events and classify DA signals.
Methods: Striatal 11C-Raclopride PET data were simulated in the presence of varying DA signals5. DA signals were modeled with differing start times, peak times, and amplitudes. Measurement variance for the NX was scaled to 10% of the HRRT, to reflect an anticipated ten-fold sensitivity improvement. Time-activity curves were fitted with lp-ntPET. The detection sensitivity to each signal was calculated. Classification thresholds were evaluated for their accuracy in separating “early”- vs. “late”-peaking, and “low”- vs. “high”-amplitude events. A 4D phantom was constructed, with each voxel containing either a null signal or a positive signal, classified as high/low and early/late. A weighted k-nearest neighbors (wkNN) algorithm incorporated the 6-, 18-, and 26-voxel neighborhoods of each voxel to reclassify it. The certainty of classification of a voxel was given by the weighted vote of its neighbors. wkNN was applied to the phantom for detection and classification. Detection sensitivities and classification accuracies were calculated for both scanners.
Results: The NX would expand the range of detectable events to signals with amplitudes as low as 200% above baseline and peaking as early as 6 min post-stimulus (Figure 1(a)), compared to the minimum of 550% and 20 min necessary for detection by the HRRT (Figure 1(b)). NX sensitivity was 72.3% and specificity was 91.5%, compared with 31.4% and 88.9% for the HRRT. Application of wkNN increased sensitivity of the NX to 91.5%, with 100% specificity. Classification of high-amplitude and early-peaking signals had a positive predictive value > 97%. All classifications had a negative predictive value > 85% (Figure 2).
Conclusion: An ultra-high sensitivity NX scanner could greatly expand the range of detectable DA signals, compared to the current state-of-the-art. Higher detection sensitivity and a novel classification algorithm will make it possible to accurately classify DA signals according to their amplitude and timing.
References
MorrisEDYoderKKWangC, et al. ntPET: A New Application of PET Imaging for Characterizing the Kinetics of Endogenous Neurotransmitter Release. Mol Imaging2005;
4: 473–489.MorrisEDKimSJSullivanJM, et al. Creating Dynamic Images of Short-lived Dopamine Fluctuations with lp-ntPET: Dopamine Movies of Cigarette Smoking. J Vis Exp2013;
78: 50358.NormandinMDSchifferWKMorrisED. A linear model for estimation of neurotransmitter response profiles from dynamic PET data. Neuroimage2012;
59: 2689–2699.KimSJSullivanJMWangS, et al. Voxelwise lp-ntPET for detecting localized, transient dopamine release of unknown timing: Sensitivity Analysis and Application to Cigarette Smoking in the PET Scanner. Hum Brain Mapp2014;
35: 4876–4891.MorrisEDFisherREAlpertNM, et al. In Vivo Imaging of Neuromodulatory Synaptic Transmission Using PET: A Review of Relevant Neurophysiology. Hum Brain Mapp1995;
3: 35–55.
2021-06
In vivo neuroimaging of histone deacetylases in multiple sclerosis using [11C]Martinostat (#24)
Mary C. Catanese, Chieh-En J. Tseng, Valeria Barletta, Constantina A. Treaba, Elena Herranz Muelas, Nicole R. Zurcher Wimmer, Anjali J. Parmar, Baileigh G. Hightower, Ambica Mehndiratta, Changning Wang, Caterina Mainero and Jacob M. Hooker
Athinoula A. Martinos Center for Biomedical Imaging, Radiology, Charlestown, MA, USA
Abstract
Introduction: Multiple sclerosis (MS) is a demyelinating disease characterized by progressive neurological impairment and lesion formation. Magnetic resonance (MR) imaging enables identification of lesions and insight into their pathology, however, there are critical gaps in our molecular understanding. Histone deacetylases (HDACs) are considered therapeutic targets based on the consistent amelioration of symptom-like behavior in preclinical models of demyelination after treatment with HDAC inhibitors.1–3 Moreover, HDAC expression is altered in postmortem lesion tissue.4 We previously demonstrated increased uptake in the HDAC-specific radiotracer, [11C]Martinostat, in white matter (WM) with age,5 which correlated with decreased WM microstructure. We hypothesized that HDACs may become dysregulated in MS, as focal lesions are found in the WM.
Methods: In a pilot study [11C]Martinostat uptake was measured in four individuals with relapsing remitting (RR) MS (2 female/2 male;53.5 ± 8.3years) and four age- and sex-matched controls (2 female/2 male;51.75 ± 9.9years). Dynamic positron emission tomography (PET) images were acquired after bolus injection (controls 5.22 ± 0.41Ci; MS5.37 ± 0.14mCi) using a combined 3-Tesla:MR-PET system. PET data were reconstructed in units of standard uptake value (SUV); six 5-min frames, 60–90min post-injection were processed using FreeSurfer and FSL. SUV was normalized to whole-brain mean (SUVR,60-90min). SUVR was registered to the corresponding T1-weighted image. Lesions segmented on Fluid-Attenuated-Inversion-Recovery (FLAIR) images (3-D-slicer-v4.2.0) were registered to the T1-weighted and SUVR images (FSL). SUVR in MS WM-lesions were compared to normal-appearing-WM (NAWM) and control-WM and MS-NAWM was compared to control-WM in GraphPad Prism (Mann-Whitney).
Results: Preliminary analysis of [11C]Martinostat uptake found no significant differences in signal comparing WM-lesions to control-WM and MS-NAWM or comparing control-WM to MS-NAWM. It is possible that differences were not found in lesions due to partial volume effects (PVE) and low sample size.
Conclusion: Our study is the first to investigate HDAC expression in MS in vivo. Preliminary data revealed that [11C]Martinostat uptake in MS-WM lesions is not different compared to control-WM and MS-NAWM. Studying molecular dysregulation underlying MS is complicated by lesion localization and by differences in lesion and disease subtype. We will account for PVE in further analyses and will increase our sample size, as well as study patients across MS-subtypes to determine whether HDACs are involved in MS-WM lesion pathology.
Acknowledgements
This research received funding from an internal pilot funding mechanism at the Athinoula A. Martinos Center for Biomedical Imaging and the Massachusetts General Hospital Research Scholar’s Program (Jacob M. Hooker).
References
FaracoGCavoneLChiarugiA.The therapeutic potential of HDAC inhibitors in the treatment of multiple sclerosis. Mol Med2011;
17: 442–447.ZhangZZhangZYWuY, et al.Valproic acid ameliorates inflammation in experimental autoimmune encephalomyelitis rats. Neuroscience2012;
221: 140–150.GöschlLPreglejTHammingerP, et al.A T cell-specific deletion of HDAC1 protects against experimental autoimmune encephalomyelitis. J Autoimmun2018;
86: 51–61.TeglaCAAzimzadehPAndrian-AlbescuM, et al.SIRT1 is decreased during relapses in patients with multiple sclerosis. Exp Mol Pathol2014;
96: 139–148.GilbertTMZürcherNRCataneseMC, et al.Neuroepigenetic signatures of age and sex in the living human brain. Nat Commun2019;
10: 1–10.
2021-07
Optimized synthesis of [18F]FPEB and preliminary neuroimaging of mGlur5 in transgenic mouse models (#25)
Cassis Varlow1, 2, Nickeisha Stephenson2, 3, Emily Murrell2 and Neil Vasdev1, 2
1Institute of Medical Science, University of Toronto, Toronto, ON, Canada
2CAMH, Azrieli Centre for Neuro-Radiochemistry, Brain Health Imaging Centre, Toronto, ON, Canada
3Department of Chemistry, The University of West Indes at MONA, Kingston, Jamaica
Abstract
Introduction: 3-[18F]fluoro-5-[(pyridin-3-yl)ethynyl] benzonitrile ([18F]FPEB) is a radiopharmaceutical selective for metabotropic glutamate receptor subtype 5 (mGluR5). The present study aims to: 1) establish the optimal synthesis of [18F]FPEB and 2) to investigate changes in mGluR5 uptake via in vitro autoradiography and preclinical PET imaging in transgenic mouse models of neurological diseases (i.e., Alzheimer’s disease (AD).
Methods:18F]FPEB was prepared through nucleophilic aromatic substitution with [18F]fluoride on an automated radiosynthesis unit (GE TRACERlab™ FX2N) using five precursors with varying leaving groups, namely: Cl, NO2, spirocyclic iodonium ylide (SCIDY) with a cyclopentyl (SPI5) and a novel adamantyl (SPIAd) auxiliary; and a sulfonium salt precursor. Preliminary PET/CT imaging studies were conducted in B6C3-Tg (APPswe,PSEN1dE9)85Dbo/J (APP/PS1) mice, and data were compared with age-matched wild-type (WT) B6C3F1/J control mice (Figure 1).
Results: Complete radiosynthesis data summarized in Table 1. The chloro- and nitro-precursors produced [18F]FPEB in low radiochemical yield (<10% RCY), whereas both the SPI5 auxiliary SCIDY precursor and the sulfonium salt precursor resulted in the highest RCY (25 ± 2% and 36 ± 6%, respectively). The new SPIAd auxiliary successfully produced [18F]FPEB in 24% RCY, however, no advantage over SPI5 was observed and HPLC separation will require further optimization. Preliminary PET imaging data with [18F]FPEB shows an increased brain uptake in the transgenic model of AD vs. the age-matched controls (10 month data shown in Figure 1).
Conclusion: Of the five precursors evaluated, the SCIDY precursor with SPI5 auxiliary and the sulfonium salt precursor appear to the be the best suited for routine radiopharmaceutical production of [18F]FPEB. Increased uptake of [18F]FPEB in the APP/PS1 mouse model of AD compared with controls is consistent with results in patients with early mild cognitive impairment. Preclinical imaging studies with [18F]FPEB in Fragile X syndrome and mouse models of autism spectrum disorder are also underway to gain mechanistic insights for mGluR5 targeting therapeutics, and will be presented.
Acknowledgements
CV Thanks Dr. Jason Holland and Dr. Steven Liang for their contributions.
Table 1. Radiofluorination of [18F]FPEB via five different precursors.
2021-08
Estimating PET partial volume full-width-half-maximum directly from human data (#30)
Douglas N. Greve1, Martin Schain2, Melanie Ganz2, Martin Nørgaard2, Claus Svarer2 and Gitte M. Knudsen2
1Massachusetts General Hospital, Martinos Center, Charlestown, MA, USA
2Rigshospitalet, CIMBI & Neurobiology Research Unit, Copenhagen, Denmark
Abstract
Introduction: PET suffers from partial volume effects (PVEs) in which limited scanner resolution causes the activity to appear to spill out of one region and into another. PVEs cause false positives and negatives and inter-scanner variability. PV correction (PVC) methods exist but require a measure of the scanner resolution, usually obtained from a point source. This research introduces Adaptive PVC (APVC) which simultaneously estimates the scanner resolution and performs PVC.
Methods: APVC is based on the GTM,1 an ROI-based PVC method in which PVE-free ROI means (b) are related to voxel-wise intensities (y) through a linear model (X): y = X*b. X = X(q) depends on a blurring function parameterized by q. Typically, q is a full-width-half-maximum (FWHM). With q measured from a point source, the GTM is inverted to compute b. In APVC, we simultaneously solve for both b and q by minimizing the residual variance of the linear model. We tested this method using 5 subjects scanned on GE Advance and Siemens HRRT using a 5HT4 tracer, the assumption being that the best method will yield closer BPnd across scanner. The blurring function was modeled as a Gaussian, with q being three FWHMs (one per dimension).
Results: The average estimated FWHMs were much larger than the nominal values: x = 8.5 mm ± 0.11 y = 9.0 mm ± 0.11 z = 8.3 mm ± 0.37 mm (Advance, nominal = 6 mm) and x = 5.1 mm ± 0.22 y = 4.5 mm ± 0.22 z = 4.7 mm ± 0.16 (HRRT, nominal = 4 mm). Figure 1 shows how the BPND for a single ROI differs between scanner/method. Without PVC, the BPNDs differ by 58%. Using the nominal FWHM, the difference drops to 31%. However, with APVC, the difference drops to only 15%. Figure 2 shows the differences for 68 cortical ROIs with NoPVC having a very large difference (41% average), nominal PVC improving to 32%, and APVC improving still further to 17%.
Conclusion: APVC estimates the scanner resolution directly from human data while correcting for PVEs. Our results show that the APVC-measured FWHM was much higher than the nominal in two scanners. APVC resulted in a much better agreement in 5HT4 BPNDs across the two scanners. While APVC uses the GTM, the output resolution measure can be used in any PVC method. Software available in FreeSurfer/PETsurfer.
References
RoussetOGMaYEvansAC.
Correction for partial volume effects in PET: principle and validation. J Nucl Med1998;
39: 904–911.
2021-09
Repurposing novel PET neuroimaging radioligands in xenograft mouse models of cancer (#32)
Amanda J. Boyle1, Junchao Tong1, Sami Goghbi2, Victor W. Pike2, Robert B. Innis2 and Neil Vasdev1 and partly submitted as a collaboration with the National Institute of Mental Health (NIMH)
1Centre for Addiction and Mental Health, Brain Health Imaging Centre, & Azrieli Centre for Neuro-Radiochemistry, Toronto, ON, Canada
2National Institute of Mental Health, Bethesda, MD, USA
Abstract
Introduction: This study aims to explore the utility of PET radioligands that target biomarkers relevant to oncology but which, to date, have only been developed as neuroimaging agents for cyclooxygenases (COX-1 and COX-2) and glycogen synthase kinase-3 (GSK-3). Herein, we evaluate [11C]PS13, [11C]MC1, and [11C]OCM-44, which target COX-1,1 COX-2,1 and GSK-3,2 respectively, in ovarian, breast, and pancreatic xenograft tumour mouse models.
Methods: Radiosyntheses of [11C]PS13, [11C]MC1, and [11C]OCM-44 were performed as previously described.2–4 The radioligands were evaluated in a panel of xenograft mouse models prepared from ovarian, breast, and pancreatic cancer cell lines, OVCAR-3, MDA-MB-231, and PANC-1, respectively, via dynamic PET imaging. Blocking studies, ex vivo biodistribution, and radiometabolite analysis of tumour homogenates were also carried out.
Results: OVCAR-3 xenografts were well visualized with [11C]PS13 (0–60 min average image; Figure 1(a)). Time-activity curves (TACs) revealed steady tumour radioactivity accumulation that plateaued from 40–60 min with an average uptake of 3.56 ± 0.81%ID/g (40–60 min), which was significantly reduced by pre-treatment with the known COX-1 inhibitor, ketoprofen, to 1.30 ± 0.18%ID/g (p = 0.0096), (Figure 1(b) and 1(c)). Preliminary radiometabolite analysis established the parent compound, [11C]PS13, to account for 65% of radioactivity in the tumour at 40 min post-injection (p.i.) of the radioligand. PANC-1 xenografts were not visualized by [11C]MC1 despite reported high expression of COX-2 but were successfully imaged with [11C]OCM-44 (Figure 2(a)), yet blocking was unsuccessful and radiometabolite analysis showed that at 40 min p.i. less than 2% of the radioactivity was parent. MDA-MB-231 xenografts were not visualized by PET imaging with [11C]PS13, [11C]MC1, or [11C]OCM-44 (Figure 2(b)), despite reported overexpression of COX-1, COX-2, and GSK-3.
Conclusion: [11C]PS13 has a favorable radiobiological profile for studying the role of peripheral COX-1 and clinical translation of this radioligand in ovarian cancer is planned. Further evaluations of [11C]MC1 and [11C]OCM-44 are underway and will also be presented.
Acknowledgements
Ian Duffy for producing these radioligands at CAMH. Cassis Varlow for her careful skills in mouse catheterization.
References
KimMJPikeVWInnisRB, et al.Evaluation of two potent and selective PET radioligands to image COX-1 and COX-2 in rhesus monkeys. J Nucl Med2018;
59: 1907–1912.Bernard-GauthierVMossineAVKnightA, et al.Structural basis for achieving GSK-3beta inhibition with high potency, selectivity, and brain exposure for positron emission tomography imaging and drug discovery. J Med Chem2019;
62: 9600–9617.Cortes-SalvaMYInnisRBPikeVW, et al.2-(4-Methylsulfonylphenyl)pyrimidines as prospective radioligands for imaging cyclooxygenase-2 with PET-synthesis, triage, and radiolabeling. Molecules2018;
23: 2850.SinghPInnisRBPikeVW, et al.3-Substituted 1,5-diaryl-1 h-1,2,4-triazoles as prospective PET radioligands for imaging brain COX-1 in monkey. Part 1: synthesis and pharmacology. ACS Chem Neurosci2018;
9: 2610–2619.
2021-10
[18F]MNI-1054, a novel PET ligand for lysine-specific histone demethylase 1A (LSD1): First-in-human validation including radiation dosimetry, kinetic modeling and test-retest variability (#39)
Adam J. Schwarz1, Christine Sandiego2, Cristian Constantinescu2, Vincent Carroll2, Olivier Barret2, Roger N. Gunn3, Akihiro Takano4, Paul McQuade1, Terry Brown1, Johannes Tauscher1 and David S. Russell2
1Takeda Pharmaceuticals, Cambridge, MA, USA
2Invicro, New Haven, CT, USA
3Invicro, London, UK
4Takeda Pharmaceuticals, Osaka, Japan
Abstract
Introduction: The enzyme lysine-specific histone demethylase 1A (LSD1), involved in the control of gene expression and cell proliferation, is widely expressed in the human brain and specifically demethylates lysine residues on histone H3K4 and H3K9. This Phase 1 study aimed to assess the LSD1-targeting radiotracer [18F]MNI-1054 in healthy volunteers in terms of: (1) brain kinetics and quantitative tracer kinetic modeling; (2) test-retest reproducibility; and (3) safety, biodistribution and radiation absorbed dose burden.
Methods: Four male subjects had two 180-minute [18F]MNI-1054 dynamic brain PET scans, with arterial blood samples, up to 2 weeks apart (303 ± 67 MBq (8.2 ± 1.8 mCi)). Additionally, two male and two female subjects underwent single whole-body PET scans to determine biodistribution (352 ± 6 MBq (9.5 ± 0.2 mCi)). All scans were performed on a Siemens HR+ PET camera. [18F]MNI-1054 binds irreversibly to the LSD1 enzyme, so the PET data were analyzed with the irreversible 2-tissue compartment model, using the metabolite-corrected arterial plasma input function, to calculate the enzyme binding parameter Ki.
Results: [18F]MNI-1054 and the study procedures were well-tolerated with no safety concerns. The highest tracer binding was observed in the cerebellum (Ki∼0.04), with lower uptake in cerebral cortices, striatum and hippocampus (Ki∼0.02), and lowest in the pons (Ki∼0.01), consistent with the expected LSD1 distribution (Figure 1(a) and (b)). The average test-retest variability in Ki across the four subjects was ∼5% for the cerebellum and ∼10-15% in other brain regions (Figure 1(c)). Analysis using only the first 120 minutes of data resulted in good agreement with that using 180 min of dynamic data (R2 = 0.88) and similar cerebellar test-retest variability of ∼5% (Figure 1(d)). The average whole-body effective dose was determined to be 0.026 ± 0.00228 mSv/MBq, or 4.18 ± 0.42 mSv per 185 MBq (5 mCi) tracer injection.
Conclusion: [18F]MNI-1054 exhibited good safety and technical characteristics suitable for repeated scans and use in occupancy studies.
2021-11
Towards the development of PET tracers for the imaging of melatonin receptors (#43)
Alexey Kostikov1, Thomas A. Singleton1, Hussein Bdair1, Karen Ross1, Min-Su Kang2, Arturo Aliaga2, Tanpreet Kaur3, Allen F. Brooks3, Peter J.H. Scott3, Gabriella Gobbi4, Chawki Benkelfat4, Marco Leyton4, Saïd Yous5, Jean-Paul Soucy1, 6 and Pedro Rosa-Neto1, 2
1McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill University, Montreal, QC, Canada
2Research Centre for Studies in Aging, McGill University, Douglas Institute, Montreal, QC, Canada
3Department of Radiology, University of Michigan Medical School, Ann Arbor, MI, USA
4Department of Psychiatry, McGill University, Montreal, QC, Canada
5Neurosciences and Cognition Research Center, University of Lille, Lille, France
6PERFORM Centre, Concordia University, Montreal, QC, Canada
Abstract
Introduction: The neurohormone melatonin modulates circadian rhythms and other physiological functions in mammals through the activation of melatonin type 1 and 2 receptors (MT1 and MT2), which are mainly expressed in the suprachiasmatic nucleus (SCN)1 of the brain. Recently, the melatonergic system emerged as a promising therapeutic target for the treatment of sleep and mood disorders and to maximize the therapeutic potential of sleep in neurodegenerative conditions2. Imaging tools for visualization and localization of melatonin receptors are highly sought after to study their role in various physiological and pathological conditions, and for the development of new drugs targeting melatonergic system.
Methods: We radiolabelled a series of four high affinity MT receptors ligands with either carbon-11 or fluorine-18 PET isotopes (Figure 1). In particular, [11C]UCM7653 and [11C]UCM10144 were labelled by 11C-methylation of the desmethoxy precursors, whereas [18F]3-fluoroagomelatine5 was produced via copper-mediated 18F-fluorination of the corresponding boronic pinacolate precursor and [18F]fluoroacetamidoagomelatine5 was labelled via nucleophilic 18F-fluorination of the bromoacetamide agomelatine derivative. All ligands were evaluated as PET tracer candidates in preclinical imaging studies in wild-type rats using the microPET R4 scanner.
Results: All four radioligands showed good to excellent brain permeability, in particular the [18F]fluoroagomelatine derivatives (SUVmax = 3.25 ± 0.31, Figure 2) and a fast washout rate in wild-type rats. However, blocking studies using selective MT ligands have not revealed reduction of the tracer uptake in the brains of rats. This could be due to the fast metabolism of these PET tracers in rodents, the small anatomical size of the SCN in rats or the nocturnal nature of these animals leading to decreased MT expression during daytime.
Conclusion: The high brain permeability of the radiolabelled ligands warrants their further investigation as PET tracer candidates for imaging of MT1 and MT2 receptors. However, rats do not appear to be suitable animal models for those further studies. Higher species, such as non-human primates, are required for subsequent imaging experiments using PET tracer candidates developed in our lab.
Acknowledgements
The authors thank the personnel of the cyclotron and radiochemistry facility at the Montreal Neurological Institute.
References
BarrenetxeJ, et al.Physiological and metabolic functions of melatonin. J Physiol Biochem2004;
60: 61–72.LiuJ, et al.MT1 and MT2 melatonin receptors: a therapeutic perspective. Annu Rev Pharmacol Toxicol2016;
56: 361–383.RivaraS, et al.
N-(substituted-anilinoethyl)amides: design, synthesis, and pharmacological characterization of a new class of melatonin receptor ligands. J Med Chem2007;
50: 6618–6626.SpadoniG, et al;
Highly potent and selective MT2 melatonin receptor full agonists from conformational analysis of 1-benzyl-2-acylaminomethyl-tetrahydroquinolines. J Med Chem2015;
58: 7512–7525.EttaoussiM, et al;
Design, synthesis and pharmacological evaluation of new series of naphthalenic analogues as melatoninergic (MT1/MT2) and serotoninergic 5-HT2C dual ligands (I).Eur J Med Chem2012;
49: 310–323.
Yasmin Zakiniaeiz1, 2, Heather Liu3, Hong Gao1, 2, Soheila Najafzadeh1, 2, Jim Ropchan1, 2, Nabeel Nabulsi1, 2, Yiyun H. Huang1, 2, Kelly P. Cosgrove2, 4 and Evan D. Morris2, 5
1Department of Radiology and Biomedical Imaging, Yale University, New Haven, CT, USA
2Yale Positron Emission Tomography (PET) Center, Yale University, New Haven, CT, USA
3Department of Biomedical Engineering, Yale University, New Haven, CT, USA
4Department of Psychiatry, Yale University, New Haven, CT, USA
5Invicro, New Haven, CT, USA
Abstract
Introduction: Tobacco smoking is a major public health concern. Smoking cessation treatments have limited success rates, including nicotine replacement therapy (NRT) which acts directly at nicotinic-acetylcholine receptors on dopamine (DA) terminals to release DA in the striatum. The goals of this study were to examine the strength of cigarette smoking-induced striatal DA release and to relate spatial extent of release to nicotine dependence in smokers under nicotine patch and placebo patch conditions. We hypothesized that nicotine patch would reduce spatial extent of release compared to placebo, and that smoking pack years would be associated with spatial extent in the striatum.
Methods: Twenty-eight tobacco smokers (13 Female) wore a nicotine patch (21mg, daily) for 1-week and a placebo patch for 1-week in a randomized, counter-balanced design. Following each condition plus an overnight abstinence, smokers participated in 90-minute [11C]raclopride PET scans. They smoked a cigarette while lying in the scanner. We used lp-ntPET,1-4 a model of tracer uptake containing a time-varying term, to identify highly localized DA transients in PET data at the voxel level. lp-ntPET was fitted (voxel-wise) to PET TACs in the pre-commissural striatum. DA responses were retained for voxels if the inclusion of the time-varying term improved the fit. “Probability of activation” maps were generated, summed by condition and divided by the number of group members.5 Participants were divided into low and high pack years groups using a median split analysis and probability of activation maps were made for each group.
Results: Thenicotine patch reduced the spatial extent of DA release and the probability of DA activation following cigarette smoking compared to placebo patch (Figure 1(a)). The high pack years group had higher spatial extent of DA release and probability of DA activation in the striatum during the placebo patch condition than the low pack years group (Figure 1(b)).
Conclusion: Consistent with our hypotheses, the nicotine patch reduced the strength of cigarette-induced striatal DA response compared to placebo patch, suggesting a potential mechanism for NRT on the rewarding response of cigarette-smoking. Pack years was associated with the strength of the striatal DA response, suggesting that nicotine dependence might contribute to DA response.
Acknowledgements
Research support provided by R01DA038709-03 (Morris) and T32DA022975 (Zakiniaeiz). We thank the Yale PET Center staff for imaging and chemistry support.
References
MorrisEDYoderKK, et al.ntPET: a new application of PET imaging for characterizing the kinetics of endogenous neurotransmitter release. Mol Imaging2005;
4: 473–489.MorrisEDKimSJ, et al.Creating dynamic images of short-lived dopamine fluctuations with lp-ntpet: dopamine movies of cigarette smoking.J Vis Exp2013;
78: 50358.NormandinMDSchifferWK, et al.A linear model for estimation of neurotransmitter response profiles from dynamic PET data. Neuroimage2012;
59: 2689–2699.KimSJSullivanJM, et al.Voxelwise lpntPET for detecting localized, transient dopamine release of unknown timing: sensitivity analysis and application to cigarette smoking in the PET scanner.Hum Brain Mapp2014;
35: 4876–4891.CosgroveKPWangS, et al.Sex differences in the brain’s dopamine signature of cigarette smoking.J Neurosci2014;
34: 16851.
2021-13
Distinct spatio-temporal patterns of putuminal dopamine processing in Parkinson’s disease: A multi-tracer positron emission tomography study (#47)
Jessie Fanglu Fu1, 2, Tilman Wegener1, 3, Ivan S. Klyuzhin4, Julia G. Mannheim1, 5, Martin McKeown4, 6, A. Jon Stoessl4, 6 and Vesna Sossi1, 6
1Department of Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada
2Massachusetts General Hospital/Harvard Medical School, Radiology, Boston, MA, USA
3Department of Medical Engineering, University of Luebeck, Luebeck, Schleswig-Holstein, Germany
4Division of Neurology, Department of Medicine, University of British Columbia, Vancouver, BC, Canada
5Department of Preclinical Imaging and Radiopharmacy, Werner Siemens Imaging Center, Eberhard-Karls University Tuebingen, Tuebingen Baden-Württemberg, Germany
6Djavad Mowafaghian Centre for Brain Health, Pacific Parkinson’s Research Centre, University of British Columbia, Vancouver, BC, Canada
Abstract
Introduction: In Parkinson’s disease (PD), different aspects of putaminal dopaminergic processing (synthesis, release and turnover) may be affected differently at different disease stages, thus following distinct spatio-temporal patterns. Differences in the patterns may reflect unique contributions from each aspect of dopaminergic processing to disease initiation, disease progression and potential compensatory mechanisms.1,2 In this work, we investigated if distinct spatio-temporal patterns exist in different aspects of presynaptic dopamine alteration as disease progresses using multi-tracer Positron Emission Tomography (PET) and a previously published data-driven approach – dynamic mode decomposition (DMD).3
Methods: We applied DMD to three presynaptic PET tracers: [11C]dihydrotetrabenazine (DTBZ, estimates dopaminergic integrity) in 70 PD patients, [11C] d-threo-methylphenidate (MP, estimates dopamine reuptake) in 74 PD patients, and [18F]fluoro-L-DOPA (FD, reflects dopamine synthesis and storage) in 61 PD patients (disease duration: 1 to 22 years). The PET images were acquired on the CTI ECAT 953B ECAT scanner4 and DMD was applied to the parametric binding maps on the less and more affected putamen in the common space.
Results: DMD modes of all three tracers showed extremely similar anterior-posterior gradients in the less and more affected putamen (Figure 1); expressions of which decreased gradually with disease progression for all tracers (Figure 2). However, the initial expressions (intercepts) were significantly different (p < 10−5) with FD showing the highest and DTBZ showing the lowest intercept. FD and MP also showed significantly higher rates of decline for the gradient expression compared to DTBZ (p < 10−12), while there was no significant difference between FD and MP.
Conclusion: With this novel approach, we showed the expression of the gradient is more sensitive to disease progression than mean putaminal binding, suggesting a differential disease impact on the anterior and posterior putamen. While DTBZ mainly reflected direct disease-induced dopamine denervation, distinct temporal progression curves obtained with MP and FD may suggest upregulation of dopamine synthesis in the anterior putamen and downregulation of dopamine reuptake in the posterior putamen in early PD, consistent with previous finding obtained with univariate analysis on the same subject cohort,2 which may be of compensatory nature. This compensatory mechanism may breakdown as the disease progresses, which can potentially contribute to progression of motor symptoms in advanced disease.
Acknowledgements
The authors thank the UBC PET scanning and the TRIUMF radio-chemistry production staff. The volunteer subjects who generously donated their time to this research are also most gratefully acknowledged. The study was financially supported by the Natural Sciences and Engineering Research Council of Canada (NSERC), Brain Canada, and Canadian Institutes of Health Research (CIHR) and Michael Smith Foundation of Health Research. AJS is supported by the Canada Research Chairs program. JF receives scholarship funding from the Isotopes for Science and Medicine program (NSERC-CREATE). TW is supported by ENEN+ project that has received funding from the EURATOM research and training Work Programme 2016–2017 (1#755576) and the PROMOS Study or Internship Abroad Scholarships which has received funding from the German Academic Exchange Service (DAAD).
References
NandhagopalRKuramotoLSchulzerM, et al.Longitudinal progression of sporadic Parkinson’s disease: a multi-tracer positron emission tomography study. Brain2009;
132: 2970–2979.LeeCS, et al.In vivo positron emission tomographic evidence for compensatory changes in presynaptic dopaminergic nerve terminals in Parkinson’s disease. Ann Neurol2000;
47: 493–503.FuJFKlyuzhinISMcKeownMJ, et al.Novel data-driven, equation-free method captures spatio-temporal patterns of neurodegeneration in Parkinson’s disease: application of dynamic mode decomposition to PET. NeuroImage Clin2020;
25: 102150.SpinksTJJonesTBaileyDL, et al.Physical performance of a positron tomograph for brain imaging with retractable septa. Phys Med Biol1992;
37: 1637–1655.
2021-14
Ketanserin can block but not displace [11C]Cimbi-36 binding in the pig brain (#68)
Hanne D. Hansen1, 2, Lene L. Donovan1, 3, Nakul R. Raval1, 3, Arafat Nasser1, Szabolcs Lehel4, Claus Svarer1, Martin Schain1 and Gitte M. Knudsen1, 3
1Neurobiology Research Unit, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark
2Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, USA
3Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
4Department of Clinical Physiology, Nuclear Medicine and PET, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark
Abstract
Introduction: Changes in cerebral target binding following an intervention are conventionally estimated by performing two PET scans. Using this experimental design, the serotonin 2A receptor (5-HT2AR) agonist PET tracer [11C]Cimbi-36 has been used to measure the occupancy of ketanserin1 and psilocybin.2 Here, we aimed to estimate the occupancy of the 5-HT2AR receptor antagonist ketanserin based on a single [11C]Cimbi-36 PET scan.
Methods: In three anesthetized pigs, [11C]Cimbi-36 was administrated using either bolus or bolus-infusion paradigms. Emission data was acquired for 120 min. Each pig underwent two measurements, one baseline and one with ketanserin (5 mg/kg) administrated intravenously ∼60 min after tracer injection. Arterial input measurements and radio-HPLC analyses were performed to allow for kinetic modelling using the 2-tissue compartment model.
Results: Unexpectedly, when ketanserin was administered 60 min after [11C]Cimbi-36 bolus injection, the [11C]Cimbi-36 time-activity curves did not differ from those obtained from the preceding baseline scan (Figure 1(a)), and the 5-HT2AR occupancy was negligible (Figure 1(b)). When we reversed the scan-order in the second animal and gave the ketanserin within-scan challenge first, the [11C]Cimbi-36 time-activity curves still did not show signs of displacement (Figure 1(c)). In the subsequent scan, however, 71% occupancy of the 5-HT2AR was observed (Figure 1(d)). In the third animal, we used a bolus-infusion protocol (Kbol of 120 min) to obtain steady-state in cortex and plasma and now the within-scan ketanserin challenge resulted in a transient increase in the cortical [11C]Cimbi-36 time-activity curve followed by noticeable wash-out of tracer from the cortex (Figure 1(e)). The corresponding cortex to plasma ratios also decreased after the ketanserin challenge (Figure 1(f)), consistent with ketanserin being able to block newly infused [11C]Cimbi-36 but not to displace already bound [11C]Cimbi-36.
Conclusion: [11C]Cimbi-36 time-activity curves are not visibly altered by within-scan ketanserin injection, suggesting that [11C]Cimbi-36 cannot be displaced from the 5-HT2AR. In contrast, the antagonist 5-HT2AR tracers [18F]MH.MZ and [18F]altanserin can be displaced by ketanserin.2,3 We speculate whether the agonist properties of [11C]Cimbi-36 causes the 5-HT2AR to internalize and undergo conformational changes, thereby rendering the receptor inaccessible for ketanserin displacement.
References
Ettrup A, da Cunha-Bang S, McMahon B, et al. Serotonin 2A receptor agonist binding in the human brain with [11C]Cimbi-36. J Cereb Blood Flow Metab 2014; 34(7): 1188–96. Madsen MK, Fisher PM, Burmester D, et al. Psychedelic effects of psilocybin correlate with serotonin 2A receptor occupancy and plasma psilocin levels. Neuropsychopharmacology 2019; 44(7): 1328–1334. Hansen HD, Ettrup A, Herth MM, et al. Direct comparison of [18F]MH.MZ and [18F]altanserin for 5-HT2A receptor imaging with PET. Synapse 2013; 67: 328–337.Pinborg LH, Adams KH, Svarer C, et al. Quantification of 5-HT2A Receptors in the Human Brain Using [18F] Altanserin-PET and the Bolus/Infusion Approach. J Cereb Blood Flow Metab 2003; 23(8): 985–996.
2021-15
Alpha-synuclein preformed fibril (PFF)-triggered synucleinopathy recapitulates neurochemical features of human PD: a PET study in rats (#71)
Vesna Sossi1, Joseph R. Patterson2, Christopher J. Kemp2, Kathryn M. Miller2, Anna C. Stoll2, Megan F. Duffy2, Kelvin C. Luk3 and Caryl E. Sortwell2
1Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada
2Translational Neuroscience, Michigan State University, Grand Rapids, MI, USA
3Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA, USA
Abstract
Introduction: Multiple progressive dopaminergic alterations in prodromal/manifest Parkinson’s disease (PD) have been identified with PET. Dopamine (DA) turnover increase (estimated as its inverse, the effective DA turnover (EDV); ∼ 60% reduction compared to controls in the putamen at symptom onset),1 and a reduction in DA transporter (DAT) function (11C-MP BPND; ∼ 55% reduction)2 have been observed. In contrast, DA synthesis and storage (18F-Fdopa Kocc) is relatively preserved (∼30% reduction), likely reflecting synthesis upregulation contributing to the delay between disease and symptom onset2 A recently developed PD model, induced by intracerebral injection of alpha-synuclein (a-syn) preformed fibrils (PFFs) to trigger accumulation of pathological a-syn inclusions showed progressive dopaminergic degeneration using immunohistochemical methods in a cross-sectional approach,3 with motor impairments only detectable at 6 months after PFF injection. Here we investigate (i) dopaminergic deficit longitudinally and (ii) its similarity to human PD using translational PET imaging.
Methods: 16 male Fischer 344 rats received unilateral intrastriatal injections of a-syn PFFs (n = 8) or vehicle (n = 8). Scans with 18F-Fdopa (EDV ratio (EDVR) for turnover and Kocc for DA synthesis)4 and 11C-MP (BPND) were performed on a Siemens Focus 120 at 2, 4, and 6 months following injection.
Results: A significant progressive decrease in DAT binding in the ipsilateral striatum (65% reduction at 6 months) compared to controls was found together with a progressive decrease in EDVR (55% reduction) with relatively preserved DA synthesis and storage (15% reduction) (Figures 1 and 2). Contralateral side values remained in the normal range.
Conclusion: This progressive rat model of PD shows remarkable similarities to human PD: dopaminergic deficit progression with relative preservation of DA synthesis and storage compared to DA turnover and DAT function and measurable dopaminergic deficit preceding detectable motor impairments. Synucleinopathy and degeneration induced in the a-syn PFF model may be appropriate/useful to investigate disease modifying treatments and possibly more complex disease mechanisms.
Acknowledgements
Supported by NS099416 (CES).
References
SossiVde la Fuente-FernandezRHoldenJE, et al.Changes of dopamine turnover in the progression of Parkinson’s disease as measured by positron emission tomography: their relation to disease-compensatory mechanisms. J Cereb Blood Flow Metabol2004;
24: 869–876.NandhagopalRKuramotoLSchulzerM, et al.Longitudinal evolution of compensatory changes in striatal dopamine processing in Parkinson’s disease. Brain2011;
134: 3290–3298.PattersonJRDuffyMFKempCJ, et al.Time course and magnitude of alpha-synuclein inclusion formation and nigrostriatal degeneration in the rat model of synucleinopathy triggered by intrastriatal alpha-synuclein preformed fibrils. Neurobiol Dis2019;
130: 104525.WalkerMDDinelleKKornelsenR, et al.In-vivo measurement of LDOPA uptake, dopamine reserve and turnover in the rat brain using [(18)F]FDOPA PET. J Cereb Blood Flow Metabol2013;
33: 59–66.
2021-16
Evaluation of MAO-B and TSPO PET radiotracers in a lipopolysaccharide rat model of neuroinflammation (#72)
Junchao Tong1, Vidya Narayanaswami1, Amanda J. Boyle1, Ashley C. Knight1, Christin Schifani1, Peter Bloomfield1 and Neil Vasdev1, 2
1Centre for Addiction and Mental Health, Azrieli Centre for Neuro-Radiochemistry, Brain Health Imaging Centre, Toronto, ON, Canada
2Department of Psychiatry, University of Toronto, Toronto, ON, Canada
Abstract
Introduction: The dynamic and multi-stage nature of neuroinflammation has prompted major efforts towards positron emission tomography (PET) radiotracer discovery targeting microglia and astrocyte activation for mechanistic and therapeutic advances of human psychiatric and neurodegenerative conditions.1 The goal of the current study is to apply our 2nd generation translocator protein 18kDa (TSPO) radiotracer, [18F]FEPPA, to characterize a lipopolysaccharide (LPS) rodent model of neuroinflammation and extend this paradigm to monoamine oxidase B (MAO-B) using [11C]L-deprenyl and [11C]SL25.1188, as putative biomarkers of astrocyte activation.
Methods: An acute model of neuroinflammation was developed by unilateral intra-striatal injection of LPS in adult rats (50µg/4µL). Longitudinal PET/MR (Mediso nanoScan PET/MR 3T) scans were acquired up to 1 month following surgery. Regional brain time activity curves (TACs) were extracted using a stereotaxic MRI atlas implemented in PMOD (v4.004). Tracer binding potentials (BP) for the ipsilateral striatum were estimated using simplified reference tissue model with the contralateral side as the reference tissue. Autoradiography studies were conducted to determine [3H]L-deprenyl binding in vitro in rat brain 6 months post intra-striatal LPS injection.
Results: TACs extracted from the dynamic PET data demonstrated significantly increased [18F]FEPPA uptake in the LPS-injected striatum (see static images Figure 1), with a peak BP of 1.1 at ∼1 week after surgery. [18F]FEPPA uptake in the injected side decreased gradually over time, with a BP of 0.37 1-month post LPS injection. In contrast, preliminary data showed that, whereas in vitro autoradiography showed increased specific [3H]L-deprenyl binding in the LPS-injected striatum (Figure 2), there was no increase in PET binding of [11C]SL25.1188; for [11C]L-deprenyl, however, a trend toward increased uptake was observed from 2–4 weeks.
Conclusion: The intra-striatal LPS rat model may serve as a preliminary screening tool for benchmarking newly developed tracers targeted towards distinct neuroinflammatory mechanisms. In this respect, [11C]SL25.1188 or [11C]L-deprenyl MAO-B PET imaging might not be of sufficient sensitivity for imaging astrocyte activation in vivo in rodents. Further studies are underway to extend the LPS preclinical neuroinflammatory paradigm for glial activation to alternate PET biomarkers, including MAO-B radioligands.
Reference
NarayanaswamiVDahlKBernard-GauthierV, et al.Emerging PET radiotracers and targets for imaging of neuroinflammation in neurodegenerative diseases: outlook beyond TSPO. Mol Imaging2018;
17: 1536012118792317.
2021-17
Clinical measures of upper motor neuron burden strongly associate with neuroimaging in amyotrophic lateral sclerosis (#73)
Meena M. Makary1, 2, Baileigh G. Hightower1, Paul M. Cernasov3, Beverly V. Reynolds3, James Chan4, Olivia R. Pijanowski3, Chieh-En J. Tseng1, Sabrina Paganoni3, Sheena Chew3, Nicole R. Zürcher1, Jacob M. Hooker1, Nazem Atassi3, 5 and Suma Babu3
1Department of Radiology, A. A. Martinos Center for Biomedical Imaging, Harvard Medical School, Massachusetts General Hospital, Charlestown, MA, USA
2System and Biomedical Engineering Department, Cairo University, Cairo, Egypt
3Department of Neurology, Sean M Healey & AMG Center for ALS, Harvard Medical School, Massachusetts General Hospital, Boston, MA, USA
4Department of Biostatistics, Harvard Medical School, Massachusetts General Hospital, Boston, MA, USA
5Sanofi Genzyme, Cambridge, MA, USA
Abstract
Introduction: Several quantitative rating scales are used to objectively measure upper motor neuron burden/dysfunction (UMNB) in patients with amyotrophic lateral sclerosis (ALS).1–4 However, their clinical relevance and relationship to neuroimaging findings are unknown. In this study, we correlated the MGH-UMNB and UPenn-UMNB clinical scales to (a) in-vivo brain expression of translocator protein 18kDa (TSPO) using [11C]PBR28-PET imaging, (b) diffusion-MR imaging, and (c) clinical disease severity using ALS Functional Rating Scale-Revised (ALSFRS-R).
Methods: Thirty ALS patients (52.12 ± 10.75 years (mean ± SD), 14 Females) underwent simultaneous PET/MR imaging. The ALSFRS-R (ranges 0–48 and measures motor, bulbar and respiratory functional ability), UPenn-UMNB (a longer and more comprehensive scale, ranges 0–32 and measures hyperreflexia, pseudobulbar affect and spasticity), MGH-UMNB (shorter scale, ranges 0–45 and measures hyperreflexia) scales were collected from all participants at a matching time-point. [11C]PBR28 standardized uptake value ratio (SUVR60-90) normalized to whole-brain mean uptake was calculated.2 Voxel-wise whole-brain correlation analysis was conducted between SUVR and clinical scales using FSL FEAT (Z > 2.3, Pcluster < 0.05). The skeletonized Fractional anisotropy (FA) was quantified using FSL-FDT and tract-based spatial statistics (TBSS) toolbox. Voxel-wise correlation analysis was then performed using nonparametric permutation (PFWE < 0.05). [11C]PBR28 SUVR60-90 extracted from the bilateral primary motor cortices (M1)/paracental lobule (PCL) and FA values extracted from corticospinal tract (CST) were correlated with each clinical score. All analyses were corrected for sex and TSPO genotype.
Results: Both MGH- and UPenn-UMNB scales significantly correlated with: (1) ALSFRS-R (Figure 1(a)); (2) [11C]PBR28 SUVR60-90 uptake within M1/PCL on voxel-wise whole-brain analyses (Figure 1(b)) and ROI analyses (Figure 1(c)). TBSS correlation analysis revealed that skeletonized FA (Figure 2(a)) and FA values extracted from CST (Figure 2(b)) are strongly correlated with UPenn- and MGH-UMNB scales.
Conclusion: In this cross-sectional ALS cohort, both UPenn- and MGH-UMNB scales correlate significantly against each other and localize UMN dysfunction to biologically relevant M1 brain region where glial activation and CNS axonal degeneration is known to occur. Both UPenn-, MGH-UMNB scales are significantly correlated with clinical disability status in ALS. Further evaluation in a larger and longitudinal sample is required to better characterize the usefulness and application of these scales for early diagnosis, responsiveness to change over time and prediction potential in ALS.
References
AlshikhoMJ, et al.Glial activation colocalizes with structural abnormalities in amyotrophic lateral sclerosis. Neurology2016; 13; 87(24): 2554–2561.EllisCM, et al.A proton magnetic resonance spectroscopic study in ALS: correlation with clinical findings. Neurology1998; 51(4): 1104–9.MitsumotoH, et al.Quantitative objective markers for upper and lower motor neuron dysfunction in ALS. Neurology2007; 24; 68(17): 1402–10.Woo JH, Wang S, Melhem ER, et al. Linear Associations between Clinically Assessed Upper Motor Neuron Disease and Diffusion Tensor Imaging Metrics in Amyotrophic Lateral Sclerosis. PLoS ONE 2014; 9(8): e105753.
2021-18
Development of novel PET ligands to image the receptor interacting protein kinase 1 (#77)
1Department of Clinical and Experimental Neuroimaging, National Center for Geriatrics and Gerontology, Morioka-machi, Obu, Japan
2Department of Pharmacy, Gifu University of Medical Science, Kani, Japan
3Department of Chemistry and Biomolecular Science, Gifu University, Gifu, Japan
Abstract
Introduction: Microglia is involved in the pathophysiology of neurodegenerative diseases including Alzheimer’s disease (AD). Receptor Interacting Protein Kinase 1 (RIPK1), a cell-death regulatory factor located at the plasma membrane as a part of complex I, is involved in the process of a phenotypic transition from the homeostatic microglia to the disease-associated microglia. Therefore, molecular imaging of RIPK1 could provide useful information on the phenotypic transition of microglia as well as potential mechanisms leading to neuronal death. The purpose of this study is to develop novel PET ligands to image RIPK1 in the brain.
Methods: We labeled RIPK1 inhibitor GSK’963 and its enantiomer GSK’9621 with [11C]CH3 to synthesize 11C-GG502 and 11C-GG503 respectively, by rapid cross-coupling, [11C]methylation of the corresponding boron precursors with [11C]CH3I. The affinities of 11C-GG502 and 11C-GG503 to RIPK1 were evaluated by binding assays using cell lysate overexpressing human RIPK1. Their pharmacokinetics were evaluated by PET imaging with metabolite analysis in healthy rats as well as rats with acute neuroinflammation.
Results: Both 11C-GG502 and 11C-GG503 showed a high affinity to human RIPK1 in binding assays. PET scans in rats showed a high initial brain uptake and rapid washout for both 11C-GG502 and 11C-GG503. No clear difference in pharmacokinetics was observed between the two ligands. Metabolite analysis of 11C-GG502 showed three major radiometabolites in the plasma, two of which also being detected in the brain. The rats with neuroinflammation showed no increased ligand uptake.
Conclusion: We developed two novel PET ligands,11C-GG502 and 11C-GG503, to image RIPK1 in the brain. Both showed high affinities in vitro. Specific binding PET signal of these two ligands was somewhat small. Based on the knowledge obtained here, however, we are hopeful to further develop RIPK1 ligands with specific binding signal sufficiently high for PET imaging.
Acknowledgements
This work was supported in part by a Grant-in-Aid for Creative Scientific Research (B) (JP18964417) from the Japan Society for the Promotion of Science (JSPS) and Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of Japan. We would like to thank Ms. K. Yamashita for her assistance to synthesize boron labeling precursors and nonradiolabeled authentic samples.
Reference
BergerSBHarrisPGoughPJ, et al.Characterization of GSK’963: a structurally distinct, potent and selective inhibitor of RIP1 kinase. Cell Death Discov2015;
1: 15009.
2021-19
[18F]NOS PET measurement of in vivo neuroinflammation in Parkinson’s disease and healthy humans (#93)
Robert K. Doot1, Anthony J. Young1, Tiffany L. Dominguez1, Zeinab Helili1, Hsiaoju Lee1, Regan Sheffer1, Suzanne M. Reichwein2, Erin K. Schubert1, Andrew D. Siderowf2, Henry R. Kranzler3, Robert H. Mach1, Ilya M. Nasrallah1 and Jacob G. Dubroff1
1Department of Radiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA
2Department of Neurology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA
3Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA
Abstract
Introduction: Parkinson’s disease (PD) is a progressive neurodegenerative disorder affecting 2% of the population over 60 years old1. PD symptoms are thought to result from abnormal protein accumulation that triggers events including neuroinflammation.1 Positron Emission Tomography (PET) imaging studies that target translocator protein (TSPO) have been used to measure neuroinflammation in Parkinson’s disease2, but TSPO radiotracers have limitations.2 [18F]6-(2-fluoro-propyl)-4-methylpyridin-2-amine ([18F]NOS) binds the inducible isoform of nitric oxide synthase, whose expression is induced by pro-inflammatory mediators.3,4 Our goal is to test the feasibility of using [18F]NOS images to measure differences in regional neuroinflammation between healthy controls and PD patients.
Methods: Ten volunteers provided informed consent, including 4 healthy controls (mean ± SD age = 73 ± 9 y, 2F+2M) and 6 PD patients (age = 64 ± 5 y, 2F+4M), and underwent a 0–60 minute dynamic brain scan in a Philips Ingenuity PET/CT following bolus injection of [18F]NOS (209 ± 21 MBq), with serial arterial blood sampling and metabolite analysis. Each participant also underwent magnetic resonance imaging for brain segmentation and co-registration of PET images to enable generation of 19 combined left & right regional brain time activity curves. Regional total distribution volumes (VT) were estimated using Logan plot5 and 1-tissue compartment (1TC) kinetic analyses and 1TC VT parameter maps were generated using Pmod v3.7 software. Two-sample, 2-tailed equal variances t-tests compared VT for controls with that of PD patients (SPSS v25).
Results: Population control and PD [18F]NOS VT parameter maps are in Figure 1. All 19 brain regions exhibited higher VT in PD patients than in controls for both Logan and 1TC models (p ≤ 0.01). Representative VT values from 11 regions are in Figure 2.
Conclusion: PD patients had higher average VT than controls in all brain regions, suggesting that PD subjects have higher levels of neuroinflammation than healthy controls.
Acknowledgements
Authors appreciate the contributions of Drs. Terence Gade and Gregory Nadolski for insertion of arterial blood lines. Research support was provided by Michael J. Fox Foundation (RM) and the United States of America’s National Institute on Drug Abuse of the National Institutes of Health under Award Numbers P30 DA046345 (RM), Foundation of the American Society of Neuroradiology Boerger Research Fund for Alzheimer’s Disease and Neurocognitive Disorders (IN), K01 DA040023 (RD), and K23 DA038726 (JD).
References
GlassCKSaijoKWinnerB, et al.Mechanisms underlying inflammation in neurodegeneration.Cell2010;
140: 918–934.VarnasKCselenyiZJucaiteA, et al.PET imaging of [11C]PBR28 in Parkinson’s disease patients does not indicate increased binding to TSPO despite reduced dopamine transporter binding.Eur J Nucl Med Mol Imaging2019;
46: 367–375.HerreroPLaforestRShoghiK, et al.Feasibility and dosimetry studies for 18F-NOS as a potential PET radiopharmaceutical for inducible nitric oxide synthase in humans.J Nucl Med2012;
53: 994–1001.HuangHJIsakowWByersDE, et al.Imaging pulmonary inducible nitric oxide synthase expression with PET.J Nucl Med, 2015;
56: 76–81.LoganJFowlerJSVolkowND, et al.Graphical analysis of reversible radioligand binding from time-activity measurements applied to [N-11C-methyl]-(-)-cocaine PET studies in human subjects. J Cereb Blood Flow Metab1990;
10: 740–747.
2021-20
Functional changes in Alzheimer’s disease evaluated by multimodality PET/MRI imaging (#96)
1Biomedical Imaging Research Center, University of Fukui, Eiheiji-cho, Japan
2Department of Neurology, University of Fukui, Eiheiji-cho, Japan
3Department of Psychiatry, University of Fukui, Eiheiji-cho, Japan
Abstract
Introduction: Multimodal image data simultaneously obtained by [11C]Pittsburgh compound-B (PiB) and [64Cu]ATSM PET/MRI in the early stage of Alzheimer’s disease (AD) were evaluated to observe pathophysiologic and functional changes, as well as alteration of volumes and connectivity in the brain. Various neurophysiologic parameters were compared with healthy controls (CTL).
Methods: Thirty patients with mild cognitive impairment and early dementia (69 ± 12y), and 17 age-matched CTL subjects (68 ± 11y) underwent [11C]PiB and [64Cu]ATSM PET/MRI with 70- and 40-min dynamic PET/MRI, respectively. PET data were reconstructed into multiple images to calculate cerebral blood flow (CBF) using the early phase data and the image-derived input function (IDIF) method, where time-activity curves of the arteries were extracted from PET images. Standardized uptake value ratio (SUVr) was calculated from the last 20-min PET data using the cerebellar cortex as a reference region. The MR images of 3D-T1WI, resting-state functional MRI (RS-fMRI), diffusion tensor image (DTI) and perfusion MRI from the arterial spin labeling (ASL) method acquired during the dynamic PET were also analyzed to evaluate various brain functions on MRI.
Results: All patients were determined as the early stage AD based on the result of [11C]PiB-PET and clinical findings. PiB-SUVr values in all AD cerebral cortices were significantly greater than those of CTL. The regional PET-CBF values of AD showed significantly lower values in the bilateral parietal and right temporal lobes compared with CTL, but not in ASL-MRI; however, SPM showed regional differences in both PET- and ASL-CBF. SPM analysis of RS-fMRI delineated differences in the anterior cingulate cortex and in the left precuneus. VBM analysis showed atrophic changes of the AD brain in the bilateral hippocampus. However, analysis of fractional anisotropy calculated from DTI did not show differences between the two groups. [64Cu]ATSM-SUVr images also showed differences between AD and CTL.
Conclusion: Multimodal analyses using PET/MRI scans showed differences in regional CBF, cortical volume and neuronal networks in different regions, indicating that pathophysiologic and functional changes in the AD brain can be observed in various aspects of neurophysiologic parameters. Assessment of multimodal images would be ideal to investigate pathophysiologic changes in patients with dementia and other neurodegenerative diseases.
Acknowledgements
This study was partly supported by JSPS (18H02763).
References
OkazawaHHigashinoYTsujikawaT, et al.Noninvasive method for measurement of cerebral blood flow using O-15 water PET/MRI with ASL correlation. Eur J Radiol2018;
105: 102–109.OkazawaHTsujikawaTHigashinoY, et al.No significant difference found in PET/MRI CBF values reconstructed with CT-atlas-based and ZTE MR attenuation correction. Eur J Nucl Med Mol Img Res2019;
9: 26.
2021-21
PET imaging of [11C]NCGG401 for colony stimulating factor 1 receptor (#111)
1Department of Clinical and Experimental Neuroimaging, National Center for Geriatrics and Gerontology, Obu, Japan
2Department of Pharmacy, Gifu University of Medical Science, Kani, Japan
3Department of Chemistry and Biomolecular Science, Gifu University, Gifu, Japan
Abstract
Introduction: Activated microglia are involved in the pathophysiology of Alzheimer’s disease and other neurodegenerative diseases. These glial cells have recently been recognized as a promising target for the treatment of these diseases. Here, positron emission tomography (PET) imaging of specific molecules for microglia could provide useful information needed for developing drugs that target these cells. We have been developing a PET ligand for imaging the colony stimulating factor 1 receptor (CSF1R). CSF1R is expressed exclusively on microglia. In this study, we evaluated the potential of this novel PET ligand for imaging CSF1R in rat brains using models of brain inflammation.
Methods: [11C]NCGG401, our PET ligand, which is modified from a specific CSF1R inhibitor BLZ945,1 was synthesized with a high radiochemical purity (>99%). We conducted 120-min dynamic PET scans with radioactive metabolite analyses in healthy rats with or without pre-administration of unlabeled NCGG401. The total distribution volume was calculated with the dual-input graphical analysis.2 Subsequently, we conducted PET scans in rats of brain inflammation to evaluate the ligand’s kinetics in the microglia activated brain. Also, the expression of CSF1R in the brain was evaluated ex-vivo and compared with the results of PET imaging.
Results: [11C]NCGG401 brain radioactivity showed a fast initial peak and good washout thereafter in the healthy rats. Pre-administration of unlabeled NCGG401 increased the rate of washout. The whole brain distribution volume decreased by 20% with blocking. The model rats of brain inflammation showed increased expression of CSF1R and two-times higher distribution volumes than did the healthy rats. However, no regional difference was observed probably because of the very low total expression of CSF1R in the rat brain.
Conclusion: Further studies appear warranted to confirm our findings in other rat inflammation models. In addition, knowledge of the characteristics of [11C]NCCC401, our initial CSF1R compound, we have obtained may be useful for further development and optimization of CSF1R radioligands for PET imaging of microglia.
Acknowledgements
We would like to thank Drs. Bin Ji, Maiko Ono, Chie Seki, Satoka Hashimoto, Yuji Nagai, Takafumi Minamimoto, Xaioyun Zhou, Makoto Higuchi from National Institutes for Quantum and Radiological Science and Technology. This work was supported in part by a Grant-in-Aid for Creative Scientific Research (B) (JP18964417) from the Japan Society for the Promotion of Science (JSPS) and Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of Japan.
References
PyonteckSMAkkariLJoyceJA, et al.CSF-1R inhibition alters macrophage polarization and blocks glioma progression. Nat Med2013;
19: 1264–1272.IchiseMFujitaMInnisRB, et al.Graphical analysis and simplified quantification of striatal and extrastriatal dopamine D2 receptor binding with [123I]epidepride SPECT. J Nucl Med1999;
40: 1902–1912.
2021-22
Development of 18F-labeled radioligands for the GluN2B subunits of NMDA receptors: Synthesis and evaluation in non-human primates (#123)
Ming-Qiang Zheng1, Hazem Ahmed2, Kelly Smart1, Yuping Xu3, Li Zhang1, Jim Ropchan1, Hong Gao1, Roger Schibli2, Gilles Tamagnan1, Richard E. Carson1, Yiyun H. Huang1 and Simon M. Ametamey2
1Yale PET Center, Yale University, New Haven, CT, USA
2Institute of Pharmaceutical Sciences, ETH Zurich, Zürich, Switzerland
3Jiangsu Atomic Institute, Wuxi, China
Abstract
Introduction: There is a great interest in developing PET radioligands for the GluN2B subunits of the N-methyl-D-aspartate (NMDA) receptors in humans. Previous in vitro and in vivo studies in rodents have shown (R)- and (S)-18F-OF-Me-NB1 to have promising binding properties to the GluN2B subunits [1]. Here we report the evaluation of these two radioligands and their corresponding demethylated analogs ( ± )18F-OF-NB1 and (R)-18F-OF-NB1 in rhesus monkeys.
Methods: Enantiopure (R)- and (S)-18F-OF-Me-NB1 were synthesized using the corresponding chiral precursors as reported previously.1 ( ± )-18F-OF-NB1 and enantiopure (R)-18F-OF-NB1 were prepared by copper-catalyzed 18F-fluorination of the protected boronic precursor followed by cleavage of the protecting groups. PET scans of up to 180 min each in rhesus monkeys were conducted on the Focus 220 scanner. Metabolite analysis was performed and arterial input function was calculated. Regional brain time activity curves were generated and analyzed with the one-tissue compartment (1TC) model and multilinear analysis-1 (MA1) method to obtain the respective regional volumes of distribution (VT).
Results: All radiotracers were prepared in > 95% radiochemical purity, and > 95% enantiomeric purity for the chiral compounds. In rhesus monkey metabolism was moderate with ∼30% of parent compound for (R)-18F-OF-Me-NB1 at 30 min post-injection, and ∼18% for (S)-18F-OF-Me-NB1. For ( ± )-18F-OF-NB1 and (R)-18F-OF-NB1, 26% and 57% of the parent radioligands were found at 30 min. All radiotracers displayed high uptake and similar pattern of regional distribution in the monkey brain. Highest uptake was observed in the putamen, hippocampus and thalamus, medium in the occipital cortex and cerebellum, and low in the white matter (centrum semiovale). Clearance from brain regions was the fastest for (R)- and (S)-18F-OF-Me-NB1, followed by ( ± )-18F-OF-NB1, and slowest for (R)-18F-OF-NB1 (Figure 1). Regional 1TC and MA1 VT values for these four radioligands are listed in the table and higher for ( ± )-18F-OF-NB1 and (R)-18F-OF-NB1 than (R)- and (S)-18F-OF-Me-NB1.
Conclusion: We have successfully synthesized and evaluated four radioligands targeting the GluN2B subunits of NMDA receptors in non-human primate. Blocking studies are ongoing to assess their binding specificity and selectivity.
Acknowledgements
Research support1. The Swiss National Science Foundation Grant Nos. 310030E_160403/1 and 310030E_182872/1. 2. NIH grant U01MH107803
References
HaiderA, et al. Identification and Preclinical Evaluation of a Radiofluorinated Benzazepine Derivative for Imaging the GluN2B Subunit of the Ionotropic NMDA Receptor. J Nucl Med2019;
60:259–266.
2021-23
Validation of PET-compatible chemogenetic tools in squirrel monkeys (#125)
Matthew Boehm1, 2, Hank Jedema1, Jordi Bonaventura1, Omar Gharbawie3, Juan Gomez1, Elliot Stein1, Charles Bradberry1 and Michael Michaelides1, 4
1National Institute on Drug Abuse, NIH Intramural Research Program, Baltimore, MD, USA
2Department of Neuroscience, Brown University, Providence, RI, USA
3Department of Neurobiology, University of Pittsburgh, Pittsburgh, PA, USA
4Department of Psychiatry & Behavioral Sciences, Johns Hopkins Medicine, Baltimore, MD, USA
Abstract
Introduction: Chemogenetic technologies offer an innovative approach for modulating activity of specific cell types in targeted brain regions. Designer Receptors Exclusively Activated by Designer Drugs (DREADDs), and the recently developed Pharmacologically Selective Actuator Modules (PSAMs), are two types of chemogenetic systems used to modulate neural activity. Current applications of these tools in primates and humans is limited in part by an inability to track the location and function of chemogenetic receptors in vivo. Positron emission tomography (PET) is a translational molecular imaging modality uniquely poised to enable in vivo confirmation of expression and function of chemogenetic systems. Recent studies have demonstrated the use of PET-reporter ligands for imaging chemogenetic receptors, namely [18F]-JHU37107 for DREADDs and [18F]-ASEM for PSAMs. In addition, newly developed actuator ligands JHU37160 and uPSEM817 claim improved selectivity and potency at DREADDs and PSAMs, respectively. Here we employ PET imaging methods to test the function of these chemogenetic ligands in squirrel monkeys (Saimiri sciureus).
Methods: Four monkeys underwent baseline scans with the PET-reporter ligands [18F]-JHU37107 and [18F]-ASEM (∼1.5mCi, bolus i.v.). Baseline scans were also performed with [18F]-fluorodeoxyglucose (FDG) as a measure of brain activity following saline or pretreatment with JHU37160 or uPSEM817 (0.1mg/kg, i.v.). Following baseline scans, monkeys were injected with AAV2/5-hSyn-HA-hM3Dq and AAV2/5-Syn1-PSAM4-GlyR in left hand/forelimb area of motor cortex. PET scans with [18F]-JHU37107, [18F]-ASEM and FDG will be repeated and compared with baseline scans to determine the location of chemogenetic receptors and their effects on brain activity.
Results: In baseline PET-reporter scans, [18F]-JHU37107 signal in the brain peaked at 10min post-injection and remained detectable after 90min, indicating some level of endogenous binding. Similarly, [18F]-ASEM signal peaked at 20min post-injection and remained detectable at 90min (n = 4). FDG-PET scans revealed rapid uptake of FDG in the brain, with over 95% occurring by 5min post-injection. In addition, scans following administration of JHU37160 or uPSEM817 showed no significant differences compared to baseline saline scans, suggesting these ligands do not produce off-target effects on brain activity (n = 4).
Conclusion: These findings support the use of recently developed PET-reporters and actuator ligands for DREADDs and PSAMs.
Acknowledgements
This research was supported [in part] by the Intramural Research Program of the NIH, NIDA
References
BonaventuraJEldridgeMAGHuF, et al.High-potency ligands for DREADD imaging and activation in rodents and monkeys. Nat Commun2019;
10: 4627.MagnusCJ, et al.Ultrapotent chemogenetics for research and potential clinical applications. Science2019;
364: eaav5282.
2021-24
Increased cerebral glucose consumption during hypoglycemia in obese patients measured using dynamic bolus-injection 18F-FDG PET/MR during hyperinsulinemic euglycemic and hypoglycemic clamp (#138)
Mark Lubberink1, Sofia Kvernby1, Niklas Abrahamsson2, Kristina Almby2, Markus Fahlstrom1, Malin Gingnell3, Sven Haller1, Johan Wikström1, Magnus Sundbom1, Anders Karlsson2 and Jan Eriksson2
1Surgical Sciences/Radiology & Nuclear Medicine, Uppsala University Hospital, Uppsala University, Uppsala, Sweden
2Medical Sciences, Uppsala University Hospital, Uppsala University, Uppsala, Sweden
3Psychiatry, Uppsala University Hospital, Uppsala University, Uppsala, Sweden
Abstract
Introduction: Obese patients develop less symptoms and hormonal response during hypoglycemia after gastric bypass surgery (GBP) than before.1 However, it is not known whether cerebral glucose consumption and/or blood flow (CBF), and their response to hypoglycemia, change in obese patients after gastric bypass surgery. The aim of the present work was to develop methodology and assess glucose metabolism and cerebral blood flow (CBF) response to hypoglycemia in obese patients using simultaneous PET/MR.
Methods: Eight non-diabetic obese subjects (BMI 35–45) underwent 120-min dynamic scans after bolus injection of 5 MBq/kg 18F-FDG on a Signa PET/MR. Plasma glucose levels were stabilized using a hyperinsulinemic-euglyclemic clamp, resulting in normal glycemia with a plasma glucose concentration of 5 mmol/L during the first 50 min of the scan. Between circa 50 and 80 min p.i., plasma glucose concentration was decreased to 2.7 mmol/L. Blood samples were taken to measure plasma glucose and radioactivity. Net uptake rate of 18F-FDG (Ki) and metabolic rate of glucose (MRglu) during euglycemia and hypoglycemia were calculated using a dual-phase basis function implementation of the irreversible two-tissue compartment (2T3k) model allowing for a change in rate constants during the glucose reduction phase, both for whole brain grey matter and at the voxel level. Changes between euglycemic and hypoglycemic CBF were assessed using simultaneous pseudo-continuous arterial spin labelling (pcASL). Accuracy and precision of the dual-phase model were assessed using numerical simulations.
Results: Mean Ki increased significantly from 0.025 ± 0.003 mL/cm3/min during normal glycemia to 0.059 ± 0012 mL/cm3/min during hypoglycemia (p = 0.08, Wilcoxon), corresponding to a significant increase of MRglu from 0.13 ± 0.01 to 0.16 ± 0.03 µmol/cm3/min (p = 0.015). No significant changes in CBF were found. Agreement between Ki values for normal glycemia based on the dual-phase model and 2T3k was high. Simulations showed a high correlation and agreement between simulated and fitted Ki values both at normal glycemia and hypoglycemia. Voxel-based results agreed well with whole brain gray matter results.
Conclusion: Simulations indicate that the dual-phase basis function method is able to robustly measure changes in glucose metabolism. Hypoglycemia appears to result in an overcompensation of glucose uptake rate in obese patients, resulting in slightly but significantly increased glucose metabolism compared to normal glycemia.
Reference
Abrahamssonet al. Gastric Bypass Reduces Symptoms and Hormonal Responses in Hypoglycemia. Diabetes2016; 65(9): 2667–2675.
2021-25
Dopamine release after fear conditioning as measured using bolus-infusion11C-raclopride PET-MRI (#139)
Mark Lubberink1, Andreas Frick2, Johannes Björkstrand3, Fredrik Åhs3 and Mats Fredriksson3
1Surgical Sciences/Radiology & Nuclear Medicine, Uppsala University Hospital, Uppsala University, Uppsala, Sweden
2Neurology, Uppsala University Hospital, Uppsala University, Uppsala, Sweden
3Psychology, Uppsala University, Uppsala, Sweden
Abstract
Introduction: Dopamine transmission is suggested to be crucial for aversive learning, as has been shown in animal studies. However, studies on the importance of dopamine release in human fear conditioning are lacking. The aim of the present study was to investigate dopamine release in striatum and amygdala during fear conditioning in healthy humans. In addition to equilibrium analysis and lp-ntPET,1 a nested version of SRTM2 was evaluated for assessment of dopamine release.
Methods: Eighteen volunteers underwent 90 min 11C-raclopride scans using a bolus-infusion protocol on a Signa PET/MR scanner (GE Healthcare). Fifty minutes after injection start, participants underwent a 20 min differential fear conditioning paradigm, pairing one cue (CS+) with an aversive electrical shock while another cue (CS-) was never paired with a shock. Changes in raclopride binding were assessed by equilibrium analysis, using lp-ntPET, and using a nested version of SRTM (SRTMnt). In this last case, BPND, R1 and k’2 were estimated for the first 45 min, and then R1 and k’2 were fixed for a fit to the last 20 min of data after conditioning. Accuracy and precision of SRTMnt were evaluated using simulations where dopamine release was described by gamma variate functions. Skin conductance response (SCR) to CS+ minus CS- served as the autonomic fear conditioning index.
Results:11C-raclopride BPND based on SRTMnt was significantly reduced following fear conditioning by 5.4, 5.1 and 13.4% in putamen, caudate and amygdala, respectively (p < 0.03). A significant correlation between changes in BPND across all three methods was found. Simulations showed that SRTMnt could accurately measure changes in BPND provided that dopamine release was persistent to the end of the scan, without requiring equilibrium, and that results were insensitive to changes in R1. Changes in BPND based on SRTMnt and equilibrium analysis in amygdala correlated with SCR (r = 0.54, p = 0.22 for SRTMnt).
Conclusion: SRTMnt provided a robust, easily implemented alternative for assessment of dopamine release induced changes in BPND provided that increased dopamine concentrations are persistent, which has previously been shown to be the case by microdialysis in amygdala in similar fear conditioning experiments in rodents. A significant correlation between dopamine release induced changes in BPND in amygdala and fear conditioning was found.
References
NormandinMDSchifferWKMorrisED.
▪▪▪. Neuroimage2012; 59(3): 2689–99.LammertsmaAAHumeSP. Simplified reference tissue model for PET receptor studies. Neuroimage 1996; 4(3 Pt 1): 153–8.
2021-26
Analysis of the expression of norepinephrine transporter and neuronal plasticity-related proteins in social isolation model rats (#141)
1Biomedical Imaging Research Center, University of Fukui, Yoshida-Gun, Japan
2Department of Nursing, Fukui Health Science University, Fukui, Japan
3Department of Neuropsychiatry, University of Fukui, Yoshida-Gun, Japan
Abstract
Introduction: Social isolation (SI) is closely associated with depression, but extended SI particularly from the early stage of the development can cause violence or aggression. As the pathogenesis of mood disorder, the dysfunction of the central norepinephrine (NE) system or the damage in neuronal plasticity is thought to be involved. Abnormality of the brain norepinephrine transporter (NET) has been reported in several psychiatric and neuronal disorders. Recently we developed radiobromine-labeled (S,S)-2-(α-(2-bromophenoxy)benzyl)morpholine ((SS)-BPBM) as a PET probe for brain NET imaging. Therefore, we applied (SS)-BPBM to SI model rats for analyzing the expression of NET in this study. In addition, we evaluated the expression of the neuronal plasticity-related proteins in SI model rats.
Methods: SI was loaded by housing each rat individually for 3 weeks or 8 weeks, and the rats of control group were housed in group for same period as SI-loaded group. Elevated pulse maze test and forced swim test were performed to evaluate behavior of rats. For analyzing the expression of NET, ex vivo autoradiography using (SS)-[77Br]BPBM were performed. NE metabolite (MHPG) in the blood was measured by HPLC combined with electrochemical detection. The expression of brain-derived neurotrophic factor (BDNF), phosphorylated cyclic AMP (cAMP) responsive element binding protein (pCREB), and post-synaptic density 95 (PSD95) was estimated by western blotting.
Results: After SI for 3 weeks, depressive-like behavior appeared. But paradoxically, the extension of SI to 8 weeks reduced depressive-like and anxiety-related behaviors than control. The expression of NET was not changed in each duration compared with control. SI for 3 weeks reduced the expression of BDNF than control in the hippocampus, and the changes of the expression of neuronal plasticity-related proteins were more prominent after SI for 8 weeks.
Conclusion: Brief SI induced depressive-like behavior and extended SI induced manic-like behavior in rats. Furthermore, accompanied by the extension of the duration of SI, the damage in neuronal plasticity became more severe, but the dysfunction of the central NE system was not observed.
Acknowledgements
This work was supported by JSPS KAKENHI Grant Number 16K10208, 18H02764, 18K15511, and 19K08030.
2021-27
Tau pathology is associated with synaptic loss and altered synaptic function: a combined [18F]flortaucipir, [11C]UCB-J and magnetoencephalography study (#145)
Emma M. Coomans1, 2, Deborah Schoonhoven2, 3, Hayel Tuncel1, Sander C.J. Verfaillie1, Emma E. Wolters1, 2, Ronald Boellaard1, Rik Ossenkoppele2, 4, Wiep Scheper2, 5, Patrick Schober6, Steven P. Sweeney7, Michael J. Ryan7, Robert C. Schuit1, Albert D. Windhorst1, Frederik Barkhof1, 8, Philip Scheltens2, Sandeep S.V. Golla1, Arjan Hillebrand3, Alida Gouw2, 3 and Bart N.M. van Berckel1
2Alzheimer Center Amsterdam, Department of Neurology, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, Netherlands
3Department of Clinical Neurophysiology and MEG Center, Department of Neurology, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, Netherlands
4Clinical Memory Research Unit, Lund University, Lund, Sweden
5Center for Neurogenomics and Cognitive Research, Department of Functional Genomics, Faculty of Science, Vrije Universiteit Amsterdam, Amsterdam, Netherlands
6Department of Anaesthesiology, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, Netherlands
7Rodin Therapeutics Inc., Boston, MA, USA
8Institute of Neurology, UCL, London, UK
Abstract
Introduction: The mechanisms contributing to synaptic loss in Alzheimer’s disease (AD) are poorly understood and may be associated with tau pathology. In this combined positron emission tomography (PET) and magnetoencephalography (MEG) study, we therefore investigated associations between tau ([18F]flortaucipir PET), synaptic density (synaptic vesicle 2A [11C]UCB-J PET) and synaptic function (MEG) in AD.
Methods: Seven amyloid-positive AD subjects (age 64.3 ± 8.2; 3/7 female; MMSE 24.1 ± 1.8) were included from the Amsterdam Dementia Cohort. All subjects underwent dynamic 130-min [18F]flortaucipir and 60-min [11C]UCB-J PET with arterial sampling. Six subjects underwent 10-min eyes-closed 306-channel MEG. [18F]flortaucipir binding potential (BPND) was determined with receptor parametric mapping (reference region: cerebellar gray matter) with partial volume correction. [11C]UCB-J BPND (DVR-1, plasma input-derived) was obtained using the centrum semi-ovale as reference region. [18F]flortaucipir BPND, [11C]UCB-J BPND and MEG spectral measures (total power (0.5–48 Hz), peak frequency, and relative delta (0.5–4 Hz), theta (4–8 Hz) and alpha (8–13 Hz) power) were calculated in 12 bilateral a priori defined temporal, parietal, frontal and occipital region-of-interests (ROIs). We used Spearman correlations and generalized estimating equations (GEE), correcting for multiple ROIs per subject, to investigate associations between [18F]flortaucipir, [11C]UCB-J and MEG. Analyses with MEG were stratified per brain lobe. Additionally, within-subject regional correlations (Spearman) between [18F]flortaucipir and [11C]UCB-J were performed.
Results: Higher [18F]flortaucipir BPND was associated with lower [11C]UCB-J BPND (β = -0.30, p < 0.001) (Figure 1(a) and (b)). Within subjects, negative associations were observed when neocortical tauwas high, while positive associations were observed when neocortical tau was low (Figure 1(c)). Additionally, higher [18F]flortaucipirwas strongly associated with oscillatory slowing, reflected by negative associations with peak frequency and alpha power, and positive associations with delta or theta power. In contrast, lower [11C]UCB-J binding was associated with lower peak frequency and alpha power, and higher delta power, most clearly in the occipital lobe (Table 1).
Conclusion: Across subjects, higher regional tau pathology was associated with synaptic loss, while within subjects, this association depended on subjects’ neocortical tau levels. Moreover, higher tau pathology and synaptic loss were associated with oscillatory slowing. This suggests that tau is associated with synaptic loss and reduced synaptic activity, yet that a tau-threshold may need to be reached before regional synaptic loss occurs.
2021-28
Effect of 5-day regular coffee consumption and subsequent abstention on A1 adenosine receptor occupancy and availability (#156)
David Elmenhorst1, Eva-Maria Elmenhorst2, Denise Lange2, Judith Fronczek-Poncelet1, Diego M. Baur3, Simone Beer1, Anna L. Pierling1, Tina Kroll1, Bernd Neumaier4, Daniel Aeschbach2, 5, Andreas Bauer1, 6 and Hans-Peter Landolt3
1Institute of Neuroscience and Medicine (INM-2), Forschungszentrum Jülich, Jülich, Germany
2German Aerospace Center, Institute of Aerospace Medicine, Cologne, Germany
3Institute of Pharmacology & Toxicology, University of Zurich, Zurich, Switzerland
4Forschungszentrum Jülich, Institute of Neuroscience and Medicine (INM-5), Jülich, Germany
5Division of Sleep Medicine, Harvard Medical School, Boston, MA, USA
6Department of Neurology, Heinrich-Heine-University Düsseldorf, Düsseldorf, Germany
Abstract
Introduction: The stimulating effects of commonly consumed caffeine, the major psychostimulant ingredient of coffee, are evoked through non-selective antagonism at adenosine receptors. [18F]CPFPX is a highly selective and affine ligand at the A1 adenosine receptor (A1AR) and has been successfully implemented as PET ligand. Here, we quantified by [18F]CPFPX PET the in vivo brain occupancy of A1AR after short-term coffee consumption during sleep restriction and subsequent abstinence in the human brain.
Methods: Nine healthy volunteers (29 ± 5 years, 4f/5m) completed an 8-day in-lab study including 3 x 110-min bolus plus constant infusion [18F]CPFPX PET experiments. After a baseline PET scan in the afternoon following 2 weeks of caffeine abstinence, participants consumed freshly brewed coffee for 5 consecutive days, while time in bed was restricted to 5 h per night. The administered coffee contained 200 mg caffeine at 7:30 h and 100 mg caffeine at 14:00 h. Subsequent PET scans were conducted at the same time of day as in baseline, roughly 7 h after final coffee intake in the morning and after ∼ 31 h of coffee abstention including an 8 h-sleep episode. Metabolite corrected blood samples were used to calculate steady-state distribution volumes (VT) (i.e., 50–100 min after start of [18F]CPFPX administration). Caffeine levels in saliva were determined regularly. Occupancy levels were calculated by applying the Lassen plot including cortical and subcortical areas, cerebellum and pons.
Results: Dependent on the resulting plasma concentrations of caffeine, [18F]CPFPX binding was reduced between 7 and 38%. The caffeine-dose to A1AR-occupancy relation was comparable to the previously estimated relation with an IC50 of 67 µM in plasma corresponding to 460 mg caffeine per 70 kg subject (approximately 4.5 cups of coffee). One day after coffee abstention, VT values did not differ from baseline.
Conclusion: Our preliminary data suggest that the consumption of 3 cups of coffee for 5 days does not alter A1AR brain availability after 24 h hours of caffeine abstention.
Acknowledgements
The work was supported by the Institute for Scientific Information on Coffee, the Swiss National Science Foundation (# 320030_163439) and respective institutional funds.
2021-29
The influence of scanning time window on 18F-FP-DTBZ PET in PD (#158)
1Department of Neurology, Xuanwu Hospital Capital Medical University, Beijing, China
2National Clinical Research Center for Geriatric Disorders, Beijing, China
3Department of Nuclear Medicine, Xuanwu Hospital Capital Medical University, Beijing, China
4Molecular Imaging Research Center, French Atomic Energy Commission, Laboratory of Neurodegenerative Diseases, Fontenay-aux-roses, France
5XingImaging, LLC, New Haven, CT, USA
Abstract
Introduction: In this study, we investigate the influence of scanning time window on the standardized uptake value ratios (SUVR) of 18F-FP-DTBZ PET in PD patients.
Methods: 191 patients with Parkinson’s disease (Table 1) were recruited from the movement disorder clinics and ward in Xuanwu Hospital between January 2017 and August 2019. Participants were diagnosed by movement disorder neurologists on the basis of the 2015 movement disorder society criteria. Participants were scanned for 15 minutes at around 50, 70, 90 or 110 minutes after an intravenous bolus injection of 18F-FP-DTBZ. The primary outcome measure was the caudate and putamen SUVR using the occipital cortex as the reference. We applied generalized linear models to control for age, gender and MDSUPDRSIII scores.
Results: There were 84 participants scanned within 60 minutes, 68 scanned between 60 to 80 minutes, 28 scanned between 80 to 100 minutes, and 11 scanned after 100 minutes post injection. We did not observe between-group difference of caudate and putamen SUVR (P > 0.05, Table 2). The SUVR correlated with Hoehn & Yahr stages and MDSUPDRSIII scores, but was not influenced by the scanning time window after adjusted for gender and MDSUPDRSIII scores (Table 3).
Conclusion: The standardized uptake value ratios of 18F-FP-DTBZ PET was not influenced by the scanning time window after 30 minutes of injection as equilibrium could be reached early in this cohort of PD patients. Further work will investigate the SUVRs of different frames after injection and possibly with arterial sampling data.
Acknowledgements
The National Key R&D Program of China, No. 2018YFC1312001, 2017YFC0840105; National Natural Science Foundation of China, No. 81901285. 81701726; Beijing Municipal Science & Technology Commission No. Z171100000117013.
2021-30
Sex-related differences in cerebral A1 adenosine receptor availability in the human brain (#160)
Anna L. Pierling1, Eva-Maria Elmenhorst2, Denise Lange2, Eva Hennecke2, Diego M. Baur3, Simone Beer1, Tina Kroll1, Bernd Neumaier4, Daniel Aeschbach2, 5, Andreas Bauer1, 6, Hans-Peter Landolt3 and David Elmenhorst1
1Institute of Neuroscience and Medicine (INM-2), Forschungszentrum Jülich, Jülich, Germany
2Institute of Aerospace Medicine, German Aerospace Center, Cologne, Germany
3Institute of Pharmacology & Toxicology, University of Zurich, Zürich, Switzerland
4Forschungszentrum Jülich, Institute of Neuroscience and Medicine (INM-5), Jülich, Germany
5Division of Sleep Medicine, Harvard Medical School, Boston, MA, USA
6Department of Neurology, Medical Faculty, Heinrich-Heine University Düsseldorf, Düsseldorf, Germany
Abstract
Introduction: Sex differences have been reported in terms of sleep duration, sleep efficiency, and sleep phases. Adenosine and its cerebral receptors, A1 adenosine receptor (A1AR) and A2A adenosine receptor (A2AAR), play an important role in homeostatic sleep-wake regulation. During wakefulness adenosine concentration increases, whereas it decreases during sleep. We investigated sex differences in the adenosine A1AR availability in human volunteers.
Methods: We used the radioligand [18F]CPFPX combined with positron emission tomography to quantify brain A1AR availability in 50 volunteers (20 female, 30 male, 28 ± 5 years). Following a one-week ambulatory sleep satiation protocol (9 hours time in bed, TIB), scans were performed under well-rested conditions after at least three nights in the sleep lab with 8 hours TIB. The A1AR availability was estimated in terms of the [18F]CPFPX binding potential (BPND) via the Logan’s reference tissue model (t* = 30 min) based on average k2’, resulting from the simplified reference tissue model. The cerebellum was used as a reference region. With independent t-tests we compared BPND between males and females.
Results: Grey matter was subdivided into 12 regions. BPND was regionally 12–29% higher in females than in males. Notably, in females, BPND was significantly higher in all brain regions indicating higher A1AR availability in females. The differences in BPND were particularly noticeable in regions which belong to the limbic system or are closely associated with it, such as anterior cingulum (0.57 ± 0.11 in females, 0.45 ± 0.11 in males), hippocampus (0.55 ± 0.08 in females, 0.43 ± 0.11 in males), and amygdala (0.51 ± 0.10 in females, 0.41 ± 0.11 in males).
Conclusion: Females compared to males have a higher A1AR availability in the human brain already under well-rested conditions, which could explain the known sex differences in habitual sleep duration.
Acknowledgements
We thank all volunteers for participating in the studies, and Sylvia Köhler-Dibowski from the Forschungszentrum Jülich and Annette von Waechter of the German Aerospace Center for their excellent technical assistance and support in study conductance.
2021-31
[18F]FDOPA PET imaging for prediction of treatment response in psychosis (#169)
Giovanna Nordio1, Rubaida Easmin1, Barbara Santangelo1, 2, Sameer Jauhar2, Enrico d’Ambrosio2, 3, Arsime Demjaha2, Hugh Salimbeni4, Jin Huajie5, Paul McCrone5, Federico E. Turkheimer1, Oliver D. Howes2 and Mattia Veronese1, 6
1Department of Neuroimaging/Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London, UK
2Department of Psychosis Studies/Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London, UK
3Psychiatric Neuroscience Group, University of Bari, Bari, Italy
4Department of Computing, Imperial College, Faculty of Engineering, London, UK
5King’s Health Economics, King’s College London, London, UK
6Department of Information Engineering, University of Padua, Padua, Italy
Abstract
Introduction: A substantial proportion of patients with psychosis shows limited or no response to antipsychotic treatment. Multiple evidences have shown that dopaminergic function measured with [18F]FDOPA PET imaging can be used to distinguish treatment responders from non-responders. However, the theragnostic potential of this method to identify patients who will not respond to first-line treatment has yet to be evaluated. In view of this, we aimed to evaluate this for the full [18F]FDOPA PET dynamic scan protocol (90 mins, dynamic acquisition) and for a simplified protocol (10 mins, static acquisition) that would be easier to implement in clinical practice.
Methods: The classification accuracy for treatment response of the standard dynamic measure of dopamine synthesis capacity (Gjedde-Patlak approach with cerebellum as reference region, Kicer) and a simplified index of FDOPA uptake (striatum to cerebellum activity ratio, SUVRc) was evaluated. Two independent datasets of patients and matched healthy controls were used. Reproducibility of SUVRc was also assessed in a different test-retest dataset of eight healthy volunteers.
Results: SUVRc calculated from a static 10-minute [18F]-FDOPA PET scan acquired after 75 minutes from tracer injection is strongly and significantly correlated with the standard Kicer (Spearman’s rho: 0.89, p < 0.0001 in whole striatum) (Figure 1). Both [18F]FDOPA PET approaches have good test-rest reproducibility (Kicer ICC:0.68–0.94, SUVRc ICC:0.76–0.91). Our classification model showed good predictive power to distinguish responders from non-responders (receiver operating curve area under the curve: Kicer = 0.80, SUVRc = 0.79) and similar sensitivity for identifying treatment non-responders with closed to 100% of specificity (Kicer:∼50%, SUVRc:40%-60%) (Figure 2).
Conclusion: Our study showed that [18F]FDOPA PET imaging provides reliable and reproducible measures of dopamine synthesis capacity and that it can be used to distinguish patients with schizophrenia who are unlikely to respond to first-line antipsychotic treatments from those who will respond at first episode. The classification accuracy of the simplified imaging method is comparable to the one obtained with a dynamic [18F]FDOPA PET scan. Preliminary economic analysis of [18F]FDOPA PET to fast-track treatment resistant patients to clozapine indicated a potential healthcare cost saving of ∼4,455 GBP per patient, supporting clinical applicability.
Acknowledgements
Supported by NIHR Maudsley BRC and Wellcome Trust Innovator Award (215747/Z/19/Z)
2021-32
Tetrazine-functionalised clearing agent to increase contrast in antibody imaging (#182)
Eva Schlein1, Johanna Rokka1, Tobias Gustavsson1, Jonas Eriksson2, 3, Stina Syvänen1 and Dag Sehlin1
1Department of Public Health and Caring Sciences, Uppsala University, Uppsala, Sweden
2PET Centre, Uppsala University, Uppsala, Sweden
3Department of Medicinal Chemistry, Uppsala University, Uppsala, Sweden
Abstract
Introduction: Antibodies, engineered to enter the brain, can reach concentrations up to 80-fold higher than unmodified antibodies, similar to those observed with small, lipophilic radioligands.1 Due to the long biological half-life of a radiolabelled antibody, imaging of brain targets with antibody-based ligands must be performed several days after injection. To overcome this problem we have used clearing agents (CA), which induce accelerated peripheral clearance via hepatic glycoprotein receptors immediately after administration.2,3 Here, we aimed to combine antibody-based imaging with a CA to allow imaging shortly after administration of an amyloid-β (Aβ) antibody.
Methods: The Aβ antibody RmAb1584 and its bispecific, brain penetrating variant RmAb158-scFv8D3 [1] were modified with mannose for direct increased clearance or with trans-Cyclooctene (TCO) for induced clearance. The mannose modified [125I]antibody was studied in wildtype mice using ex vivo biodistribution 24 h after injection. Transgenic (tg-ArcSwe,5 Aβ pathology model) and wildtype mice were administered with [125I]RmAb158-TCO and SPECT/CT scanned 3 days later. Immediately after, CA was administered and additional SPECT/CT scans were taken after 1 h and 1 day post CA injection. The tetrazine-functionalised CA reacts quickly with the TCO-modified antibody (invers electron-demand Diels-Adler reaction) to induce radioligand clearance from blood. Blood was sampled at defined time points after antibody injection for pharmacokinetic analysis. The perfused brain tissue was cryosectioned for subsequent ex vivo autoradiography analysis.
Results: Both strategies cleared efficiently, with a decrease of up to 90% of RmAb158. The bispecific RmAb158-scFv8D3 could not be cleared from the blood, likely due to its binding to blood cells. Hence we used RmAb158 for imaging, in combination with the tetrazine-functionalized CA, which was more efficient and has the advantage of inducible clearance. [125I]RmAb158 SPECT/CT images obtained 3 days after antibody injection showed immediate liver accumulation upon CA administration, which substantially reduced antibody concentration in blood (Figure 1). Contrast of the brain derived signal increased after 1 h CA administration and further improved after 24 h (Figure 2).
Conclusion: Results indicate that the method works as intended and that the use of clearing agents may be a promising strategy for antibody-based imaging in combination with more short-lived radionuclides.
Acknowledgements
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 813528.
References
HultqvistG, et al.Bivalent brain shuttle increases antibody uptake by monovalent binding to the transferrin receptor.Theranostics2017;
7: 308–318.RossinR, et al.Diels−Alder reaction for tumor pretargeting: in vivo chemistry can boost tumor radiation dose compared with directly labeled antibody. J Nucl Med2013;
54: 1989–1995.LeeS-P, et al.Noninvasive imaging of myocardial inflammation in myocarditis using 68Ga-tagged mannosylated human serum albumin positron emission tomography. Theranostics2017;
7: 413–424.EnglundH, et al.Sensitive ELISA detection of amyloid-β protofibrils in biological samples. J Neurochem2007;
103: 334–345.LordA, et al.The Arctic Alzheimer mutation facilitates early intraneuronal Aβ aggregation and senile plaque formation in transgenic mice. Neurobiol Aging2006;
27: 67–77.
2021-33
Measure of metabolic and electric rat brain changes induced by kainic acid, an EEG and 18FDG-PET studio (#184)
Arturo Avendaño-Estrada1, 2, Iñigo Aguirre-Aranda3, Miguel Ángel Ávila-Rodríguez2, Camilo Rios3, Roberto Olayo1, Juan Morales1 and Araceli Díaz-Ruíz3
Introduction: The kainic acid (KA) rat model is useful to evaluate new treatments and to understand biological processes involved in epilepsy. Nevertheless, metabolic processes the first 24 h in this animal model are still uncharacterized; of 2-deoxy-2-(18F)fluoro-á´-glucose positron emission tomography (18FDG-PET) is suitable to in vivo evaluate glucose brain consumption non-invasively, allowing to follow up and characterize processes robustly and quantitatively. In the KA model, the gold standard to evaluate new and established treatments is electroencephalography (EEG), nevertheless it is an invasive technique that limits experimentation. This work aimed to relate metabolic and electric brain activity the first 24 h after KA administration in rat brains.
Methods: Sixteen healthy male Wistar rats underwent 18FDG-PET scans at baseline and after kainic acid (KA) administration (1 h, 2 h, 3 h, 4 h, 21 h, 22 h, 23 h, and 24 h). Brain PET images were normalized to an atlas and segmented to locate the hippocampal regions, hippocampus-to-pons standardized uptake value ratios (SUVr) were obtained at the evaluated point times. Besides, EEGs of 5 ratas were acquired to compare electrical and metabolic activity.
Results: PET data showed hypermetabolism in the whole hippocampus (mean SUVr = 1.61) at 2 h after KA, and a hypometabolism at 24 h after KA (mean SUVr = 1.01) compared with the basal value (mean SUVr = 1.14) (see Figures 1 and 2). EEG also showed atypical values at 2 h (mean spectral power = 433216.6 µV2 and mean frequency = 4.42 Hz).
Conclusion: PET data can measure metabolic brain changes of the KA rat model the first 24 h, providing an accurate tool to in vivo and non-invasively evaluate new therapies at this phase, furthermore, PET data is more robust than EEG data. In this work, metabolic and electric changes after KA administration was characterized the first 24 h, showing PET potential to be used to evaluate in vivo, the rat model without damage the animal as EEG technique do.
Acknowledgements
The authors would like to acknowledge Victoria López and Dafne Garduño for microPET imaging; Fernando Trejo, Gabriela Contreras, Héctor Gama, Armando Flores, Adolfo Zárate, Juan C. Manríquez, Ruben Tecuapetla, Ulises Rabadan and Efraín Zamora for radiopharmaceutical production. To the Universidad Autónoma Metropolitana for the fellowship PRODEP number 740.2019.
References
KandrataviciusLAlves BalistaPLopes-AguiarC, et al.Animal models of epilepsy: use and limitations. Neuropsychiatr Dis Treat2014;
10: 1693–1705.LévesqueMAvoliMBernardC.
Animal models of temporal lobe epilepsy following systemic chemoconvulsant administration. J Neurosci Methods2016;
260: 15–52.GuoYGaoFWangS, et al.In vivo mapping of temporospatial changes in glucose utilization in rat brain during epileptogenesis: an 18F-fluorodeoxyglucose-small animal positron emission tomography study. Neuroscience2009;
162: 972–979.GoffinKVan PWDupontP, et al.Longitudinal microPET imaging of brain glucose metabolism in rat lithium-pilocarpine model of epilepsy. Exp Neurol2009;
217: 205–209.
2021-34
Simultaneous PET/fMRI reveals differential D1 and D2 receptor trafficking induced by amphetamine in NHP (#185)
Hanne D. Hansen1, 2, Martin Schain1, Helen P. Deng2, Joseph B. Mandeville2, 3, Bruce R. Rosen2, 3 and Christin Y. Sander2, 3
1Neurobiology Research Unit, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark
2Department of Radiology, Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, USA
3Harvard Medical School, Boston, MA, USA
Abstract
Introduction: Amphetamine is powerful dopamine releaser and has been shown to induce D2 receptor internalization1,2 on timescales relevant to repeated drug use. Differences in D1 vs. D2 receptor trafficking have been suggested in vitro3 but remain largely unknown in vivo. Here, we use PET/fMRI to investigate the effects of repeated amphetamine administrations on dopamine receptor trafficking.
Methods: Three anesthetized rhesus macaques were scanned twice each in a PET/MR scanner. [11C]Raclopride was administered as a bolus+infusion with a scan duration of 2 h. Amphetamine (0.6 mg/kg) was given intravenously at 40 min as a within-scan challenge, while functional MRI with iron oxide contrast agent was acquired simultaneously. A second amphetamine injection was given after 3 h using the same PET/fMRI acquisitions. In an analogous set of imaging sessions, the antagonist SCH23390 was administered to block D1 receptors prior to radiotracer injection. FMRI data were analyzed with a general linear model and converted to relative changes in cerebral blood volume (CBV). Receptor availability (BPND) was quantified using a modified SRTM with a time-dependent binding term.4,5
Results: Amphetamine caused a prolonged reduction in BPND (peak occupancy: 28 ± 8% (mean ± SD), n = 6, Figure 1(a)), together with a short-lived decrease in CBV (Figures 1(c) and 2(a)). Repeated amphetamine administration after 3 h caused a further reduction in BPND (peak occupancy: 19 ± 9%, n = 3, Figure 1(b)). In contrast, the second CBV response differed entirely from the first, showing a long-lasting increase in CBV (Figures 1(d) and 2(b)). Administration of SCH23390 did not alter the overall shape of the amphetamine-induced changes in CBV at 0 h (Figures 1(c) and 2(c)). However, the long-lasting positive CBV response at 3 h was reduced to a short-lived positive response (Figures 1(d) and 2(d)).
Conclusion: These results suggest that D1 receptors recycle and are functionally available within three hours, whereas D2 receptor availability stays decreased for 24–48 h after amphetamine exposure (1). Overall, this study provides novel insight into the differentially lasting effects of amphetamine exposure on D1 vs. D2 receptor dynamics in vivo. It highlights a potentially important role of not only D2 but also D1 receptor internalization mechanisms involved in substance use disorders.
References
NarendranRSlifsteinM, et al.Amphetamine-induced dopamine release: duration of action as assessed with the D2/3 receptor agonist radiotracer (––)-N-[11C]propyl-norapomorphine ([11C]NPA) in an anesthetized nonhuman primate. Synapse2007;
61: 106–109.SkinbjergMLiowJ-S, et al.D2 dopamine receptor internalization prolongs the decrease of radioligand binding after amphetamine: a PET study in a receptor internalization-deficient mouse model. Neuroimage2010;
50: 1402–1407.BartlettSEEnquistJ, et al.Dopamine responsiveness is regulated by targeted sorting of D2 receptors. Proc Natl Acad Sci2005;
102: 11521–11526.IchiseMLiowJS, et al.Linearized reference tissue parametric imaging methods: application to [11C]DASB positron emission tomography studies of the serotonin transporter in human brain. J Cereb Blood Flow Metab2003;
23: 1096–1112.SanderCYHookerJM, et al.Neurovascular coupling to D2/D3 dopamine receptor occupancy using simultaneous PET/functional MRI. Proc Natl Acad Sci USA2013;
110: 11169–11174.
2021-35
In vivo evidence for disrupted association between synaptic and glutamatergic markers in depression – A combined [11C]UCB-J and [18F]FPEB study (#188)
Sophie Holmes1, Mika Naganawa2, Jean-Dominique Gallezot2, Takuya Toyonaga2, Robert H. Pietrzak1, Richard E. Carson2 and Irina Esterlis1
1Psychiatry Department, Yale University, New Haven, CT, USA
2Radiology Department, Yale University, New Haven, CT, USA
Abstract
Introduction: Converging evidence implicates a loss of synaptic connections in depression.1 A related pathophysiological mechanism is glutamate dysfunction, and the metabotropic glutamate receptor 5 (mGluR5) has been specifically implicated in symptoms of depression.2 In two separate Positron Emission Tomography (PET) studies, we demonstrated lower synaptic density in individuals with major depressive disorder (MDD),3 and that modulation of mGluR5 can alleviate depressive symptoms.4 However, the relationship between synaptic density and the glutamatergic system has not been investigated in vivo. We examined the association between the densities of presynaptic SV2A and postsynaptic mGluR5 in the same subjects in both MDD and healthy controls (HC) groups.
Methods: Twelve healthy controls (mean ± SD = 44 ± 16yrs; 5 women) and 20 individuals with major depressive disorder (MDD; 40 ± 11yrs; 16 unmedicated; 10 women) participated in one [11C]UCB-J and one [18F]FPEB scan (measuring SV2A density and mGluR5 availability, respectively). Outcome measure was volume of distribution (VT) for both scans. Correlations between [11C]UCB-J and [18F]FPEB VT were assessed using Pearson’s rin 11 ROIs across the brain. Differences between correlations were assessed using Fisher’s r to z transformation.
Results: We observed multiple significant positive correlations between [11C]UCB-J and [18F]FPEBVTin the HCs across regions (Figure 1(a)). The strongest correlations converged on dlPFC, cerebellum and posterior cingulate cortex (r’s > 0.60,p’s < 0.05). Conversely, in individuals with MDD, there were no significant correlations between SV2A and mGluR5 in any ROIs (all r’s < 0.1; Figure 1(b)). The difference in correlations between HC and MDD groups was significant across multiple ROI pairs, converging on the dlPFC and cerebellum (Figure 2).
Conclusion: We demonstrate a disconnect between synaptic density and glutamatergic tone in individuals with MDD across multiple ROIs. The significant correlation between SV2A and mGluR5 in HCs is in line with research indicating a relationship between mGluR5 and synaptic plasticity.5 The lack of association in MDD could reflect a disconnect between pre- and post-synaptic function, or an aberrant association between mGluR5 and synaptic plasticity in depression. Further work should evaluate whether synaptogenic treatments such as ketamine ameliorate coupling between synaptic markers and mGluR5.
References
DumanRSAghajanianGKSanacoraG, et al.Synaptic plasticity and depression: new insights from stress and rapid-acting antidepressants. Nature Med2016;
22: 238.KrystalJHMathewSJD’SouzaDC, et al.Potential psychiatric applications of metabotropic glutamate receptor agonists and antagonists. CNS Drugs2010;
24: 669–693.HolmesSEScheinostDFinnemaSJ, et al.Lower synaptic density is associated with depression severity and network alterations. Nature Commun2019;
10: 1529.EsterlisIDellaGioiaNPietrzakRH, et al.Ketamine-induced reduction in mGluR5 availability is associated with an antidepressant response: an [11 C] ABP688 and PET imaging study in depression. Mol Psychiatry2018;
23: 824.PiersTKimDHKimB, et al.Translational concepts of mGluR5 in synaptic diseases of the brain. Front Pharmacol2012;
3: 423.
2021-36
Translocator protein in occupational post-traumatic stress disorder: Preliminary findings using the [18F]FEPPA PET radioligand (#189)
Sarah E. Watling1, 2, Talwinder Gill1, 2, Erin Gaudette1, 2, Junchao Tong1, 3, Tina McCluskey1, 3, Jeffrey Meyer1, 4, Jerry Warsh1, 4, Rakesh Jetly5, Don Richardson6, 7, Pablo M. Rusjan1, 8, Romina Mizrahi1, 8, Sylvain Houle1, 4, Stephen Kish1, 4, Michael G. Hutchison9, Neil Vasdev1, 10, Shawn Rhind11 and Isabelle Boileau1, 4
1Research Imaging Centre, Centre for Addiction and Mental Health, Toronto, ON, Canada
2Institute of Medical Sciences, University of Toronto, Toronto, ON, Canada
3Campbell Mental Health Research Institute, Centre for Addiction and Mental Health, Toronto, ON, Canada
4Department of Psychiatry, University of Toronto, Toronto, ON, Canada
5Department of National Defence, Canadian Forces Health Services, Ottawa, ON, Canada
6Department of Psychiatry, University of Western ON, London, ON, Canada
7McDonald Franklin OSI Research Centre, Parwood Institute, London, ON, Canada
8Department of Pharmacology and Toxicology, University of Toronto, Toronto, ON, Canada
9Faculty of Kinesiology and Physical Education, University of Toronto, Toronto, ON, Canada
10Azrieli Centre for Neuro-Radiochemistr, Centre for Addiction and Mental Health, Toronto, ON, Canada
11Defence Research and Development Canada, Toronto Research Centre, Toronto, ON, Canada
12Parkwood Operational Stress Injury Clinic, London ON, Canada
Abstract
Introduction: Inflammation, and specifically, the 18 kDa mitochondrial translocator protein (TSPO, a marker of gliosis) has been implicated in the pathophysiology of Post-traumatic stress disorder (PTSD). To date, two positron emission tomography (PET) studies have investigated brain TSPO, reporting decreased TSPO in patients diagnosed with PTSD1 and increased TSPO in 9/11 first responders experiencing PTSD symptoms.2 Therefore, more research is required to characterize immune-dysregulation in PTSD. Accordingly, the purpose of this study was to utilize PET of the TSPO probe N-acetyl-N-(2-[18F]fluoroethoxyben-zyl)-2-phenoxy-5-pyridinamine ([18F]FEPPA) to investigate TSPO binding in humans with occupational related PTSD.
Methods: TSPO binding was measured with PET and arterial sampling in 15 patients diagnosed with PTSD and 17 healthy controls (HC). A magnetic resonance image was acquired for delineation of regions of interest (ROIs). A repeated-measures analysis of covariance (ANCOVA) was employed to evaluate group differences within 6 ROIs in the limbic-striatum (prefrontal cortex [PFC], anterior cingulate cortex [ACC], insula, striatum, hippocampus, amygdala), controlling for the TSPO polymorphism (rs6971).
Results: The PTSD (∼ 45 years) and HC (∼ 31 years) groups did not differ in sex, TSPO polymorphism, ethnicity, or body mass index (p > 0.05); however PTSD patients were significantly older than HCs (p = 0.005). A repeated-measures ANCOVA with TSPO polymorphism as a covariate revealed no significant difference in TSPO binding between PTSD and HCs (F(1,26) = 0.220, p = 0.643) and no group by ROI interaction (F(2.543,66.109) = 1.203, p = 0.313). Interestingly, there was a marginally significant group by genotype interaction (F(1,23) = 2.87, p = 0.101), whereby TSPO binding was 34% lower in PTSD middle affinity binders (MABs) compared to HC MABs, and 46% higher in PTSD high affinity binders (HABs) compared to HC HABs. TSPO binding in the amygdala was correlated with Beck Depression Inventory scores (p = 0.023, R = 0.7).
Conclusion: The current study investigating TSPO binding in a cohort of patients with occupational related PTSD did not replicate the recent finding of decreased TSPO in PTSD. This finding highlights the importance of characterising type of trauma in study samples. Furthermore, the group by genotype interaction was an unexpected finding and merits further investigation to better understand the potential role this genotype may or may not have in PTSD.
References
BhattSHillmerATGirgentiMJ, et al.PTSD is associated with neuroimmune suppression: evidence from PET imaging and postmortem transcriptomic studies.Nat Commun2020;
11: 1–11.DeriYCloustonSADeLorenzoC, et al.Neuroinflammation in World Trade Center responders at midlife: a pilot study using [18F]-FEPPA PET imaging.Brain Behav Immun Health2021;
16: 100287.
2021-37
The PET radioligands ([11C]NR2B-X; X = Me, R or S; X = SMe, S) are selective for binding to the NR2B subunit of NMDA over the s1 receptor in rat in vivo (#191)
Lisheng Cai, Jeih-San Liow, Cheryl Morse, Sanjay Telu, Riley Davies, Emily Feigen, Robert B. Innis and Victor W. Pike
Molecular Imaging Branch, NIMH, Bethesda, MD, USA
Abstract
Introduction: The NR2B subunit within the NMDA complex and sigma receptors (σ1, σ2) are physically intertwined.1 In our effort to develop PET radioligands for PET imaging of the NR2B receptor, we found that some ligands have quite high affinities for s-receptors.2 Here we used our candidate radioligands targetting NR2B ([11C]NR2B; X-Me, R- or S-enantiomer; X = SMe; S-enantiomer) and an established radioligand targetting s1 receptors ([18F]FTC146)3 with various challenge agents to investigate whether s1 receptors influence the PET imaging of NR2B in rat brain.
Methods: PET imaging of brain was performed after intravenous administration of an [11C]NR2B-X radioligand or [18F]FTC146 to rats at baseline and after intravenous administration of ligands putatively selective for NR2B, such as NR2B-SMe and Ro 25 6981, or σ1, such as FTC146, and BD1047 at a range of doses (each typically at 0.01 − 3 mg/kg), at 10 min before radioligand injection. For each radioligand, a dose-response curve was established for each blocking agent.
Results: FTC146 gave an ED50 value of 46 nmol/kg body for self-blockade of [18F]FTC146 binding in rat brain, whereas ED50 values were far lower for blockade of the three tested NR2B radioligands depending on radioligand (i.e., 2571, 725, or > 1000 nmol/kg). Similarly, BD1047 gave an ED50 of 169 nmol/kg for blockade of [18F]FTC146, and far lower ED50 values for blockade of the three NR2B radioligands (>1000, 555, or 900 nmol/kg). NR2B-SMe, gave a weak ED50 of 1064 nmol/kg against [18F]FTC146 and a strong ED50 value of 9.5 nmol/kg against [11C](S)-NR2B-SMe (Figure 1).
Conclusion: The ED50 values of FTC146, BD1047, and NR2B-SMe depended on the radioligand used in PET imaging of rat brain. Both FTC146 and BD1047 potently blocked rat brain s1 receptors, affirming the reported high s1 selectivity of [18F]FTC146. FTC146 was extremely weak at blocking the NR2B radioligands. (S)-NR2B-SMe potently blocked the test NR2B radioligand [11C](S)-NR2B-SMe, but not [18F]FTC146. Taken together the data indicate that the influence of s1 receptors on the PET imaging of NR2B receptors with the tested [11C]NR2B radioligands is likely negligible.
Acknowledgements
Intramural Research Program of the National Institutes of Health (NIMH).
References
PabbaMSibilleE. Sigma-1 and N-Methyl-d-Aspartate Receptors: A Partnership with Beneficial Outcomes. Mol Neuropsychiatry2015;
1: 47.CaiL, et al. Evaluation of 11C-NR2B-SMe and Its Enantiomers as PET Radioligands for Imaging the NR2B Subunit Within the NMDA Receptor Complex in Rats. J Nucl Med2020;
61: 1212–1220.JamesML, et al. New positron emission tomography (PET) radioligand for imaging σ-1 receptors in living subjects. J Med Chem2012;
55: 8272.
2021-38
Generation of a normative dopamine neuroreceptor template from [11C]PHNO PET imaging modelling population variability (#195)
Alessio Giacomel1, 2, Ottavia Dipasquale1, Robert Mccutcheon3, Tarik Dahoun3, Matthew Nour3, Oliver D. Howes3, Alessandra Bertoldo2, Federico E. Turkheimer1 and Mattia Veronese1
1Department of Neuroimaging, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London, UK
2Department of Information Engineering, Padova University, Padova, Italy
3Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London, UK
Abstract
Introduction: Brain normative templates are becoming more and more popular for detecting structural and functional abnormalities in brain disorders,1 as well as to characterise pharmacological response2 in imaging studies. In this work we tested and validated a non-hierarchical multiple linear modelling approach for the generation of a normative template for the dopamine neuroreceptor system as measure with [11C]PHNO PET imaging. The aim was to investigate the importance of modelling the population variability when comparing individual data with normative templates.
Methods: 90 minutes [11C]PHNO Brain PET scans (age: 26.09 ± 7.17 years, BMI: 28.28 ± 2.63 kg/m2, 19/25 Females/Males) of forty-four healthy controls were reanalysed from previous studies of our internal database.3 Individual BPND generated with SRTM were normalised into standard MNI space via individual MRI transformations. Regional BPND values in cortical and subcortical regions (Striatum, Thalamus, Midbrain, Substantia Nigra and Hippocampus) we correlated with individual covariates (age, gender and BMI) that are known to be associated with dopamine function.4,5 Based on these analyses and these covariates we formulated seven BPND population models using linear-mixed effect modelling. We compared the predicted parametric BPND maps of these models with the average maps using a leave-one-out cross validation approach. Mean absolute error (MAE) computed from the distribution of voxel residuals was used as a performance index.
Results: Regional BPND estimates were significantly associated to age (r > 0.46, Cortex and Striatum) and BMI (r = 0.30 Cortex) (Figure 1), while gender did not reach statistical significance (highest correlation in the Striatum, r = 0.28, p = 0.07). The best performing model used a combination of all these variables as described by BPND = θ+αAge+βBMI+δSex, where the effects of BMI and gender were modulated depending on the voxel anatomical location (Figure 2). Cross-validation showed that the model compared to the simple average improved prediction by 32% ± 30% in the whole brain, with regional improvement ranging from 4% ± 63% in hippocampus to 23% ± 18% in the Striatum.
Conclusion: We showed that including covariates in the creation of normative templates improves the prediction of individual data as compared to simple population mean. The same approach can be extended to other targets in order to obtain normative templates of different neurotransmitters.
References
BeliveauV, et al.A high-resolution in vivo atlas of the human brain’s serotonin system. J Neurosci2017;
37: 120–128.DipasqualeOSelvaggiPVeroneseM, et al.Receptor-Enriched Analysis of functional Connectivity by Targets (REACT): a novel, multimodal analytical approach informed by PET to study the pharmacodynamic response of the brain under MDMA. Neuroimage2019;
195: 252–260.NODE-PET, https://www.maudsleybrc.nihr.ac.uk/research/precision-psychiatry/neuroimaging/node-pet/ (accessed 9 November 2021).MatuskeyD, et al.Age-related changes in binding of the D2/3 receptor radioligand [11C](+)PHNO in healthy volunteers. Neuroimage2016;
130: 241–247.GaiserEC, et al.Elevated dopamine D2/3 receptor availability in obese individuals: a PET imaging study with [11C](+)PHNO. Neuropsychopharmacology2016;
41: 3042–3050.
2021-39
mGluR5 as a biomarker for suicidal behavior in trauma related disorders: Evidence from in vivoPET imaging studies (#200)
Margaret T. Davis, Ansel T. Hillmer, Sarah Debonee, Olivia Wilson, Nabeel Nabulsi, David Matuskey, Gustavo A. Angarita, Richard E. Carson and Irina Esterlis
Department of Psychiatry, Yale School of Medicine, New Haven, CT, USA
Abstract
Introduction: Suicide remains a pressing public health concern, yet few pharmacological interventions are capable of reducing suicide risk.1 Recent findings implicate dysregulation of mGluR5 – a metabotropic glutamate receptor – in the pathophysiology of suicidal behavior.2,3 Using positron emission tomography (PET) with the tracer 18F-FPEB, we quantified mGluR5 availability in vivo in three clinical populations: post-traumatic stress and major depressive disorders (PTSD, MDD; Study 1), and borderline personality disorder (BPD; Study 2).
Methods: Study 1:Participants [(n = 29 PTSD, 14 with scan-day suicidal ideation (SI); n = 29 MDD, 15 with SI]. Study 2: [(n = 14 MDD, 8 with suicide attempt (SA) history; n = 14 BPD, 6 with SA history] completed a single 18F-FPEB PET scan to quantify mGluR5 availability in vivo. Radiotracer was injected as bolus plus constant infusion and subjects were scanned during steady state (90–120mins post-injection). All participants completed a comprehensive battery of psychiatric rating scales and cognitive testing. Volume of distribution (VT: ratio of parent radioligand concentration in tissue relative to that in plasma) in grey matter regions was computed using a venous input function. Analyses focused on 5 fronto-limbic brain regions.
Results:Study 1: MANOVA with mGluR5 availability (VT) in specified regions of interest as dependent variable (DV) and SI as independent variable (IV) revealed a main effect of SI for those with PTSD (p = .009) but not MDD. Post hoc analyses indicated PTSD-SI group had higher mGluR5 availability in all ROIs (p’s < .007, 21–29% difference; Figure 1). Study 2: MANOVA with VT as DV, SA as IV revealed a main effect of SA in BPD; higher mGluR5 availability was associated with history of SA (p’s < .042, 19–23% difference; Figure 2) in BPD but not MDD. Relationships between SI and mGluR5 in Study 2 were non-significant.
Conclusion: Higher mGluR5 availability was associated with scan day SI in PTSD, and SA history in BPD. Observed differences in mGluR5 availability may reflect differences in glutamate neurotransmission between those who develop suicidal behavior in trauma-related disorders. Findings underscore the importance of continued investigation of mGluR5 as a target for intervention and suicide risk management in psychiatry.
Acknowledgements
This study was funded in part by the VA National Center for PTSD, and NIH grants R01MH104459 (PI: Esterlis) and 1K08MH117351-01 (PI: Davis).
References
LutzPMechawarNTureckiG.
Neuropathology of suicide: recent findings and future directions.Mol Psychiat2017;
22: 1395–1412.ChandleyMJSzebeniASzebeniK, et al.Elevated gene expression of glutamate receptors in noradrenergic neurons from the locus coeruleus in major depression. Int J Neuropsychopharmacol2014;
17: 1569–1578.ZhaoJVerwerRvan WamelenD, et al.Prefrontal changes in the glutamate-glutamine cycle and neuronal/glial glutamate transporters in depression with and without suicide. J Psychiat Res2016;
82: 8–15.
2021-40
Neuroinflammation in epilepsy: A systematic review of microglial activation imaging studies (#203)
Abhishekh H. Ashok1, 2 and Tomasz Matys1, 2
1Department of Radiology, University of Cambridge, Cambridge, UK
2Department of Radiology, Cambridge University Hospital NHS Foundation Trust, Cambridge, UK
Abstract
Introduction: Magnetic resonance imaging (MRI) and FDG- Positron emission tomography (PET) is part of a standardized presurgical epilepsy imaging protocol in several centres. FDG-PET do not accurately predict seizure foci in a substantial number of patients, suggesting the need for novel imaging method. Preclinical studies have implicated neuroinflammation and microglial activation in epileptogenesis and pathophysiology of epilepsy. PET radiotracers which bind to the Translocator Protein (TSPO), an outer mitochondrial membrane protein expressed by microglia, have been used to evaluate neuroinflammation in vivo.1 In this systematic review, we evaluated the studies which measured microglial activation in animal models and patients diagnosed with epilepsy.
Methods: Systematic review was conducted in PubMed/Medline with the following keyword: (Positron Emission Tomography OR PET OR Single photon emission tomography OR SPET OR Single Photon Emission Computed Tomography OR SPECT OR autoradiography) AND (epilepsy OR seizure) AND (microglia* OR microglia* activation OR TSPO OR Translocator protein OR peripheral benzodiazepine receptor OR peripheral benzodiazepine binding site). Out of the 60 articles, 14 preclinical, 4 human in-vitro and 3 human PET studies met the inclusion criteria.
Results: All of the preclinical studies consistently reported elevation of TSPO signal. Four in-vitro human studies conducted using post-mortem or surgically resected specimen showed elevated microglial activity. Three in-vivo TSPO studies detected increased uptake of radioactivity in the ipsilateral brain region to the seizure focus in patients with temporal lobe epilepsy. Two of these studies reported increased TSPO signal in patients compared to healthy controls. These studies were conducted with [11C]PBR28 radiotracer with a ratio of distribution volume to free fraction (VT/fP) as outcome parameter.
Conclusion: Preclinical studies consistently reported elevation of TSPO activity in epilepsy. In-vivo TSPO PET studies suggest a promising role of TSPO imaging in the evaluation of seizure foci. Future large-scale studies are needed to confirm these findings.
Acknowledgements
Dr Ashok would like to thank National Institute of Health Research (NIHR), UK for funding.
References
ScottGMahmudMOwenDR, et al.Microglial positron emission tomography (PET) imaging in epilepsy: applications, opportunities and pitfalls. Seizure2017;
44: 42–47.
2021-41
Towards an opioid receptor atlas of the human brain (#206)
Douglas N. Greve1, Hsiao-Ying Wey1, Lauri Tuominen2, Tomi Karjalainen3 and Lauri Nummenmaa3
1Massachusetts General Hospital, Martinos Center, Charlestown, MA, USA
2Royal’s Institute of Mental Health Research, Ontario, ON, Canada
3Turku PET Centre and Department of Psychology, University of Turku, Turku, Finland
Abstract
Introduction: The opioid system plays a fundamental role in human behavior and experience including addiction, depression, and anxiety.1 This abstract describes an effort to create a publicly available atlas of the human opioid system to be made available through FreeSurfer/PETsurfer (surfer.nmr.mgh.harvard.edu/fswiki/PetSurfer).
Methods: Two opioid tracers were evaluated. 18 subjects (43 ± 11y, 8M/9F) were scanned using carfentanil (CFN, GE Discovery 690) and 7 subjects (3M/4F, 24.1 ± 2.7y) were scanned using diprenorphine (DPN, Siemens 3T-BrainPET). All subjects also had T1-weighted MRIs. All data were analyzed in FreeSurfer (FS)/PETsurfer.2,3 After registering the PET to the MRI, the time activity curves (TACs) were mapped onto the common surface (fsaverage) where they were surface-smoothed; TACs were also mapped to the MNI152 through a nonlinear volume registration where a subcortical GM mask was applied prior to volume smoothing. TACs were also averaged in FS-defined ROIs. After mapping and smoothing or ROI-averaging, BPND was computed using MRTM1 using the individual’s FS-defined calcarine sulcus as the reference region.
Results: Figure 1 shows the cross-subject averages for each tracer on the surface (A and B) and in the volume with the subcortical results merged with the cortical maps. Generally, the maps have the typical spatial distribution expected from the m-opioid system (low in primary sensory areas and high in frontal, limbic, and basal ganglia). CFN and DPN are highly correlated across cortex (Figure 2) though there are some clear differences (e.g., middle temporal). Some differences are expected because DPN has some affinity for kappa and delta in addition to m receptors whereas CFN is exclusive to mu.
Conclusion: Brain atlases are a useful as a neuroscience resource.4,5 The opioid atlases will consist of mean BPND maps in a common space (cortical surface and volume) as well as tables of average cortical and subcortical ROIs. They will be made publicly available through FreeSurfer/PETsurfer. It is our intention to gather opioid PET data from all available sources for all opioid receptor types to provide a comprehensive picture of the spatial distribution of opioid receptors in the human brain.
References
NummenmaaLTuominenL.
Opioid system and human emotions. Br J Pharmacol2018;
175: 2737–2749.GreveDN, et al.Cortical surface-based analysis reduces bias and variance in kinetic modeling of brain PET data. Neuroimage2014;
92: 225–236.GreveDN, et al.Different partial volume correction methods lead to different conclusions: an (18)F-FDG-PET study of aging. Neuroimage2016;
132: 334–343.BeliveauV, et al.A high-resolution in vivo atlas of the human brain’s serotonin system. J Neurosci2017;
37: 120–128.HawrylyczMJ, et al.An anatomically comprehensive atlas of the adult human brain transcriptome. Nature2012;
489: 391–399.
2021-42
DaTscan-based progression subtypes for Parkinson’s disease (#213)
Yuan Zhou1, Sule Tinaz2 and Hemant Tagare1, 3
1Department of Radiology and Biomedical Imaging, Yale University, New Haven, CT, USA
2Department of Neurology, Yale University, New Haven, CT, USA
3Department of Biomedical Engineering, Yale University, New Haven, CT, USA
Abstract
Introduction: Parkinson’s disease (PD) is known to progress heterogeneously.1 We propose to use DaTscan images from the Parkinson’s Progression Marker Initiative (PPMI) to find PD progression subtypes using a mixture of linear dynamical systems (MLDS) model (see Figure 1).
Methods: We analyzed DaTscan images from 365 PPMI subjects scanned at baseline, and approximately at years 1,2,4,5 with some missing data (Male: 237, Female:128, Age: 62.6 ± 9.8 yrs.). Mean striatal binding ratios (SBRs)2 for the left caudate (LC), left putamen (LP), right putamen (RP), and right caudate (RC) were extracted from the images. The MLDS model with K progression subtypes assumes that each progression subtype is a multi-variate differential equation with a fixed transition matrix, which models the coupled progression of the four SBRs (see Figure 2). A t-distributed model fitting error allows for outliers.3 The model is fitted to the image data using Bayesian inference with collapsed Gibbs sampling.4 This allows the identification of the number of subtypes as well as the transition matrix of each subtype. Finally, the total movement scores (TMS, sum of MDS-UPDRS Part III scores) within each progression subtype are evaluated for comparison.
Results: Bayesian model selection5 (as well as cross validation) identifies three progression subtypes in the PPMI dataset. The transition matrices of all subtypes have leading skew-symmetric eigenvectors and trailing symmetric eigenvectors. The leading eigenvalues in the three subtypes are -0.71, -0.37, -0.37 and the trailing eigenvalues are -0.20, -0.09, -0.02. Thus the subtypes are identified as fast (14.6% of the subjects), medium (61.2%), and slow (24.2%) progressing. The corresponding TMS progression rates of these subtypes are 4.11, 2.48 and 2.42 per year.
Conclusion: The MLDS model identifies three DaTscan-based progression subtypes from the PPMI dataset. In each subtype, the skew-symmetric eigenvector identifies the fastest progression time constant. This has the potential to serve as a reliable disease progression marker. The TMS of the subtypes also show analogous progression speeds providing clinical corroboration for the validity of our approach.
Acknowledgements
Funding source: R01NS107328.
References
ThenganattMAJankovicJ.Parkinson disease subtypes. JAMA Neurol2014;
71: 499–504.InnisRBCunninghamVJDelforgeJ, et al.Consensus nomenclature for in vivo imaging of reversibly binding radioligands.J Cereb Blood Flow Metabol2007;
27: 1533–1539.PeelDMcLachlanGJ.
Robust mixture modelling using the t distribution.Stat Comput2000;
10: 339–348.MurphyKP. Machine learning: a probabilistic perspective.
Cambridge, MA:
MIT Press, 2012.McLachlanGJRathnayakeS.On the number of components in a Gaussian mixture model.Wiley Interdiscip Rev Data Min Knowl Discov2014;
4: 341–355.
2021-43
A single session of theta burst stimulation alters metabolic activity in the core depression network (#230)
Lauri Tuominen1, Ines Jani1, Cecelia Shvetz1, Abir Gebara1, Juho Joutsa2 and Sara Tremblay1
1The Royal’s Institute of Mental Health Research, Ottawa, ON, Canada
2Turku PET Centre, Turku University Hospital, Turku, Finland
Abstract
Introduction: Intermittent theta burst stimulation (iTBS) to the left dorsolateral prefrontal cortex (DLPFC) has emerged as a promising new treatment for depression.1 Effects of left DLPFC iTBS in humans however are not well understood. In this sham controlled double blind randomized cross-over study, we tested the effects of a single session of iTBS on [18F]FDG uptake and resting state fMRI (rsfMRI) in healthy individuals.
Methods: Planned unblinding and interim analyses were carried out on 8 healthy individuals (6 females, mean ± sd age 27.5 ± 8.2). All subjects underwent a standard iTBS stimulation (80% active motor threshold; 600 pulses) and a sham stimulation on separate days. iTBS was targeted to the DLPFC using neuronavigation (MNI coordinates -38, 44, 26).1 [18F]FDG was injected as a bolus immediately after both stimulations. Between the bolus and data collection, subjects rested for 40 minutes eyes open in a quiet and dimly lit room. [18F]FDG emission data was collected for 30 minutes simultaneously with rsfMRI using a Siemens PET/MRI scanner. Difference between iTBS and sham in standardized uptake value ratio images and rsfMRI connectivity was calculated, and effects of stimulation were analyzed using voxel-wise one sample t-tests (Figure 1). All tests were thresholded at voxel-wise p-values < 0.01, uncorrected.
Results: Preliminary results suggest that compared to sham stimulation, iTBS increased [18F]FDG uptake in the pregenual anterior cingulate cortex (pgACC), in the anterior insula and in the left caudate. iTBS decreased [18F]FDG uptake in the somatosensory cortex. RsfMRI connectivity between the stimulation site and the pgACC was lower after iTBS than after sham stimulation.
Conclusion: To our knowledge this is the first PET/MRI study on iTBS. Our interim analysis suggests that a single session of iTBS is sufficient to modulate the metabolic activity of the core depression network. These results are in line with previous studies that have used a conventional repetitive transcranial magnetic stimulation. If these results are confirmed in the final analysis of a larger sample, they can be used to optimize and design more personalized iTBS treatments in clinical populations.
Acknowledgements
This work is supported by the eRIMh award to Dr. Tuominen.
Reference
BlumbergerDM, et al.Effectiveness of theta burst versus high-frequency repetitive transcranial magnetic stimulation in patients with depression (THREE-D): a randomised non-inferiority trial. Lancet 391.10131 (2018): 1683–1692.
2021-44
MRI quantification of brain oxygenation and relationship with cerebrovascular reactivity in Moyamoya disease using simultaneous [15O]-water PET/MRI (#242)
Audrey P. Fan1, 2, David Y.-T. Chen1, 3, David D. Shin4, Moss Y. Zhao1, Jun-Hyung Park1, Bin Shen1, Mohammad M. Khalighi1, Dawn Holley1, Kim Halbert1, Gary K. Steinberg1 and Greg Zaharchuk1
1Department of Radiology, Stanford University, Stanford, CA, USA
2Department of Biomedical Engineering and Neurology, University of California, Davis, Davis, CA, USA
3Department of Radiology, Taipei Medical University, Taipei, Taiwan
4GE Healthcare, CA, USA
5Department of Neurosurgery, Stanford University, Stanford, CA, USA
Abstract
Introduction: We compared two MRI approaches based on tissue R2’ relaxation and vein magnetic susceptibility, respectively, to measure oxygen extraction fraction (OEF) abnormalities in Moyamoya disease,1 a steno-occlusive disorder of intracranial arteries. MRI-derived OEF was compared for different stenosis severities, and correlated to [15O]-water PET of cerebral blood flow (CBF) using hybrid PET/MRI.
Methods: Simultaneous 3T PET/MRI was acquired in 15 Moyamoya patients (ages 32–62 years, 9 female). PET perfusion imaging with [15O]-water (550–925 MBq) was performed at baseline and after vasodilation with acetazolamide to observe the perfusion change (ΔCBF).2 For baseline MRI of oxygenation, we calculated multi-parametric maps of R2’, which are proportional to tissue OEF.3,4 These scans included fast spin echo with 8 echoes (28–122ms) and multi-echo gradient echo with 10 echoes (11.1–42.4ms). We also acquired gradient echo scans (0.41x0.41x0.70 mm3) with flow compensation for quantitative susceptibility mapping (QSM) in veins.5 CBF, ΔCBF, and R2’ were assessed within 10 vascular territories per hemisphere; QSM values were also evaluated in cortical veins draining these territories.
Results: Figure 1(a) shows PET/MRI in a 32-y.o. male Moyamoya patient with right middle cerebral artery (MCA) occlusion and impaired PET cerebrovascular reactivity after vasodilation (green arrows). Higher vein density and susceptibility on QSM maps (indicating elevated OEF), and higher white matter R2’ were observed in the right hemisphere. In contrast, the 33-y.o. male patient in Figure 1(b) had healthy baseline PET perfusion and preserved reactivity. The corresponding QSM maps were normal, with symmetric vessel caliber and susceptibility, as were the R2’ oxygenation maps. Across patients, [15O]-water PET revealed 22.5% reduction in baseline CBF and 55.6% reduction in ΔCBF for MCA areas with severe stenosis compared to normal territories (Figure 2). Relative OEF assessed by vessel susceptibility was 20.4% higher for severe stenosis compared to normal areas (P = 0.007); and inversely correlated with PET baseline CBF and ΔCBF (P < 10−4). In contrast, R2’ showed smaller abnormality (9.6%) in stenosed areas, and weaker correlations to PET perfusion.
Conclusion: Advanced MRI revealed abnormal, elevated OEF in Moyamoya disease, which inversely correlated with CBF and reactivity measured by simultaneous PET. Susceptibility MRI detected more robust OEF increases in areas of pathology than R2’, with stronger correlation to perfusion deficits.
Acknowledgements
This study is funded by NIH grants 5K99NS102884-02 and R01EB025220-02.
References
NiWWChristenTRosenbergJ, et al.Imaging of cerebrovascular reserve and oxygenation in Moyamoya disease. J Cereb Blood Flow Metab2017;
37: 1213–1222.IshiiYThammTGuoJ, et al.Simultaneous phase‐contrast MRI and PET for noninvasive quantification of cerebral blood flow and reactivity in healthy subjects and patients with cerebrovascular disease. J Magn Reson Imag2020;
51: 183–194.StoneAJBlockleyNP.
A streamlined acquisition for mapping baseline brain oxygenation using quantitative BOLD. NeuroImage2017;
147: 79–88.KaczmarzSGöttlerJZimmerC, et al.Characterizing white matter fiber orientation effects on multi-parametric quantitative BOLD assessment of oxygen extraction fraction. J Cereb Blood Flow Metab2020;
40: 760–774.FanAPKhalilAAFiebachJB, et al.Elevated brain oxygen extraction fraction measured by MRI susceptibility relates to perfusion status in acute ischemic stroke. J Cereb Blood Flow Metab2020;
40: 539–551.
2021-45
In vivo assessment of astroglial activation in cognitively impaired subjects using 11C-BU99008 PET and its relationship with amyloid load (#247)
Paul Edison
Imperial College London, London, UK
Abstract
Introduction: Glial activation may play a major role in the cognitive function in Alzheimer’s disease. 11C-BU99008 is a novel PET tracer selective for imidazoline I2 binding sites (BS), which are found mainly in astroglial cells. The aim of this pilot study was to explore 11C-BU99008 uptake in cognitively impaired subjects stratified as amyloid-positive or negative using 18F-florbetaben PET.
Methods: 11 cognitively impaired patients and 9 healthy controls (HC) underwent 3T MRI, 18F-florbetaben PET and 11C-BU99008 PET. 11C-BU99008 PET was analysed quantitatively using an arterial plasma input function while 18F-florbetaben uptake (referenced to that in the cerebellum) was used to define amyloid-positivity
Results: The 8/11 cognitively impaired subjects who were amyloid-positive had higher 11C-BU99008 tracer uptake relative to the HC across the whole brain, but particularly in the frontal, occipital and medial temporal lobes. A positive voxel-wise correlation between 11C-BU99008 and 18F-florbetaben uptake was observed. Autoradiography using 3H-BU99008 conducted in post mortem AD brains also showed an increase binding in AD patients compared with controls and this binding was not displaced by PiB or Florbetaben
Conclusion: Our results demonstrate increased I2 receptor availability for 11C-BU99008 in cognitively impaired amyloid-positive subjects neuroanatomically localised to cortical regions of highest amyloid deposition. Based on prior literature, the increased uptake is believed to reflect astroglial activation predominantly. This novel proof of principle study thus provides novel in vivo evidence for pathological astroglial activation with neurodegeneration and demonstrates 11C-BU99008 PET as a potential new clinical research tool for its assessment.
Acknowledgements
The authors thank Invicro Centre for Imaging Sciences for the provision of 11C-BU99008, scanning and blood analysis equipment. The authors also thank Piramal Life Sciences/Life Molecular Imaging for providing the 18F-florbetaben and permission to acquire unlabelled Florbetaben. We thank Dementia Platform UK (DPUK) and GSK for the generous funding for this project. This research was co-funded by the NIHR Imperial Biomedical Research Centre, and was supported by the NIHR Imperial Clinical Research Facility. The views expressed are those of the authors and not necessarily those of NHS, the NIHR or the Department of Health.
2021-46
Relations of EEG and α2-adrenoceptor availability in patients with Parkinson’s disease (#252)
Albert Gjedde1, Adam F. Kemp1, Martin B. Kinnerup2 and Adjmal Nahimi3
1Department of Nuclear Medicine, University of Southern Denmark, Odense, Denmark
2Department of Nuclear Medicine and PET Center, Aarhus University, Aarhus, Denmark
3Restorative Parkinson Unit, Lund University, Lund, Sweden
Abstract
Introduction: Alterations of electroencephalography (EEG) may be correlated to cognitive impairment and to predict dementia in patients with Parkinson’s disease. There is reason to believe that the cognitive impairment is related to degeneration of locus coeruleus (LC) as the primary source of noradrenergic innervation (NE) in CNS.1 NE is essential for cognition and memory, and degeneration of LC unrelated to age has been detected in PD patients. Here, we claim that the alteration of EEG and the cognitive impairment both result from degeneration of LC.
Methods: We recruited 19 patients aged 53–80 years with a mean of 65 years (13 men) from the Department of Neurology, Aarhus University Hospital, Denmark, and 13 healthy control (HC) subjects aged 52–76 years with a mean of 66 years (6 men) by newspaper advertisement in Aarhus, Denmark. We used magnetic resonance imaging (MRI) and positron emission tomography (PET) with the α2 adrenoceptor ligand [11C]yohimbine to determine α2 adrenoceptor expression in thalamus and frontal cortex (FC) of 17 patients and the 13 HC subjects, and we calculated α2 adrenoceptor density from binding potentials, specific activity, and the affinity of binding of the tracer.2,3 We correlated the α2 adrenoceptor density with EEG theta power bands in FC and thalamus by linear regression (Figure 1).
Results: On the day of tomography. the patients had UPDRS scores of 41 (range 17–55) in the "off" state with Hoehn & Yahr stage 2.5 and mean L-DOPA equivalent doses of 1052 (range 360–2057). Patients had more power in the EEG theta and delta bands compared to control subjects. Among the patients, the EEG theta power correlated with the α2 adrenoceptor availability in both FC and thalamus . Age, gender and background rhythm frequency (BRF) correlated with α2 adrenoceptor density in FC, and BRF correlated with α2 adrenoceptor density in thalamus in HC subjects. BRF and theta power both correlated with α2 adrenoceptor density in FC and thalamus when HC subjects and patients were analyzed as a single group.
Conclusion: The EEG theta power was related to α2 adrenoceptor expression in frontal cortex and thalamus among patients with Parkinson’s disease. The evidence is consistent with the hypothesis of degeneration in LC as responsible for non-motor symptoms in PD, with EEG theta power predictive of some non-motor symptoms of PD.
Acknowledgements
Thanks to the staffs of Aarhus University Hospital’s Department of Nuclear Medicine and PET Center, and to Odense University Hospital’s Department of Nuclear Medicine.
References
LandauAMDyveSJakobsenS, et al.Acute vagal nerve stimulation lowers α2 adrenoceptor availability: possible mechanism of therapeutic action. Brain Stimul2015;
8: 702–707.NahimiAJakobsenSMunkOL, et al.Mapping α2 adrenoceptors of the human brain with with 11C-yohimbine. J Nucl Med2015;
56: 392–398.PhanJALandauAMJakobsenS, et al.Radioligand binding analysis of α2 adrenoceptors with [11C]yohimbine in brain in vivo: extended inhibition plot correction for plasma protein binding. Scientif Rep2017;
7: 15979.
2021-47
Electroconvulsive stimulation differentially affects [11C]MDL100907 binding to cortical and subcortical 5HT2A receptors in porcine brain (#253)
Albert Gjedde1, 2, Anne M. Landau3, 4, Michael Winterdahl5, Ove Noer4, Aage O. Alstrup4, Hélène Audrain4, Poul Videbech6, 7, Gregers Wegener3, Arne Møller8 and Doris J. Doudet9
1Department of Nuclear Medicine, University of Southern Denmark, Odense, Denmark
2Department of Neurology and Neurosurgery, McGill University, Montreal, QC, Canada
3Translational Neuropsychiatry Unit, Aarhus University, Aarhus, Denmark
4Department of Nuclear Medicine & PET Center, Aarhus University Hospital, Aarhus, Denmark
5Department of Nuclear Medicine and PET Center, Aarhus University, Aarhus, Denmark
6Center of Mental Health, Glostrup, Denmark
7University of Copenhagen, Copenhagen, Denmark
8Center of Functionally Integrative Neuroscience, Aarhus University, Aarhus, Denmark
9Department of Medicine & Neurology, University of British Columbia, Vancouver, BC, Canada
Abstract
Introduction: Electroconvulsive therapy is an effective therapy of depression. On the basis of past work in non-human primates,1 we hypothesized that the beneficial effects are mediated partly by decreased serotonin receptor availability in the cortex. To test the hypothesis, we used positron emission tomography with the serotonin 5HT2A receptor radioligand [11C]MDL100907 to determine serotonin receptor availability in response to electroconvulsive stimulation (ECS), a design derived from previously reported primate studies.
Methods: Seven female Göttingen minipigs aged 16–18 months were deeply anaesthetized with intravenous pentobarbital and imaged at baseline before the onset of ECS with a clinical device, and at 1–2 and 8–10 days after the end of a clinical course of ECS, consisting of 10 sessions over 3.5 weeks, and post-ECS values were compared to baseline. Animals were prone during the experiments as previously described.2 The ECS used 70 Hz frequency, 0.5 ms pulse width, and 0.9 A current with a maximum charge of 504 mC. Electrodes were placed bilaterally on each side of the head. One additional minipig was anaesthetized over 10 sessions without ECS, as a control. We analysed images in terms of binding potentials with the multilinear reference tissue method applied to the tracer binding in cortex and hippocampus, followed by whole-brain analysis by statistical non-parametric mapping.3
Results: We found significantly increased binding potential of [11C]MDL100907 in the cortex and hippocampus 1–2 days after ECS, consistent with increased serotonin receptor availability compared to baseline. By 8–10 days after the final ECS, the average tracer binding had returned towards baseline. However, we also found significantly decreased tracer binding in the subcortical regions of olfactory bulb, pons, thalamus and striatum (Figures 1 and 2).
Conclusion: With ECS, minipigs, like rodents, have higher availability at cortical and hippocampal 5HT2A receptors. Decreased tracer binding is consistent with reduced serotonin receptor availability as a differential effect of ECS on 5HT2A receptors in subcortical regions of minipig brain. The cortical changes are consistent with previous studies in rodents but are opposite changes observed in human and non-human brains. The differences may account for differences of serotonergic modulation of subcortical versus cortical brain.
Acknowledgements
We are grateful for technical support from staff at the Aarhus University Hospital PET Center and from the staff at the Aarhus University Farm for help with handling and stimulation of the animals.
References
LandauAMChakravartyMMClarkCM.Electronconvulsive therapy alters dopaminergic signaling in the striatum of non-human primates. Neuropsychopharmacology2011;
36: 511–518.LandauAMAlstrupAKONoerO, et al.Electroconvulsive stimulation differentially affects [C-11]MDL100907 binding to cortical and subcortical 5HT(2A) receptors in porcine brain. J Psychopharmacol2019;
33: 714–721.GjeddeAWongDFRosa-NetoP.Mapping neuroreceptors at work: On the definition and interpretation of binding potentials after 20 years of progress. Int Rev Neurobiol2005;
63: 1–20.
2021-48
Network integration derived non-invasive biomarkers for early prediction of Alzheimer’s disease (#254)
Paule-Joanne Toussaint and Alan C. Evans
Department of Neurology and Neurosurgery, Faculty of Medicine, Montreal Neurological Institute and Hospital, McGill University, Montreal, QC, Canada
Abstract
Introduction: Development of preventive therapies and the design of disease-modifying treatments for neurodegenerative diseases such as Alzheimer’s (AD) require early diagnosis. Accurate identification of predictive patterns in the evolution of AD is essential, and recent evidence suggests that the degenerative process develops along anatomically well-defined functionally connected networks. We have devised a technique to identify early markers of conversion to AD in groups of cognitively healthy controls and individuals with mild cognitive impairment (MCI), to serve in a diagnostic procedure based on multivariate approaches.
Methods: Positron emission tomographic (PET) data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI; www.adni-info.org) were used. Participants from the ADNI1 with baseline 18F-FDG PET scans were selected (N = 399) and grouped based on their diagnosis at 18 months post- inclusion.1 Spatial independent component analysis (ICA) was applied on the baseline PET maps of glucose metabolism to obtain common regions of greatest coherent activity between groups of controls, individuals with mild cognitive impairment (MCI), and patients with AD. Independent components were extracted using an InfoMax algorithm, followed by dimension reduction with principal component analysis and Bayesian model selection (NetBrainWork2).
Results: Patterns of correlated FDG-PET regional activity suggest that regions such as the cingulate and temporo-parietal areas (Figure 1) could be more susceptible to neurodegeneration linked to AD, as they show damage early in the progression of the pathology.
Conclusion: Networks of regional activity extracted from FDG-PET data show significant co-occurring changes in radioactivity patterns between controls, individuals with MCI, and patients with AD. These distinctive patterns can serve as biomarkers of the initial stages of AD in a predictive model based on anatomo-functional networks. The method provides novel information on the functional and anatomical features underlying the disturbed neuronal connectivity and altered vascular and cellular mechanisms, and assesses the extent to which these features are early predictors of neuronal circuitry dysfunction.
Acknowledgements
This work was supported by the Canadian Institutes of Health Research (CIHR) grant MOP-37754, and by the National Institutes of Health (NIH) operating grant 248216.
References
ToussaintPJPerlbargVBellecP, et al.Resting-state FDG-PET functional connectivity as an early biomarker of Alzheimer’s disease using conjoint univariate and independent component analyses. NeuroImage2012;
63: 936–946.PerlbargVBellecPAntonJL, et al.CORSICA: correction of structured noise in fMRI by automatic identification of ICA components.Magn Reson Imaging2007;
25: 35–46.
2021-49
Sucrose lowers µ-opioid and D2/3 dopamine receptor availability of porcine brain (#256)
Albert Gjedde1, 2, Michael Winterdahl3, Ove Noer3, Dariusz Orlowski4, Anna M. Schacht3, Steen Jakobsen3, Aage O. Alstrup3 and Anne M. Landau5
1Department of Nuclear Medicine, University of Southern Denmark, Odense, Denmark
2Department of Neurology and Neurosurgery, McGill University, Montreal, QC, Canada
3Department of Nuclear Medicine and PET Center, Aarhus University, Aarhus, Denmark
4Department of Neurosurgery, Aarhus University, Aarhus University Hospital, Aarhus, Denmark
5Translational Neuropsychiatry Unit, Aarhus University, Aarhus, Denmark
Abstract
Introduction: The intake of sucrose as a palatable substance is known to release dopamine and induce dependency in rodents.1 Excessive sucrose consumption elicits addiction-like craving that may therefore underpin the obesity epidemic. Opioids and dopamine mediate the rewarding effects of drugs of abuse, and of natural rewards from stimuli such as palatable food. We determined the effects of repeated intermittent access to sucrose on opioid and dopaminergic neurotransmission.
Methods: We investigated the effects of sucrose using positron emission tomography (PET) with [11C] carfentanil (µ-opioid receptor agonist) and [11C]raclopride (dopamine D2/3 receptor antagonist) in seven anesthetized female Göttingen minipigs. We gave the seven minipigs access to a sucrose solution for one hour on 12 consecutive days and completed positron emission tomography again 24 hours after the final sucrose access. In a smaller sample of five minipigs, we completed an additional [11C]carfentanil PET session after the first sucrose exposure. We calculated voxel-wise binding potentials (BPND) using the cerebellum as a region of non-displaceable (ND) binding, then analyzed differences with statistical non-parametric mapping, and carried out an analysis of separate regions of the brains.
Results: We obtained average parametric maps of [11C]carfentanil and [11C]raclopride binding potentials, as shown in Figure 1. After 12 days of sucrose access, the BPND of both tracers had declined significantly in striatum, nucleus accumbens, thalamus, amygdala, cingulate cortex, and prefrontal cortex, a result consistent with down-regulation of receptor densities, as increased saturation was judged to be unlikely (Figure 2). After a single exposure to sucrose, we found decreased binding of [11C]carfentanil in nucleus accumbens and cingulate cortex, consistent with opioid release.
Conclusion: The results clearly demonstrate that sucrose affects reward mechanisms in a manner similar to that of drugs of abuse. The lower receptor availability of opioid and dopamine receptors may explain the addictive potential associated with intake of sucrose. Thus, excessive consumption of palatable food may both cause, and become the result of, addiction with direct consequences for health by obesity.
Acknowledgements
An Aarhus University "AU Ideas Project Development" grant to Anne M. Landau funded the study. We are grateful for the technical support from the staff at the Aarhus University Hospital PET Center and the Aarhus University Farm.
References
AvenaNMRadaPHoebelBG.Evidence for sugar addiction: Behavioral and neurochemical effects of intermittent excessive sugar intake.Neurosci Biobehav Rev2008;
32: 20–39.WinterdahlMNoerOOrlowskiD, et al.Sucrose intake lowers µ-opioid and dopamine D2/3 receptor availability in procine brain.Scientif Rep2019;
9: 16918.
2021-50
Nicotine attenuates age-related memory and learning impairment in D-galactose-induced senescence in mice (#258)
Albert Gjedde1, 2, Alireza Majdi3, Saeed Sadigh-Eteghad3, Mahnaz Talebi3, Fereshteh Farajdokht3, Marjan Erfani3 and Javad Mahmoudi3
1Department of Nuclear Medicine, University of Southern Denmark, Odense, Denmark
2Department of Neurology and Neurosurgery, McGill University, Montréal, QC, Canada
3Neurosciences Research Center, Tabriz University of Medical Sciences, Tabriz, Iran
Abstract
Introduction: Brain aging is a multifaceted process that results from changes of specific molecular pathways that may lead to cognitive dysfunction. Nicotine has been found to possess procognitive properties in animals and humans that may attenuate the changes.1–3
Methods: Here, we tested the hypothesis that nicotine ameliorates learning and memory impairment in aging brains of mice, induced by d-galactose (DGal). We administered DGal at 500 mg/kg doses by subcutaneous (s.c.) injection for 6 weeks to induce the impairment. As treatment, we administered nicotine at doses of 0.1, 0.5 or 1 mg/kg by the s.c route, or at 0.1 mg/kg by the intranasal (i.n.) route into the aged animals. We evaluated animal anxiety and withdrawal signs by counting the number of somatic signs, and examining thermal hyperalgesia, elevated plus maze and open field tests. Also, we assessed spatial and recognition memories by means of novel object recognition and Barnes maze tasks. We measured reactive oxygen species, mitochondrial membrane potential, caspase-3, Bax, Bcl-2, cytochrome C, brain-derived neurotrophic factor, and nerve growth factor levels in the brain tissue of the aged mice.
Results: The results revealed that treatment of aged mice with nicotine at 0.5 mg/kg s.c,. and 0.1 mg/kg i.n., doses remarkably ameliorated spatial and episodic memory decline, compared with the decline of vehicle-injected aged mice. Nicotine at these doses also decreased mito-oxidative damage as well as apoptosis, and it increased neurotrophic factors in the aged animals, compared with the levels of vehicle-injected mice. We detected withdrawal signs only at the 1 mg/kg s.c. dose of nicotine, compared to the vehicle-injected mice. Figure 1 shows the means ± SEM, (n = 12), including significant differences yielded by two- (A) or one-way ANOVA followed by Tukey’s post hoc test; labeled as *P < 0.05 or **P < 0.01, compared to the control group, #P < 0.05 or ##P < 0.01, compared to the DG+NS group, where NS is normal saline; DG is D-galactose; Nic is nicotine; s.c. is subcutaneous; i.n. is intranasal; MWM is Morris water maze; and NOR is novel object recognition
Conclusion: Nicotine at the examined doses and routes had the potential to ameliorate age-induced cognitive decline and affect related downstream pathways, i.e., mito-oxidation, apoptosis, and neurotrophic factors, without causing apparent development of dependence.
Acknowledgements
This research was in part supported by a grant from Neurosciences Research Center (NSRC), Tabriz University of Medical Sciences (TUOMS; grant number: 5/88/1523) to SS-E and grants from Department of Nuclear Medicine, Odense University Hospital, and from Alzheimer Foundation of Denmark to AG.
References
MajdiAKamariFVafaeeMS, et al.Revisiting nicotine’s role in the ageing brain and cognitive impairment. Rev Neurosci2017;
28: 767–781.ArendashGWSanbergPRSengstockGJ.Nicotine enhances the learning and memory of aged rats. Pharmacol Biochem Behav1995;
52: 517–523.MajdiAKamariFSadigh-EteghadS, et al.Molecular insights into memory-enhancing metabolites of nicotine in brain: a systematic review. Front Neurosci2019;
12: 1002.MajdiASadigh-EteghadSTalebiM, et al.Nicotine modulates cognitive function in D-galactose-induced senescence in mice. Front Aging Neurosci2019;
10: 194.
2021-51
Anesthetics differentially influence [11C]MDL100907 binding to 5HT2A receptors in porcine brain (#259)
Albert Gjedde1, 2, Anne M. Landau3, 4, Ove Noer3, Aage O. Alstrup3, Hélène Audrain3, Gregers Wegener4, Doris J. Doudet5 and Michael Winterdahl3
1Department of Nuclear Medicine, University of Southern Denmark, Odense, Denmark
2Department of Neurology & Neurosurgery, McGill University, Montreal, QC, Canada
3Department of Nuclear Medicine and PET Center, Aarhus University, Aarhus, Denmark
4Translational Neuropsychiatry Unit, Aarhus University, Aarhus, Denmark
5Division of Neurology, Department of Medicine, University of British Columbia, Vancouver, BC, Canada
Abstract
Introduction: We previously considered data collected in Göttingen minipigs to compare the effects of two commonly used anesthetics routinely used in large animal positron emission tomography (PET), the inhalant isoflurane and the injectable ultrashort acting nonbarbiturate anaesthetic propofol, on the binding of two radioligands of the monoaminergic receptor systems. We evaluate the effects of the same two anesthetics on the binding of [11C]MDL100907 to serotonin 5HT2A receptors.
Methods: We imaged 14 Isoflurane- and propofol-anesthetized Göttingen minipigs with [11C]MDL100907, an antagonist of 5HT2A receptors. PET imaging was performed on the high-resolution research tomograph (HRRT, CTI/Siemens). After a brief transmission scan, the minipigs were imaged with [11C]MDL100907. Parametric maps of the tracer’s binding potentials relative to non-displaceable reference activity (BPND) used a multilinear reference tissue kinetic approach and a white matter mask as reference region. The same kinetic modelling approach was applied to time-activity curves of specific regions of interest including the striatum, cortical areas, hippocampus and amygdala.
Results: In the striatal regions of the averaged parametric maps of brains of minipigs anesthetized with isoflurane, values of BPND of [11C]MDL100907 exceeded those of minipigs anaesthetized with propofol (Figure 1). We performed region-of-interest analysis of primary data, and two-way ANOVA revealed a significant effect of type of anaesthetic (P < 0.001) and region (P < 0.0001). Post hoc Bonferroni testing revealed significantly higher binding potentials in striatum of minipigs anesthetized with isoflurane rather than with propofol (P < 0.01) (Figure 2). In order to evaluate the contribution of cerebral blood flow (CBF) changes to the higher binding potential observed in the striatum of the isoflurane anesthetized animals, we surrogate flow values and found significantly higher values with isoflurane than with propofol.
Conclusion: We found significant differences in striatal binding of [11C]MDL100907 to 5HT2A receptors in minipig brain under propofol compared to isoflurane anesthesia, and we attribute these differences, at least in part, to increases in blood flow inherent to the isoflurane condition. The effects of anesthesia on the serotonin system as well as interactions between radioligands and anesthesia must be considered when designing and interpreting experiments. The choice of anesthetic may mask potential differences or effects due to experimental manipulations (pharmacologic or behavioural) in a particular study or may exaggerate differences.
Acknowledgements
The Lundbeck Foundation and Danish Medical Research Council funded these experiments. We are grateful for the technical support from staff at the Aarhus University Hospital PET Center.
References
WaabenJHusumBHansenAJ, et al.Hypocapnia prevents the decrease in regional cerebral metabolism during isoflurane-induced hypotension. J Neurosurg Anesthesiol1989;
1: 29–34.SchlünzenLVafaeeMSColdGE, et al.Effects of subanaesthetic and anaesthetic doses of sevoflurane on regional cerebral blood flow in healthy volunteers. A positron emission tomographic study. Acta Anaesthesiol Scand2004;
48:1268–1276.GjeddeAJohannsenPColdGE, et al.Cerebral metabolic response to low blood flow: possible role of cytochrome oxidase inhibition.J Cereb Blood Flow Metab2005;
25: 1183–1196.LandauAMNoerOAstrupAKO, et al.Type of anaesthetic influences [11C]MDL100907 Binding to 5HT2A receptors in porcine brain. Mol Imag Biol2020;
22: 797–804.
2021-52
Transcranial photoacoustic imaging of NMDA-evoked focal circuit dynamics in rat hippocampus (#260)
Dean F. Wong1, Albert Gjedde2, 3, Shilpa D. Kadam4, Joshua S. Elmore5, Brennan J. Sullivan4, Heather Valentine1, Adarsha P. Malla6, Maged M. Harraz6, Arman Rahmim1, Jin U. Kang7, 8, Leslie M. Loew9, Michael Baumann10, Anthony A. Grace11, Emad M. Boctor7 and Jeeun Kang1, 7
1Department of Radiology and Radiological Science, Johns Hopkins University, Baltimore, Maryland, USA
2Department of Nuclear Medicine, University of Southern Denmark, Odense, Denmark
3Department of Neurology and Neurosurgery, McGill University, Montreal, QC, Canada
4Johns Hopkins Medical Institutions, Hugo W. Moser Research Institute, Baltimore, Maryland, USA
5Dept of Radiology & Radiological Science, Johns Hopkins University, Baltimore, Maryland, USA
6Solomon H. Snyder Dept of Neuroscience, Johns Hopkins University, Baltimore, Maryland, USA
7Whitting School of Engineering, Johns Hopkins University, Baltimore, Maryland, USA
8Dept of Electrical and Computer Engineering, Johns Hopkins University, Baltimore, Maryland, USA
9Dept of Cell Biology, University of Connecticut Health, Farmington, Connecticut, USA
10Intramural Research Program, National Institute of Drug Abuse, Baltimore, Maryland, USA
11Depts of Neuroscience, Psychatry and Psychology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
Abstract
Introduction: Transcranial functional photoacoustic (fPA) voltage-sensitive dye (VSD) imaging promises to overcome current temporal and spatial limitations of current neuroimaging modalities. The technique previously distinguished global seizure activity from control neural activity in groups of rats. To validate the focal specificity of transcranial fPA with VSD (IR780 perchlorate) imaging in vivo, we now present proofs-of-concept that the results differentiate N-methyl-D-aspartate (NMDA) evoked neural activity in rat hippocampus. Concurrent quantitative EEG (qEEG) and microdialysis recorded real-time circuit dynamics and glutamate concentration change, respectively. We hypothesized that location-specific fPA VSD contrast would identify the neural dynamics in hippocampus with the correlation to NMDA evoked focal glutamate release and time-specific EEG signals.
Methods: To test the hypothesis, we infused 0.3, 1, and 3 mM NMDA at 2 µl/min over 60 min via an implanted microdialysis probe. The dialysate samples collected every 20 minutes during the infusion were analyzed for focal changes in extracellular glutamate release and quantified by high-performance liquid chromatography (HPLC).
Results: Transcranial fPA VSD imaging provided NMDA evoked VSD responses at the contralateral side of the microdialysis probe, presenting positive correlation with glutamate increase during 3 mM NMDA infusion. On the other hand, insignificant VSD response and glutamate release were obtained during 0.3 mM NMDA infusion. Quantitative EEG (qEEG) successfully confirmed induction of focal seizure activity or low response during 3 mM and 0.3 mM NMDA infusion, respectively. This graded response suggests all-or-none gating system in the dentate gyrus (DG) in hippocampus.
Conclusion: We conclude that transcranial fPA VSD imaging distinguished graded DG gatekeeping functions, based on the VSD redistribution mechanism sensitive to electrophysiologic membrane potential. The results suggest the potential future use of this emerging technology in clinics and science as an innovative and significant functional neuroimaging modality.
Acknowledgements
This work was supported by the NIH Brain Initiative under Grant No. R24 MH106083-03 (DFW, AG, AR, AAG, EMB, HV, JE) and the NIH National Institute of Biomedical Imaging and Bioengineering under Grant No. R01EB01963. NIH National Institute of Child Health and Human Development (NICHD) for R01HD090884 (SDK). NIH National Institute of Heart, Lung and Blood (NHLBI) under grant number R01HL139543 (Jeeun K, APM, EMB). National Cancer Institute (NCI) under grant number R21CA202199 and its equipment supplement. Funding of the PA equipment via resources of Jin K and EB NSF Career award #1653322. Jeeun K was partially supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education #2018R1A6A3A03011551.
References
KangJKadamSDElmoreJ, et al.Transcranial photoacoustic imaging of NMDA-evoked focal circuit dynamics in rat hippocampus. J Neural Eng 2020; 17(2): 025001.KangJZhangHKKadamS, et al.Transcranial recording of electrophysiological neural activity in the rodent brain in vivo using functional photoacoustic imaging of near-infrared voltage-sensitive dye. Front Neurosci2019;
13: 1–14.PakRWKangJValentineH, et al.Voltage-sensitive dye delivery through the blood brain barrier using adenosine agonist Regadenoson. Biomed Opt Express2018;
9: 3915–3922.ZhangHKYanPKangJ, et al.Listening to membrane potential: photoacoustic voltage-sensitive dye recording. J Biomed Opt2017;
22: 045006
2021-53
Intrinsic connectivity of the human brain provides scaffold for tau aggregation in Alzheimer’s disease (#262)
Joseph Therriault, Tharick A. Pascoal, Melissa Savard and Pedro Rosa-Neto
McGill, H4C 0B5, H4C 0B5 Montreal, QC, Canada
Abstract
Introduction: Preclinical models suggest that tau pathology spreads intracellularly, thought to explain the topographical distribution of tau observed in Alzheimer’s disease (AD). However, these findings have yet to be extended to humans.
Methods: We assessed 131 cognitively unimpaired elderly and 63 AD individuals who underwent amyloid-PET with [18F]AZD4694, tau-PET with [18F]MK6240, structural MRI, fMRI and diffusion-weighted MRI. Of the subjects with AD, 11 had behavioural/dysexecutive AD, 18 had PCA and 11 had Logopenic variant PPA, while 23 subjects had an amnestic presentation (all were Ab+/Tau+). A voxelwise multivariate regression model was employed to determine the peak difference in [18F]MK6240 SUVR between each AD variant and CU elderly, with each clinical diagnosis entered as a different categorical variable and correcting for age, gender and MMSE score. Within each AD variant, the peak voxels derived from the regression model were used to compute the correlation between [18F]MK6240 SUVR in the seed voxel and [18F]MK6240 SUVR in every voxel, thus generating an [18F]MK6240 covariance network for each AD group. The same seeds were also employed in functional connectivity and diffusion tractography analyses in the CU elderly group. To determine whether the topographical distribution of tau pathology is related to connectivity properties of the human brain, we applied regression models to assess the relationship between functional/structural connectivity values from the CU population and [18F]MK6240 in each variant of AD.
Results: Tau organization differed between AD groups, reflecting clinical phenotypes and organizing within distinct brain networks. Furthermore, structural (Figure 1) and functional (Figure 2) connectivity patterns of the human brain predicted in vivo [18F]MK6240 SUVR across the cerebral cortex in each variant of AD.
Conclusion: These results support a framework in which the intrinsic connectivity of the human brain provides a scaffold for tau pathology to spread to anatomically distant regions.
2021-54
Correspondence between gene expression and neurotransmitter receptor and transporter density in the human cortex (#275)
Justine Y. Hansen1, Ross D. Markello1, Nicola Palomero-Gallagher2, 3, Alain Dagher1 and Bratislav Misic1
1Neurology and Neurosurgery, McGill University, Montreal, QC, Canada
2Institute of Neuroscience and Medicine (INM-1), Research Centre Julich, Julich, Germany
3C. and O. Vogt Institute for Brain Research, Heinrich Heine University Dusseldorf, Dusseldorf, Germany
Abstract
Introduction: Neurotransmitter receptors and transporters modulate neuronal excitability and firing rate, and are therefore critical for intercellular communication. As such, neurotransmitter receptors and transporters play a key role in shaping brain connectivity and dynamics.1,2 Due to the lack of comprehensive open-source neurotransmitter receptor/transporter density datasets, microarray gene expression is often used as a proxy for these protein densities. Despite the frequent replacement of receptor/transporter densities with gene expression, the assumed correlation between gene expression and receptor/transporter density has yet to be comprehensively and formally tested across multiple neurotransmitter systems.
Methods: Here we investigate whether microarray gene expression can be used to estimate neurotransmitter receptor/transporter densities in the cortex. We use the Allen Human Brain Atlas for the expression levels of specific genes which code for specific neurotransmitter receptors or transporters.3,4 Additionally, we use both PET- and autoradigoraphy-derived estimates of neurotransmitter receptor densities, which we collected for a total of 24 neurotransmitter receptors and transporters across 8 different neurotransmitter systems.5 We study the association between gene expression and neurotransmitter receptor densities at the level of the cortex, and across five bins along the unimodal-transmodal functional hierarchy. To ensure results are not biased by methodological choices, we repeat our analyses in a separate parcellation resolution and use a conservative spatial autocorrelation-preserving null model.
Results: We find poor spatial correspondences between gene expression and density for all neurotransmitter receptors except for three single-protein metabotropic receptors (5-HT1A, D2, and MOR; Figure 1). We find that genes with the greatest stability across brain samples are most closely associated with receptor density (Figure 2). Finally, we find variability in system-wide associations, with unimodal brain regions showing stronger gene-receptor associations and transmodal brain regions showing weaker gene-receptor associations.
Conclusion: Future efforts to map neurotransmitter receptor and transporter profiles to brain structure and function should verify the expression-density association when using microarray gene expression in place of receptor and transporter density. Altogether, we recommend using direct measures of receptor and transporter density when relating neurotransmitter systems to brain structure and function.
References
ShineJM.Neuromodulatory influences on integration and segregation in the brain. Trends Cognitive Sci2019;
23: 572–583.ZillesKBacha-TramsMPalomero-GallagherN, et al.Common molecular basis of the sentence comprehension network revealed by neurotransmitter receptor fingerprints. Cortex2015;
63: 79–89.HawrylyczMJLeinESGuillozet-BongaartsAL, et al.An anatomically comprehensive atlas of the adult human brain transcriptome. Nature2012;
489: 391–399.MarkelloRArnatkeviciuteAPolineJB, et al.Standardizing workflows in imaging transcriptomics with the abagen toolbox. bioRxiv2021;
▪: ▪.ZillesKPalomero-GallagherN. Multiple transmitter receptors in regions and layers of the human cerebral cortex. Frontiers Neuroanat2017;
11: 78.
2021-55
Mapping neurotransmitter receptor and transporter distributions to the connectivity and dynamics of the human neocortex (#276)
Justine Y. Hansen1, Golia Shafiei1, Ross D. Markello1, Sylvia Cox10, Kelly Smart11, Etienne Aumont2, Stijn Servaes9, Stephanie Scala10, Gabriel Wainstein3, Gleb Bezgin9, Thomas Funck7, Marc-Andre Bedard1, 2, R. Nathan Spreng1, Jean-Paul Soucy1, Synthia Guimond6, Jarmo Hietala5, Marco Leyton1, 10, Pedro Rosa-Neto1, 9, Richard E. Carson11, Nicola Palomero-Gallagher7, 8, Lauri Tuominen6, James M. Shine3, Alain Dagher1 and Bratislav Misic1
1Neurology and Neurosurgery, McGill University, Montreal, QC, Canada
2UQAM, Cognitive Pharmacology Research Unit, Montreal, QC, Canada
3Brain and Mind Centre, University of Sydney, Sydney, Australia
4McGill University, Douglas Hospital, Montreal, QC, Canada
5University of Turku, Psychiatry, Turku, Finland
6Institute of Mental Health Research, University of Ottawa, Ottawa, ON, Canada
7Institute of Neuroscience and Medicine (INM-1), University of Julich, Julich, Germany
8C. and O. Vogt Institute for Brain Research, Heinrich Heine University Dusseldorf, Dusseldorf, Germany
9Research Centre for Studies in Aging, McGill University, Montreal, QC, Canada
Introduction: The patterns of neural activity demonstrated by billions of neurons ultimately manifest as whole-brain connectivity and dynamics. Modulatory influences on neural activity occur due to the receptors that span the surface of each cell. The heterogeneous distribution of receptor densities across the cortex suggests a diversity of modulatory influence, and therefore also of signal integration, neural dynamics, and connectivity.1,2
Methods: We used two recent state-of-the-art datasets (PET and autoradiography) of neurotransmitter receptor densities across the neocortex, which include a total of 27 receptors, receptor binding-sites, and transporters.3 First, we mapped receptor distributions to structural and functional connectivity, thereby profiling how receptors may influence whole-brain communication. Second, we used multiple linear regression models to predict magnetoencephalography power.4 Finally, we conducted a partial least squares analysis between receptor densities and Neurosynth-derived measures of functional association of 123 cognitive terms to uncover a latent axis of covariation across the cortex.5
Results: We find that the similarity of receptor profiles between two regions is greater for more proximal regions. Moreover, regions show more similar receptor distributions if they are physically connected with one another, and receptor distributions are highly similar in very densely connected brain regions. In terms of functional connectivity, we find that receptor distributions are more similar within regions of the same functional network rather than between regions of different functional networks. We also find that receptor similarity is significantly correlated to functional connectivity. We further find that receptor distributions augment structure-function coupling, particularly in the ventral temporal lobe. Interestingly, we find that receptor densities strongly predict all six MEG power bands. Finally, we extracted a significant axis of covariation between receptor densities and Neurosynth cognitive association, which indicates receptor densities and mood-related processes covary in the limbic and insular cortices.
Conclusion: Here, we show that the heterogeneous distribution of receptors across the human neocortex reflects structural and functional network architectures, shapes neural dynamics, and spatially covaries with neurocognitive functional activation. Altogether, we uncover the neurochemical infrastructure that shapes the brain’s connectivity and dynamics by comprehensively mapping receptor distributions to the structure and function of the human brain.
References
ShineJM.Neuromodulatory influences on integration and segregation in the brain. Trends Cognitive Sci2019;
23: 572–583.ZillesKBacha-TramsMPalomero-GallagherN, et al.Common molecular basis of the sentence comprehension network revealed by neurotransmitter receptor fingerprints. Cortex2015;
63: 79–89.ZillesKPalomero-GallagherNMultiple transmitter receptors in regions and layers of the human cerebral cortex. Frontiers Neuroanatomy2017;
11, 78.Van EssenDCSmithSMBarchDM, et al.; HCP Consortium.
The WU-Minn human connectome project: an overview. Neuroimage2013;
80: 62–79.YarkoniTPoldrackRANicholsTE, et al.Large-scale automated synthesis of human functional neuroimaging data. Nature Methods2011;
8: 665–670.
2021-56
A possible link between Fragile X mental retardation protein and metabotropic glutatmate receptor subtype 5 in men with fragile X syndrome (#277)
James R. Brasic1, Jack A. Goodman2, David S. Russell3, 4, Danna Jennings3, 5, Olivier Barret3, 6, Ayon Nandi1, Anil K. Mathur1, Thomas W. Sedlak1, 7, Keith Slifer7, 8, Samuel D. Martin1, 9, Elizabeth M. Berry-Kravis10, John P. Seibyl3, 4, Dean F. Wong1, 11 and Dejan B. Budimirovic7, 12
1The Russell H. Morgan Department of Radiology and Radiological Science/Division of Nuclear Medicine and Molecular Imaging/Section of High Resolution Brain Positron Emission Tomography Imaging, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
2Frank H. Netter MD School of Medicine, Quinnipiac University, North Haven Connecticut, USA
3Institute for Neurodegenerative Disorders, New Haven Connecticut, USA
4Invicro, New Haven Connecticut, USA
5Denali Therapeutics, South San Francisco California, USA
6Commissariat à l’Énergie Atomique et aux Énergies Alternatives (CEA)/Université Paris-Saclay, Laboratoire des Maladies Neurodégénératives, Molecular Imaging Research Center (MIRCen)/Institut de Biologie François Jacob/Centre National de la Recherche Scientifique (CNRS), Fontenay-aux-Roses, France
7Johns Hopkins University School of Medicine, Department of Psychiatry and Behavioral Sciences, Baltimore, Maryland, USA
8Johns Hopkins Medical Institutions, Department of Behavioral Psychology/Kennedy Krieger Institute, Baltimore, Maryland, USA
9Department of Neuroscience/Zanvyl Krieger School of Arts and Sciences, Johns Hopkins University, Baltimore, Maryland, USA
10Departments of Pediatrics, Neurological Sciences, and Biochemistry, Rush University Medical Center, Chicago Illinois, USA
11Department of Radiology/Laboratory of Central Nervous System (CNS) Neuropsychopharmacology and Multimodal Imaging (CNAMI)/Mallinckrodt Institute of Radiology, Washington University School of Medicine, Saint Louis Missouri, USA
12Department of Psychiatry/Kennedy Krieger Institute, Johns Hopkins Medical Institutions, Baltimore, Maryland, USA
Abstract
Introduction: Although animal models of fragile X syndrome (FXS) suggest a link between Fragile X Mental Retardation Protein (FRMP) and metabotropic glutamate receptor subtype 5 (mGluR5) in the pathogenesis and pathophysiology of fragile X syndrome (FXS), numerous animal and human studies of FMRP and mGluR5 provide inconsistent or conflicting findings about those relationships. We aimed to (A) compare and contrast FMRP levels and mGluR5 expression by positron emission tomography (PET) with 3-[18F]fluoro-5-(2-pyridinylethynyl)benzonitrile ([18F]FPEB) in FXS and typical development (TD)1–4 (Figure 1) and (B) show the preliminary value of measurement of FRMP and mGluR5 expression for the application of precision medicine for the diagnosis and treatment of individuals with FXS and related conditions.
Methods: Two teams of investigators at the Institute for Neurodegenerative Disorders (IND), New Haven, Connecticut, and the Johns Hopkins University (JHU), Baltimore, Maryland, independently administered 3-[18F]fluoro-5-(2-pyridinylethynyl)benzonitrile ([18F]FPEB)4 to men with FXS [N = 7, age 27.67 + 4.28, range (22.3, 33.6)] and participants of both sexes with TD [N = 18, age 33,15 + 14.06, range (19, 62.4)].1 Participants with FXS had a diagnosis of FXS based on FMR1 DNA gene testing by polymerase chain reaction (PCR)/Southern Blot on peripheral venous blood samples supplemented by clinical neurobehavioral profiling.1–3
Results: The FMRP concentration (nanograms per microgram total protein) and the mGluR5 uptake ([18F]FPEB BPND) in the cortical regions of all participants is plotted in Figure 2 (Graphpad Prism 9 for Mac, GraphPad Software, San Diego, California USA, www.graphpad.com). The FMRP value of 0.78 nanograms per microgram total protein, the average value of ten healthy people with typical development (TD) with normal CGG repeat sizes [range (20,37)] utilizing the same analyses as other participants, was used as the value for all participants with TD since none had undergone analysis for FMRP. There appears to be a positive correlation between mGluR5 uptake in the cortical regions and FMRP concentrations in the cohorts of men with FXS and the cohorts of participants with TD (Figure 2).
Conclusion: This protocol to measure FMRP and mGluR51–5 may facilitate future clinical trials of individuals with FXS and related conditions.
Acknowledgements
This research was made possible by a Radiology BRidge/Development Funding Initiative to STimulate and Advance Research (RAD BriteStar Bridge) Award from the Johns Hopkins University School of Medicine, Baltimore, Maryland to J.R.B. with the assistance of D.F.W.; and the Intellectual & Developmental Disabilities Research Center (U54 HD079123), Kennedy Krieger Institute, and Johns Hopkins Medical Institutions, Baltimore, Maryland, to J.R.B.
The authors thank the patients and families for their participation and dedication to these studies; they are the inspiration for our efforts at improving treatments. The authors thank the FORWARD Database and Registry of the National Fragile X Foundation (NFXF) funded by the Centers for Disease Control and Prevention (CDC), Atlanta, Georgia, for referral of participants.
The authors thank the teams of the Institute for Neurodegenerative Disorders and the Positron Emission Tomography (PET) Radiotracer Service Center of the Johns Hopkins University School of Medicine for conducting the scans. The authors thank Hiroto Kuwabara for PET analysis.
Earlier versions of this article were presented at the Undergraduate Research Symposium, Johns Hopkins University, Baltimore, Maryland, 28 October 2020, and the Society for Neuroscience Global Connectome, Virtual, 12 January 2021 (Martin et al.3).
References
BrasicJRMartinSDGoodmanJA, et al. Fragile X Mental Retardation Protein and cerebral expression of metabotropic glutamate receptor subtype 5 in men with fragile X syndrome, Zenodo, v1, Geneva, Switzerland: CERN European Organization for Nuclear Research, 2021https://doi.org/10.5281/zenodo.5154217.BrasicJRNandiARussellDS, et al.Cerebral expression of metabotropic glutamate receptor subtype 5 in idiopathic autism spectrum disorder and fragile X syndrome: a pilot study. Int J Mol Sci2021;
22: 2863.MartinSDBerry-KravisERusselD, et al. Correlation between the Fragile X Mental Retardation Protein (FMRP) and the cerebral expression of the metabotropic glutamate receptor subtype 5 (mGluR5) in fragile X syndrome’, Program No. 027–24. 2021 Neuroscience Meeting Planner. Chicago, IL: Society for Neuroscience, 2021. Online. January 12, 2021.WongDFWaterhouseRKuwabaraH, et al.18F-FPEB, a PET radiopharmaceutical for quantifying metabotropic glutamate 5 receptors: a first-in-human study of radiochemical safety, biokinetics, and radiation dosimetry. J Nucl Med2013;
54: 388–396.ModyMPetibonYHanP, et al.In vivo imaging of mGluR5 receptor expression in humans with fragile X syndrome towards development of potential biomarker. Sci Rep2021;
11: 15897.
2021-57
A translational investigation of morphine-induced neuroimmune signaling: Implications for opioid use disorder (#278)
Eric A. Woodcock1, 2, Gustavo A. Angarita2, 3, David Matuskey2, 3, Jim Ropchan3, Nabeel Nabulsi3, Yiyun H. Huang3, Ansel T. Hillmer2, 3, Richard E. Carson3 and Kelly P. Cosgrove2, 3
1Psychiatry and Behavioral Neurosciences, Wayne State University, Detroit, Michigan, USA
2Psychiatry, Yale University, New Haven, Connecticut, USA
3Yale PET Center, Yale University, New Haven, Connecticut, USA
Abstract
Introduction: Preclinical studies indicate that opioids evoke pro-inflammatory responses in the periphery and brain. These pro-inflammatory signals influence appetitive and dysphoric addiction processes and thus, may influence development of opioid use disorder (OUD) and/or perpetuate continued use. Here, we investigated the neuroimmune effects of morphine using Positron Emission Tomography (PET) imaging with [11C]PBR28. [11C]PBR28 binds to the 18kDa translocator protein (TSPO), a marker that is sensitive to pro-inflammatory neuroimmune stimulation.
Methods: Study #1: Eight healthy individuals with prior medical opioid exposure (3F; 34.3yrs [range = 26-49yrs]; BMI = 24.8 [range = 20-30]) completed two 120-minute [11C]PBR28 scans: before and 2-hours after intramuscular morphine (0.04mg/kg or 0.07mg/kg [‘low’ vs. ‘high’ dose]). Total volume of distribution (VT; TSPO availability) was estimated in 12 brain regions of interest (ROIs) using multilinear analysis-1 (t* = 30) and the metabolite-corrected arterial input function. Linear mixed models were used to evaluate morphine’s effect on TSPO with rs6971 Genotype, Morphine Dose, and Time as fixed factors and regional [11C]PBR28 VT as the within-subject repeated factor. Subjective effects were measured via visual analogue scales. Study #2: Two adult male rhesus macaques completed identical neuroimaging procedures as Study #1 before/after intramuscular morphine (1mg/kg). One macaque was pre-treated with intravenous (+)-naloxone (1mg/kg), a Toll-like Receptor-4 antagonist (TLR4) with negligible binding affinity for opioid receptors, to investigate the putative neurobiological mechanism. Studies were approved by the Yale University IRB and IACUC.
Results: Study #1: Morphine increased TSPO availability by 25–32% across ROIs, F(1,203) = 282.2, p < .001, (Figure 1) and evoked a subjective ‘high’, F(4,24) = 3.06, p = .036 (Figure 2(a)). Controlling for morphine dose, the increase in TSPO availability in the caudate was positively correlated with subjective ‘high’ at post-scan (R2 = 0.57, p = .03; Figure 2(b)). Study #2: Without pretreatment, morphine increased regional TSPO availability by 24–54%. Pretreatment with (+)-naloxone attenuated morphine’s effect on TSPO by 43%, on average, across ROIs.
Conclusion: Our findings indicate that morphine evokes a neuroimmune response in people and that morphine-induced ‘high’ is linearly related to neuroimmune signaling in the caudate. Given its well-established role in mesolimbic reward circuitry, our data suggest that neuroimmune signaling in the caudate may modulate opioid-induced euphoria. Finally, our non-human primate findings highlight TLR4 as a target for medication development.
Acknowledgements
The authors acknowledge the staff at the Yale University PET center, Jon Mikael Anderson, Aleksandra Rusowicz, Brittany LeVasseur, Dr. Stephen Baldassarri, Olivia Wilson, Nicole DellaGioia, Ryan Cool, Sarah DeBonee, Elizabeth Yanac, Daniel Holden, Krista Fowles, and Dr. Kenner Rice.
Funding
Funding was generously provided by the National Institute on Drug Abuse (NIDA K99/R00 DA048125; EAW), the National Institute of Mental Health (R01 MH110674; KPC), and the Veterans Affairs National Center for PTSD (KPC). (+)-Naloxone was provided by Dr. Kenner Rice in the Drug Design and Synthesis Section at NIDA. The work of the Drug Design and Synthesis Section, MTMDB, NIDA, and NIAAA was supported by the NIH Intramural Research Programs of the National Institute on Drug Abuse (NIDA) and the National Institute of Alcohol Abuse and Alcoholism (NIAAA).
2021-58
[124I]IBETA PET/CT studies in Alzheimer’s disease 5XFAD mouse model of β-amyloid (Aβ) Plaques (#279)
Christopher Liang, An N. Nguyen, Eunice Lee and Jogeshwar Mukherjee
Radiology, University of California, Irvine, California, USA
Abstract
Introduction: Several high affinity PET radiotracers for imaging Aβ plaques are currently being in the diagnosis of Alzheimer’s disease (AD).1,2 Radioiodinated pyridyl benzofuran derivatives for SPECT imaging of Aβ plaques using 123I and 125I have been reported.3 Because of the long half life of iodine-124 and use in extended PET imaging,4 we have prepared the benzofuran, [124I]I-IBETA (Ki = 2.36 ± 0.533) for use in PET imaging. Here we report our preliminary findings on [124I]I-IBETA in the 5XFAD AD mouse model.
Methods: [124I]Sodium Iodide (from 3D Imaging LLC) was used to prepare [124I]I-IBETA by electrophilic substitution of the tributyltin derivative (Figure 1) and purified by chromatography. Female hemizygous 5XFAD (n = 3) 7-month old mice obtained from MMRRC JAX were used for in vitro and in vivo study. Horizontal brain slices were incubated with [124I]I-IBETA and analyzed using Optiquant. In 5XFAD mice, [124I]I-IBETA was injected retroorbitally (0.9 MBq) under 2% isoflurane anesthesia. All mice underwent 15-minute PET scans (90 minutes and 24 hours post-injection) in a supine position with an accompanying 7-minute CT scan for attenuation correction. An Inveon Multimodality scanner was used for CT acquisitions in combined PET/CT experiments.
Results: Iodine-124 radiolabeling proceeded smoothly and purified [124I]I-IBETA was found to be stable for in vitro and in vivo studies. In vitro binding of [124I]I-IBETA in 5XFAD brain slices was high in thalamus, cortex and hippocampus, while cerebellum revealed very low levels of binding (Figure 2(b)). Ratio of regions versus cerebellum were > 10 (Figure 2(C) and (D)) suggesting high levels of Aβ plaques in these regions. In vivo PET/CT [124I]I-IBETA scans, uptake was non-uniform, but showed uptake in the thalamus and frontal cortex region as shown in Figure 2(e). At later time points, significant amount of activity was detected in the thyroid region, suggesting that some deiodination of [124I]I-IBETA in vivo.
Conclusion: Preliminary PET/CT scans and mouse brain slice autoradiography in 5XFAD mice have shown binding of [124I]I-IBETA to Aβ plaques consistent with 5XFAD mice model.5 There is potential for radioiodinated pyridyl benzofurans to be used as PET tracers to detect Aβ plaques.3 Further studies are planned to investigate drug effects and reversibility of binding of [124I]I-IBETA on Aβ plaques in vivo.
Acknowledgements
NIH/NIA RF1 AG029479 (JM), UCI UROP (AN)
References
PanMLMukherjeeMTPatelHH, et al.Evaluation of [11C]TAZA for amyloid Ab plaque imaging in postmortem Alzheimer’s disease brain region and whole body distribution in rodent PET/CT. Synapse2016;
70: 163–176.KaurHFelixMRLiangC, et al.Development and evaluation [18F]Flotaza for Ab plaque imaging in post-mortem Alzheimer’s disease brain. Bioorg Med Chem Lett2021;
46: 128164 .OnoMChengYKimuraH, et al.Development of novel 123I-labeled pyridyl benzofuran derivatives for SPECT imaging of b-amyloid plaques in Alzheimer’s disease.PloS One2021;
8: e74104.PandeySKVenugopalAKantR, et al.124I-Epidepride: a high affinity and selective PET radiotracer with potential for extended imaging of dopamine D2/D3 receptors. Nucl Med Biol2014;
41: 426–431.OakleyHColeSLLoganS, tet al.Intraneuronal beta-amyloid aggregates, neurodegeneration, and neuron loss in transgenic mice with five familial Alzheimer’s disease mutations: potential factors in amyloid plaque formation.J Neurosci2006;
26: 10129–10140.
2021-59
Examining kinetic spectrum of extracerebral signal and its contributions to reference regions of 18F-MK6240 PET (#280)
Jessie Fanglu Fu1, 2, Cristina Lois2, 3, Justin Sanchez3, Alex Becker2, 3, Zoe Rubinstein3, Emma Thibault3, Andrew Salvatore1, Hasan Sari1, 2, Michelle Farrell2, Marc Normandin2, 3, Nicolas J. Guehl2, 3, Georges El Fakhri2, 3, Keith Johnson2, 3 and Julie C. Price1, 2
1Athinoula A. Martinos Center for Biomedical Research, Massachusetts General Hospital, Charlestown Massachusetts, USA
2Department of Radiology, Harvard Medical School, Boston, Massachusetts, USA
3Gordon Center for Medical Imaging, Massachusetts General Hospital, Division of Nuclear Medicine and Molecular Imaging, Boston Massachusetts, USA
Abstract
Introduction: Off-target 18F-MK-6240 binding to melanocytes (extracerebral signal (EC), Figure 1(a))1,2 can contaminate cerebellar gray matter (CerGM) reference uptake and impact the detection of emergent tau signal in Alzheimer’s disease (AD)3,4 by reference tissue methods. We compared 18F-MK6240 kinetic spectra quantified in EC, CerGM and alternative reference regions to inform reference region selection, across the AD spectrum.
Methods: Thirteen subjects (8 controls (CN)–63 ± 12yrs; 5 mild cognitive impairment (MCI) or AD–75 ± 9yrs) underwent 120 min 18F-MK6240 PET, MRI, and arterial blood sampling. Masks for EC (Figure 1(b) and (c)) and reference regions (CerGM, eroded CerGM 3 mm, inferior CerGM (InfCer), cerebral white matter (WM) and pons) were generated (Freesurfer). Region-level spectral analysis (SA)5 was applied to decompose total 18F-MK6240 uptake into spectral components with frequency band amplitude a. SA kinetic spectra were compared between EC and reference regions (T-tests, p < 0.05).
Results: SA indicated 2 reversible components in reference regions (Figure 2(a) and (b)): 1) b ∼0.03min−1 (slow, non-displaceable uptake) and 2) b ∼0.18min−1 (fast, tracer delivery). Compared to CerGM, EC SUV was 30% higher and constantly increasing in late frames (Figure 1(d)) and only exhibited a fast component with 90% lower amplitudes (P < 10E-5, Figure 2(d)). A trapping/irreversible component (b = 10E-5min−1) was detected in all regions, with amplitudes significantly higher in EC (∼3-fold) than other regions (P < 0.05) except InfCer (Figure 2(e)). Compared to CerGM, trapping amplitudes were higher in InfCer (50%), but lower for WM and pons (60%) and eroded CerGM (18%).
Conclusion: SA supports a 2-tissue compartment configuration in reference regions. EC signal uniquely exhibited 1 reversible component and largest trapping amplitudes. The trapping component in reference regions likely reflects EC contamination, which was most evident in InfCer and less prevalent in WM and pons. Further studies will quantify the EC signal longitudinally and examine model-based methods to account for the EC signal as an alternative to partial-volume-correction.
References
BetthauserTJCodyKAZammitMD, et al.In vivo characterization and quantification of neurofibrillary tau PET radioligand 18F-MK-6240 in humans from Alzheimer disease dementia to young controls. J Nucl Med2019;
60: 93–99.AgueroCDhaynautMNormandinMD, et al.Autoradiography validation of novel tau PET tracer [F-18]-MK-6240 on human postmortem brain tissue. Acta Neuropathol Commun2019;
7: 37.PascoalTAShinMKangMS, et al.In vivo quantification of neurofibrillary tangles with [18F]MK-6240. Alzheimer’s Res Ther2018;
10: 74.GuehlNJWootenDWYokellDL, et al.Evaluation of pharmacokinetic modeling strategies for in-vivo quantification of tau with the radiotracer [18F]MK6240 in human subjects. Eur J Nucl Med Mol Imaging2019;
46: 2099–2111.VeroneseMRizzoGBertoldoA, et al.Spectral analysis of dynamic PET studies: a review of 20 years of method developments and applications. Comput Math Methods Med. 2016; Article ID 7187541.
2021-60
The relationship between glutamate, dopamine receptors, dopamine release and cortical grey matter: A simultaneous PET-MR study (#281)
Antoine Rogeau, Giovanna Nordio, Mattia Veronese, Oliver D. Howes and Robert McCutcheon
King’s College London, Department of Neuroimaging, Institute of Psychiatry, Psychology & Neuroscience, London, UK
Abstract
Introduction: The relationship between cortical grey matter, striatal dopamine signalling and cortical glutamate levels is of interest given their frequent simultaneous disruption in neuropsychiatric disease. Simultaneous measurement of these systems is challenging given the different imaging modalities required for accurate assessment. The aim of this study is to find out whether correlations exist between these neural systems.
Methods: In the current study, twenty-eight healthy subjects were enrolled to undergo 2 simultaneous [11C]-PHNO PET-MRI scans after double-blind randomised placebo or amphetamine administration in a cross-over design. This allowed measurement of striatal dopamine release capacity, striatal dopamine receptor availability, anterior cingulate glutamine & glutamate levels, and cortical grey matter volumes (GMV). A multiple regression model was used with voxel-based morphometry to seek associations between neurochemical measures and GMV.
Results: Left associative striatum dopamine receptor availability demonstrated a significant positive correlation with the volume of a grey matter cluster in the right prefrontal cortex (pFWE corrected = 0.022 – Figure 1). In addition, a negative interaction effect between sensorimotor dopamine receptor availability and glutamate+glutamine (Glx) correlated with a right cluster in BA6 (pFWE corrected = 0.047 – Figure 2).
Conclusion: These results extend previous findings that showed association between prefrontal grey matter volume and striatal dopamine synthesis capacity in schizophrenia to demonstrate that even in healthy individuals striatal dopamine function and prefrontal GMV are linked. It also highlights an interaction effect of ACC glutamate and striatal dopamine on prefrontal GMV. This brings new insights to the relationship between dopaminergic and glutamatergic physiology, and their relationship with cortical structure.
Acknowledgements
Supported by NIHR Maudsley BRC and Wellcome Trust Innovator Award (215747/Z/19/Z).
2021-61
Peripheral red blood cell (RBC) docosahexaenoic acid (DHA) and serum triglyceride levels influence [11C]PBR28 binding to TSPO in the brain (#282)
Savannah Tollefson1, Michael Himes1, Katelyn Kozinski1, Brian Lopresti1, Neale S. Mason1 and Rajesh Narendran1, 2
1School of Medicine, Department of Radiology, University of Pittsburgh, Pittsburgh Pennsylvania, USA
2School of Medicine, Department of Psychiatry, University of Pittsburgh, Pittsburgh Pennsylvania, USA
Abstract
Introduction: DHA, an omega-3 fatty acid, exhibits anti-inflammatory effects in the brain in basic investigations. DHA’s ability to modulate brain inflammation is of interest in Alzheimer’s disease and traumatic brain injury. We hypothesized that binding to the 18KDa translocator protein (TSPO), a proxy for M1-type microglia, will be higher in individuals with low relative to high DHA levels.
Methods: RBC DHA was measured in 320 healthy males. [11C]PBR28 was used to measure TSPO binding in 38 and 32 males in the lowest and highest RBC DHA quartiles. Volumes of distribution expressed relative to total plasma ligand concentration (VT) was derived using a two-tissue compartment kinetic analysis in fourteen brain regions (ROIs). Stress-resilience and cognitive assessments were conducted to correlate with VT.
Results: Age, weight and body mass index (BMI) were not different between-groups. RBC DHA in the high and low groups were 3.27 ± 0.56% and 2.13 ± 0.35%. Trend-level negative associations between DHA and serum triglycerides was present.
Between-group differences in [11C]PBR28 injected dose and mass, fp and CL were not present. VT was significantly lower (by 12% and 20% in C/T and C/C rs6971 genotypes) in the low compared to high RBC DHA group (Linear Mixed Model: DHA group, p = 0.047; ROI, p < 0.001; genotype, p < 0.001). VT was positively and negatively associated with high-density lipoproteins and triglycerides, respectively. Negative correlations between VT and triglycerides (5/14 ROIs), but not high-density lipoproteins survived the Bonferroni correction. No relationships between VT and BMI, or clinical assessments were present.
Conclusion: Contrary to our hypothesis, we found lower TSPObinding in low relative to high DHA subjects. It is unclear as to whether low TSPO binding reflects differences in microglia and/or triglyceride metabolism in this study. These results underscore the need to consider lipid parameters as a factor when interpreting TSPO PET clinical findings.
Funding
W81XWH-15-PRMRP-IIRA/Log Number: PR150716 from the US Department of Defense (DoD)
2021-62
Choroid plexus enlargement is associated with neuroinflammation and reduction of blood brain barrier permeability in depression (#283)
Noha S. Althubaity1, 2, Julia J. Schubert1, Daniel Martins1, Tayyabah Yousaf1, Maria A. Nettis3, Valeria Mondelli3, Carmine Pariante3, Neil A. Harrison4, 5, Edward T. Bullmore6, 7, Danai Dima1, 9, Federico E. Turkheimer1 and Mattia Veronese1, 10 and Neuroimmunology of Mood Disorders and Alzheimer’s Disease Consortium11
1Department of Neuroimaging, King’s College London, IoPPN, London, UK
2Radiological Sciences, College of Applied Medical Science, King Saud bin Abdulaziz University for Health Sciences, Riyadh, Saudi Arabia
3Department of Psychological Medicine, King’s College London, IoPPN, London, UK
4Brain Research Imaging Centre, Cardiff University, Cardiff, UK
5Department of Neuroscience, Brighton and Sussex Medical School, University of Sussex, Brighton, UK
6Department of Psychiatry, School of Clinical Medicine, University of Cambridge, Cambridge, UK
7Cambridgeshire and Peterborough NHS Foundation Trust, Cambridge, UK
8GlaxoSmithKline R&D, Immuno-Psychiatry, Immuno-Inflammation Therapeutic Area Unit, Stevenage, UK
9Department of Psychology, School of Arts and Social Sciences, City University of London, London, UK
10Department of Information Engineering, University of Padua, Padua, Italy
11Wellcome Trust Consortium for the Neuroimmunology of Mood Disorders and Alzheimer’s Disease, UK, UK
Abstract
Introduction: Depression is often associated with elevations of peripheral inflammatory markers1 but the mechanisms linking peripheral inflammation and changes in the central nervous system are still under investigation. Recent studies have shown that choroid plexuses (CP) may be involved in the neuro-immune axes, playing a role in brain homeostasis, and mediating the interaction between central and peripheral inflammation.2,3 Here we aimed to investigate CP volume alterations in depression and their associations with central brain inflammation.
Methods: 51 depressed participants (HDRS score > 13) and 25 age- and sex-matched healthy controls (HCs) from the Wellcome Trust NIMA consortium were re-analysed for the study. All the participants underwent full peripheral cytokine profiling and simultaneous [11C]PK11195 PET/structural MRI imaging for measuring neuroinflammation and blood-to-CSF radiotracer exchange parameters, and CP volume, respectively. We leveraged transcriptomic data from the Allen Human Brain Atlas to explore possible associations between the brain map depicting the correlations between CP volume and TSPO with functional brain pathways.
Results: We found a significantly greater CP volume in depressed subjects compared to HCs (t(76) = +2.17, p = 0.03) that was positively correlated with [11C]PK11195 binding in the anterior cingulate cortex (r = 0.28, p = 0.02), prefrontal cortex (r = 0.24, p = 0.04), and insular cortex (r = 0.24, p = 0.04) (Figure 1). The CP volume exhibited a negative association with the blood-to-CSF radiotracer exchange parameters (r = -0.28, p = 0.02). Integration of transcriptomic data with CP volume/TSPO imaging correlation map showed significant gene enrichment for several pathways involved in the neuroinflammatory response.
Conclusion: These results support the hypothesis that changes in brain barriers may cause reduction in solute exchanges between blood and CSF, disturbing brain homeostasis and ultimately contributing to inflammation in depression. Given that CP anomalies have been recently detected in other brain disorders4,5 these results may not be specific to depression and might extend to other conditions with a peripheral inflammatory component.
Acknowledgements
The Wellcome Trust and NIHR Maudsley Biomedical Research Centre.
References
MillerAHMaleticVRaisonCL.
Inflammation and its discontents: the role of cytokines in the pathophysiology of major depression. Biol Psychiatr2009;
65: 732–741.TurkheimerFE, et al.Increased serum peripheral C-reactive protein is associated with reduced brain barriers permeability of TSPO radioligands in healthy volunteers and depressed patients: implications for inflammation and depression. Brain, Behav, Immun2020;
▪: ▪.SchwartzMBaruchK.
The resolution of neuroinflammation in neurodegeneration: leukocyte recruitment via the choroid plexus. EMBO J, 2014.
33: 7–22.LizanoP, et al.Association of choroid plexus enlargement with cognitive, inflammatory, and structural phenotypes across the psychosis spectrum. Am J Psychiatr2019;
176: 564–572.TadayonE, et al.Choroid plexus volume is associated with levels of CSF proteins: relevance for Alzheimer’s and Parkinson’s disease. Neurobiol Aging2020;
89: 108–117.
2021-63
Evidence of blood-to-cerebrospinal fluid alterations in traumatic brain injury (#284)
Julia J. Schubert1, Mattia Veronese1, Gregory Scott2, Oliver Cousins1, Richard J. Greenwood3, 4, Anil F. Ramlackhansingh2, David J. Sharp2 and Federico E. Turkheimer1
1Department of Neuroimaging, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London, UK
2Department of Medicine, Imperial College London, London, UK
3University College London, London, UK
4MedTech West at Sahlgrenska University Hospital, Gothenburg, Sweden
Abstract
Introduction: The brain relies on cerebrospinal fluid (CSF) for delivery, movement, and removal of solutes.1 CSF is mostly produced by choroid plexuses (CPs) lining the ventricular system and forming part of the blood-to-CSF barrier (BCSFB).2 Reduced lateral ventricle (LV) PET signal has been observed in Alzheimer’s disease (AD),3 possibly attributed to reduced CSF-mediated tissue clearance. Given that traumatic brain injury (TBI) exhibits similar pathological features to AD,4,5 we investigated whether blood-to-CSF dynamic alterations exist in TBI using existing PET data by comparing LV 11C-PBR28 signal, perfusion rates, and CP volumes of TBI subjects and healthy controls (HC).
Methods: 14 moderate-severe TBI subjects and 44 HC underwent 90-minute dynamic 11C-PBR28 PET with arterial input function and structural MRI. LV and CP segmentations were generated. LV 11C-PBR28 signal was quantified as standard uptake value ratios (SUVRs) from 50–70 minutes using cerebellum grey matter reference. Compartmental modelling of LV 11C-PBR28 signal was also performed to quantify input from plasma (K1), clearance to plasma and the remaining ventricular system (kClearance), and rates of specific binding on (kon) and off (koff) LV binding sites (Figure 1). CP volume was calculated using FSL (FMRIB, Oxford UK).
Results: Significantly lower LV SUVRs (F(1,54) = 7.78, P = 0.007; Figure 2(a)) and K1 (F(1,52) = 8.82, P = 0.004; Figure 2(b)) and significantly higher CP volumes (F(1,54) = 39.74, P < 0.001; Figure 2(c)) were observed in TBI compared to HC. No significant differences were observed in kClearance, kon, or koff.
Conclusion: We observed reduced LV signal and perfusion rates from plasma to LV in TBI, similar to previous results in AD.3 We also observed larger CP volume in TBI, demonstrating that the CP contribution of the BCSFB is altered in TBI. Blood-to-CSF transfer rate alterations may lead to reduced passage of small molecules into and through the CSF in TBI, contributing to early development of neurodegenerative disease.
Acknowledgements
Hammersmith Imanet provided radiotracers and scanning facilities used for data acquisition.
We gratefully acknowledge all study participants.
References
IliffJJWangMLiaoY, et al.A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid β. Sci Transl Med, 2012;
4, 147ra111.OreškovićDKlaricaM.
The formation of cerebrospinal fluid: Nearly a hundred years of interpretations and misinterpretations. Brain Res Rev, 2010;
64: 241–262.SchubertJJVeroneseMMarchitelliL, et al.Dynamic 11 C-PiB PET shows cerebrospinal fluid flow alterations in Alzheimer’s disease and multiple sclerosis. J Nucl Med, 2019;
60: 1452–1460.IliffJJChenMJPlogBA, et al.Impairment of glymphatic pathway function promotes tau pathology after traumatic brain injury. J Neurosci2014;
34: 16180–16193.EdwardsGZhaoJDashPK, et al.Traumatic brain injury induces tau aggregation and spreading. J Neurotrauma2020;
37: 80–92.
2021-64
Comparison of Lewy body distribution and monoamine oxidase A in anterior cingulate of postmortem Parkinson’s disease brains (#286)
Reisha M. Ladwa, Rommani Mondal, Christopher Liang and Jogeshwar Mukherjee
Radiological Sciences, University of California, Irvine California, USA
Abstract
Introduction: Lewy bodies (LB) in anterior cingulate (AC) play a significant role in Parkinson’s disease (PD).1 Imaging agents for LB or surrogate markers for LB will assist in earlier and accurate diagnosis of PD and LB dementia. The goal in this study is to evaluate LB using anti-ubiquitin immunohistochemistry (UIHC) in AC2 (Figure 1) and correlate them to monoamine oxidase A (MAO-A) expression using the selective PET agent, [18F]FAZIN3.3
Methods: Human post-mortem brain slices of anterior cingulate and corpus callosum (controls (CN), n = 6; age 81–90 LB = 0 and PD, n = 6, age 77–89, LB = III-IV) were used. Brain slices were UIHC stained for LB and [18F]FAZIN3 for MAO-A.3 The UIHC image was analyzed using QuPath to create a heatmap of the image.4 Spatially aligned UIHC and autoradiograph images were processed in MATLAB to generate normalized image intensity plots through the cortical layers for quantitation. Autoradiography image intensity corresponded to [18F]FAZIN3 binding to MAO-A and QuPath image intensity corresponded to the presence of LB.
Results: Figure 2 shows relative intensity profiles of UIHC and autoradiography of one PD subject indicating a correlation of the increase in [18F]FAZIN3 binding with LB. This increased [18F]FAZIN3 binding and UIHC staining was observed in all PD subjects in the AC compared to the CN subjects. Outer cortical layers (I-III) had 21%, inner layers (IV-VI) had 75% and white matter (WM) had < 1% UIHC. In the CN subjects all grey matter (GM) and white matter (WM) had < 1% UIHC. [18F]FAZIN3 ratio in PD Figure 2(a) was GM/WM = 3.57 while CN subjects was GM/WM = 2.24. Increased LB in PD brains resulted in increased [18F]FAZIN3 binding to MAO-A.
Conclusion: A novel method QuPathRad (QPR) has been developed which allows quantitative comparative image analysis of IHC with autoradiography. QPR confirmed the increased [18F]FAZIN3 binding in the PD with the presence of LB in the inner cortical layers of AC. This suggests increased levels of MAO-A in LB, and possibly increased mitochondria in LB.5 Thus, MAO-A imaging using [18F]FAZIN3 may serve as an early diagnostic tool for PD.
Acknowledgements
NIH/NIA RF1 AG029479 (JM), UCI UROP (RML, RM). Banner Health Research Institute for brain samples and UCI Pathology for immunostaining.
References
KovariEGoldGHerrmannFR, et al.Lewy body densities in the entorhinal and anterior cingulate cortex predict cognitive deficits in Parkinson’s disease.Act Neuropathol2003;
106: 83–88.LennoxGLoweJMorrellK, et al.Anti-ubiquitin immunocytochemistry is more sensitive than conventional techniques in the detection of diffuse Lewy body disease. J Neurol Neurosurg Psychiatry1989;
52: 67–71.MukherjeeJLiangCSyedAU, et al.Monoamine oxidase-A is upregulated in post-mortem human Alzheimer’s disease and Parkinson’s disease brain. J Nucl Med2021;
62(Suppl 1): 1609.BankheadP, et al.
QuPath: open source software for digital pathology image analysis. Scient Rep2017;
7: 16878.Mahul-MellierA-LBurtscherJMaharajanN, et al.The process of Lewy body formation, rather than simply a-synuclein fibrillization, is one of the major drivers of neurodegeneration.Proc Natl Acad Sci2020;
117: 4971–4982.
2021-65
A multiple regression modelling approach to investigate the coupling between [18F]fluorodeoxyglucose positron emission tomography and resting-state functional MRI (#287)
Tommaso Volpi1, 2, Erica Silvestri1, 3, Marco Aiello4, Maurizio Corbetta1, 2 and Alessandra Bertoldo1, 3
1Padova Neuroscience Center, University of Padova, Padova, Italy
2Department of Neuroscience, University of Padova, Padova, Italy
3Department of Information Engineering, University of Padova, Padova, Italy
4IRCCS SDN, Napoli, Italy
Abstract
Introduction: Brain glucose metabolism as measured by [18F]fluorodeoxyglucose ([18F]FDG) positron emission tomography (PET) is expected to be related to spontaneous activity and functional connectivity (FC) derived from resting-state functional MRI (rs-fMRI)1, but their coupling model is not thoroughly understood.
Methods: Employing simultaneous acquisitions on a Siemens Biograph mMR scanner in 26 healthy individuals (59.8 ± 10.8 yrs),2,3 we related [18F]FDG standard uptake value ratio (SUVR, relative to whole-brain average uptake) to 50 rs-fMRI features pertaining to 1) signal and local properties, 2) hemodynamic response function (HRF), 3) static FC (sFC), 4) time-varying FC (tvFC), and 5) phase synchronization (PC). Both [18F]FDG and rs-fMRI data were parcellated using the Schaefer cortical atlas4 supplemented by subcortical parcels (218 regions). To assess which rs-fMRI variables better describe SUVR across regions at the group median level, we performed feature selection using sparse regression approaches.
Results: Feature selection led to the choice of 9 rs-fMRI predictors out of 50. The combination of the selected predictors explained a significant proportion of the group-level SUVR variance across regions (R2 = 0.41), with local rs-fMRI features (i.e., signal complexity, local coherence, number of high-amplitude events etc.) as more relevant with respect to large-scale network information, i.e. sFC and tvFC (Figure 1(a)). The region-wise model residuals are shown in Figure 1(b).
Conclusion: Using multiple regression modelling, we highlight that local rs-fMRI metrics are more tightly linked to SUVR than large-scale FC, possibly because of their inherent blood flow (CBF) information. Future work on [18F]FDG-fMRI coupling should aim at regressing out the CBF component from the rs-fMRI signal5, and performing PET kinetic modelling to disentangle influx of [18F]FDG (K1) from its actual phosphorylation (k3).
References
TomasiDWangGJVolkowND.
Energetic cost of brain functional connectivity. Proc Natl Acad Sci USA2013;
110: 13642–13647.RiedlVBienkowskaKStrobelC, et al.Local activity determines functional connectivity in the resting human brain: a simultaneous FDG PET/fMRI study. J Neurosci2014;
34: 6260–6266.AielloMSalvatoreECachiaA, et al.Relationship between simultaneously acquired resting-state regional cerebral glucose metabolism and functional MRI: a PET/MR hybrid scanner study. NeuroImage2015;
113: 111–112.SchaeferAKongRGordonEM, et al.Local-global parcellation of the human cerebral cortex from intrinsic functional connectivity MRI.Cerebral Cortex2018;
28: 3095–3114.TongYLindseyKPHockeLM, et al.Perfusion information extracted from resting state functional magnetic resonance imaging. J Cereb Blood Flow Metab2017;
37: 564–576.
2021-66
Comparative sensitivity of PET radioligands to partial inhibition of P-glycoprotein at the blood-brain barrier (#288)
Louise Breuil1, 2, Solène Marie1, 3, Sébastien Goutal1, Sylvain Auvity1, 4, Charles Truillet1, Wadad Saba1, Oliver Langer5, Fabien Caillé1 and Nicolas Tournier1
1Laboratoire d’Imagerie Biomédicale Multimodale (BIOMAPS), Université Paris-Saclay, CEA, CNRS, Inserm, Service Hospitalier Frédéric Joliot, Orsay, France
2Pharmacy Department, Robert-Debré Hospital, AP-HP, Université de Paris, Paris, France
3Pharmacy Department, Bicêtre Hospital, AP-HP, Université Paris-Saclay, Le Kremlin-Bicêtre, France
4Pharmacy Department, Necker Hospital, AP-HP, UMR-S 1144, Université de Paris, Paris, France
5Department of Clinical Pharmacology, Medical University of Vienna, Vienna, Austria
Abstract
Introduction: P-glycoprotein (P-gp, ABCB1) is a major efflux transporter at the blood-brain barrier (BBB). Radiolabeled substrates used for quantitative determination of P-gp function at the BBB using PET were initially selected based on their ability to generate maximal contrast in response to complete P-gp inhibition. However, only partial deficiency or inhibition is likely to occur in pathophysiological situations or drug-drug interactions. This raises questions regarding the sensitivity of available PET probes to detect moderate but clinically relevant changes in P-gp function at the BBB.
Methods: The P-gp-mediated transport characteristics of the radiolabeled substrates 11C-verapamil, 11C-N-desmethyl-loperamide and 11C-metoclopramide were compared in vitro and in vivo. Substrate uptake in MDCKII-MDR1 cells overexpressing human P-gp was compared in the increasing concentrations presence of the P-gp inhibitor tariquidar (0–200 nM, n = 4). Then, the impact of increasing doses of tariquidar (i.v, 1–8 mg/kg) on the brain exposure to selected substrates was assessed using PET imaging in rats.
Results:In vitro, the half-maximum inhibitory concentration (IC50) of tariquidar was significantly different between 11C-verapamil (44 nM [33–53 nM]), 11C-N-desmethyl-loperamide (19 nM [13–25 nM]), and 11C-metoclopramide (4 nM [2–8 nM]) (Figure vitro). Maximal increase in cellular uptake was 1.4 ± 0.1, 3.9 ± 0.7 and 1.4 ± 0.1-fold for 11C-verapamil, 11C-N-desmethyl-loperamide and 11C-metoclopramide, respectively (p < 0.001). In vivo, half-maximum inhibition of P-gp-mediated transport of 11C-metoclopramide at the BBB was achieved using 1 mg/kg [0.78–1.5 mg/kg] of tariquidar (p < 0.05, n = 4), corresponding to an IC50 of 82 nM tariquidar in plasma and resulting in a 2.1-fold increase in brain exposure (Figure vivo). This dose increased the brain exposure of 11C-verapamil to a similar extent (2.4-fold increase, p < 0.05, n = 4) but did not significantly increase the brain exposure of 11C-N-desmethyl-loperamide (n = 4) compared with baseline.
Conclusion: This comparative in vitro/in vivo study showed differences in the “vulnerability” to P-gp inhibition among radiolabeled substrates, which was apparently unrelated to their maximal response to P-gp inhibition. We advocate that partial inhibition of transporter function, in addition to maximal inhibition, should be a primary criterion of evaluation regarding the sensitivity of radiolabeled substrates to detect moderate but physiologically-relevant changes in transporter function in vivo.
Acknowledgements
We thank Maud Goislard, Thierry Lekieffre, Christine Coulon and Kevin Pansavath for helpful technical assistance. Louise Breuil received funding from the joined AP-HP/CEA grant. This work was performed on a platform partially funded by the France Life Imaging network (grant ANR-11-INBS-0006).
References
KalvassJCPolliJWBourdetDL, et al.Why clinical modulation of efflux transport at the human blood–brain barrier is unlikely: the ITC evidence-based position. Clin Pharmacol Therapeut2013;
94: 80–94.KannanPJohnCZoghbiSS, et al.Imaging the function of P-glycoprotein with radiotracers: pharmacokinetics and in vivo applications. Clin Pharmacol Therapeut2009;
86: 368–377.TournierNBauerMPichlerV, et al.Impact of P-glycoprotein function on the brain kinetics of the weak substrate 11C-metoclopramide assessed with PET imaging in humans. J Nucl Med2019;
60: 985–991.
2021-67
Multimodal investigation of the synaptic and metabolic basis of low-frequency oscillations in the human brain: A [11C]UCB-J, [18F]BCPP-EF and resting state fMRI study (#289)
Ekaterina Shatalina1, Gaia Rizzo2, Robert A. Comley6, Hideo Tsukada7, Oliver D. Howes1, 5, Eugenii A. Rabiner2, 5 and Matthew Wall2, 3
1MRC London Institute of Medical Sciences, Imperial College London, London, UK
2Invicro LLC, London, UK
3Clinical Psychopharmacology Unit, University College London, London, UK
4Faculty of Medicine, Imperial College London, London, UK
5Institute of Psychiatry, Psychology and Neuroscience, King’s College London, London, UK
6AbbVie, Chicago Illinois, USA
7Hamamatsu Photonics K.K., Hamamatsu, Japan
Abstract
Introduction: The neuronal basis of low-frequency blood oxygen level dependent (BOLD) signal oscillations in fMRI is not fully understood but may reflect synaptic firing or underlying metabolic activity. This multimodal imaging study combined [11C]UCB-J and [18F]BCPP-EF PET and resting state fMRI to investigate the biological basis of BOLD fluctuations. [11C]UCB-J, a purported marker of synaptic density, binds to synaptic vesicle protein 2A (SV2A), while [18F]BCPP-EF binds to mitochondrial complex I (MC1) and is assumed to be a marker of cellular metabolism. FMRI measures used were fractional Amplitude of Low Frequency Fluctuations (fALFF) and dual-regression analyses of functional connectivity.
Methods: 23 healthy subjects (12M,11F; age range 22–78, mean age 55.7 years) completed a resting state fMRI scan, a [18F]BCPP-EF scan, and a [11C]UCB-J scan (n = 22). Eight standard resting-state networks (RSN) were used as regions of interest across all imaging modalities. FMRI data were processed using standard methods, with fALFF values and dual-regression analyses carried out subsequently. Parametric volume of distribution (VT) images were generated using a one-tissue compartment model for [11C]UCB-J and the Logan Graphical Analysis (LGA) for [18F]BCPP-EF. Average VT values for the eight RSNs were estimated for both tracers. Data from all modalities was transformed into MNI152 space.
Results: FALFF values for three networks (default mode, ventral attention, and frontoparietal right) were significantly correlated with UCB-J VT (multiple-comparisons corrected; all r > 0.5, p < 0.01). [18F]BCPP-EF VT was not significantly correlated with fALFF in any RSN. Neither PET measure showed significant correlations with the dual-regression data.
Conclusion: Individual differences in synaptic density may contribute to BOLD low-frequency oscillations, however mitochondrial density may only have a secondary role. FALFF may also have potential as a useful proxy measure of synaptic activity in fMRI data. This is the first study investigating the synaptic and mitochondrial origins of BOLD fMRI signals in humans, with important implications for interpreting pathology-specific group differences seen in resting state fMRI
2021-68
Receptor distribution associations to mRNA gene expression patterns in the human cerebral cortex (#290)
Matej Murgaš, Paul Michenthaler, Gregor Gryglewski and Rupert Lanzenberger
Department of Psychiatry and Psychotherapy, Medical University of Vienna, Wien Wien, Austria
Abstract
Introduction: Investigation of potential links between molecular mechanisms and neuroreceptor distributions in the human cerebral cortex could help to enable in vivo analysis of multi-receptor systems and therefore lead to more effective clinical applications. In this work, we present a multi-receptor comparison of mRNA gene expression patterns to neuroreceptor distribution associated with them in the human cerebral cortex.
Methods: To this end, we utilized publicly available proteomic autoradiography data1 and transcriptomic data – normalized microarray (mA) and RNA sequencing (RNA-seq) data, published by Allan Brain institute2 and interpolated mRNA expression patterns (int-mA).3 Spatial distribution of 15 receptors listed in literature1 and 64 gene expression patterns associated with at least one of the receptors were represented by average regional values for 38 cytoarchitectonicaly distinct areas defined Julich-Brain atlas.4 Spearman’s rank correlation, together with the permutation test establishing significance, was used to measure the similarity of investigated data sets
Results: Solid positive associations between autoradiography and gene expression data sets were found for e.g. glutamate ionotropic receptor kainate type subunit 2 (int-mA: r = 0.63, mA: r = 0.60, RNA-seq: r = 0.56) or serotonin 1A receptor (int-mA: r = 0.71, mA: r = 0.60, RNA-seq: r = 0.54). Additionally, negative associations were found for e.g. GABA A receptor subunit beta 1 (int-mA: r = −0.58, mA: r = −0.54), GABA A receptor subunit theta (int-mA: r = −0.50, RNA-seq: r = −0.61) as well as for GABA A receptor subunit alpha 3 (int-mA: r = −0.64, mA: r = −0.62, RNA-seq: r = −0.49). However, most of the receptors showed low correlation values (int-mA: 60%, mA: 74%, RNA-seq: 76% of r < 0.35).
Conclusion: Comparison of mRNA gene expression patterns and proteomic data unveiled few strong positive links between neuroreceptor spatial distributions and molecular processes. Counterintuitively, most of the relations showed negative or no correlation, highlighting the importance of further studies on splicing variants and the potential role of mRNA markers associated with post-transcriptional and post-translational modifications.
Acknowledgements
M. Murgaš is funded by the Austrian Science Fund FWF, DOC 33-B27.
References
ZillesKPalomero-GallagherN .
Multiple transmitter receptors in regions and layers of the human cerebral cortex. Front Neuroanat2017;
11: 78.HawrylyczMJLeinESGuillozet-BongaartsAL, et al.An anatomically comprehensive atlas of the adult human brain transcriptome. Nature2012;
489: 391–399.GryglewskiGSeigerRJamesGM, et al.Spatial analysis and high resolution mapping of the human whole-brain transcriptome for integrative analysis in neuroimaging. Neuroimage2018;
14: 617–631.AmuntsKMohlbergHBludauS, et al.Julich-Brain: a 3D probabilistic atlas of the human brain’s cytoarchitecture. Science2020;
369: 988–992.
2021-69
Altered neuroepigenetics in autism spectrum disorder: a [11C]Martinostat PET brain imaging study (#292)
Chieh-En J. Tseng1, 2, Baileigh G. Hightower1, Anjali J. Parmar1, Rachel E. Marcus1, 3, Jennifer E. Mullett3, Christopher J. McDougle2, 3, Jacob M. Hooker1, 2 and Nicole R. Zurcher Wimmer1, 2
1Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Department of Radiology, Charlestown Massachusetts, USA
2Harvard Medical School, Boston Massachusetts, USA
3Lurie Center for Autism, Massachusetts General Hospital, Lexington Massachusetts, USA
Abstract
Introduction: Prenatal exposure to histone deacetylase (HDAC) inhibitors leads to autism-like behavior in rodent models and has been associated with autism spectrum disorder (ASD) in humans,1 suggesting that epigenetic HDAC enzymes may play a role in the etiology of ASD. Moreover, changes in HDACs have been demonstrated to affect core symptoms and comorbidities of ASD in preclinical studies2,3 and HDACs interact with several risk genes for ASD.4 Postmortem work has shown altered acetylation in genes implicated in ASD in frontal and temporal cortices in children and adults with ASD.5 No study has yet investigated HDAC expression in individuals with ASD in vivo.
Methods: The radiotracer [11C]Martinostat binds to HDAC1, 2, 3 and 6. Eight participants with ASD (5M/3F, age = 27.8 ± 7.6years) and eight controls (CON) matched for sex and age (5M/3F, age = 27 ± 5.6years) completed a [11C]Martinostat PET-MRI scan on a 3T Siemens TIM Trio with PET insert at the A. A. Martinos Center for Biomedical Imaging. [11C]Martinostat (ASD:5.3 ± 0.3mCi; CON:4.8 ± 0.4mCi) was injected as a bolus. A high-resolution T1-weighted anatomical scan with prospective motion correction was collected. PET data were reconstructed using the 3D ordinary Poisson ordered-subset expectation maximization (OP-OSEM) algorithmin units of standard uptake value (SUV), motion-corrected, and processed with FreeSurfer and FSL. SUV maps were normalized to the whole brain mean from 60–90 minutes post-injection (SUVR60-90). Whole brain voxelwise analysis was performed using an unpaired t-test in FSL’s FEAT with ordinary least squares mixed-effects modeling, with age and sex as covariates (Z > 2.3, Pcluster < 0.05).
Results: Lower [11C]Martinostat uptake was found in the bilateral orbitofrontal cortex, and right anterior cingulate cortex, right insula and right amygdala. These regions are associated with socio-cognitive processing and have previously been implicated in ASD in functional MRI studies.6 Additionally, higher [11C]Martinostat uptake was found in the bilateral visual cortex in ASD compared to CON (Figure 1).
Conclusion: Preliminary findings show altered HDAC brain expression in ASD compared to CON. [11C]Martinostat PET-MRI data from a larger sample size will be required to determine if HDAC alterations represent a common feature in ASD. This pilot study was conducted using SUVR60-90 given the challenges associated with obtaining arterial blood in individuals with ASD. Future work will investigate whether regional alterations in HDAC levels are linked to ASD-related symptoms.
Acknowledgements
We wish to thank the physicians Dr. Christopher Keary and Dr. Michelle Palumbo for obtaining informed consent, Dr. Lisa Nowinski for neuropsychological assessments, nuclear medicine technologists Shirley Hsu, Regan Butterfield and Grae Arabasz for radiotracer injection and assistance with PET-MRI scans, Judit Sore, Phill Nielson, Daniela Bernales and the radiopharmacy team for radioligand production, and the nurse practitioners Amy Kendall and Natacha Nortelus for medical coverage. Funding for this study was provided by the Transatlantic Research Scholar and pilot funding from the Athinoula A. Martinos Center for Biomedical Imaging awarded to N.R.Z. and the Phyllis and Jerome Lyle Rappaport MGH Research Scholar awarded to J.M.H. This research was carried out at the Martinos Center, using resources provided by the Center for Functional Neuroimaging Technologies, P41EB015896, a P41 Biotechnology Resource Grant supported by the National Institute of Biomedical Imaging and Bioengineering (NIBIB), NIH. This work also involved the use of instrumentation supported by the NIH Shared Instrumentation Grant Program S10RR017208-01A1, S10RR026666, S10RR022976, S10RR019933, S10RR023043, S10RR023401, and 1S10OD023517-01A1.
References
ChomiakTTurnerNHuB.
What we have learned about autism spectrum disorder from valproic acid. Patholog Res Int2013. https://doi.org/10.1155/2013/712758.SahASotnikovSKharitonovaM, et al.Epigenetic mechanisms within the cingulate cortex regulate innate anxiety-like behavior. Int J Neuropsychopharmacol2019;
22: 317–328.QinLMaKWangZ-J, et al.Social deficits in Shank3-deficient mouse models of autism are rescued by histone deacetylase (HDAC) inhibition. Nat Neurosci2018; https://doi.org/10.1038/s41593-018-0110-8.SatterstromFKKosmickiJAWangJ, et al.Large-scale exome sequencing study implicates both developmental and functional changes in the neurobiology of autism. Cell2020;
180: 568–584.e23.SunWPoschmannJCruz-Herrera del RosarioR, et al.Histone acetylome-wide association study of autism spectrum disorder. Cell2016;
167: 1385–1397.e11.
2021-70
Preclinical imaging of [11C]colchicine, [11C]verubulin, and [11C]HD-800 for imaging microtubules (#293)
Anton Lindberg1, Andrew V. Mossine2, Arturo Aliaga3, Robert Hopewell4, Gassan Massarweh4, Pedro Rosa-Neto3, Xia Shao2, Vadim Bernard-Gauthier1, 5, Peter J.H. Scott2 and Neil Vasdev1, 5
1CAMH, Brain Health Imaging Centre, Toronto, ON, Canada
2University of Michigan Medical School, Department of Radiology, Ann Arbor Michigan, USA
4McGill University, McConnell Brain Imaging Centre, Montreal, QC, Canada
5University of Toronto, Department of Psychiatry, Toronto, ON, Canada
Abstract
Introduction: [11C]Verubulin (a.k.a.[11C]MCP-6827), [11C]HD-800 and [11C]colchicine have been developed for imaging microtubules (MTs) with PET.1,2 The objective of this work was to conduct an in vivo comparison of [11C]verubulin for MT imaging in mouse and rat brain, as well as an in vitro study with this radiotracer in rodent and human Alzheimer’s Disease (AD) tissue. We also conducted the first comparative in vivo PET imaging study with all three of these radiotracers in a non-human primate (NHP).
Methods: [11C]Colchicine, [11C]verubulin, and [11C]HD-800 were labeled by O-11C-methylation from [11C]CH3OTf. [11C]Verubulin was used for autoradiography in rat and AD and healthy control (HC) tissues with verubulin (50 mL, 500 µM) co-administrated in blocking studies. [11C]Verubulin was used in PET measurements in rodents with verubulin (5 mg/kg) administered 20 min prior to radiotracer. [11C]verubulin, [11C]HD-800, and [11C]colchicine were compared in preliminary baseline PET scans in rhesus monkey.
Results: [11C]Verubulin, [11C]HD-800 and [11C]colchicine were obtained in 0.5–1.5% uncorrected radiochemical yields, 99% radiochemical purity and molar activitiesof 159–389 GBq/µmol. Autoradiography with [11C]verubulin in rat tissue showed higher binding compared to blocking measurements and higher uptake in AD patient tissue was seen compared with HC. In PET imaging studies, [11C]verubulin had decreased brain uptake in mouse, while increased brain uptake in rat was measured when pretreated with verubulin. In NHP PET imaging, [11C]verubulin and [11C]HD-800 entered the brain, and as expected [11C]colchicine did not cross the blood-brain barrier.
Conclusion: Preliminary PET imaging studies of [11C]verubulin in rodents revealed contradictory results between mouse and rat brain uptake under pretreatment conditions. [11C]verubulin showed an unexpected higher uptake in AD patient tissue compared with HC. Both [11C]Verubulin and [11C]HD-800 require pharmacokinetic modeling and quantification studies to understand the role of how these radiotracers bind to MTs in the central nervous system before translation to human use.
Acknowledgements
The authors would like to thank Ashley Knight and Dr. Shil Patel for their input with [3H]colchicine binding assays, as well as Jenelle Stauff, Janna Arteaga and Phillip Sherman for assistance with preclinical imaging studies. The authors also thank members of the CAMH Brain Health Imaging Centre and the University of Michigan PET Center for support.
References
KumarDHinesJNormanS, et al.▪▪▪. J Nucl Med2019;
60: ▪.LevchenkoAMehtaBMLeeJB, et al.▪▪▪. J Nucl Med2000;
41: 493–501.
2021-71
Task-specific dynamics of dopamine synthesis during monetary gain and loss (#294)
Andreas Hahn1, Murray B. Reed1, Verena Pichler2, 3, Lucas Rischka1, Godber M. Godbersen1, Lukas Nics2, Marcus Hacker2 and Rupert Lanzenberger1
1Medical University of Vienna, Department of Psychiatry and Psychotherapy, Vienna Wien, Austria
2Medical University of Vienna, Department of Biomedical Imaging and Image-guided Therapy, Division of Nuclear Medicine, Vienna Wien, Austria
3University of Vienna, Department of Pharmaceutical Sciences, Division of Pharmaceutical Chemistry, Vienna, Austria
Abstract
Introduction: The competition model is well-established for the investigation of dopamine action. To overcome disadvantages inherent to the assessment of cognitive processes, we combine functional PET (fPET) and 6-[18F]FDOPA imaging into a novel approach. The technique is based on the dynamic regulation of dopamine synthesis by neuronal firing,1 indicating a direct association with dopamine release2 (Figure 1).
Methods: 16 healthy subjects (24.8 ± 4.8 years, 7 female) underwent one 6-[18F]FDOPA fPET (GE Advance, 5.5 MBq/kg as bolus+infusion, 70x43 s frames). Subjects completed the monetary incentive delay (MID) task with separate assessment of monetary gain and loss (4x5 min task blocks). Task effects were assessed by the general linear model and dopamine synthesis was quantified with the Gjedde-Patlak plot after correction for the 3-OMFD component.
Results: In men monetary gain induced stronger increases in dopamine synthesis than loss, but this pattern was reversed in women (all p < 0.001, Figure 2), resulting in a sex difference for the contrast gain vs. loss (p < 0.001). Task-specific changes ranged between Ki = 0.009 ± 0.004/min and Ki = 0.019 ± 0.004/min, corresponding to increases of 105 ± 31% to 165 ± 64% from baseline Ki. Individual changes in DA synthesis were associated with behavioral parameters of reward sensitivity in men (rho = −0.67, p = 0.059) and punishment sensitivity in women (rho = 0.79, p < 0.05).
Conclusion: Assuming that task effects were driven by k3, presumably reflecting AADC activity, and considering DOPA fractions available for dopamine synthesis,3,4 reward processing seems to yield 56–107% of additionally synthesized dopamine, matching physiological values in rodents. Our approach enables the assessment of dopamine action during cognitive processing in a single scan with high temporal resolution and robust task-specific effect size.5 The findings provide a biological rationale for the well-known behavioral sex differences in reward and punishment processing. This may have important implications in psychiatric conditions with sex-specific prevalence rates, altered reward processing and dopamine signaling.
Acknowledgements
This research was supported by a grant from the Austrian Science Fund to A. Hahn (FWF KLI 610). L. Rischka and M.B. Reed are recipients of a DOC Fellowship of the Austrian Academy of Sciences at the Department of Psychiatry and Psychotherapy, Medical University of Vienna. The scientific project was performed with the support of the Medical Imaging Cluster of the Medical University of Vienna.
References
MorgenrothVH, et al.Tyrosine hydroxylase: activation by nerve stimulation. PNAS1974;
71: 4283–4287.KehrW, et al.Interaction of haloperidol and gamma-butyrolactone with (þ)-amphetamine-induced changes in monoamine synthesis and metabolism in rat brain. J Neural Transm1977;
40: 129–147.CummingP, et al.[3H]DOPA formed from [3H]tyrosine in living rat brain is not committed to dopamine synthesis. JCBFM1998;
18: 491–499.BestJA, et al.Homeostatic mechanisms in dopamine synthesis and release: a mathematical model.Theor Biol Med Model2009;
6: 21.HahnA, et al.Functional dynamics of dopamine synthesis during monetary reward and punishment processing. JCBFM2021. Epub ahead of print.
2021-72
Interdependent adaptations of glucose metabolism and functional connectivity elicited through learning (#295)
Andreas Hahn1, Sebastian Klug1, Godber M. Godbersen1, Lucas Rischka1, Wolfgang Wadsak2, 3, Verena Pichler2, 4, Marcus Hacker2 and Rupert Lanzenberger1
1Medical University of Vienna, Department of Psychiatry and Psychotherapy, Vienna Wien, Austria
2Medical University of Vienna, Department of Biomedical Imaging and Image-guided Therapy, Division of Nuclear Medicine, Vienna Wien, Austria
3Center for Biomarker Research in Medicine (CBmed), Graz Steiermark, Austria
4University of Vienna, Department of Pharmaceutical Sciences, Division of Pharmaceutical Chemistry, Vienna, Austria
Abstract
Introduction: Learning elicits numerous neuronal adaptations, but it is poorly understood how they act together to support improvements in cognitive processing. We employed metabolic connectivity mapping (MCM) to investigate learning-induced changes in the association of functional connectivity (FC) and the underlying glucose metabolism (CMRGlu), thereby enabling the computation of directional connectivity.1
Methods: 41 healthy subjects (23.2 ± 3.3 years, 21 female) underwent two PET/MRI scans (Siemens mMR), while performing the video game Tetris® in two levels of difficulty (easy, hard). The training group (n = 21) practiced the task for 53.6 ± 5.2 min/day for 20.9 ± 1.5 days between the two scans. [18F]FDG was applied as bolus+infusion (5.1 MBq/kg, 104x30 s frames). Task effects were assessed by the general linear model and CMRGlu was quantified with the Gjedde-Patlak plot.1,2 Simultaneous BOLD acquisition (TE/TR = 30/2000 ms, 6 min per condition) was used to compute FC at rest and during task performance.
Results: Compared to the control group, learning altered the directional connectivity (MCM) from the salience network (insula and dorsal anterior cingulate) to the occipital cortex, with increases during rest and decreases during task execution (group*time interactions, all p < 0.05 corrected, Figure 1). A higher divergence in MCM between rest and task was associated with better cognitive performance after learning. Simulations revealed that MCM changes at rest were driven by CMRGlu, but those during the task execution were dependent on FC.
Conclusion: Metabolism-driven increases in the association of CMRGlu and FC at rest potentially reflect learning-induced anchoring of glutamatergic AMPA receptors,3 which doubles the postsynaptic energy consumption.4 Retrieval of this metabolically expensive skill engram during task execution seems to enable the minimization of prediction errors5 between neuronal task representations between higher- (salience network) and lower-order (occipital cortex) brain regions. Disentangling the contribution of CMRGlu and FC in deficits of learning and cognition may provide novel insights in numerous patient populations.
Acknowledgements
This research was supported by a grant from the Austrian Science Fund to A. Hahn (FWF KLI 610). L. Rischka is recipient of a DOC Fellowship of the Austrian Academy of Sciences at the Department of Psychiatry and Psychotherapy, Medical University of Vienna. The scientific project was performed with the support of the Medical Imaging Cluster of the Medical University of Vienna.
References
HahnA, et al.Reconfiguration of functional brain networks and metabolic cost converge during task performance. eLife2020;
9: e52443.RischkaL, et al.Reduced task durations in functional PET imaging with [18F]FDG approaching that of functional MRI.NeuroImage2018;
181: 323–330.RedondoRL, et al.Making memories last: the synaptic tagging and capture hypothesis. Nat Rev Neurosci2011;
12: 17–30.HarrisJJ, et al.Synaptic energy use and supply. Neuron2012;
75: 762–777.FeldmanH, et al.Attention, uncertainty, and free-energy. Front Human Neurosci2010;
4: 215
2021-73
Harmonization of neuroimages acquired on two brain PET scanners (#296)
Jocelyn Hoye, Yasmin Zakiniaeiz, Takuya Toyonaga, Michelle Hampson and Evan D. Morris
Yale University, Radiology and Biomedical Imaging, New Haven Connecticut, USA
Abstract
Introduction: Two separate imaging studies were carried out with [11C]Raclopride to study striatal dopamine release in humans. Each human cohort was scanned with a different PET scanner. The purpose here was to develop an inter-scanner harmonization technique to allow for cohort comparison.
Methods: To develop the harmonization technique (experiment 1), the Iida brain phantom was filled with F-18 solution and scanned on two scanners (HRRT,HR+, Siemens/CTI). The HRRT images were filtered with isotropic gaussian filters of varying FWHM (0.5 mm-10 mm). Filtered HRRT images were matched to HR+ images by calculating the squared sum of the voxel-wise error (SSE) across voxels (using whole-brain or a mid-brain region alone). The gaussian filter that minimized SSE was deemed the optimal filter. To evaluate the harmonization on real data (experiment 2), inter-scanner variability was calculated using [11C]Raclopride scans of 2 human subjects on both the HRRT and HR+. Binding potential (BPND) in striatum was calculated for HR+ and HRRT (with various filters). Percent difference (PD) in striatal BPND was calculated between HR+ and HRRT (with various filters). Finally (experiment 3), PD was calculated for intra-scanner test-retest of striatal BPND (T/RT) for 8 human subjects scanned only on the HR+.
Results: Experiment 1 resulted in optimal whole-brain and mid-brain filters of 4.5 mm and 3.5 mm, respectively (Figure 1). Experiment 2 resulted in 15.89% PD for unfiltered HRRT, 9.25% for HRRT filtered with 3.5 mm gaussian, and 4.45% for HRRT filtered with 4.5 mm (Figure 2). Experiment 3 yielded 5.24% intra-scanner T/RT for HR+.
Conclusion: The mean inter-scanner difference for the HRRT with 4.5 mm filter was less than the mean intra-scanner T/RT for the HR+. The results indicate that BPND estimates from the HRRT can be combined with those from the HR+, with minimal addition to the overall variance, by using optimal gaussian smoothing of HRRT images.
2021-74
PET simulations investigating the effect of off-target binding in [F-18]MK-6240 PET scans for the detection of early-stage Alzheimer’s disease (#298)
Andrew McVea1, 2, Alexandra DiFilippo1, 2, Max McLachlan1, 2, Yangchun Xin2, Sterling Johnson3, Tobey Betthauser3 and Bradley Christian1, 2
1University of Wisconsin – Madison, Medical Physics, Madison Wisconsin, USA
2Waisman Center, Madison Wisconsin, USA
3Wisconsin Alzheimer’s Disease Research Center, Madison Wisconsin, USA
Abstract
Introduction: PET imaging of AD-related tauopathies has made substantial contributions towards the understanding of spatial and temporal changes of this pathological feature. [F-18]MK-6240 is a next-generation radiotracer with favorable imaging characteristics for detecting early, pre-symptomatic accumulation of tau in the entorhinal cortex (ERC). However, off-target (OT) binding in the meninges and sinus may confound the ability to accurately quantify specific MK-6240 binding due to “spill-in" signal to the reference and target regions of the brain. The goal of this work is to simulate regional uptake of MK-6240 signal to examine the effects of OT binding on PET outcome measures.
Methods: Simulations were generated with PET-SORTEO (Reilhac, 2004), which uses Monte-Carlo methods to generate PET images based on input time-activity curves (TACs) for designated brain regions. Ten realizations were created for each set of four 5-minute frame (70–90 minutes) scans simulating the ECAT HR+ scanner (i.d. = 10mCi). The input TACs were derived from human MK-6240 scans at our center and used to generate emission maps based upon the Harvard-Oxford atlas with custom regions for early-stage elevated tau in the ERC. Simulation cohorts included: no meninges or sinus uptake (OT_0), low MK-6240 signal in both (OT_L) and high MK-6240 signal in both (OT_H) with separate tau(+) and tau(-) scans for each. The SUVR outcome measure used the inferior cerebellar grey matter reference region and anterior parahippocampal gyrus to represent the ERC.
Results: For OT_0 the average ERC SUVR was 0.94 ± 0.02 and 1.27 ± 0.02 for tau(-/+) groups, respectively. The ERC SUVR was 0.92 ± 0.02 and 0.93 ± 0.01 for tau(-) in OT_L and OT_H, respectively, and 1.23 ± 0.03 and 1.20 ± 0.02 for tau(+). In the tau(+) images compared to OT_0 there was an average cerebellum SUV increase of 11.47 ± 0.68% and 17.52 ± 0.93% for OT_L and OT_H and an ERC SUV increase of 10.54 ± 3.51% and 15.02 ± 2.33%.
Conclusion: A significant elevated spill-in signal from off-target uptake has been simulated for the reference and early Braak regions and may offset each other for the SUVR outcome. Investigation is ongoing to examine a greater variety of OT uptake patterns and the performance of partial volume correction methods for calculating accurate outcome measures.
2021-75
Visual memory scores are associated with lateralization of tau in the medial temporal lobe (#299)
Jaime Fernández Arias1, 2, Yi-Ting Wang1, 2, Firoza Z. Lussier1, 2, Cécile Tissot1, 2, Joseph Therriault1, 2, Tharick A. Pascoal3, 2, Mira Chamoun2, Min-Su Kang1, 2, Gleb Bezgin2, Stijn Servaes2, Nina M. Poltronetti2, Jenna Stevenson2, Nesrine Rahmouni2, Pedro Rosa-Neto1, 2 and Serge Gauthier1, 2
2Translational Neuroimaging Laboratory, McGill University Research Centre for Studies in Aging, Verdun, QC, Canada
3University of Pittsburgh, Psychiatry and Neurology, Pittsburgh Pennsylvania, USA
Abstract
Introduction: Aggie Figures Learning Test (AFLT) is a visual memory test that was conceived as an analogue of the widespread Rey Auditory Verbal Learning Test (RAVLT), which tests verbal memory. Previous research has indicated that performance may rely on the left medial temporal lobe (lMTL) for verbal memory and on the right medial temporal lobe (rMTL) for nonverbal memory, although evidence is inconclusive. The present study looks into the association between delayed recall (DR) scores in both tests and tau binding in the brain.
Methods: Tau PET ([18F]-MK6240) was acquired for 130 individuals for analysis involving AFLT, and 139 for analysis involving RAVLT. Demographic data is shown on Tables 1 and 2. MRI were segmented into probabilistic grey (GM) and white (WM) maps, non-linearly registered to the ADNI template using Dartel and smoothed with an 8 mm FWHM gaussian kernel. Voxel-wise linear regression models were applied, using VoxelStats, with either DR AFLT or DR RAVLT as dependent variables and tau as a predictor. We corrected for age, sex, diagnosis, apoe genotype, years of education, difference between scan date and neuropsychological assessment date, and amyloid load.
Results: We found negative associations between tau binding and DR RAVLT in the MTL bilaterally. Interestingly, we found a negative correlation between tau binding and DR AFLT scores that is remarkable in the rMTL and disappears in the lMTL
Conclusion: Our findings provide evidence in support of the lateralization of memory in the brain based on the learning modality. This dissociation is in line with previous findings; namely, that the rMTL is responsible for visual memory. However, these findings are not entirely in line with previous research since associations with RAVLT are found bilaterally. Of note, most of the previous research has been made in the context of epilepsy and/or patients with anterior MTL lesions. Here, however, we are investigating a totally different population and we are looking at a parameter that is specific to this population.
Figure 1. Demographic data. Demographics for the AFLT subsample (above) and for the RAVLT subsample (below)
References
GleissnerUHelmstaedterC andElgerCE.
Right hippocampal contribution to visual memory: a presurgical and postsurgical study in patients with temporal lobe epilepsy. J Neurol NeurosurgPsychiatr1998;
65: ▪. https://doi.org/10.1136/jnnp.65.5.665.MilnerB.
Disorders of learning and memory after temporal lobe lesions in man. Neurosurgery1972;
19(CN_suppl_1). https://doi.org/10.1093/neurosurgery/19.CN_suppl_1.421RedobladoMAGraysonSJ andMillerLA.
Lateralized-temporal-lobe-lesion effects on learning and memory: examining the contributions of stimulus novelty and presentation mode. J Clin Exp Neuropsychol2003;
25.FosterPSDragoVHarrisonDW.
Assessment of nonverbal learning and memory using the design learning test. J Psychol2009;
143. https://doi.org/10.3200/JRLP.143.3.245–266.MaassALockhartSNHarrisonTM, et al.Entorhinal tau pathology, episodic memory decline, and neurodegeneration in aging. J Neurosci2018;
38. https://doi.org/10.1523/JNEUROSCI.2028-17.2017
2021-76
Investigating tauopathy in military occupational blast: a [18F]flortaucipir positron emission tomography study in Canadian armed forces members (#302)
Shamantha J. Lora2, 3, Sarah E. Watling3, Jerry Warsh2, Oshin Vartananian5, Neil Vasdev2, 7, Iain Vergie6, Isabelle Vallee6, Tina McCluskey2, Carmela Tartaglia4, Shawn Rhind5 and Isabelle Boileau1, 2
1University of Toronto, Psychiatry, Toronto, ON, Canada
2Centre for Addiction and Mental Health, Campbell Mental Health Research Institute, Toronto, ON, Canada
3University of Toronto, Institute of Medical Sciences, Toronto, ON, Canada
4University of Toronto, Tanz Centre for Research in Neurodegenerative Diseases, Toronto, ON, Canada
5Toronto Research Centre, Defence Research and Development Canada, Toronto, ON, Canada
6Canadian Armed Forces, Canadian Special Operation Forces Command, Ottawa, ON, Canada
7Centre for Addiction and Mental Health, Azrieli Centre for Neuro-Radiochemistry, Toronto, ON, Canada
Abstract
Introduction: Chronic traumatic encephalopathy (CTE), a tauopathy, is suspected to occur as a result of repetitive exposure to low-intensity blast overpressure during military training and operations. Although animal models putatively link explosive blast exposure with tau aggregation, studies in humans exposed to repeated low level blasts are limited. We used positron emission tomography (PET) imaging of [18F]flortaucipir (a.k.a [18F]TAUVID™, [18F]AV-1451; [18F]T807) to evaluate tau levels in Canadian Armed Forces (CAF) operators with significant career exposure to blast.
Methods: CAF members (5 males; 44.6 ± 6.2 years old) exposed to blast completed an MRI and a PET [18F]flortaucipir scan. Standardized Uptake Value ratio (SUVr) were calculated with the cerebellum as reference tissue. Participants performed a test of executive function (Stroop) and completed mood and clinical questionnaires.
Results: [18F]flortaucipir uptake was highest in midbrain/substantia nigra (SUVr range: 0.8 to 1.6), basal ganglia (SUVr range: 1.0 to 1.3), temporal (SUVr range: 0.8 to 1.13) and frontal cortices (SUVr range: 0.8 to 1.11). [18F]flortaucipir SUVr values were positively correlated with years of exposure to explosives and to years of breaching (r = 0.9; p-uncorrected < 0.05). Greater [18F]flortaucipir SUVr values in prefrontal regions were related to poorer performance on the Stroop test of executive function (r = 0.9; p < 0.05). [18F]flortaucipir SUVr did not correlate with mood or clinical symptoms.
Conclusion: Thesepreliminary results suggest that while [18F]flortaucipir uptake in CAF operators regularly exposed to low-intensity blast is within the range reported for normal mid-aged controls,1 our study found that greater tau deposition is associated with years of exposure to blast, as noted previously.2 Studies in a larger cohort should aim to model tau in order to understand what constitutes safe exposure to blast.
Acknowledgements
This work was funded by the Canadian Institute for Military and Veteran Health Research (CIMVHR) and the Canadian Department of National Defence (DND).
References
MaassALandauSBakerSL, et al.
Comparison of multiple tau-PET measures as biomarkers in aging and Alzheimer’s disease. Neuroimage2017;
157: 448–463.RobinsonMEMcKeeACSalatDH, et al.Positron emission tomography of tau in Iraq and Afghanistan Veterans with blast neurotrauma. NeuroImage2019;
21: 101651.
2021-77
In vivo imaging of microglial activation by positron emission tomography with 11C-ER176 in a mouse model for post-infectious autoimmune encephalitis (#304)
Zeljko Tomljanovic1, 2, Aubrey Johnson1, 2, Charlotte Wayne2, Tyler Cutforth2, Maryann Platt2, Mikhail Doubrovin3, Dritan Agalliu1, 4 and William C. Kreisl1, 2
1Columbia University, The Taub Institute for Research on Alzheimer’s Disease and the Aging Brain, New York New York, USA
2Columbia University, Department of Neurology, New York New York, USA
3Columbia University, Department of Radiology, New York New York, USA
4Columbia Universit, Department of Pathology and Cell Biology, New York New York, USA
Abstract
Introduction: Group A Streptococcus (GAS) infections are associated with basal ganglia encephalitis (BGE), a syndrome producing movement and psychiatric sequelae. Human and animal studies have shown that aberrant cellular and humoral adaptive immune responses target the brain and elicit neurovascular damage. 11C-ER176 is a third generation PET radioligand for the 18kDa translocator protein (TSPO) – a marker of activated microglia. We sought to evaluate the ability of 11C-ER176 to detect differences in TSPO binding in an animal model for post-infectious BGE as an independent method to assess the degree of neuroinflammation.
Methods: Twenty-three four-week-old C57BL/6 female mice were infected intranasally weekly with either GAS-2W streptococci (treated) or PBS (control) for five weeks. Twenty-four hours after the last inoculation, mice were injected intravenously with 11C-ER176 (110.11 ± 24.67 uCi) and PET images were acquired 0–60 minutes post-injection. Images were co-registered to CT and normalized using PMOD 4.2. Standardized uptake values ratios (SUVR) were calculated with rostral thalamus (rThal) as a “pseudo-reference” region. Statistical analysis was performed using frames from 30–60 min post-injection. Microglia activation was assessed in serial sections from the olfactory bulb (OB) in the subset of treated and untreated mice by immunofluorescence (IF) for Iba1.
Results: Treated mice (n = 12; 53.83 ± 4.55 days; weight 15.42 ± 2.61g) showed greater binding of 11C-ER176 than control mice (n = 11; 52.18 ± 2.56 days; weight 17.82 ± 0.98g) in the OB (11.60%, p = 0.0274), and amygdala (16.33%, p = 0.0002) consistent with previous studies in this mouse model. IF image analysis of activated (Iba1+) microglia in OB revealed higher mean number of cells in treated (n = 4; 6.57 ± 1.05) than control (n = 4; 4.25 ± 0.79, p = 0.0137) mice. However, microglia activation by IF did not correlate with 11C-ER176 SUVRs in any CNS region tested in the study.
Conclusion: GAS-treated C57BL/6 mice show increased binding of 11C-ER176 in olfactory bulb and amygdala, demonstrating that TSPO can serve as an inflammatory biomarker in post-infectious BGE. Lack of correlation between PET and IF results may be due to small sample size; alternatively, there may not be a one-to-one relationship between the number of activated microglia and the density of TSPO in this model.
References
DileepanTSmithEDKnowlandD, et al.Group A Streptococcus intranasal infection promotes CNS infiltration by streptococcal-specific Th17 cells.J Clin Investig2016;
126: 303–317.PlattMPBoldingKAWayneCR, et al.Th17 lymphocytes drive vascular and neuronal deficits in a mouse model of postinfectious autoimmune encephalitis.Proc Natl Acad Sci2020;
117: 6708–6716.JamesMLBelichenkoNPShuhendlerAJ, et al.[18F] GE-180 PET detects reduced microglia activation after LM11A-31 therapy in a mouse model of Alzheimer’s disease. Theranostics2017;
7: 1422.
2021-78
Development of a novel carbon-11 radiotracer for brain PET imaging of sulfonylurea receptors 1 (#305)
Fabien Caillé, Wadad Saba, Sébastien Goutal, Louise Breuil, Maud Goislard, Bertrand Kuhnast and Nicolas Tournier
Université Paris-Saclay, CEA/CNRS/INSERM, Orsay, France
Abstract
Introduction: Sulfonylurea receptors 1 (SUR-1, ABCC8) are almost absent of the healthy brain but overexpressed in neurons and glial cells under cerebral ischemic or injury conditions and are therefore a potential biomarker for brain trauma PET imaging. Kharade et al. described VU0071063, a ligand with affinity and selectivity for SUR-1, able to cross the BBB.1 Herein we report the isotopic labeling of VU0071063 and the in vivo PET imaging in the healthy rat brain.
Methods: The labeling precursor of VU0071063 was synthesized in one step. Radiomethylation with [11C]CH3I was realized in dimethylformamide with potassium carbonate at 100°C for 3 minutes using a TRACERlab® FX CPro module and the compound was purified by semi-preparative HPLC. Quality control including identity, radiochemical purity and molar activity was realized by analytical HPLC. Healthy male Wistar rats (365 ± 12 g, n = 4) were injected with [11C]VU0071063 (36 ± 5 MBq) in the tail vein and brain PET images were recorded for 60 minutes using a µPET Inveon®. Images were analyzed using the Pmod software to derive the time-activity curves expressed in % ID/cc over time in selected brain regions.
Results: The precursor was synthesized in one step from 3-methylxanthine in quantitative yield. Automated radiosynthesis of the precursor was realized within 35 minutes to afford ready-to-inject [11C]VU0071063 in 12 ± 2% (n = 5) decay-corrected radiochemical yield. [11C]VU0071063 was obtained in > 99% radiochemical purity with a molar activity of 85 ± 10 GBq/µmol. High and homogenous brain uptake was observed at early time points (0.95 ± 0.13% ID/cc from 0–10 min.) suggesting that [11C]VU0071063 readily crosses the BBB. A rapid elimination of [11C]VU0071063 was then observed with a low uptake at later time points (0.26 ± 0.03% ID/cc from 30–60 min.), consistent with the low expression of SUR-1 in the healthy brain.
Conclusion: A new radioligand targeting SUR-1 in the brain was designed and successfully radiolabeled with carbon-11. Preliminary study of this tracer in the healthy rat brain prompt us to pursue further investigation in animal models of brain injury with increased expression of SUR-1.
Acknowledgements
The authors thank the Atomic Energy Commission (CEA) for financial support.
Reference
KharadeSVSanchez-AndresJVFultonMG, et al.Structure-activity relationships, pharmacokinetics, and pharmacodynamics of the Kir6.2/SUR1-specific channel opener VU0071063s.J Pharmacol Exp Ther2019;
370: 350–359.
2021-79
Imaging the endogenous opioid response to acute cannabis smoking in humans (#307)
Kelly Smart1, Jon Mikael Anderson2, Patrick D. Skosnik2, David Matuskey1, Jim Ropchan1, Zachary Felchner1, Nabeel Nabulsi1, Yiyun H. Huang1, Kelly P. Cosgrove2 and Ansel T. Hillmer1, 2
1Yale School of Medicine, Yale PET Center, New Haven Connecticut, USA
2Yale School of Medicine, Department of Psychiatry, New Haven Connecticut, USA
Abstract
Introduction: The major psychoactive compound in cannabis, Δ9-tetrahydrocannabinol (THC), acutely increases brain levels of endogenous in preclinical models, which may contribute to both analgesic and reinforcing effects. To translate these findings to human studies, our objective was to assess if endogenous opioid release is detected in healthy volunteers following cannabis smoking using positron emission tomography (PET) with the selective mu opioid receptor (MOR) agonist radioligand [11C]carfentanil.
Methods: Eight healthy volunteers (29.3 ± 8.8 years, 4M/4F) with limited past recreational cannabis use completed PET scans with bolus injections of [11C]carfentanil at baseline (529 ± 177 MBq) and 16 ± 8 minutes after smoking cannabis (524 ± 140 MBq; 5.6 ± 0.57% THC, 0.01 ± 0.00% cannabidiol). Ratings of subjective drug response were collected, and the Cold Pressor Task was performed to assess pain response pre- and post-cannabis.
[11C]Carfentanil binding potential (BPND) was determined by the simplified reference tissue model in frontal cortex, anterior cingulate, insula, caudate, putamen, amygdala, and thalamus. Global BPND was determined as the average value across these regions. Change in BPND between baseline and post-cannabis scans was assessed as an index of endogenous opioid release. Secondary analyses explored correlations between baseline BPND, drug response, and pain tolerance.
Results: The post-cannabis change in [11C]carfentanil BPND ranged from -2.2 ± 6.9% in frontal cortex to +4.5 ± 9.7% in anterior cingulate. This change was not statistically significant in any region (ps > 0.11).
Baseline global [11C]carfentanil BPND was negatively correlated with ratings of ‘high’ post-cannabis (r = −0.79, p = 0.035), such that people with lower MOR availability at baseline had stronger subjective intoxication responses. This was driven by relationships in amygdala, caudate, thalamus, and insula (ps < 0.1). Baseline global BPND was positively correlated at trend-level with pain tolerance at baseline (p = 0.08) and post-cannabis (p = 0.07).
Conclusion: In healthy volunteers with low levels of recreational cannabis use, we did not find evidence of endogenous opioid release following smoked cannabis using [11C]carfentanil PET. The relatively low THC content may have limited the ability to detect such effects. Baseline MOR availability in limbic brain regions, reflecting receptor expression or endogenous endorphin levels, may be linked to subjective cannabis response and pain tolerance.
2021-80
Comparative analysis of immunohistochemical and autoradiographic images of Tau in postmortem human Alzheimer’s disease brain (#308)
Rommani Mondal, Reisha M. Ladwa, Christopher Liang and Jogeshwar Mukherjee
University of California, Irvine, Preclinical Imaging, Radiological Sciences, Irvine California, USA
Abstract
Introduction: High resolution scans of immunohistochemical (IHC) stains of Alzheimer’s disease (AD) brain slices and autoradiography binding assays both give us information about the distribution of Tau protein. Accurate assessment of the amount and regional location of Tau is essential to understand AD pathology.1 Our goal is to develop a quantitative method for analysis of coregistered IHC-autoradiography images. Here we report our results of AD hippocampus (HP) labeled with [125I]IPPI and total Tau IHC.
Methods: Postmortem human HP from AD subjects were IHC stained for total Tau and [125I]IPPI autoradiography for Tau as reported.2 QuPath and a pixel thresholder was created to outline the IHC image.3 Adobe Photoshop was used to spatially align and crop the IHC and autoradiography images. A heatmap was generated in QuPath using SLIC superpixel method.3 The co-registered autoradiography and QuPath heatmap image were loaded into MATLAB, converted to grayscale intensity images with a range of 0–255 (Fig ure 1). Intensity quantification in both images was indicative of the presence of higher regional Tau.
Results: Figure 2(a) and (b) shows a detailed profile of [125I]IPPI binding in the cortical fold showing four distinct peaks (1 to 4). Similarly, four peaks were observed in Tau QuPath in Fig 2(c) and (d). Peaks 1 and 3 were consistent in magnitude (>80%) and location, but intensity of peaks 2 and 4 were higher in QuPath. Intensity of Tau and IPPI binding on the slice using AutoPath were compared with the 2-D correlation coefficient matrix, image correlation coefficient (0.7317), ratio of peak locations (1.0006), and a visual graph of the intensity profiles. The variability of intensity between the two modalities may be due to the limited pixels measured by the line in Figure 2(a) and (c) and potential differences between total Tau measured in QuPath versus phosphorylated Tau (pTau) measured by [125I]IPPI.4
Conclusion: Compared to semi-quantitative histopathological scoring, our novel method QuPathRad (QPR) has the potential for quantitative image analysis of IHC with autoradiography for Tau in AD. Since various pTau are present in the AD brain HP,5 efforts are underway to identify aggregate pTau as well as size differentiation of Tau using the QPR technique.
Acknowledgements
NIH/NIA RF1 AG029479 (JM), UCI UROP (RM, RML). Banner Health Research Institute for brain sections of hippocampus and UCI Pathology for immunostaining.
References
DunnWDGearingMParkY, et al.Applicability of digital analysis and imaging technology in neuropathology assessment. Neuropathology2016;
36: 270–282.MondalRLadwaRMLiangL, et al. [125I]IPPI: A novel Tau radioligand for assessing neurofibrillary tangles in human hippocampus of postmortem Alzheimer’s disease subjects. In: Annual meeting of Society of Neuroscience, Chicago, Illinois, 13–17 November 2021, Abstract # 7939, Session # P941.BankheadPLoughreyMBFernándezJA, et al.
QuPath: open source software for digital pathology image analysis. Sci Rep2017;
7: 16878.SjorenMDavidssonPTullbergM, et al.Both total and phosphorylated tau are increased in Alzheimer’s disease. J Neurol Neurosurg Psychiatr2001;
70; 624–630.CherryJDEsnaultCDBaucomZH, et al.Tau isoforms are differently expressed across the hippocampus in chronic traumatic encephalopathy and Alzheimer’s disease. Acta Neuropathol Commun2021;
9: 86.
2021-81
Evaluation in rat, monkey, and humans of [18F]PF-06445974, a PET radioligand for phosphodiesterase 4B (#309)
Yuichi Wakabayashi1, Per Stenkrona3, Jeih-San Liow1, Kevin P. Maresca2, Thomas A. Chappie2, Lei Zhang2, Zoë A. Hughes2, Christopher J. Schmidt2, Shawn D. Doran2, Ryosuke Arakawa3, Andrea Varrone3, Akihiro Takano3, Paolo Z. Fregonara1, Sami S. Zoghbi1, Maarten Ooms1, Cheryl Morse1, Victor W. Pike1, Christer Halldin3 and Robert B. Innis1
1NIH – Bethesda, MD, Molecular Imaging Branch, Bethesda Maryland, USA
2Pfizer Inc, Worldwide Research, Development & Medicine, New York New York, USA
3Karolinska Institute and Stockholm County Council, Department of Clinical Neuroscience, Centre for Psychiatry Research, Stockholm, Sweden
Abstract
Introduction: Phosphodiesterase-4 (PDE4), which metabolizes the second messenger cyclic adenosine monophosphate (cAMP), has four isozymes: A, B, C, and D. PDE4B and PDE4D have the highest expression in brain and are associated with neuropsychiatric diseases such as major depressive disorder. This study evaluated the properties of the newly developed PDE4B-selective radioligand [18F]PF-06445974 in the brain of rats, rhesus monkeys, and humans
Methods: Three monkeys from two different institutions (NIMH and Karolinska Institute) were scanned with 180-min dynamic PET images and metabolite-corrected arterial input function. Five healthy volunteers from both institutions were scanned with 120-min dynamic PET and input function. [18F]PF-06445974 binding was quantified as total distribution volume (VT) using a two-tissue compartment model. Radiochromatograms were performed in rat brain tissue 180 minutes after injection to assess the presence of radiometabolites.
Results: [18F]PF-06445974 readily distributes in the monkey brain, the highest binding being in thalamus. VT was stable over time (<10% difference from 90 to 180 minutes). Similarly, [18F]PF-06445974 readily entered the human brain (Figure 1) and its VT matched the distribution of mRNA expression derived from the Allen Brain Atlas. However, on average, VT (value over the whole brain 9.5 ± 2.4 mL/cm−3) showed linear increase until the end of the scans (Figure 2). Radiochromatograms of rat brains found that 98.7% of the activity was due to the parent compound.
Conclusion: [18F]PF-06445974 successfully imaged PDE4B in monkey and human brain. VT was stable over time in monkeys, but it linearly increased after 120 minutes in humans. Despite the lack of radiometabolites in the rat brain, they may accumulate in the human brain. This radioligand will be further assessed to understand whether the lack of time stability in humans is due to the presence of radiometabolites or to a slow equilibration time in the human brain.
Acknowledgements
Laboratories of Dr. Pike and Dr. Innis
NIMH veterinary staffs
Karolinska Institute scientists
2021-82
Perfusion heterogeneity in extra-tumoural brain in grades III and IV glioma measured using positron emission tomography with radiolabelled water (#311)
Ana Jorge Gonçalves1, Moudi Alotaibi1, Alice R. Carlin1, Yaonan Chen1, Constance Y.W. Wong1, Ibrahim K. Djoukhadar2, Georgios Krokos1, Alan Jackson1, David Coope2 and Marie-Claude Asselin1
1The University of Manchester, Manchester, UK
2Salford Royal NHS Foundation Trust, Manchester, UK
Abstract
Introduction: The impact of the disrupted neo-vasculature in glioma to the extra-tumoural brain has clinical interest since radiotherapy and drug delivery rely on adequate perfusion and perfusion may be predictive of areas of infiltration and recurrence. The aim of this study is to assess heterogeneity of cerebral blood flow (CBF) with distance from the tumour with a novel image sampling method.
Methods: Twelve glioma (5 grade III, 7 grade IV) patients (3 females) aged 51.0 ± 12.8 years underwent dynamic [15O]H2O PET on the HRRT before treatment.1 Parametric maps of absolute CBF were generated using generalised linear least squares2 for the 1-tissue reversible compartment model with a sampled arterial blood input function. Tumour regions were manually delineated on MR images by an experienced neuroradiologist (ID). Extra-tumoural white matter (WM) and gray matter (GM) were segmented from the T1-weighted images and coregistered to the PET images using FSL. Concentric GM and WM shells growing away from the tumour edge were generated as explained in Figure 1. The asymmetry index (AI) was calculated to assess hemispherical CBF differences. The 95% confidence intervals of CBF and AI were determined from fifteen healthy volunteers (HV, 8 females) aged 43.5 ± 5.7 years.
Results: CBF was significantly different compared to HVs only in grade IV glioma (Figure 2(a)), being reduced in WM ipsilaterally by 18.5% (p = 0.01716) and contralaterally by 16.2% (p = 0.02128), but not in GM. In both grades, perfusion in GM and WM was not significantly different between ipsilateral and contralateral sides. Close to the tumour edge of several grade IV, CBF was very asymmetric (AI > 30%), either elevated or depleted relative to the contralateral side, and gradually became more symmetric moving away from the tumour (Figure 2(b)). This was most marked in WM but also observed in GM.
Conclusion: Gliomas can cause extra-tumoural alterations in CBF even on the normal-appearing contralateral hemisphere. The shells sampling method revealed sub-hemispheric heterogeneity, with asymmetry increasing towards the tumour edge. Future work aims to create similarly sized shell regions in the HVs to better estimate the normal variability and define the asymmetry patterns in the healthy brain.
Acknowledgements
The research leading to these results was financially supported by the Europeen Community’s Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 201380 and by the Cancer Research UK and EPSRC Cancer Imaging Centre in Cambridge and Manchester under grant agreement C8742/A18097. AJG is funded by the EU H2020 MSCA-ITN-2018: INtegratingMagnetic Resonance SPectroscopyand Multimodal Imaging for Research and Education in MEDicine (INSPiRE-MED), funded by the European Commission under Grant Agreement #813120.
References
WalkerMDFeldmannMMatthewsJC, et al.Optimization of methods for quantification of rCBF using high-resolution [15O]H2O PET images. Phys Med Biol2012;
57: 2251–2271.FengDWangZHuangSC.
A study on statistically reliable and computationally efficient algorithms for generating local cerebral blood flow parametric images with positron emission tomography. IEEE Trans Med Imaging1993;
12: 182–188.
2021-83
Expression of transduced estrogen receptor-α fragment fusion protein in rhesus monkey brain using [18F]FES as a PET reporter probe (#312)
Palak Wadhwa1, Bing Li1, Walter Lerchner1, Paolo Zanotti-Fregonara1, Jeih-San Liow1, Mark A. G. Eldridge1, Xuefeng Yan1, Michael Michaelides2, Sami S. Zoghbi1, Barry J. Richmond1 and Robert B. Innis1
1National Institutes of Health, Intramural Research Programs of NIMH, Bethesda Maryland, USA
2National Institutes of Health, Biobehavioral Imaging & Molecular Neuropsychopharmacology Unit, NIDA, Baltimore, Maryland, USA
Abstract
Introduction: Positron emission tomography (PET) reporter gene systems have contributed to progress in cell/gene therapy, and have improved our understanding of the molecular pathology of neurological diseases by providing an in vivo molecular imaging tool.1,2,3,4 PET reporter gene systems comprise of: (1) a PET reporter gene which is linked to a therapeutic gene of interest, packaged within a viral vector that can transduce the targeted neural cell population and (2) a complementary PET reporter probe that can image the expression of the reporter proteins within the transduced cells.2,3,4 This PET study is designed to assess whether [18F]FES can be used as a PET reporter probe to image an estrogen receptor-α (ER- α) fragment expressed from a fusion construct comprising Channelrhodopsin2 and the ER – α ligand binding domain (ChRERa) in the rhesus monkey brain.
Methods: Lentivirus expressing ChRERa under the control of a human synapsin promoter (Lenti-ChRERa) and Adeno-associated virus expressing ChRERa under the control of a human synapsin promoter (AAV-ChRERa) were injected intracerebrally within the right and left anterior caudate, respectively, of two rhesus monkeys.[18F]FES baseline PET scans were performed before and after the viral transduction in both rhesus monkeys, with measurement of the arterial input function. Specific binding was measured as the difference between the baseline scan and that blocked by estradiol (0.1 mg/kg). Volume of distribution (VT) was calculated using Logan graphical analysis.
Results: Prior to viral transduction, [18F]FES uptake in the brain was similar between baseline and estradiol-blocked scans, showing that ER are not significantly expressed in the normal brain. High (and blockable) uptake was seen only in the hypophysis (a region known to have high ER density5). Viral transduction induced a focal tracer uptake at the Lenti-ChRERa injection site (with an estimated increase of at least 70% in plasma-free fraction corrected distribution volume, VT/fp, compared to baseline), but not at the AAV-ChRERa injection site.
Conclusion: The rhesus monkey brain has no significant baseline expression of ER, thus [18F]FES is a promising tracer to be used as a PET reporter probe. Intracerebral injection of Lenti-ChRERa successfully induced a detectable focal uptake of [18F]FES.
References
YaghoubiSSCampbellDORaduCG, et al.Positron emission tomography reporter genes and reporter probes: gene and cell therapy applications. Theranostics2012;
2: 374–391.HaywoodTBeinatCGowrishankarG, et al.Positron emission tomography reporter gene strategy for use in the central nervous system. Proc Natl Acad Sci2019;
116: 11402–11407.SerganovaIBlasbergRG.
Molecular imaging with reporter genes: has its promise been delivered?J Nucl Med2019;
60: 1665–1681.MassoudTFSinghAGambhirSS.
Noninvasive molecular neuroimaging using reporter genes: part I, principles revisited. Am J Neurorad2008;
29: 229–234.Moraga-AmaroRvan WaardeADoorduinJ, et al.Sex steroid hormones and brain function: PET imaging as a tool for research. J Neuroendocrinol2018;
30: e12565.
2021-84
Quantification of cerebral nicotinic α7 acetylcholine receptors (α7 nAChRs) under gastric stimulation of the vagus nerve in piglets (#313)
Michael Rullmann1, Georg A. Becker1, Anton Antonov2, Bernhard Sattler1, Tatjana Sattler3, 4, Winnie Deuther-Conrad5, Stefan Schimpf6, Marianne Patt1, Philipp M. Meyer1, Rodrigo Teodoro5, Barbara Wenzel5, Matthias Scheunemann5, Swen Hesse1, Peter Brust5, Marco Leitzke2 and Osama Sabri1
1University of Leipzig, Department of Nuclear Medicine, Leipzig, Germany
2Helios Clinics, Department of Anaesthesiology, Leisnig, Germany
3University of Leipzig, Deparment of Claw Animals, Leipzig, Germany
4University of Leipzig, Albrecht-Daniel-Thaer-Institute of Agricultural Sciences, Leipzig, Germany
5Helmholtz-Zentrum Dresden-Rossendorf, Institute of Radiopharmaceutical Cancer Research, Department of Neuroradiopharmaceuticals, Research Site Leipzig, Leipzig, Germany
6Dräger Medical GmbH, Lübeck, Germany
Abstract
Introduction: Electrical gastric vagus nerve stimulation (VNS) shifts the sympathetic-vagal balance toward a parasympathetic predominance.1 We aim to assess central changes in α7 nAChR-mediated transmission and hypothesize that VNS changes the parasympathetic tone by changing the α7 nAChR availability in the nucleus tractus solitarii (NTS) and distinct cortical and subcortical regions.2
Methods: Following a standard 35-frames, 120-min protocol for dynamic brain PET imaging, data from eight piglets (15.6 ± 3.2 kg, ∼6 weeks) were acquired post injection of 194.8 ± 9.4 MBq of [18F]DBT-103 followed by T1-MPRAGE MRI: three baseline, two with infusion of the acetylcholine esterase inhibitor physostigmine (0.04 mg/kg, 1 ml/min, at 10 min prior to tracer injection; 0.24 mg/kg, 1 ml/min, at tracer injection/scan start over 120 min) and three with VNS in repeated sequences of 0.5 Hz over 5 min, 5 min pause started at scan start. TACs were analyzed using a 2-tissue compartment model involving a metabolite-corrected arterial input function to generate individual total distribution volumes (VT) as receptor parameter.
Results: Compared to baseline, we observed an increase of the mean VT after physostigmine infusion (61%) as well as after VNS (28%) without major alterations of K1 in the NTS (Figures 1 and 2).
Conclusion: These initial data indicate blood-flow-independent changes under VNS as compared with baseline suggesting an increase in α7 nAChR availability although the changes appear more heterogeneous in VNS as compared with physostigmine. The finding is in contrast to our hypothesis expecting lower α7 nAChR-availability as a result of increased acetylcholine release following VNS. We speculate that higher VT under VNS may reflect an increase in affinity of the α7 nAChR or result in an upregulation of the α7 nAChR. Increase of VT following physostigmine administration could be related to a positive allosteric effect on the α7 nAChR. Furthermore, the results of this study allow sample size estimations for further preclinical and clinical studies.
References
LeitzkeMSchimpfSAltrockM, et al.Afferent vagal stimulation via gastric electrical stimulation alters sympathetic-vagal balance in house pigs – a pilot trial. J Biol Regul Homeost Agents2021;
35:11–24.SherEChenYSharplesTJ, et al.Physiological roles of neuronal nicotinic receptors subtypes: new insights on the nicotinic modulation of neurotransmitter release, synaptic transmission and plasticity. Curr Top Med Chem2004;
4: 283–297.HillmerATZhengMQLiS, et al.PET imaging evaluation of [18F]DBT-10, a novel radioligand specific to α7 nicotinic acetylcholine receptors, in nonhuman primates. Eur J Nucl Med Mol Imaging2016;
43: 537–547.SaikaliSMeuricePSauleauP, et al.A three-dimensional digital segmented and deformable brain atlas of the domestic pig. J Neurosci Methods2010;
192: 102–109.
2021-85
Regional analysis demonstrates asymmetric binding with [11C]-(R)-PK11195 positron emission tomography in normal brain (#315)
Bandar Q. Alfaifi1, 2, Daniel Lewis3, 4, Alan Jackson1, 2, David Coope3, 4 and Rainer Hinz1, 2
1University of Manchester, Wolfson Molecular Imaging Centre, Manchester, UK
2University of Manchester, Division of Informatics, Imaging & Data Sciences, Manchester, UK
3University of Manchester, Division of Neuroscience and Experimental Psychology, Manchester, UK
4University of Manchester, Geoffrey Jefferson Brain Research Centre, Manchester, UK
Abstract
Introduction: An understanding of the normal translocator protein (TSPO) distribution in healthy brain is essential if physiological variation is to be distinguished from pathology associated changes in TSPO expression. In this study we investigated the distribution and expression of TSPO in human brains using [11C]-(R)-PK11195 positron emission tomography (PET), evaluating hemispheric asymmetry in TSPO expression in one of the largest TSPO PET datasets of healthy subjects.
Methods: A total of 50 healthy subjects (mean ± SD = 41 ± 15) underwent MR and dynamic [11C]-(R)-PK11195 PET scanning. [11C]-(R)-PK11195 PET scans were acquired for 60-min post-injection using a HRRT scanner.1 Summed [11C]-(R)-PK11195 PET images were created and used for PET/MRI co-registration. Brain Hammers atlas2 was used to permit quantification of [11C](R)-PK11195 BPND with atlas-defined brain regions. Fourteen regions of interest were included (Table 1). Parametric images of [11C]-(R]-PK11195 binding potential (BPND) were generated using the simplified reference tissue model and bilateral grey-matter cerebellar time-activity curve as an input function. Paired-t-test was performed to assess right/left asymmetry. At regional level, multiple comparison correction (Bonferroni-Dunn’s) was applied (P < 0.01 was considered significant).
Results: Significant asymmetry in mean [11C]-(R)-PK11195 BPND values was observed, with both total right hemisphere (GM+WM) and right hemispheric GM showing higher [11C]-(R)-PK11195 BPND (p < 0.001) values compared to the left side (Table 1). At the regional level, paired t-tests with Bonferroni multiple comparison correction showed that the mean [11C]-(R)-PK11195 BPND in right hemispheric regions were significantly higher (p < 0.001) than left regions in all sampled regions except for the occipital cortex.
Conclusion: [11C]-(R)-PK11195 BPND asymmetry was observed with right-brain regions showing higher BPND in all normal brain regions except of the occipital cortex. Only one reference tissue input function was considered here although similar asymmetry was found with alternative reference-tissue input functions. This identified asymmetry should be considered and accounted for when evaluating TSPO PET datasets in pathological conditions.
Acknowledgements
Bandar Alfaifi is a scholarship holder from AL JOUF University, Saudi Arabia.
References
SuZHerholzKGerhardA, et al.[11C]-(R) PK11195 tracer kinetics in the brain of glioma patients and a comparison of two referencing approaches. Eur J Nucl Med Mol Imaging2013;
40: 1406–1419.HammersAAllomRKoeppMJ, et al.Three‐dimensional maximum probability atlas of the human brain, with particular reference to the temporal lobe. Human Brain Mapping2003;
19: 224–247.
2021-86
Impact of cerebral blood flow and amyloid load on SUVR bias (#316)
Fiona Heeman1, Maqsood Yaqub1, Janine Hendriks1, Bart N.M. van Berckel1, Lyduine E. Collij1, Katherine R. Gray2, Richard Manber2, Robin Wolz2, Valentina Garibotto3, 4, Catriona Wimberley5, Craig Ritchie5, Frederik Barkhof1, 6, Juan Domingo Gispert López7, 8, David Vállez García1, Isadora Lopes Alves1 Adriaan A. Lammertsma1 and on behalf of the AMYPAD consortium
3Geneva University, NIMTLab, Faculty of Medicine, Geneva, Switzerland
4Geneva University Hospitals, Division of Nuclear Medicine and Molecular Imaging, Geneva, Switzerland
5University of Edinburgh, Edinburgh Imaging, Queen’s Medical Research Institute, Edinburgh, UK
6UCL, Institutes of Neurology and Healthcare Engineering, London, UK
7Barcelonaβeta Brain Research Centre, Pasqual Maragall Foundation, Barcelona, Spain
8Centro de Investigación Biomédica en Red de Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Madrid, Spain
9Universitat Pompeu Fabra, Department of Experimental and Health Sciences, Barcelona, Spain
10IMIM (Hospital del Mar Medical Research Institute), Barcelona, Spain
Abstract
Introduction: Despite the widespread use of static amyloid PET scans for clinical and research purposes, it is known that the standardized uptake value ratio (SUVR), calculated from a static scan, may be biased compared with the distribution volume ratio (DVR) derived from a dynamic scan.1 This bias may be partially explained by changes in cerebral blood flow (CBF) and is likely to be also dependent on the severity of the underlying amyloid-β (Aβ) pathology.2,3 To date, most Alzheimer’s disease (AD) studies have compared SUVR and DVR only at a diagnostic group level and not as a function of underlying Aβ pathology. The purpose of the present study was to compare SUVR with DVR and to evaluate the effects of underlying Aβ pathology and CBF on bias in SUVR in a group of mainly cognitively unimpaired participants.
Methods: Participants (N = 121) were scanned according to a dual-time window protocol,4 with either [18F]flutemetamol (N = 90) or [18F]florbetaben (N = 31). The validated voxel-based implementation of the two-step simplified reference tissue model was used to derive DVR and R1 and SUVR was calculated for a 90–110 min post-injection uptake window, all with the cerebellar grey matter as reference tissue.5 First, linear regression and Bland-Altman analyses were used to compare SUVR with DVR. Then, Generalized Linear Models were applied to evaluate whether (bias in) SUVR relative to DVR could be explained by R1 for four regions (i.e. global cortical average (GCA), precuneus, posterior cingulate, and orbitofrontal cortex).
Results: Parameter distributions for both tracers are depicted in Figure 1. Despite high correlations (GCA: R2≥0.85), large overestimation and proportional bias of SUVR relative to DVRwas observed (regression line of the proportional bias: GCA: [18F]flutemetamol R2 = 0.69, slope = 0.56 and intercept = −0.48; [18F]florbetaben: R2 = 0.65, slope = 0.31 and intercept = −0.20). Figure 2. Negative associations were observed between both SUVR or SUVRbias and R1, albeitnon-significant.
Conclusion: The present findings demonstrate that bias in SUVR relative to DVRis primarily due to underlying Aβ pathology. Furthermore, in a cohort consisting mainly of cognitively unimpaired individuals, the effect of CBF on bias in SUVR appears limited.
Acknowledgements
The authors would like to thank all staff of the various centres for skilful acquisition of the scans.
This work has received support from the EU-EFPIA Innovative Medicines Initiatives 2 Joint Undertaking (grant No 115952). This communication reflects the views of the authors and neither IMI nor the European Union and EFPIA are liable for any use that may be made of the information contained herein.
References
CarsonREChanningMABlasbergRG, et al.Comparison of bolus and infusion methods for receptor quantitation: application to [18F]Cyclofoxy and positron emission tomography. J Cerebral Blood Flow Metab1993;
13:24–42.Berckel vanBNMOssenkoppeleRTolboomN, et al.Longitudinal amyloid imaging using 11C-PiB: methodologic considerations. J Nucl Med2013;
54:1570–1157Lopes AlvesIHeemanFCollijLE, et al.Strategies to reduce sample sizes in Alzheimer’s disease primary and secondary prevention trials using longitudinal amyloid PET imaging. Alzheimers Res Ther2021;
13:82.HeemanFYaqubMLopes AlvesI, et al.Optimized dual-time-window protocols for quantitative [18F]flutemetamol and [18F]florbetaben PET studies. EJNMMI Res. 2019;
9:32.HeemanFYaqubMHendriksJ, et al.Parametric imaging of dual-time window [18F]flutemetamol and [18F]florbetaben studies. NeuroImage2021;
234:117953.
2021-87
[18F]Flotaza and [124I]IBETA, two new β-amyloid PET imaging agents in post-mortem human Alzheimer’s disease brain (#317)
An N. Nguyen, Christopher Liang, Harsimran Kaur and Jogeshwar Mukherjee
University of California, Irvine, Radiological Sciences, Irvine, USA
Abstract
Introduction: PET studies of amyloid β (Aβ) accumulation in Alzheimer’s disease (AD) have shown clinical utility.1 Several fluorine-18 labeled PET radiotracers are being clinically used. The aim of this study was to develop and evaluate in vitro effectiveness of new fluorine-18 radiotracer, [18F]Flotaza2 and long half-life iodine-124 radiotracer [124I]IBETA3 (Figure 1), for Aβ plaque imaging in postmortem human AD brain slices.
Methods: Molecular docking was carried out using Autodock-Chimera on amyloid fibrils.4 Nucleophilic fluorine-18 was used to label tosylateprecursor.1 [124I]Sodium Iodide was used to prepare [124I]IBETA by electrophilic substitution of the tributyltin derivative.5 Human AD (n = 6) post-mortem brain tissues consisting of anterior cingulate (AC) and corpus callosum (CC) were used. Brain slices (10 µm thick) were treated with [18F]Flotaza or [124I]IBETA in 40% ethanol PBS buffer pH 7.4 at 25°C for 1.25 hr. The slices were then washed with cold PBS buffer, 50% ethanolic PBS buffer twice (90% ethanolic PBS buffer twice for [124I]IBETA), PBS buffer and cold water. The brain sections were air-dried, exposed overnight on phosphor films and the extent of binding of [18F]Flotaza and [124I]IBETA was measured in DLU/mm2.
Results: Both [18F]Flotaza and [124I]IBETA exhibit nanomolar affinities for Aβ plaques. Preferred binding sites of [18F]Flotaza and [124I]IBETA on Aβ fibril were similar (Figure 1). For Aβ plaques, slices from all subjects were positively immunostained with anti-Aβ (Figure 2). High ratios (>100) of [18F]Flotaza in AC to CC was observed (Figure 2(b)) in all the 6 subjects while ratios with [124I]IBETA (Figure 2(c)) were lower (<10). Very little white matter binding was seen when alcohol was used for washing. [18F]Flotaza and [124I]IBETA binding in the AC strongly correlated with anti-Aβ immunostains (Figure 2(d)). In preliminary PET/CT studies, [18F]Flotaza was found to be stable in vivo in transgenic 5XFAD mice models and no bone uptake confirmed lack of defluorination, while some deiodination of [124I]IBETA was observed.
Conclusion: [18F]Flotaza and [124I]IBETA are new β-amyloid PET imaging agents which can be made in one-step radiosynthesis followed by HPLC purification. Both radiotracers exhibited high binding to Aβ plaques in postmortem human AD brains. Higher AC/CC ratios of [18F]Flotaza compared to [124I]IBETA may be attributed to PEG functionality.
Acknowledgements
NIH/NIA RF1 AG029479 (JM), UCI UROP (AN, HK)
References
UzuegbunamBCLibrizziDYousefiBH.
PET radiopharmaceutical for Alzheimer’s disease and Parkinson’s disease diagnosis, the current and future landscape. Molecules2020;
25: 977.KaurHFelixMRLiangC, et al.Development and evaluation [18F]Flotaza for Ab plaque imaging in post-mortem Alzheimer’s disease brain. Bioorg Med Chem Lett2021;
46: 128164.OnoMChengYKimuraH, et al.Development of novel 123I-labeled pyridyl benzofuran derivatives for SPECT imaging of b-amyloid plaques in Alzheimer’s disease. PloS One2021;
8: e74104.GremerLScholzelDSchenkC, et al.
Fibril structure of amyloid-beta (1–42) by cryo-electron microscopy. Science2017;
358: 116–119.PandeySKVenugopalAKantR, et al.124I-Epidepride: a high affinity and selective PET radiotracer with potential for extended imaging of dopamine D2/D3 receptors. Nucl Med Biol2014;
41: 426–431.
2021-88
Multi-tracer PET joint correlation analysis reveals disease-specific patterns in both Parkinson’s disease and asymptomatic LRRK2 mutation carriers compared to healthy controls (#318)
Julia G. Mannheim1, 2, Jessie Fanglu Fu1, 3, Tilman Wegener1, 4, Ivan S. Klyuzhin5, A. Jon Stoessl5, 6 and Vesna Sossi1, 6
1University of British Columbia, Department of Physics and Astronomy, Vancouver British Columbia, Canada
2Eberhard-Karls University Tuebingen, Department of Preclinical Imaging and Radiopharmacy, Tuebingen Baden-Württemberg, Germany
3Massachusetts General Hospital, Athinoula A. Martinos Center for Biomedical Research, Charlestown Massachusetts, USA
4University of Luebeck, Department of Medical Engineering, Luebeck Schleswig-Holstein, Germany
5University of British Columbia, Division of Neurology, Department of Medicine, Vancouver British Columbia, Canada
6University of British Columbia & Vancouver Coastal Health, Djavad Mowafaghian Centre for Brain Health, Pacific Parkinson’s Research Centre, Vancouver British Columbia, Canada
Abstract
Introduction: Parkinson’s disease (PD) is a neurodegenerative disorder associated with dopaminergic terminal loss.1 We previously demonstrated the benefit of a joint pattern approach based on canonical correlation analysis (CCA) to extract PD-progression related spatial patterns.2 This study extends the analysis to extract PD- and leucine-rich repeat kinase 2 (LRRK2) mutation-specific dopaminergic patterns in relation to healthy controls (HCs).
Methods: CCA was applied to: (i)sporadic PD (n = 40, disease duration:7.5 ± 5.9y), (ii)HC (n = 27) and (iii)asymptomatic LRRK2 mutation carriers (UC, n = 11) scanned with [11C]dihydrotetrabenazine (DTBZ, VMAT2 marker) and [11C]d-threo-methylphenidate (MP, DAT marker). DTBZ and MP binding potentials (BPND) were calculated for the caudate and putaminal (anterior-middle-posterior) regions on the less/more affected side for PD, and dominant/non-dominant side for HCs and UCs. CCA was applied to HCs pooled with UCs and to HCs pooled with UCs and PDs.
Results: CCA of pooled HCs and UCs revealed disease-associated patterns (component 1: overall denervation & anterior-posterior gradient, component 2: asymmetry, component 3: residual gradient for MP, Figure 1(a)) similar to PD-progression related patterns as previously described.2 Subject scores for component 1 separated HCs from UCs for DTBZ and MP (Figure 2(a), p < 0.01). A significant age dependency was detected for both HC and UC subject scores along component 1 for MP (p < 0.05). Pooling all three cohorts revealed similar patterns (Figure 1(b)) with significantly different subject scores between the three cohorts along component 1 (Figure 2(b), p < 0.01). Interestingly, subject scores of UCs while consistent with HCs of younger age approached those of the PD subjects with increasing age, likely indicating preclinical disease.
Conclusion: CCA extracted patterns of topological dopaminergic denervation. Spatial patterns identified for UC revealed similarities compared to PD patterns, suggesting topological denervation similarity between preclinical and clinical disease stage. CCA-derived component 1 of the three-group analysis revealed a stronger separation of UCs and HCs compared to univariate analysis (data not shown) along with a correlation with disease risk/age for LRRK2 and HC, i.e. as age increases, the subject scores for HCs and especially UCs trend towards those obtained for PDs. These data may highlight age-related alterations in the dopaminergic system that may increase the risk of developing PD.
Acknowledgements
The authors would like to thank the UBC PET scanning team and the TRIUMF radio-chemistry production staff. All volunteer subjects, who generously donated their time to this research, are also most gratefully acknowledged.
References
StoesslAJ.
Neuroimaging in Parkinson’s disease. Neurotherapeutics2011;
8: 72–81.FuJF, et al.Joint pattern analysis applied to PET DAT and VMAT2 imaging reveals new insights into Parkinson’s disease induced presynaptic alterations. Neuroimage Clin2019;
23: 101856.
2021-89
Quantification of cerebral blood flow by a non-invasive hybrid PET/MR method for extracting the 15O-water image-derived input function free of partial volume errors (#319)
Lucas Narciso1, 2, Tracy Ssali1, 2, Linshan Liu1, Sarah Jesso1, 3, Justin W. Hicks1, 2, Udunna Anazodo1, 2, Elizabeth Finger1, 3 and Keith St Lawrence1, 2
1Lawson Health Research Institute, London, ON, Canada
2Western University, Department of Medical Biophysics, London, ON, Canada
3Western University, Department of Clinical Neurological Sciences, London, ON, Canada
Abstract
Introduction: Efforts to avoid invasive arterial blood sampling required for imaging cerebral blood flow (CBF) by positron emission tomography (PET) generally focus on obtaining an image-derived input function (IDIF).1 Hybrid PET/magnetic resonance imaging (MRI) offers an alternative in which MRI measurements of whole-brain (WB) CBF are used to calibrate 15O-water-PET.2,3 Originally, a double integration method (PMRFlowDIM) was proposed; however, PMRFlowDIM does not account for activity from the arterial cerebral blood volume (CBVa). Here, we present a modified version (PMRFlowIDIF) that extracts the IDIF (Figure 1(a) and (b)) from the WB 15O-water time-activity curve, is free of partial volume errors and generates CBVa images.
Methods: Data were collected in a PET/MR scanner (Siemens Biograph mMR) and included 5-min dynamic 15O-water-PET and phase-contrast MRI4 measurements. The accuracy of PMRFlowIDIF was assessed in a porcine model (n = 12) by comparison to PET-only CBF measurements (weighted non-linear least squares fitting routine). CBF maps from healthy individuals (n = 13) were generated with both PMRFlow approaches to investigate the impact of CBVa.
Results: CBF estimates from PMRFlowIDIF (Figures 1(c) and 2(a)) were in excellent agreement with the corresponding PET estimates in the porcine experiments. In general, good agreement was observed between CBF measurements from the two PMFlow methods (Figure 1(d); e.g., grey matter CBF = 45.1 ± 6.6 and 44.6 ± 6.4 mL/100g/min for PMRFlowDIM and PMRFlowIDIF, respectively). However, closer inspection of the images revealed regional discrepancies were proportional to the regional-to-WB CBVa differences (Figure 2(b); e.g. the cerebellum and insula).
Conclusion: The primary source of error of PMRFlowIDIF is inaccuracies in WB CBF measured by MRI. As expected in the healthy brain, differences in regional CBF between the PMRFlow techniques were typically small (<10%) with the largest found in highly vascularized regions. Consequently, we would recommend applying PMRFlowIDIF, rather than the simpler PMRFlowDIM, when significant variations in regional CBVa are expected, such as due to cerebrovascular disease.
References
Zanotti-FregonaraPChenKLiowJS, et al.Image-derived input function for brain PET studies: many challenges and few opportunities.J Cereb Blood Flow Metab2011;
31: 1986–1998.SsaliTAnazodoUCThiessenJD, et al.A noninvasive method for quantifying cerebral blood flow by hybrid PET/MRI. J Nucl Med2018;
59: 1329–1334.NarcisoL, et al.A noninvasive method for quantifying cerebral metabolic rate of oxygen by hybrid PET/MRI: validation in a porcine model.J Nucl Med2021.JainVLanghamMCWehrliFW.MRI estimation of global brain oxygen consumption rate.J Cereb Blood Flow Metab2010;
30: 1598–1607.
2021-90
Synaptic density measured by [18F]SynVesT-1 PET in ageing mice (#321)
Mengfei Xiong1, Sahar Roshanbin1, Dag Sehlin1, Jonas Eriksson2, 3, Johanna Rokka1 and Stina Syvänen1
1Uppsala University, Public Health and Caring Science/Molecular Geriatrics, Uppsala, Sweden
2Uppsala University, Medicinal Chemistry, Uppsala, Sweden
3Uppsala University Hospital, PET Centre, Uppsala, Sweden
4Uppsala University, Department of Public Health and Caring Sciences, Uppsala, Sweden
Abstract
Introduction: Synaptic vesicle 2A protein (SV2A) imaging with positron emission tomography (PET) has linked the loss of functional synapses to many neurodegenerative diseases by comparing diseased subjects with healthy controls. However, synaptic density changes in healthy ageing populations remain unclear. A recent PET study reported no decline in synaptic density with age in healthy human brains, but a similar study has not been conducted preclinically. Thus, in the present study, SV2A PET ligand [18F]SynVesT-1 was used to quantify synaptic density in healthy mice of different ages.
Methods: Three age groups (4–5 months: n = 7, 12–14 months: n = 11, 17–19 months: n = 8) of C57BL/6J mice were PET scanned with [18F]SynVesT-1 for 60 min followed by a 3 min CT examination. Images were subsequently aligned with an MRI-based brain atlas.1 Regions of interests (ROI) in the brain were extracted to obtain time-activity curves (TACs), and an image-derived input function (IDIF) was extracted from an ROI placed in the left ventricle of the heart.2 Brain retention of [18F]SynVesT-1 was displayed as the model-independent AUCbrain/blood ratio and the volume of distribution (VT). The area under the curves (AUC) were generated from TACs, whereas VT was obtained through kinetic modelling with either one- or two-tissue compartment model (1TCM or 2TCM).3
Results: VT from 1TCM suggested that brain uptake in the two younger groups were significantly higher than in the oldest group (Figure 1(e)). Interestingly, the highest brain retention appeared in mice aged 12–14 months. All age groups displayed the same blood concentrations. Thus, the difference in brain exposure was not due to differences in blood exposure. The 2TCM VT was in excellent agreement with 1TCM VT (Figure 1(f)) and AUCbrain/blood (r = 0.9929, p < 0.0001).
Conclusion: Consistent with human PET data,4 our data suggest that the synaptic density in healthy ageing mice remains stable during the majority of the mouse life span and does not decrease until very old age. Even then, the decrease is moderate. This information is important as preclinical PET may be used to test novel drug compounds. Therefore, that data is also likely to be translatable to higher species.
Acknowledgements
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 813528.
References
LastnMaYHofPRGrantSC, et al.A three-dimensional digital atlas database of the adult C57BL/6J mouse brain by magnetic resonance microscopy. Neuroscience2005;
135: 1203–1215.DelorenzoCStaelensSVerhaegheJ, et al.Noninvasive relative quantification of [11C]ABP688 PET imaging in mice versus an input function measured over an arteriovenous shunt. Front Neurol2018;
9: 516.NaganawaMLiSNabulsiNB, et al.First-in-human evaluation of 18 F-SynVesT-1, a novel radioligand for PET imaging of synaptic vesicle protein 2A. J Nucl Med2021;
62: 561–567.MichielsLDelvaAAalstJ Van, et al.Synaptic density in healthy human aging is not influenced by age or sex : a 11C-UCB-J PET study. Neuroimage2021;
232: 117877.
2021-91
The positron emission tomography brain imaging data structure (PET-BIDS) extension: A new standard for sharing PET data (#322)
Martin Norgaard1, 2, Granville J. Matheson3, 4, Hanne D. Hansen1, 5, Adam Thomas6, Graham Searle7, Gaia Rizzo7, Mattia Veronese8, 9, Alessio Giacomel8, Maqsood Yaqub10, Matteo Tonietto11, Thomas Funck12, Ashley Gillman13, Hugo Boniface14, Alexandre Routier15, Jelle R. Dalenberg16, Tobey Betthauser17, Franklin Feingold2, Christopher J. Markiewicz2, Krzysztof J. Gorgolewski2, Ross W. Blair2, Stefan Appelhoff18, Remi Gau19, Taylor Salo20, Guiomar Niso21, Cyril Pernet1, Christophe Phillips22, Robert Oostenveld23, 24, Jean-Dominique Gallezot25, Richard E. Carson25, Gitte M. Knudsen1, Robert B. Innis26 and Melanie Ganz1, 27
1Univ. Copenhagen, Neurobiology Research Unit, Rigshospitalet, and Institute of Clinical Medicine, Copenhagen, Denmark
2Stanford University, Department of Psychology, Stanford California, USA
3Columbia University, Department of Psychiatry, New York New York, USA
4Karolinska Institutet and Stockholm Health Care Services, Centre for Psychiatry Research, Department of Clinical Neuroscience, Stockholm, Sweden
5Athinoula A. Martinos Center for Biomedical Imaging, MGH/HST, Charlestown Massachusetts, USA
6Intramural Research Program, NIMH, Bethesda, USA
7Imperial College London, Invicro and Division of Brain Sciences, London, UK
8King’s College London, Centre for Neuroimaging Sciences, London, UK
9University of Padua, Department of Information Engineering, Padua, Italy
10Amsterdam UMC, location VUmc, department of radiology and nuclear medicine, Amsterdam, Netherlands
11Universite Paris-Saclay, CEA, CNRS, Inserm, BioMaps, Service Hospitalier Frederic Joliot, Orsay, France
13Commonwealth Scientific and Industrial Research Organisation, Australian eHealth Research Centre, Townsville, Australia
14CEA, Centre d’Acquisition et de Traitement des Images, Paris, France
15Inria, Aramis project-team, Sorbonne Universite, Institut du Cerveau – Paris Brain Institute – ICM, Inserm, CNRS,AP-HP, Hospital de la Pitie Salpetriere, Paris, France
16University of Groningen, University Medical Center Groningen, Department of Neurology, Groningen, Netherlands
17University of Wisconsin-Madison School of Medicine and Public Health, Wisconsin Alzheimer’s Disease Research Center, Division of Geriatrics, Department of Medicine, Madison Wisconsin, USA
18Max Planck Institute for Human Development, Center for Adaptive Rationality, Berlin, Germany
19Université catholique de Louvain, Institute of psychology, Louvain la Neuve, Belgium
20Florida International University, Department of Psychology, Miami Florida, USA
21Indiana University, Psychological Brain Sciences, Bloomington Indiana, USA
22University of Liege, GIGA Cyclotron Research Centre in vivo imaging, Liege, Belgium
23Radboud University, Donders Institute for Brain, Cognition and Behaviour, Nijmegen, Netherlands
24Karolinska Institutet, NatMEG, Stockholm, Sweden
25Yale University, Department of Radiology and Biomedical Imaging, New Haven, USA
26National Institutes of Health, Molecular Imaging Branch, Bethesda, USA
27University of Copenhagen, Department of Computer Science, Copenhagen, Denmark
Abstract
Introduction: The neuroimaging community has increasingly recognized the importance of data sharing. Reasons for this are the poor replicability of findings, the need for appropriate quality control, the greater statistical power provided by larger samples, and the higher scientific impact of multilateral collaborations. In addition, funding bodies and scientific journals increasingly require that data be shared. The Brain Imaging Data Structure (BIDS)1 initiative has sought to establish consensus on how to organize and share data obtained in neuroimaging experiments. BIDS is being developed in an ongoing and inclusive community effort; many neuroscientists consult to ensure that BIDS covers most common neuroimaging experiments. While BIDS was originally based on functional and structural MRI acquisitions, it has now been extended to cover a wide range of different modalities.2 In addition, a growing number of software tools (see http://bids-apps.neuroimaging.io/) can read data organized according to the BIDS structure.3
Methods: The new PET community data standard is now part of the BIDS specification and in detail described in literature.4 The data standard follows the community guidelines for the reporting of PET experiments5 and describes storage of all image and metadata needed to quantify PET data using uptake imaging or reference tissue modelling (RTM). It also specifies how to store blood data and how to convey the meta information related to it in a standardized fashion.
Results: Example data following the BIDS standard have been added to OpenNeuro to demonstrate feasibility. The interface is shown in Figure 1. Furthermore, we are currently developing converter software to facilitate easy conversion to PET BIDS based on different type of scanner files, e.g. dcm or ecat, that are distributed in an open-source GitHub repository for use by the community.
Conclusion: BIDS is intuitive and easy to adopt. The specification was intentionally based on existing file formats and intuitive folder structures to reflect current lab practices and make data machine-readable. The rapid and widespread adoption of this standard by the wider PET community will facilitate not only data sharing and aggregation, but also improve the applicability of software designed to accommodate this standard.
References
GorgolewskiKJAuerTCalhounVD, et al.The brain imaging data structure, a format for organizing and describing outputs of neuroimaging experiments. Sci Data2016;
3: 160044.https://bids.neuroimaging.io/specification.htmlGorgolewskiKJAlfaro-AlmagroFAuerT, et al.BIDS apps: improving ease of use, accessibility, and reproducibility of neuroimaging data analysis methods. PLoS Comput Biol2017;
13: 1–16.NorgaardMMathesonGJHansenHD, et al.PET-BIDS, an extension to the brain imaging data structure for positron emission tomography. bioRxiv2021.KnudsenGMGanzMAppelhoffS, et al.Guidelines for the content and format of PET brain data in publications and archives: a consensus paper. J Cerebral Blood Flow Metab2020;
40: 1576–1585.
2021-92
Development of a novel PET ligand for receptor-interacting protein kinase 1 in brain (#323)
1National Center for Geriatrics and Gerontology, Department of Clinical and Experimental Neuroimaging, Obu, Japan
2Nagoya University of Economics, Department of Human Life and Sciences, Nagoya, Japan
3Gifu University of Medical Science, Department of Pharmacy, kani, Japan
Abstract
Introduction: Microglia are the immune cells in the brain. Microglia play a critical role in neuroinflammation associated with neurodegenerative diseases, such as Alzheimer’s disease (AD). Recently, efforts in drug discovery for treatment of AD progression have focused on modifying the function of microglia. For example, one such effort has been focusing on receptor-interacting protein kinase 1 (RIPK1), which is a multifaceted kinase that controls cell death and inflammation. Inhibition of RIPK1 expressed in microglia is known to prevent phenotypic changes of microglia from the resting to disease-associated states. RIPK1 inhibitors are therefore considered to be of therapeutic values in treating AD progression. To monitor the effectiveness of this treatment strategy, in vivo imaging biomarker of RIPK1 would be of potential usefulness.
Methods: We chose one RIPK1 inhibitor, (3S)-3-(2-Benzyl-3-chloro-7-oxo-2,4,5,7-tetrahydro-6H-pyrazolo-[3,4-c]pyridin-6-yl)-5-methyl-4-oxo-2,3,4,5-tetrahydro-1,5-benzoxa-zepine-8-carbonitrile (1), from several candidates with a high RIPK1 inhibitory potency that would possess optimal characteristics for PET imaging of the brain. To radiolabel 1 with11C, a precursor, N-desmethylated 1 (Dm-GG511), was synthesized from starting materials related to the 1 structure. Then, 11CH3 group was introduced to Dm-GG511 to the final product, 11C-1 (11C-GG511), by optimizing methylating processes. To evaluate if 11C-GG511 has adequate properties as a PET ligand for imaging RIPK1, PET imaging was performed in normal and inflammation model rats including the evaluation of the presence of any of 11C-GG511 specific binding by conducting in vivo blocking and in vitro brain tissue autoradiography experiments.
Results: In the normal rats, 11C-GG511 showed overall good reversible brain uptake with a relatively higher uptake in the cerebellum, but otherwise no other regional uptake differences. In the inflammation model rats, no obvious 11C-GG511 uptake differences were observed between the inflamed and contralateral striata or any healthy brain regions. The cerebellar uptake in the normal rats was minimally blocked, suggesting the presence of specific binding of 11C-GG511 if any. However, the presence of any specific binding was not evident on autoradiography.
Conclusion:11C-GG511 shows a good overall brain uptake, but no clear evidence of specific binding in rats. Further studies are warranted to modify this ligand or try other related ligands for more optimal PET imaging of RIPK1.
Acknowledgements
We would like to thank Takeda Pharmaceutical Company Limited for providing chemical substance {Ethyl 1-Benzyl-5-chloro-4-(2-oxoethyl)-1H-pyrazole-3-carboxylate}, standard samples of labeling precursors (Dm-GG511) and nonradiolabeled authentic samples (1).
2021-93
Positron emission tomography imaging of alpha-synuclein?In vitro and in vivo evaluation of MODAG-005 (#324)
Ran Sing Saw1, Sabrina Buss1, Laura Kuebler1, Sergey Ryazanov2, 3, Andrei Leonov2, 3, Federica Bonanno1, Felix Schmidt2, Viktoria Ruf4, Daniel Bleher1, Marilena Poxleitner1, Ann-Kathrin Grotegerd1, Bernd J. Pichler1, Gregory D. Bowden1, Andreas Maurer1, Christian Griesinger4, Armin Giese2, 4 and Kristina Herfert1
1Eberhard Karls University of Tübingen, Werner Siemens Imaging Center, Department of Preclinical Imaging & Radiopharmacy, Tübingen Baden-Württemberg, Germany
3Max Planck Institute for Biophysical Chemistry, Göttingen Lower Saxony, Germany
4Ludwig Maximilians University, Center for Neuropathology and Prion Research, Munich Bavaria, Germany
Abstract
Introduction: Positron emission tomography (PET) imaging of alpha-synuclein (αSYN) aggregates would be invaluable for the non-invasive diagnosis of synucleinopathies as well as facilitating the development of novel treatment strategies. Here, we report the in vitro and in vivo evaluation of MODAG-005 as a promising αSYN PET tracer.
Methods:In vitro binding of [3H]MODAG-005 was determined using human recombinant αSYN, amyloid-beta1-42 (Aβ)andtau fibrils. (Micro)autoradiography was performed in postmortem human brain tissues from multiple system atrophy (MSA), Parkinson’s disease (PD), Alzheimer’s disease (AD), progressive supranuclear palsy (PSP), healthy controls, and in the transgenic αSYN(A30P) mouse model. Subsequently, the binding affinity to αSYN in human brain was determined in MSA tissues using autoradiography. In vivo pharmacokinetics and metabolism of [11C]MODAG-005 were studied in healthy mice and rats, and in rats intrastriatally injected with αSYN fibrils. Time-activity curves were generated from dynamic 60-minute PET for the striata (target region) and cerebellum (reference), and the binding potential (BPND) values were calculated using Logan’s reference tissue model.
Results: Saturation binding assays revealed Kd values for [3H]MODAG-005 of 0.2 nM for αSYN, 7 nM for tau and > 100 nM for Aβ fibrils. Autoradiography in human brain tissues confirmed the high affinity binding of [3H]MODAG-005 to αSYN with a Kd of 0.25 nM in MSA, while showing a specific binding in the αSYN(A30P) mouse model. [11C]MODAG-005 was obtained at high molar activities of > 200 GBq/µmol. The tracer showed an excellent blood-brain barrier penetration and a fast clearance from the brain. We observed one metabolite in the brain, with 96% and 79% of the parent compound remaining at 5 and 15 minutes post-injection. Increased tracer binding was detected in the fibril-injected striatum compared to the sham-injected striatum, whereas no difference was detected in non-injected rats.
Conclusion: Here, we present a novel PET tracer targeting αSYN in the human brain. MODAG-005 possesses a very high affinity to αSYN in MSA tissues, in line with the αSYN fibril binding assay. It also exhibits excellent brain availability, good kinetics, and sufficient signal-to-noise ratio in fibril-injected rats. Despite the lack of full selectivity, MODAG-005 is currently one of the most promising αSYN-targeting PET tracers.
Acknowledgements
We thank Ramona Stremme, Elena Kimmerle and Johannes Kinzler for the radiosynthesis. We also thank the technical assistants Linda Schramm, Maren Harant, Stacy Huang, Miriam Owczorz, Natalie Hermann and Isabel Sehnke for their experimental support. Additionally, we acknowledge Dr. Julia Mannheim, Dr. Rebecca Rock, Dr. Neele Hübner, Dr. Andreas Dieterich, Hans Jörg Rahm, Dr. Carsten Calaminus and Funda Cay for their administrative support.
In vitro evaluation of [3H]MODAG-005 NOT FOR PUBLICATION
In vivo evaluation of [11C]MODAG-005 NOT FOR PUBLICATION
2021-94
Simultaneous Multifactor Bayesian Analysis (SiMBA) of PET Time Activity Curve Data (#325)
Granville J. Matheson1, 2 and R. Todd Ogden1, 2
1Columbia University, Molecular Imaging and Neuropathology Division, New York New York, USA
2Columbia University, Department of Biostatistics, New York New York, USA
Abstract
Introduction: Positron emission tomography (PET) is essential for studying the neurochemical pathophysiology of psychiatric and neurological disease, however its high cost and exposure of participants to radiation limit the feasibility of large sample sizes. In clinical research, the major shortcoming of PET imaging has therefore been its lack of statistical power for studying clinically-relevant research questions. Here, we introduce a new method for performing PET quantification and analysis called SiMBA to make more effective use of PET time activity curve data.
Methods: SiMBA makes use of hierarchical, multifactor Bayesian modelling to exploit similarities both between individuals and among regions within individuals while applying the two-tissue compartment model. In this way, the model borrows strength across the whole dataset to improve stability and robustness to measurement error, as well as parameter identifiability and estimation accuracy, without sacrificing model interpretability. Furthermore, parameter estimation and statistical analysis are performed simultaneously using all measured data and its associated uncertainty, in contrast with the conventional approach of using point estimates. We tested this approach in simulated data in comparison with conventional nonlinear least squares modelling and subsequent statistical analysis.
Results: In simulated [11C]WAY100635 data, this approach greatly improves both statistical power and the consistency of effect size estimation, equivalent to effectively doubling and quadrupling the sample size respectively, without affecting the false positive rate. This comes at the cost of increased computational expense, taking in the order of hours to days to fit for typical clinical datasets.
Conclusion: This method has the potential to make it possible to test clinically-relevant hypotheses which could not be studied before given the practical constraints. Furthermore, because this method does not require any additional information, it makes it possible to re-examine data which has already previously been collected. In the absence of dramatic advancements in PET image data quality, radiotracer development, or data sharing, PET imaging has been limited in the scope of research hypotheses which could be studied. This method, especially combined with the recent steps taken by the PET imaging community to embrace data sharing, will make it possible to further improve the research possibilities and clinical relevance of PET neuroimaging.
Acknowledgements
The work reported here has been partially supported by US NIH grants 5 P50 MH090964 and 5 R01 EB024526, and by the Hjärnfonden Postdoctoral Fellowship.
2021-95
Investigating fatty acid amide hydrolase in mild traumatic brain injury; A PET study of [11C]CURB in occupational posttraumatic stress disorder (#327)
Khadija N. Brouillette1, 2, Erin Gaudette2, 3, Sarah E. Watling2, 5, Jerry Warsh2, 3, Laura M. Best2, 5, Duncan Green2, 5, Tina McCluskey2, 3, Shawn Rhind6, Rachel Tyndale3, 4, Don Richardson7, 8, Neil Vasdev3, 10, Rakesh Jetly9, Stephen Kish2, 3 and Isabelle Boileau2, 3
1McMaster University, School of Interdisciplinary Sciences, Hamilton, ON, Canada
2Centre for Addiction and Mental Health, Brain Health Imaging Centre, Toronto, ON, Canada
3Centre for Addiction and Mental Health, Campbell Mental Health Research Institute, Toronto, ON, Canada
4University of Toronto, Pharmacology and Toxicology, Toronto, ON, Canada
5University of Toronto, Institute of Medical Sciences, Toronto, ON, Canada
6Toronto Research Centre, Defense Research and Development Canada, Toronto, ON, Canada
7St. Joseph’s Operational Stress Injury Clinic, London, ON, Canada
8University of Western Ontario, Department of Psychiatry, London, ON, Canada
9University of Ottawa, The Royal’s Institute of Mental Health Research and Department of Psychiatry, Ottawa, ON, Canada
10Centre for Addiction and Mental Health, Azrieli Centre for Neuro-Radiochemistry, Toronto, ON, Canada
Abstract
Introduction: Concussive head trauma or mild traumatic brain injuries (mTBI) are prevalent and impactful in individuals with occupational posttraumatic stress disorder (PTSD) and may affect the course and severity of disease. Higher levels of the major endocannabinoid anandamide have been shown in preclinical models of neuronal damage; putatively suggesting a possible neuroprotective role for anandamide. This study aimed to use positron emission tomography (PET) to test the hypothesis that levels of Fatty Acid Amide Hydrolase (FAAH), the enzyme that catabolizes anandamide, are lower in individuals with occupational PTSD with mTBI relative to those without.
Methods: Participants with occupational PTSD (e.g., Canadian Armed Forces members, veterans and first responders) (n = 16; 8 with an mTBI) completed a PET scan with arterial blood sampling following injection of the FAAH tracer [11C]CURB. All participants were genotyped for the FAAH C385A polymorphism which affects [11C]CURB binding. Clinical symptoms and cognitive function were assessed. A Repeated Measures ANCOVA (brain region of interest (ROI)1 x GROUP2 controlling for FAAH C385A) was conducted to assess whether FAAH levels were different in the brains of those with PTSD alone versus those with comorbid PTSD and mTBI.
Results: There were no group differences in age, sex, body mass, smoking status, FAAH C385A genotype or medication and clinical symptom scores were also not different (p’s > 0.94). However, PTSD with mTBI had marginally poorer performance on tests of executive function and verbal memory (trails B: p = 0.01; Hopkin’s Verbal Learning Test: p = 0.1). [C-11]CURB binding in all 10 ROIs was 12%-20% lower in the mTBI group (F(1,13) = 4.53, p = 0.05) with the most robust effects noted in the hippocampus and amygdala (∼20%; p(uncorrected) = 0.03). [C-11]CURB binding in the 10 ROIs did not correlate with clinical measures (p’s > 0.99).
Conclusion: These preliminary findings suggest that mTBI in PTSD may be associated with lower FAAH binding in the brain and lend support to preclinical evidence suggesting increased anandamide after neuronal damage. Future studies should systematically investigate the consequences of mTBI on endocannabinoid levels and their key hydrolytic enzymes, including FAAH, to gain a better understanding of their roles in recovery from trauma and the potential utility of endocannabinoid focused therapeutics (e.g., FAAH inhibitors).
Acknowledgements
This work was funded by the Canadian Institute for Health Research, Canadian Institute for Military and Veteran Health Research.
References
MayoLMRabinakCAHillMN, et al.Targeting the endocannabinoid system in the treatment of PTSD: a promising case of preclinical-clinical translation?Biol Psychiatr2021.MechoulamRShohamiE.
Endocannabindoids and truamatic brain injury.Mol Neurobiol2004;
36: 68–74.
2021-96
5HT1A receptor co-expression based polygenic score associates with 5HT1A predicted binding potential in frontal regions (#328)
Danusa M. Arcego1, 4, Eve Peraldi3, Irina Pokhvisneva1, 4, Zihan Wang1, 4, Sachin Patel1, 4, Barbara Barth3, 4, Michael J. Meaney1, 4, Andreas Hahn2, Rupert Lanzenberger2 and Patricia P. Silveira1, 4
1Douglas Mental Health University Institute/McGill university, Department of Psychiatry, Montréal, QC, Canada
2Medical University of Vienna, Department of Psychiatry and Psychotherapy, Vienna, Austria
3McGill University, Montreal, QC, Canada
4McGill University, Ludmer Center for Neuroinformatics and Mental Health, Montreal, QC, Canada
Abstract
Introduction: Serotonin is an important modulatory neurotransmitter with distinctive neuroplastic capabilities. Serotoninergic neuroplasticity dysfunctions can impact gray matter volume, which is associated with disruption of cognition and emotional functions. We aimed at evaluating if variations in the main inhibitory serotonin receptor (5HT1A) expression-based polygenic score (ePRS) in the frontal cortex associates with variations in the predicted binding potential for this receptor in participants from the UK Biobank.
Methods: The predicted binding potential was calculated based on a 5HT1A binding-gray matter reference atlas (measured by PET scan,1). This information was used to predict 5HT1A binding potential in the UK Biobank individuals using the corresponding gray matter volume information for this sample. The ePRS was based on 5HT1A co-expression network using human RNA-sequencing data from BrainEAC (http://www.braineac.org) in the frontal cortex. Single-nucleotide polymorphisms mapped on the network genes (144 genes) were weighted by the association between genotype and gene expression using GTEx (http://www.gtexportal.org) (Figure 1).
Results: Regression models adjusted by population stratification, age and sex were used to investigate the association between the predicted 5HT1A binding potential and the frontal cortex 5HT1A ePRS, both calculated for UK Biobank participants. We observed that the frontal 5HT1A ePRS was significantly correlated with serotonin predicted binding potential in specific cortical regions (frontal orbital cortex, FDR adjusted p-value = 0.03, ß = 11.25; insular cortex, FDR = 0.03, ß = 7.45; precentral gyrus FDR = 0.03, ß = −24.72; cingulate gyrus FDR = 0.04, ß = −7.43; frontal pole, FDR = 0.04, ß = −32.85, N = 17,181, Figure 2).
Conclusion: We observed that frontal 5HT1A ePRS (based on a co-expression gene network) strongly associated with a functional brain measure (binding potential) specifically for this receptor in a large sample of individuals. This confirms the ability of the polygenic score to reflect biological differences in neurotransmitter systems in a brain region-specific manner.
Acknowledgements
JPB Foundation, Hope for Depression Research Foundation, and Canadian Institutes of Health Research (CIHR).
Reference
KrausCHahnASavliMKranzGSBaldingerPHöflichASpindeleggerCUngersboeckJHaeuslerDMitterhauserMWindischbergerC.2012.
Serotonin-1A receptor binding is positively associated with gray matter volume—a multimodal neuroimaging study combining PET and structural MRI. Neuroimage,
63: 1091–1098.
2021-97
[124I]IPPI: Molecular modeling and autoradiographic study for binding to Taupathies (#329)
Taylor R. Moran, Reisha M. Ladwa, Rommani Mondal, Christopher Liang and Jogeshwar Mukherjee
University of California, Irvine, Preclinical Imaging, Radiological Sciences, Irvine California, USA
Abstract
Introduction: Tau isoforms 3R and 4R are present in Alzheimer’s disease (AD, 3/4R), Pick’s disease (PiD, 3R) and corticobasal degeneration (CBD, 4R).1 Selective imaging agents for 3R and 4R will help in the development of differential diagnostic methods of various taupathies. We have docked the azaindole IPPI in Tau molecular stuctures to assess binding sites and developed [124I]IPPI as a potential PET imaging and evaluated [125I]IPPI in autoradiographic studies of human taupathies.
Methods: Molecular docking was carried out using Autodock-Chimera on AD Tau, PiD Tau and CBD Tau2 [Figure 1]. [124I]Sodium Iodide was used to prepare [124I]IPPI by electrophilic substitution of the tributyltin derivative3 [Figure 2(a)]. Human AD post-mortem brain tissues consisting of anterior cingulate (AC) were used (n = 6). Brain slices (10 µm thick) were treated with [124I]IPPI as reported previously with [125I]IPPI.3 Preliminary autoradiographic studies were also carried out in PiD (n = 2) and CBD (n = 1) using [125I]IPPI. Nonspecific binding was measured using 10 µM MK-6240. The brain sections were exposed overnight on a phosphor film and extent of binding [124/125I]IPPI was measured in DLU/mm2. Adjacent sections were immunostained with anti-Tau antibody in AD brains (Figure 2(d)).
Results: IPPI exhibits nanomolar affinities for AD Tau.3 Four separate binding sites were found for AD Tau,3 CBD Tau and PiD Tau4 (Figure 1). Binding energies were found to be favorable for site 1 for each of the Tau (-9.7, -9.6, -10.8 Kcal/mol for AD, CBD, PiD). This suggested that IPPI is likely to bind to the different Tau forms in the three tauopathies. High binding of [124IIPPI in AD brain slices was observed in AC in all subjects (Figure 2(c)) which was consistent with adjacent slice positively immunostained with anti-Tau (Figure 2(d)). This is similar to our reported findings with [125I]IPPI.3 Preliminary results of [125I]IPPI in PiD and CBD suggest some binding which was not displaceable by MK-6240. This is consistent with the finding that [18F]MK-6240 does not bind to PiD.5
Conclusion: [124I]IPPI has been successfully prepared for PET imaging of taupathies. Additional studies are underway to confirm its binding to different forms of human Tau. Transgenic P301S mice models of Tau are currently being used for PET/CT imaging of [124I]IPPI.
Acknowledgements
NIH/NIA RF1 AG029479 (JM), UCI UROP (TM, RML, RM). Banner Health Research Institute and UCI MIND for tissue samples and UCI Pathology for immunostaining.
References
KatsumotoATakeuchiHTanakaF.
Tau pathology in chronic traumatic encephalopathy and Alzheimer’s disease: similarities and differences. Front Neurol2019;
10: 980.Protein Data Bank, www.rcsb.org (accessed ▪).MukherjeeJLiangCPatelKK, et al.Development and evaluation [125I]IPPI for tau imaging in post-mortem human Alzheimer’s disease brain. Synapse2021;
74: e22183.Arul MuruganNNordbergAÅgrenH.
Cryptic sites in tau fibrils explain the preferential binding of the AV-1451 PET tracer toward Alzheimer’s tauopathy. ACS Chem Neurosci2021;
12: 2437–2447.AgueroCDhaynautMNormandinMD, et al.Autoradiography validation of novel tau PET tracer [F-18]-MK-6240 on human postmortem brain tissue. Acta Neuropathologica Commun2019;
7: 37.
2021-98
Increased Tau, Aβ amyloid, and monoamine oxidase-A in post-mortem human Alzheimer’s disease anterior cingulate (#330)
Amina U. Syed, Christopher Liang, Tonya Mukherjee, Krystal Patel and Jogeshwar Mukherjee
University of California, Irvine, Preclinical Imaging, Radiological Sciences, Irvine California, USA
Abstract
Introduction: Successful multitarget approaches are currently underway towards the accurate and early diagnosis of Alzheimer’s disease (AD).1 Our ongoing multitarget approach involves Tau imaging with [125I]IPPI,2 Aβ plaque imaging with [18F]Flotaza,3 and [18F]FAZIN3 for imaging monoamine oxidase A (MAO-A).4 The role of monoamine oxidases (MAOs) in neurotransmitter breakdown, increased production of reactive oxygen species, and possible roles in microglia activation and inflammation have attracted much attention. To support role of MAO-A in AD, we reported [18F]fluoroethylharmol ([18F]FEH).5 Using anterior cingulate (AC) regions of the brain, we now correlate the three biomarkers for AD (Figure 1).
Methods: [125I]IPPI, [18F]Flotaza, [18F]FAZIN3, and [18F]FEH were prepared as previously reported. For in vitro studies, human post-mortem brain tissues, consisting of AC and corpus collosum (CC) regions, from Banner Health, Sun City, Arizona were used. Brain slices (10 µm thick) were obtained on a Leica 1850 cryotome. Brain sections were incubated with [18F]FAZIN3 or [18F]FEH (approx. 1 µCi/cc) in PBS (pH 7.4) buffer at 25°C for 60 min and subsequently washed with PBS buffer, while [125I]IPPI and [18F]Flotaza were washed with alcohol-PBS buffer. Using the Optiquant program, regions of interest were drawn in digital light units (DLU)/mm2 to quantify the percentage binding of the radiotracers. Immunostaining with anti-Tau and anti-Aβ was carried on adjacent sections of all subjects.
Results: Autoradiography and immunostaining results support that Aβ, MAO-A, and Tau are upregulated in the AD brain (Figure 2). The gray matter (GM) were clearly delineated versus white matter (WM) in the AD autoradiographic images of [18F]FEH with an average GM/WM > 2 (Figure 2(c)) and [18F]FAZIN3 with an average GM/WM > 3 (Figure 2(b)) for the binding to MAO-A. MAO-A binding positively correlated with the high binding of [125I]IPPI to Tau (GM/WM > 4) (Figure 2(e)) and [18F]Flotaza to Aβ plaques (GM/WM > 100) (Figure 2(h)), consistent with cortical binding corresponding to anti-Aβ (Figure 2(i)) and anti-Tau (Figure 2(f)). All cognitively normal (CN) subjects exhibited lower binding in the GM compared to AD subjects in autoradiography studies. Nonspecific binding of [18F]FEH was higher than [18F]FAZIN3.
Conclusion: Based on our results, increased [18F]FEH and [18F]FAZIN3 binding in AD brains suggests upregulation of MAO-A. Additionally, with increased [125I]IPPI binding to Tau and [18F]Flotaza binding to Aβ in AD, our findings support a potential multitarget approach to AD determination.
Acknowledgements
NIH/NIA RF1 AG029479 (JM), UCI UROP (AS). Banner Health Research Institute and UCI MIND for tissue samples and UCI Pathology for immunostaining.
References
PascoalTABenedetALAshtonNJ, et al.Microglial activation and tau propagate jointly across Braak stages. Nat Med2021;
▪: 1–8.MukherjeeJLiangCPatelKK, et al.Development and evaluation [125I]IPPI for tau imaging in post-mortem human Alzheimer’s disease brain. Synapse2021;
74: e22183.KaurHFelixMRLiangC, et al.Development and evaluation [18F]Flotaza for Ab plaque imaging in post-mortem Alzheimer’s disease brain. Bioorg Med Chem Lett2021;
46: 128164.MukherjeeJLiangCSyedAU, et al.Monoamine oxidase-A is upregulated in post-mortem human Alzheimer’s disease and Parkinson’s disease brain. J Nucl Med2021;
62: 1609.SyedAULiangCMukherjeeJ.
[18F]Fluoroethylharmol: improved radiosynthesis and evaluation of monoamine oxidase-A in human Alzheimer’s and Parkinson’s disease. J Nucl Med2021;
62: 1628.
2021-99
A three-factor model of common early-onset psychiatric disorders: Temperament, adversity, and dopamine (#331)
Maisha Iqbal1, Sylvia Cox2, Natalia Jaworska3, Maria Tippler1, Natalie Castellanos-Ryan6, Sophie Parent6, Alain Dagher1, Frank Vitaro4, Mara Brendgen5, Michel Boivin7, Robert Pihl8, Sylvana Cote4, Richard Tremblay4, Jean Seguin4 and Marco Leyton1, 2
1McGill University, Neurology and Neurosurgery/Irving Ludmer Psychiatry Research and Training Building, Montreal, QC, Canada
3University of Ottawa Institute of Mental Health Research, Ottawa, ON, Canada
4CHU Ste-Justine Research Center, Montreal, QC, Canada
5Université de Québec à Montréal, Department of Psychology, Montreal, QC, Canada
6Université de Montréal, School of Psychoeducation, Montreal, QC, Canada
7Tomsk State Universit, Institute of Genetic, Neurobiological and Social Foundations of Child Development, Tomsk, Russia
8McGill University, Department of Psychology, Montreal, QC, Canada
Abstract
Introduction: Commonly comorbid early-onset psychiatric disorders might reflect the varying expression of overlapping risk factors.1 The mediating processes remain poorly understood, but three factors show some promise: adolescent externalizing (EXT) traits, early life adversity,2 and midbrain dopamine autoreceptors.3,4 Here, we investigated whether these features acquire greater predictive power when combined.
Methods: Participants were transitional aged youth who had been followed since birth and lived in the area of Montreal and Quebec City. They were invited to participate in the current study based on EXT trait scores between the ages of 10 to 16,5 as measured with the Social Behavioural Questionnaire. In early adulthood (age 18.5 ± 0.6 y.o.) participants were assessed with the Structured Clinical Interview for DSM-5, completed the Childhood Trauma Questionnaire (CTQ), and had a 90-min high-resolution positron emission tomography scan with [18F]fallypride. Fifty-two participants (30F, 22M) completed the study. Follow-up interviews were conducted 1, 2 and 3 years later. Binomial logistic regression analyses tested whether midbrain [18F]fallypride BPND values, EXT and CTQ scores predicted the presence of lifetime DSM-5 diagnoses. All analyses were run for diagnoses obtained at the time of or prior to the PET scan and again including diagnoses obtained during the follow-up interviews.
Results: The three-factor model predicted their presence with an overall accuracy of 90.4% (p = 2.4 x 10−5) and explained 91.5% of the area under the receiver operating characteristic curve [95% CI: .824, 1.000]. The model remained significant upon addition of new diagnoses that developed during the follow-up period (p = 3.5 x 10−5).
Conclusion: A combination of EXT traits, early life adversity, and poorly regulated dopamine transmission might increase the risk for diverse early-onset psychiatric disorders. The data reported here raise the possibility that these features can predict susceptibility prospectively.
References
CaspiAMoffittTE.
All for one and one for all: mental disorders in one dimension. Am J Psychiatr2018;
175: 831–844.McLaughlinKAColichNLRodmanAM, et al.Mechanisms linking childhood trauma exposure and psychopathology: a transdiagnostic model of risk and resilience. BMC Med2020;
18: 96.BuckholtzJWTreadwayMTCowanRL, et al.Dopaminergic network differences in human impulsivity.Science2010;
329: 532–532.JaworskaNCoxSMLTipplerM, et al.Extra-striatal dopamine2/3 receptor availability in youth at-risk for addictions. Neuropsychopharmacology2020;
45: 1498–1505.CoxSMLCastellanos-RyanNParentS, et al.Externalizing risk pathways for adolescent substance use and its developmental onset: a Canadian Birth Cohort study. Can J Psychiatr2021. doi: 070674372098242.
2021-100
[18F]2-fluoro-2-deoxy-sorbitol PET imaging for quantitative estimation of blood-brain barrier permeability in vivo (#332)
Gaëlle Hugon1, Sébastien Goutal1, Ambre Dauba1, Louise Breuil1, Benoit Larrat2, Alexandra Winkeler1, Anthony Novell1 and Nicolas Tournier1
1Université Paris-Saclay, CEA, CNRS, Inserm, BioMaps, Orsay, France
2Université Paris-Saclay, CNRS, CEA, DRF/JOLIOT/NEUROSPIN/BAOBAB, Gif-sur-Yvette, France
Abstract
Introduction: The non-transported and non-metabolized sorbitol derivative [18F]-2-fluoro-2-deoxy-sorbitol ([18F]FDS) can be straightforwardly obtained from chemical reduction of commercial 18F-2-deoxy-2-fluoro-D-glucose ([18]FDG). Biodistribution 18F-FDS shows negligible baseline brain uptake. For the first time, [18F]FDS was evaluated as a small molecule (paracellular) marker of blood-brain barrier (BBB) integrity for PET imaging.
Methods: Five mice underwent focused ultrasound (FUS) to generate spatially controlled BBB disruption in the right brain hemisphere and three mice were used as a control without FUS protocol. Five min after FUS, the BBB integrity marker Evan’s blue (EB) and 18F-FDS (4.2 ± 0.7 MBq) were injected i.v, followed by 60 min brain microPET acquisition. Animals were then sacrificed and brain removed to visually check EB extravasation as a post-mortem marker of BBB disruption. Time-activity curves (TACs) were measured in the right and left hemisphere for each group. Area Under the TACs (AUC) and AUCR using the non-sonicated hemisphere as a reference region were calculated. PET kinetics of [18F]FDS in each brain hemisphere were described by a 1-tissue compartment model using an image derived input function.
Results: Brain PET signal was visually increased in the sonicated brain region. Brain AUC was significantly higher in the hemisphere with disrupted BBB compared to contralateral hemisphere (2.2 ± 0.5-fold increase, p < 0.01); or corresponding (right) hemisphere of control mice (p < 0.05). A 2.4 ± 0.8-fold increase was observed in the brain distribution (VT, p < 0.01) of [18F]-FDS in the sonicated area compared with the contralateral hemisphere in the FUS group but not in the control group, consistent with ex vivo brain distribution of EB extravasation. Enhanced brain uptake was associated with an increase in the influx transfer rate K1 (+1.4 ± 0.7-fold, p < 0.05) and a decrease in the efflux transfer rate k2 (-1.7 ± 0.4-fold, p < 0.01).
Conclusion: Thanks to the quantitative performance of PET compared with other neuroimaging techniques, [18F]FDS PET and kinetic modelling provides a readily available and sensitive method for non-invasive determination of BBB permeability in vivo.
2021-101
Relative cerebral blood flow, amyloid burden and cognition in individuals with subjetive cognitive decline are closely associated (#333)
Jarith Ebenau1, Denise Visser2, Sander C.J. Verfaillie2, Tessa Timmers2, Mardou van Leeuwenstijn1, Frederik Barkhof2, 3, Philip Scheltens1, Niels Prins1, 4, Ronald Boellaard2, Albert D. Windhorst2, Wiesje M. van der Flier1, 5 and Bart N.M. van Berckel1, 2
1Amsterdam UMC, Alzheimer Center, Amsterdam, Netherlands
2Amsterdam UMC, Deparment of Radiology & Nuclear Medicine, Amsterdam, Netherlands
3UCL, London, UK
4Brain Research Center, Amsterdam, Netherlands
5Amsterdam UMC, Department of Epidemiology & Biostatistics, Amsterdam, Netherlands
Abstract
Introduction: The role of cerebral blood flow (CBF) in the early stages of Alzheimer’s disease (AD) is complex and studies provide conflicting results regarding its relationship with AD biomarkers. We investigated the cross-sectional and longitudinal relationships between amyloid burden, relative CBF (rCBF) and cognition, in individuals with subjective decline (SCD).
Methods: We included 187 individuals with SCD from the SCIENCe cohort (64 ± 8y, 39%F, MMSE 29 ± 1). Each underwent a dynamic [18F]florbetapir PET and T1-weighted MRI scan, enabling calculation of mean binding potential (BPND) and R1 (measure of rCBF) in frontal, temporal, parietal, occipital and composite regions-of-interest. Eighty-three individuals underwent a second [18F]florbetapir PET (mean difference 2.6 ± 0.7y). Individuals annually underwent neuropsychological testing (follow-up time 3.8 ± 3.1y).
Results: Spearman’s correlation analyses showed a positive association between baseline BPND and R1 in occipital regions, but not in other regions (Figure 1). Linear mixed models (LMM) showed that a low baseline R1 was associated with an increase in BPND in frontal, temporal and composite regions. Vice versa, a high baseline BPND was associated with a decline in R1 in all regions (Table 1). Furthermore, a low baseline R1 was associated with a worse cognitive trajectory for tests for memory, attention and global cognition (range betas 0.01 to 0.28, p < 0.05). High BPND was associated with a worse cognitive trajectory for tests covering all domains (range betas -0.01 to -0.70, p < 0.05). When both predictors were simultaneously added to the model, associations remained comparable.
Conclusion: BPND and R1 were both associated with cognitive test performance over time in individuals with SCD. Furthermore, their longitudinal trajectories were closely associated. Although our results do not provide evidence for a causal relationship, they suggest two pathways in the development of AD, starting with low rCBF or high amyloid burden, which are influenced by each other, and could eventually lead to further deterioration.
2021-102
Genetically identical twins show comparable tau PET load and spatial distribution (#334)
Emma M. Coomans1, 2, Jori Tomassen2, Rik Ossenkoppele2, 3, Sandeep S.V. Golla1, Marijke E. den Hollander1, Lyduine E. Collij1, Emma Weltings1, Sophie van der Landen2, Emma E. Wolters1, Albert D. Windhorst1, Frederik Barkhof1, 4, Eco J.C. de Geus5, Philip Scheltens2, Pieter Jelle Visser2, 6, Bart N.M. van Berckel1 and Anouk den Braber2, 5
1Vrije Universiteit Amsterdam, Amsterdam UMC, Department of Radiology & Nuclear Medicine, Amsterdam Neuroscience, Amsterdam, Netherlands
2Vrije Universiteit Amsterdam, Amsterdam UMC, Alzheimer Center Amsterdam, Department of Neurology, Amsterdam Neuroscience, Amsterdam, Netherlands
3Lund University, Clinical Memory Research Unit, Lund, Sweden
4UCL, Institute of Neurology, London, UK
5Vrije Universiteit Amsterdam, Department of Biological Psychology, Amsterdam, Netherlands
6Maastricht University, Alzheimer Center Limburg, School for Mental Health and Neuroscience, Maastricht, Netherlands
7Karolinska Institutet, Department of Neurobiology, Care Sciences and Society, Division of Neurogeriatrics, Stockholm, Sweden
Abstract
Introduction: Tau accumulation starts during the preclinical phase of Alzheimer’s disease and is closely associated with cognitive decline. For preventive purposes, it is important to identify factors associated with tau accumulation and spread. Studying genetically identical twin-pairs may give insight into genetic and environmental contributions to tau pathology, as similarities in identical twin-pairs largely result from genetic factors, while differences can be attributed to non-shared, environmental factors. This study aimed to examine genetic and environmental contributions to 1) tau load and 2) spatial distribution of tau in a sample of cognitively unimpaired genetically identical twins.
Methods: We included 78 twins (39 pairs), aged 73.4 ± 5.9 years (51.3% female), who underwent dynamic [18F]flortaucipir-PET. We extracted binding potentials (BPND) in entorhinal (Braak-I), temporal (Braak-III-IV), neocortical (Braak-V-VI) and global (Braak-I-VI) regions, and examined within-pair similarities in BPND using intra-class correlations (Figure 1(a)). Furthermore, we correlated each participant’s voxel-by-voxel [18F]flortaucipir spatial distribution to that of every other participant using Spearman correlation models and tested, using an independent t-test, whether the average correlation coefficient obtained for twin-pairs was significantly higher (i.e. more similar spatial distribution) than for non-twin pairings of participants (Figure 2(a)). Last, we explored whether environmental (e.g. physical activity, obesity) risk factors could explain observed within-pair differences in [18F]flortaucipir-BPND.
Results: [18F]flortaucipir-BPND was correlated within twin-pairs in entorhinal (r = 0.40; p = 0.01), neocortical (r = 0.59; p < 0.01) and global (r = 0.56; p < 0.01) regions, but not in the temporal region (r = 0.20; p = 0.10) (Figure 1(b)). The [18F]flortaucipir distribution pattern was significantly more similar between twins of a pair (mean r = 0.27; SD = 0.09) than between non-twin pairings of participants (r = 0.01; SD = 0.10) (p < 0.01) (Figure 2(b)). Finally, within-pair differences in [18F]flortaucipir-BPND were associated with within-pair differences in depressive symptoms (0.37 < β < 0.56), physical activity (-0.41 < β < −0.42) and social activity (-0.32 < β < −0.36) (all p < 0.05).
Conclusion: Overall, identical twin-pairs were comparable in tau load and spatial distribution, highlighting the important role of genetic factors in the accumulation and spreading of tau pathology. Considering also the presence of dissimilarities in tau pathology in identical twin-pairs, our results additionally support a role for (potentially modifiable) environmental factors in the onset of Alzheimer’s disease pathological processes, which may be of interest for future prevention strategies.
2021-103
Assessing neurobiological correlates of [11C]WAY-100635 two-tissue compartment model parameters using SiMBA (#335)
Granville J. Matheson1, 2 and R. Todd Ogden1, 2
1Columbia University, Molecular Imaging and Neuropathology Division, New York New York, USA
2Columbia University, Department of Biostatistics, New York New York, USA
Abstract
Introduction: We have introduced a new method called SiMBA, which exploits similarities between both individuals and regions within individuals while applying the traditional two-tissue compartment (2TC) model in order to improve the identifiability and estimation of model parameters. Another important advantage of this approach is that estimation of model parameters is performed at the levels of individuals, regions, and individual TACs, allowing us to study the influence of different levels in isolation.
Methods: We applied SiMBA to a sample of 97 individuals measured using [11C]WAY100635 using ten regions, and extracted regional estimates of measurement error, as well as blood volume fraction (vB) and non-displaceable distribution volume (VND) from the 2TC. We assessed correlations between these estimated parameters and regional neurobiological attributes derived from average region size, as well as using the defined regions in MNI space to assess mean regional vascular partial volume from the VENAT atlas of vascular neuroanatomy1 and myelin concentrations from the MWF atlas.2
Results: Firstly, it is widely known that smaller regions have greater measurement error, and that the magnitude of this relationship is useful for regional weighting. We observe a strong linear association between the natural logarithms of region size and estimated measurement error (β = −0.23, r = −0.96, p < 0.001), estimating that for every doubling of region size, the measurement error decreases by 15%. Next, we show a significant positive correlation between mean VENAT (1) regional vasculature partial volume and 2TC estimates of vB (r = 0.70, p = 0.026). Lastly, examining 2TC estimates of VND, we observe a two-fold variance across regions (σ = 23%, 95% CI: 16–34%), with highest values in the parahippocampus (150% of the mean), and the lowest values in the cerebellum (72%). We show a significant positive correlation between MWF (2) regional myelin concentrations and VND estimates (r = 0.85, p = 0.008).
Conclusion: The current investigation not only sheds light on the neurobiological interpretation of the parameters of the 2TC in [11C]WAY100635, but also serves to validate the performance and estimates of SiMBA using external sources.
Acknowledgements
The work reported here has been partially supported by US NIH grants 5 P50 MH090964 and 5 R01 EB024526, and by the Hjärnfonden Postdoctoral Fellowship.
References
HuckJWannerYFanAP, et al.High resolution atlas of the venous brain vasculature from 7 T quantitative susceptibility maps. Brain Struct Funct2019;
224: 2467–2485.DvorakAVSwift-LaPointeTVavasourIM, et al.An atlas for human brain myelin content throughout the adult life span. Sci Rep2021;
11: 269.
2021-104
[18F]FLUDA – A novel radiotracer for PET imaging of the adenosine A2A receptor (A2AR) (#337)
Thu Hang Lai1, 2, Magali Toussaint1, Rodrigo Teodoro1, Daniel Gündel1, Friedrich-Alexander Ludwig1, Sladjana Dukić-Stefanović1, Barbara Wenzel1, Susann Schröder2, Bernhard Sattler3, Rareş-Petru Moldovan1, Osama Sabri3, Winnie Deuther-Conrad1 and Peter Brust1
1Helmholtz-Zentrum Dresden-Rossendorf, Institute of Radiopharmaceutical Cancer Research, Department of Neuroradiopharmaceuticals, Leipzig Saxony, Germany
2ROTOP Pharmaka GmbH, Department of Research and Development, Dresden Saxony, Germany
3University Hospital Leipzig, Department of Nuclear Medicine, Leipzig Saxony, Germany
Abstract
Introduction: Selective A2AR antagonists have emerged as potential therapeutics for multiple diseases. With regard to Parkinson’s disease, adjunctive treatment of A2AR antagonists potentially reduces adverse effects of long-term L-DOPA treatment. Therefore, imaging of receptor availability during the A2AR-tailored therapy is of utmost importance. We recently developed [18F]FLUDA as an novel A2AR-specific PET radiotracer.1
Methods: [18F]FLUDA was synthesized by an automated procedure Biological evaluation was performed in healthy mice and piglets. In vitro autoradiography was performed with brain cryosections. In vivo metabolism was analysed by radio-HPLC of plasma and brain homogenate. Pharamcokinetics and biodistribution was assessed by dynamic PET imaging under control and blocking conditions (2.5 mg/kg tozadenant and/or 1.0 mg/kg istradefylline). SUV ratio (SUVr) of striatum-over-cerebellum was used as a metric for specific uptake. A single dose acute toxicity study was performed in Wistar rats according to the ICH guideline M3(R2). Radiation dosimetry was investigated in piglets.
Results: In vitro autoradiography revealed an A2AR affinity (KD) of 4.3 and 0.7 nM and an A2AR density (Bmax) of 556 and 218 fmol/mg in the striatum of mice and piglets. No radiometabolites were detected in the mouse brain at 15 min p.i., whereas radiometabolites were found in piglet plasma but are assumed to not cross the blood-brain barrier. PET demonstrated high specific binding of [18F]FLUDA in both species (Figure 1). Toxicity studies revealed no adverse effects up to a dose of 30 µg/kg (∼4000-fold of expected human dose). The ED to humans is 16.4 µSv/MBq, which is in the range of other 18F-labeled radiotracers.2
Conclusion: We have demonstrated that [18F]FLUDA is suitable for determination of the A2AR availability in the striatum. No safety concerns are expected upon administration of [18F]FLUDA according to toxicity and dosimetry data. These results encourage the clinical translation of [18F]FLUDA.
Acknowledgements
This work (Project No. 100226753) has been funded by the European Regional Development Fund (ERDF) and Sächsische Aufbaubank (SAB).
References
LaiTHToussaintMTeodoroR, et al.Improved in vivo PET imaging of the adenosine A2A receptor in the brain using [18F]FLUDA, a deuterated radiotracer with high metabolic stability. Eur J Nucl Med Mol Imaging2021;
48: 2727–2736.SattlerBKranzMLaiTH, et al.Preclincal incorporation dosimetry of [18F]FLUDA – a novel 18F-labeled tracer for PET imaging of the expression of the adenosine A2A receptor (A2AR).J Nucl Med2020;
61 (supplement 1): 1014.
2021-105
Development of a non-invasive PET/MRI method for quantifying cerebral glucose kinetics (#338)
Lucas Narciso1, 2, Alaa Taha1, Praveen Dassanayake1, 2, Tommaso Volpi3, 4, Linshan Liu1, Andrea Soddu5, Udunna Anazodo1, 2, Alessandra Bertoldo3, 6 and Keith St Lawrence1, 2
1Lawson Health Research Institute, London, ON, Canada
2Western University, Department of Medical Biophysics, London, ON, Canada
3University of Padua, Padova Neuroscience Centre, Padua, Italy
4University of Padua, Department of Neuroscience, Padua, Italy
5Western University, Department of Physics, London, ON, Canada
6University of Padua, Department of Information Engineering, Padua, Italy
Abstract
Introduction: Quantifying the kinetics of glucose metabolism by dynamic [18F]FDG PET is not commonly performed because of invasive nature of measuring the arterial input function (AIF).1 Extracting an image-derived input function (IDIF) by vessel segmentation is one alternative but is sensitive to partial volume errors (PVEs).1 We present a simultaneous estimation (SIME) approach that avoids PVEs by deriving the AIF from multiple brain time-activity curves (TACs). Unlike previous SIME methods that model the IDIF as a series of exponentials,2 our method derives the IDIF from the two-compartment FDG model, which reduces the number of fitting parameters, and incorporates late timepoints from the superior sagittal sinus (SSS) TAC to constrain the optimization. The IDIF was combined with a Variational Bayesian (VB) method to generate images of the microparameters.3
Methods: Sixty min of dynamic [18F]FDG data were acquired from 10 healthy individuals on a Siemens Biograph mMR system. PET images were reconstructed using OSEM. Each IDIF was extracted from the whole-brain TAC by applying SIME to 10 TACs obtained by k-means clustering (500 iterations, 10 replicates; same clusters used for VB). SSS activity was obtained by MRI vessel segmentation,4 and one venous blood sample was collected at 60 min for comparison. Only SSS timepoints ≥ 45 min were used to ensure arterial-to-venous [18F]FDG equilibrium.5 Parametric maps of FDG microparameters were obtained from VB analysis using a lumped constant of 0.65 to calculate the cerebral metabolic rate of glucose (CMRGlu).
Results: No significant difference was found between venous blood and SSS activity (3.82 ± 0.57 vs 4.09 ± 1.02 kBq/ml). IDIFs were successfully extracted for all subjects. Figure 1 includes average microparameter values obtained by SIME. Figure 2 presents parametric maps of the rate constants and CMRGlu.
Conclusion: In this preliminary analysis, microparameter estimates from SIME were in good agreement with literature values,1 and parametric maps were successfully generated with the VB approach. However, discrepancies were found between the SIME and VB estimates, indicating further work is required regarding optimal priors and clustering for VB analysis, as well as possible influence of noise in the IDIFs.
References
Zanotti-FregonaraPChenKLiowJS, et al.Image-derived input function for brain PET studies: many challenges and few opportunities. J Cereb Blood Flow Metab2011;
31: 1986–1998.WongKPFengDMeikleSR, et al.Simultaneous estimation of physiological parameters and the input function – in vivo PET data. IEEE Trans Inf Technol Biomed2001;
5: 67–76.CastellaroM, et al.A variational Bayesian inference method for parametric imaging of PET data. Neuroimage2017;
150: 136–149.DassanayakeP, et al.CALIPER: a software for blood-free parametric Patlak mapping using PET/MRI input function. bioRxiv2021; 2021.07.08.451713.ChenK, et al.Noninvasive quantification of the cerebral metabolic rate for glucose using positron emission tomography, 18F-fluoro-2-deoxyglucose, the Patlak method, and an image-derived input function. J Cereb Blood Flow Metab1998;
18: 716–723.
2021-106
Positron emission tomography with [18F]ROStrace reveals increased oxidative stress in a mouse model of alpha synuclein aggregation (#339)
Evan Gallagher1, Yi Zhu2, Meagan McManus2, Kelvin C. Luk3 and Robert H. Mach4
1University of Pennsylvania, Neuroscience Graduate Group, Philadelphia Pennsylvania, USA
2Children’s Hospital of Philadelphia, Center for Mitochondrial and Epigenomic Medicine, Philadelphia Pennsylvania, USA
3Perelman School of Medicine, Department of Pathology and Laboratory Medicine, Philadelphia Pennsylvania, USA
4Perelman School of Medicine, Department of Radiology, Philadelphia Pennsylvania, USA
Abstract
Introduction: Parkinson’s disease (PD) is a debilitating neurodegenerative disorder characterized in part by the progressive accumulation of insoluble intraneuronal inclusions called Lewy Bodies (LBs). LBs are largely composed of oxidatively modified alpha-synuclein (aSyn), and oxidative stress is thought to be both a mediator and a consequence of aSyn aggregation in PD. Here, we aim to determine whether the recently developed positron emission tomography (PET) radioligand [18F]ROStrace1 is capable of detecting increases in oxidative stress in vivo in a well-characterized mouse model of aSyn aggregation.
Methods: 3 month-old transgenic mice overexpressing the A53T mutant form of human aSyn (’A53T mice’)2 were stereotactically injected with preformed aSyn fibrils (PFFs) in the right dorsal striatum to induce aSyn aggregation3. Groups of PFF-injected mice were then scanned with ROStrace at 1, 2, or 3 months post-injection (MPI) alongside age- and sex-matched saline-injected controls. Following the PET scans, the mice were euthanized, and the brains were extracted and histologically examined for aSyn aggregates and other markers of oxidative damage.
Results: PFF-injected animals showed significantly higher whole-brain average ROStrace signal at 2MPI compared to saline-injected animals (Figure 1). This increase in signal at 2MPI was accompanied by widespread formation of pathological aSyn aggregates in PFF-injected but not saline-injected mice (Figure 2). In PFF-injected mice, the aSyn aggregates were predominantly found in neurons, including dopaminergic neurons in the substantia nigra (white arrows in Figure 2).
Conclusion: In this study, we show that PFF-injected A53T mice display increased oxidative stress in brain compared to saline-injected mice, as measured by the recently developed PET probe [18F]ROStrace. These results support the idea that oxidative stress is associated with disease progression in PD and suggest that ROStrace could develop into a valuable tool for studying redox biology and antioxidant drug efficacy in vivo.
References
ChuWChepetanAZhouD, et al.Development of a PET radiotracer for non-invasive imaging of the reactive oxygen species, superoxide, in vivo. Org Biomol Chem2014;
12: 4421–4431.GiassonBIDudaJEQuinnSM,
Neuronal α-synucleinopathy with severe movement disorder in mice expressing A53T human α-synuclein. Neuron2002;
34: 521–533.LukKCKehmVMZhangB, et al.Intracerebral inoculation of pathological α-synuclein initiates a rapidly progressive neurodegenerative α-synucleinopathy in mice. J Exp Med2012;
209: 975–986. https://doi.org/10.1084/jem.20112457
2021-107
Visual memory test equal to commonly used verbal memory test in predicting Tau in the medial temporal lobe (#340)
Nina M. Poltronetti1, 2, Jaime Fernandez Arias7, 2, Vanessa Pallen1, 2, Nesrine Rahmouni7, 2, Jenna Stevenson7, 2, Joseph Therriault7, 2, Sulantha Mathotaarachchi2, Yi-Ting Wang7, 2, Andréa L. Benedet4, 2, Cécile Tissot7, 2, Tharick A. Pascoal5, 2, Firoza Z. Lussier3, 7, Gleb Bezgin3, Serge Gauthier7, 2 and Pedro Rosa-Neto7, 3
Introduction: The Aggie Figures Learning Test (AFLT) is a visual memory assessment tool, constructed as an analog to the Rey Auditory Verbal Learning Test (RAVLT). The AFLT includes three sequences of abstract drawings as stimuli –a primary set which is repeated across five trials, an interference set, and a recognition set. After about twenty minutes, the test assesses free delayed recall (DR) and delayed recognition of the learned material. This test is very similar to the RAVLT in terms of administration; therefore, the present study will evaluate whether the predictive value of DR scores from both tests is comparable in relation to Tau PET.
Methods: Tau PET ([18F]-MK6240) was acquired for 130 individuals for analysis involving only AFLT, 139 for analysis involving only RAVLT, and 127 individuals for analysis including both AFLT and RAVLT. Demographic data is shown on Tables 1, 2, and 3. MRI were segmented into probabilistic grey (GM) and white (WM) maps, non-linearly registered to the ADNI template using Dartel and smoothed with an 8 mm FWHM gaussian kernel. Voxel-wise linear regression models were applied, using VoxelStats, with Tau PET as the dependent variable and either DR AFLT or DR RAVLT as predictors. We ran three analyses where one included both DR scores. Additionally, we corrected for age, sex, diagnosis, APOE, and amyloid load. All other variables that were tested did not significantly contribute to predict Tau PET.
Results: We found negative associations between tau binding in the medial temporal lobe and both DR scores when looking at each of them separately. More importantly, we found that DR AFLT scores lose their predictive value when including DR RAVLT in the same model.
Conclusion: Our results support the use of both tests interchangeably when testing memory in elderly populations, even though they tap into two different modalities. This piece of evidence confirms what was reported by the authors who published the original AFLT article. Nonetheless, that article dealt with a sample whose demographics and clinical presentation strongly differed from those of our sample.
Figure 1. MK delayed recall on RAVLT & AFLT.
MK ∼ DR_AFLT + DR_RAVLT + amyloid + age + sex + apoe + diagnosis
MK ∼ DR_RAVLT + DR_AFLT + amyloid + age + sex + apoe + diagnosis
Figure 2. MK delayed recall on AFLT & RAVLT.
MK ∼ DR_AFLT + amyloid + age + sex + apoe + diagnosis
MK ∼ DR_RAVLT + amyloid + age + sex + apoe + diagnosis
References
SziklasVJones-GotmanM.
RAVLT and nonverbal analog: French forms and clinical findings. Can J Neurol Sci/J Can Des Sci Neurolog2008;
35: 323–330.MajdanASziklasVJones-gotmanM.
Performance of healthy subjects and patients with resection from the anterior temporal lobe on matched tests of verbal and visuoperceptual learning. J Clin Exp Neuropsychol1996;
18: 416–430.AriasJFPascoalTABenedetAL, et al.Amyloid (a), tau (t) and voxel-based morphometry (n) correlates of visual memory performance.Alzheimer’s Dement2020;
16: e040689.
2021-108
3D reconstruction of 20 neurotransmitter receptor atlases from 2D autoradiographs (#341)
Thomas Funck1, Konrad Wagstyl2, Claude Lepage3, Paule-Joanne Toussaint3, Mona Omidyeganeh3, Karl Zilles1, Alexander Thiel4, Alan C. Evans3 and Nicola Palomero-Gallagher1, 5
1Forschungszentrum Jülich, INM-1, Julich North Rhine-Westphalia, Germany
2University College London, Wellcome Center for Human Neuroimaging, London, UK
3McGill University, Montreal Neurological Institute, Montreal, QC, Canada
4McGill University, Lady Davis Institute, Jewish General Hospital, Montreal, QC, Canada
5Heinrich-Heine-University, C. & O. Vogt Institute for Brain Research, Dusselforf, Germany
Abstract
Introduction: Quantitative atlases of neurotransmitter receptor densities in healthy and pathologic human brains are important for characterizing normal and pathologic brain function and behaviour. We present a 3D reconstruction pipeline for 2D autoradiographs1 that will allow for the creation of the first ever set of ultra-high resolution (50µm) 3D atlases for 20 different neurotransmitter binding sites in the human brain. This pipeline was designed to overcome significant challenges in the data, including: non-linear deformations in the brain tissue, intensity variations between autoradiographs, variability in autoradiograph acquisitions, a large number of slices lost in sectioning, and non-orthogonal brain sections.
Methods: The reconstruction pipeline (Figure 1) is composed of 4 major processing stages: 1) automated cropping of the autoradiographs, 2) inter-autoradiograph 2D alignment, 3) iterative multi-resolution 3D volumetric followed by 2D section-wise alignment of autoradiographs to the donor’s MRI, 4) surface-based interpolation of receptor densities.2–5 Two methods have been implemented to validate the quantitative accuracy of the reconstructed ligand values. Pixel intensities of sections in the reconstructed ligand volumes were compared to the raw autoradiographs. his is done to verify that the pipeline preserves the ligand binding densities. The second validation technique applies the surface-based interpolation algorithm within acquired autoradiographs.
Results: The reconstruction pipeline was successful in aligning the 2D sections to the donor’s MRI (Figure 2(a)). Some sections appear to extend over the cortical surfaces, indicating that some sections may be suboptimally aligned. The 3D cortical maps of receptor density are seen in Figure 2(b). Initial results indicate ∼90% accuracy for a flumazenil binding density volume at 0.6 mm resolution.
Conclusion: The methods presented here provide all the necessary processing steps to reconstruct a dataset of multiple neurotransmitter receptor atlases at 50µm for a full human brain. Moreover, as both hemispheres of 3 donor brains were acquired in Ref.1, we will therefore reconstruct 20 neurotransmitter binding site atlases for each of the 3 human brains. This will provide a novel and unparalleled set of atlases for the neuroscience research community that will help better elucidate the functional anatomy of the human brain.
References
ZillesK.
Transmitter receptors and functional anatomy of the cerebral cortex. J Anat2004;
205: 417–432.LepageC, et al. Human MR evaluation of cortical thickness using CIVET v2.1. OHBM (poster 4166), Vancouver, 2017.FunckT, et al.Surface-based partial-volume correction for high-resolution PET. Neuroimage2014;
102: 674–687.AvantsBB, et al.Symmetric diffeomorphic image registration with cross-correlation: evaluating automated labeling of elderly and neurodegenerative brain. Med Image Anal2008;
12: 26.RenkaRJ.
ALGORITHM 773: SSRFPACK: interpolation of scattered data on the surface of a sphere with a surface under tension. ACM Trans Math Software1997;
23: 437–439.
2021-109
Neuroimaging of M4 muscarinic acetylcholine receptors using [11C]MK-6884 in rhesus macaques (#342)
Vasily Belov1, Nicolas J. Guehl1, Sridhar Duvvuri2, Philip Iredale2, Sung-Hyun Moon1, Maeva Dhaynaut1, Peter A. Rice1, Daniel L. Yokell1, John Renger2, Georges El Fakhri1 and Marc D. Normandin1
1Massachusetts General Hospital, Harvard Medical School, Gordon Center for Medical Imaging, Boston Massachusetts, USA
2Cerevel Therapeutics, LLC, Boston Massachusetts, USA
Abstract
Introduction: Quantitative imaging of M4 muscarinic acetylcholine receptors (mAChR) is important due to implications of mAChR in both neurodegenerative and psychiatric disorders.1-3 Recently, a novel 11C-radiolabeled PET ligand ([11C]MK-6884), selectively binding to M4 mAChR as a positive allosteric modulator (PAM), was developed.4 Further detailed characterization of PET imaging and quantitative performance is required. The current study pursued this aim using a M4 PAM drug candidate (CVL-231) as a blocking agent.
Methods: Two rhesus macaques underwent six paired baseline-blocking 90-min PET scans. [11C]MK-6884 was synthesized with high molar activity (187.2 ± 68.7 GBq/µmol) and radiochemical purity (100%). CVL-231 i.v. administration started 10 min prior the blocking scan and lasted until the end of scan. Arterial blood samples were serially collected for radiometabolites and radioactivity concentration analysis. Both blood-based (1T2k, 2T4k, Logan plot, MA1) and reference region-based (SRTM, MRTM2 and Logan DVR) modelling methods were applied to calculate distribution volume (VT) and binding potential (BPND) in various cerebral regions (caudate nucleus, cerebellar grey matter, entire cortex, hippocampus, putamen, central white matter).5
Results: Blood data indicated the tracer’s fast bi-exponential plasma clearance (t1/2,fast = 2.3 ± 0.5 min and t1/2,slow = 66.7 ± 13.7 min), moderate metabolic stability (44.9 ± 7.6% parent in plasma at 30 min), and moderate binding to plasma proteins (plasma free fraction = 34 ± 10%). Estimates of regional VT by different blood-based methods were highly correlated (R2 > 0.95) and estimates of BPND by reference region methods were in strong agreement (R2 = 1.00). Only caudate nucleus and putamen displayed statistically significant and reproducible tracer uptake (striatum VT = 3.2 ± 0.5 ml/cc, BPND = 0.7 ± 0.3) and dose-dependent blockade by CVL-231. Truncation of the time interval to 60 min improved data stability and fitting robustness of all kinetic models. The data supported the use of the cerebellar grey matter as a reference region. Receptor occupancy determined by changes in BPND estimated by SRTM was in good agreement with Lassen plot for different blood-based methods (R2∼0.8), with dependence on the CVL-231 injected dose and plasma concentration described by a classical Hill dose-response function (ID50 = 1.1 ± 0.1 mg/kg and IC50 = 581.4 ± 54.9 ng/ml).
Conclusion: Kinetic modeling of [11C]MK-6884 data using arterial and reference region methods can quantify M4 mAChR availability in the brain. Receptor occupancy estimates demonstrate dose-dependent blockade by the structurally distinct PAM CVL-231.
References
Sowa DumondARGrossHKBohnenNI, et al.Classics in neuroimaging: imaging the cholinergic system with positron emission tomography. ACS Chem Neuroci2021;
12: 1472–1479.JeonWJDeanBScarrE, et al.The role of muscarinic receptors in the pathophysiology of mood disorders: a potential novel treatment?Curr Neuropharmacol2015;
13: 739–749.MoranSPMaksymetzJConnPJ.
Targeting muscarinic acetylcholine receptors for the treatment of psychiatric and neurological disorders. Trends Pharmacol Sci2019;
40: 1006–1020.TongLLiWLoMM-C, et al.Discovery of [11C]MK-6884: a positron emission tomography (PET) imaging agent for the study of M4 muscarinic receptor positive allosteric modulators (PAMs) in neurodegenerative diseases. J Med Chem2020;
63: 2411–2425.InnisRB, et al.Consensus nomenclature for in vivo imaging of reversibly binding radioligands. J Cereb Blood Flow Metab2007;
27: 1533−1539.
2021-110
Population-based input function for [11C]PBR28 quantification in non-human primates (#343)
Lucero G. Aceves-Serrano1, Vesna Sossi2 and Doris J. Doudet1
1University of British Columbia, Department of Medicine/Neurology, Vancouver British Columbia, Canada
2University of British Columbia, Department of Physics and Astronomy, Vancouverr British Columbia, Canada
Abstract
Introduction: A significant caveat of [11C]PBR28 studies is the need to compute an arterial input function (AIF) needing arterial cannulation for data quantification. Canulation makes scanning burdensome on the staff, uncomfortable for subjects, and very challenging in a clinical setting. Population-based input function (PBIF) can, in theory, be used to eliminate the need for arterial canulation. In this work, we evaluated the validity of different PBIF functions.
Methods: A total of 11 [11C]PBR28 scans were acquired from 7 non-human primates. We acquired arterial blood samples from 10 scans and venous blood samples from one scan. Two venous blood curves from a different [11C]PBR28 study were used for comparison between arterial and venous samples. AIFs were averaged to create a population template input function (TIF). PBIFs were created by scaling the TIF with injected activity per body weight (PBIF) or unmetabolized tracer activity in blood at 15-,30- and 60-minutes post-injection (PBIF15, PBIF30, PBIF60). Another population-derived function was created using total activity in plasma (in anticipation of an image-derived input function) and a template parent fraction curve scaled using metabolism at t = 30 (TPF30-IF). Total distribution volumes (VT) estimates were calculated using PBIF, PBIF30, PBIF15, PBIF60, TPF30-IF, and the individual AIF (VTAIF).
Results: VT-s computed using PBIF15 and PBIF30 showed the best correlation with VTAIF (r > 0.90), while VT derived from PBIF60 and TPF30-IF yield lower correlation (r = 0.76 and r = 0.69). The blood-free scaled PBIF showed the worst correlation (r = 0.46). Venous and arterial metabolite fractions showed comparable values (Figure 1).
Conclusion: Population-based input functions scaled with a single blood sample might be a useful alternative to using AIF to compute [11C]PBR28 binding in healthy NHPs. These results are in line with previous human work,1, 2 where authors demonstrated that a PBIF scaled using a blood sample (at 27min post-injection) outperformed the PBIF scaled using activity dose/weight.1 Taken together this suggests, single blood samples could be used instead of multiple sampling, reducing the invasiveness of [11C]PBR28 scanning. Single samples could be acquired using an arterial stick, which is significantly less invasive and technically demanding than arterial cannulation.
Acknowledgements
We thank the UBC/TRIUMF PET program staff for their contribution to this work: this study would not have been possible without Carolyn English and Siobhan McCormick’s assistance for the metabolite analyses radiochemistry laboratories at TRIUMF. Special thanks are due to Julian Kaye and the UBC Animal Care Facilities personnel for their animals’ outstanding care.
References
BuchertR, et al.Reliable quantification of 18F-GE-180 PET neuroinflammation studies using an individually scaled population-based input function or late tissue-to-blood ratio.Eur J Nucl Med Mol Imaging2020;
47: 2887–2900.Zanotti-FregonaraP, et al.Population-based input function and image-derived input function for [11C](R)-rolipram PET imaging: methodology, validation and application to the study of major depressive disorder. Neuroimage2012;
63: 1532–1541.
2021-111
[11C]MC1 has adequate sensitivity to measure low density cyclooxygenase 2 (COX-2) in healthy human brain (#344)
Xuefeng Yan, Andrea Zhang, Sami S. Zoghbi, Jeih-San Liow, Cheryl Morse, Min-Jeong Kim, Maria Ferraris Araneta, Jose A. Montero Santamaria, Lester Manly, Madeline Jenkins, Maia Van Buskirk, Bruny Kenou, Sara Rubovits, William Miller, Victor W. Pike, Paolo Zanotti-Fregonara and Robert B. Innis
National Institutes of Health, Molecular Imaging Branch/National Institute of Mental Health, Bethesda Maryland, USA
Abstract
Introduction: Cyclooxygenase enzymes (COXs) are important targets for neuroinflammation. [11C]MC1, a selective and high-affinity radioligand for COX-2, can detect elevated COX-2 in nonhuman primates after intracerebral lipopolysaccharide (LPS) injection and in patients with rheumatoid arthritis at the symptomatic joints1. This study investigated whether COX-2 can be measured by [11C]MC1 in healthy human brain.
Methods: Ten healthy participants were injected with [11C]MC1 (730 ± 48 MBq). Each participant received two 120-minute brain PET scans, two hours apart—a baseline scan in the morning followed by a blockade scan with celecoxib (600 mg p.o.), a preferential COX-2 inhibitor. Both scans were performed with concurrent arterial sampling. Enzyme binding was calculated as total distribution volume corrected for free parent fraction in plasma (VT/fp) using a two-tissue compartment model (2TCM). Receptor occupancy was determined via Lassen plot.
Results: After [11C]MC1 injection, the concentration of brain radioactivity peaked at 4.0 SUV at ∼2.8 minutes and declined to 16% of the peak at 120 minutes. The brain time-activity curve was reasonably well-fitted by a 2TCM. VTs were stable after 40 minutes. However, the measured VT values after 60 minutes were higher than the fitted values, possibly reflecting accumulation of a small amount of radiometabolites in brain. Nine participants showed specific uptake (VS), which was only ∼20% of total uptake. Based on the Lassen plot, celecoxib occupied 72% of available COX-2 in the brain. The occipital cortex, insula, and prefrontal cortex had the highest VS. The distribution of VS correlated well with regional mRNA transcripts of the COX-2 gene.
Conclusion: [11C]MC1 uptake in healthy human brain reflected specific binding to COX-2 based on blockade by celecoxib and on the correlation between regional VS values and the gene transcript. Together, the results demonstrate that [11C]MC1 has adequate sensitivity to measure the low density of constitutively expressed COX-2 in healthy participants.
Acknowledgements
Intramural Research Program at the National Institute of Mental Health
2021-112
Combining plasma p-Tau181 and p-Tau231 enhances Alzheimer’s disease in vivo classification (#346)
Pâmela C. Lukasewicz Ferreira1, Wagner S. Brum2, João Pedro Ferrari-Souza1, 2, Cécile Tissot3, Bruna Bellaver1, 2, Joseph Therriault3, Nicholas J. Ashton4, Andréa L. Benedet3, 5, Stijn Servaes3, Firoza Z. Lussier3, Mira Chamoun3, Jenna Stevenson3, Nesrine Rahmouni3, Serge Gauthier3, Eugeen Vanmechelen3, Henrik Zetterberg4, Kaj Blennow6, Eduardo R. Zimmer2, Thomas Karikari1, 5, Pedro Rosa-Neto3 and Tharick A. Pascoal1
1University of Pittsburgh, Department of Psychiatry, Pittsburgh Pennsylvania, USA
2Universidade Federal do Rio Grande do Sul (UFRGS), Graduate Program in Biological Sciences: Biochemistry, Porto Alegre, Brazil
3McGill University, Montreal, QC, Canada
4University of Gothenburg, Wallenberg Centre for Molecular and Translational Medicine, Gothenburg, Sweden
5University of Gothenburg, Department of Psychiatry and Neurochemistry, Institute of Neuroscience and Physiology, The Sahlgrenska Academy, Gothenburg, Sweden
6Sahlgrenska University Hospital, 11 Clinical Neurochemistry Laboratory, Mölndal, Sweden
Abstract
Introduction: Pathophysiological detection of Alzheimer’s disease (AD) is commonly made using cerebrospinal fluid (CSF) and positron emission tomography (PET). However, these methods are invasive and expensive, limiting their use in clinical settings. Blood-based biomarkers, mainly measuring phosphorylated tau (p-Tau) epitopes, have shown promising results to detect AD in a scalable way. However, it is still unknown whether the use of different plasma p-Tau epitopes offers overlap or complementary information. Here, we tested the hypothesis that plasma p-Tau181 and p-Tau231 markers provide complementary information to each other to identify AD pathophysiology.
Methods: Plasma p-Tau181, p-Tau231, [18F]AZD4694 amyloid-PET, [18F]MK6240 tau-PET, Magnetic resonance imaging (MRI) and cognitive assessment from 284 individuals [30 CU young adults, 155 CU, 54 MCI and 38 AD] were obtained from the McGill TRIAD cohort. The individuals with AD or MCI clinical diagnosis were classified as CI. Cut-points (p-tau181, 11.1 ng/mL; p-tau231, 11.8 ng/mL) were derived using young CU controls as reference. Individuals were group based on these two biomarkers. Then, imaging and cognition measurements were compared across groups.
Results: Plasma p-Tau231 and p-Tau p-Tau181 showed moderate correlation with each other (r = 0.64, p < 0.0001) (Figure 1). The p-Tau231+/p-Tau181+ group presented higher tau PET and amyloid PET uptake compared to groups negative for both or only one p-tau biomarker (all p < 0.01) (Figure 2(a) and (b)). Similarly, p-Tau231+/p-Tau181+ group showed worse cognitive performance and hippocampal atrophy than individuals with at least one p-tau biomarker negative (Figure 2(c) and (d)).
Conclusion: Individuals positive for both p-Tau231 and p-Tau181 showed worse cognitive performance, lower hippocampal volume, higher amyloid and tau burden than individuals with only one abnormal marker. Our findings show that both p-Tau epitopes are promising biomarkers and seem to provide complementary information and could be used to enhancing AD in vivo detection, with potential applications to trial screening and recruitment.
Acknowledgements
The authors thank all participants of the study and staff of the McGill Center for studies in Aging.
2021-113
Disrupted association between Mu-opioid receptor levels and resting state activity in patients with schizophrenia: Multimodal imaging study with [11C]-carfentanil PET and resting state fMRI (#347)
Ekaterina Shatalina1, Abhishekh H. Ashok3, 4, Matthew Wall5, 6, Jim Myers6, Tiago Reis Marques2, 1, Eugenii A. Rabiner5, 2 and Oliver D. Howes1, 2
1Imperial College London, MRC London Institute of Medical Sciences, London, UK
2Kings College London, Institute of Psychiatry, Psychology and Neuroscience, London, UK
3Addenbrooke’s Hospital, Cambridge University Hospital NHS Foundation Trust, Department of Radiology, Cambridge, UK
4University of Cambridge, Department of Radiology, Cambridge, UK
5Invicro LLC, London, UK
6Imperial College London, Faculty of Medicine, London, UK
7Kings College London, Department of Neuroimaging, Institute of Psychiatry, Psychology and Neuroscience, London, UK
Abstract
Introduction: Mu opioid receptors (MOR) are involved in hedonic processing and their levels have been shown to be reduced in patients with schizophrenia, who have deficits in hedonic processing and aberrant hedonic network activity at rest. Particularly, preclinical studies have shown that activation of MOR system in amygdala can improve aversive or defensive behavioural states induced by stress or social rejection. However, precise role of MOR levels in amygdala and its association with resting neuronal activity in unknown. We set out to test whether MOR density may mediate spontaneous neural activity of the hedonic network at rest measuring by resting state functional MRI (fMRI) in patients with schizophrenia and in healthy controls.
Methods: 19 healthy controls and 18 patients with a DSM-5 confirmed diagnosis of schizophrenia underwent a [11C]-carfentanil PET scan to measure MOR availability and an 8-minute resting state fMRI scan. Dynamic PET data were corrected for attenuation, scatter, and motion, and were rigid body coregistered to the structural MRI with a neuroanatomical atlas applied to the PET image by non-linear deformation parameters. [11C]-carfentanil binding potential (BPND) values were quantified using the simplified reference tissue model (SRTM) with occipital lobe grey matter as the reference. FMRI data were pre-processed using standard methods followed by amplitude of low frequency fluctuations (ALFF) analyses. ALFF Z-values for regions comprising the hedonic network (striatum, amygdala, insular cortex, anterior cingulate cortex, orbitofrontal cortex) were compared between groups and correlated with [11C]-carfentanil BPND for corresponding regions.
Results: Both [11C]-carfentanil BPND and ALFF Z-values were comparable between patients and controls across all 5 regions (p > 0.01, adjusted for multiple comparisons). In the amygdala [11C]-carfentanil BPND was significantly negatively associated with ALFF in healthy controls (r = −0.623, p = 0.004), but not in patients (r = –0.032, p = 0.899). No other associations were seen in any other region in either group
Conclusion: Our results show that MOR density in the amygdala is negatively associated with spontaneous neural activity in this region at rest. Given the inhibitory role of MOR, our results may indicate that reduced spontaneous firing of the amygdala may be associated with increased MOR-mediated inhibition. In patients with schizophrenia, this mechanism may be disrupted by disease pathophysiology.
2021-114
Individual-level molecular connectivity of GABAA receptors: assessing the similarity of [11C]Ro15-4513 kinetics across brain regions (#348)
Tommaso Volpi1, 2, Erica Silvestri1, 3, Alexander Hammers4 and Alessandra Bertoldo1, 3
1University of Padova, Padova Neuroscience Center, Padova, Italy
2University of Padova, Department of Neuroscience, Padova, Italy
3University of Padova, Department of Information Engineering, Padova, Italy
4King’s College London & Guy’s and St Thomas’ PET Centre, School of Biomedical Engineering & Imaging Sciences, London, UK
Abstract
Introduction: Molecular connectivity (MC) is an emerging topic in neuroscience. Neurotransmitter MC is conventionally assessed as the across-subject correlation of positron emission tomography (PET) measurements, which leads to a population-level estimate1,2. Single-subject connectivity networks can also be calculated as the similarity between PET time-activity curves (TACs), but this has only been done for [18F]fluorodeoxyglucose ([18F]FDG) so far3. Here, we extract single-subject MC networks from [11C]Ro15-4513, a PET ligand with high affinity for the α5 subunit of the GABAA receptor.
Methods: 4 healthy males (41.5 ± 4.4 years), who were scanned on two separate days, underwent 90-min [11C]Ro15-4513 dynamic PET acquisitions (Siemens ECAT EXACT HR+, reconstruction grid: 4 × 15 s, 4 × 60 s, 2 × 150 s, 10 × 300 s, 3 × 600 s)4. PET data were parcellated using the grey matter-masked Hammers atlas (60 regions)5. The region-wise TACs were interpolated and standardized by: 1) z-scoring across regions and demeaning across time, 2) demeaning across regions and z-scoring across time, 3) dividing by mean TAC across regions3. Single-subject MC was calculated on standardized TACs as A) Pearson’s correlation, B) Euclidean distance, C) Cosine Similarity. Test-retest reproducibility of MC matrices was assessed via Pearson’s correlation.
Results: Figure 1 shows the median MC matrices for each standardization and estimation method: “networks” along the main diagonal (e.g., connections between temporal lobe regions, where there is high [11C]Ro15-4513 activity), and inter-hemispheric homotopic links (secondary diagonals) are evident. Despite some similarities across methods, it emerges that the choice of the standardization strategy is clearly non-trivial. The within-subject test-retest reproducibility is very high for all methods (Pearson’s R: 0.90–0.99).
Conclusion: We extracted single-subject [11C]Ro15-4513 MC, testing different standardization and estimation methods. While requiring validation to understand its functional underpinnings, this approach provides a simple way to assess a PET tracer’s kinetic structure and similarity across regions without performing kinetic modelling, and may hold promise as a biomarker of disease or drug response.
References
TuominenLNummenmaaLKeltikangas-JärvinenL, et al.Mapping neurotransmitter networks with PET: an example on serotonin and opioid systems. Hum Brain Mapp2014;
35: 1875–1884.VeroneseMMoroLArcolinM, et al.Covariance statistics and network analysis of brain PET imaging studies. Sci Rep2019;
9: 2496.TomasiDGShokri-KojoriEWiersCE, et al.Dynamic brain glucose metabolism identifies anti-correlated cortical-cerebellar networks at rest. J Cereb Blood Flow Metab2017;
37: 3659–3670.McGinnityCJRiaño BarrosDARossoL, et al.Test-retest reproducibility of quantitative binding measures of [11C]Ro15-4513, a PET ligand for GABA A receptors containing alpha5 subunits. NeuroImage2017;
152: 270–282.HammersAAllomRKoeppMJ, et al.Three-dimensional maximum probability atlas of the human brain, with particular reference to the temporal lobe. Hum Brain Mapp2003;
19: 224–247.
2021-115
The pandemic brain: neuroinflammation in healthy, non-infected individuals during the COVID-19 pandemic (#349)
Ludovica Brusaferri1, Zeynab Alshelh1, Daniel Martins2, 3, Minhae Kim1, Akila Weerasekera1, Hope Housman1, Erin J. Morrisey1, Paulina C. Knight1, Kelly A. Castro-Blanco1, Daniel Albrecht1, Chieh-En J. Tseng1, Nicole R. Zürcher1, Eva-Maria Ratai1, Oluwaseun Johnson-Akeju1, 5, Nathaniel D. Mercaldo1, Nouchine Hadjikhani1, 6, Mattia Veronese2, 4, Federico E. Turkheimer2, 3, Bruce R. Rosen1, Jacob M. Hooker1 and Marco L. Loggia1, 5
1Harvard University, Athinoula A. Martinos Center for Biomedical Imaging, Boston Massachusetts, USA
2King’s College London, Department of Neuroimaging, London, UK
3NIHR Maudsley Biomedical Research Centre, London, UK
4University of Padua, Department of Information Engineering, Padua, Italy
5Massachusetts General Hospital, Department of Anesthesia, Boston Massachusetts, USA
6University of Gothenburg, Sahlgrenska Academy, Gothenburg, Sweden
Abstract
Introduction: The impact of COVID-19 on human health extends beyond the morbidity and death toll directly caused by the SARS-CoV-2 virus. In fact, accumulating evidence indicates a global increase in the incidence of fatigue, brain fog and depression, including among non-infected, since the pandemic onset.1 Motivated by previous research linking these symptoms to neuroimmune activation in other pathological contexts,2–5 we hypothesized that subjects examined after the enforcement of lockdown/stay-at-home measures would demonstrate increased neuroinflammation.
Methods: We performed simultaneous brain Positron Emission Tomography/Magnetic Resonance Imaging in healthy volunteers either before (n = 57) or after (n = 15) the 2020 Massachusetts lockdown, using [11C]PBR28, a radioligand for the glial marker 18 kDa translocator protein. In a subset (n = 13 pre-lockdown; n = 11 post-lockdown), we also quantified brain (thalamic) levels of myoinositol (mIns), another putative glial marker. First, we compared [11C]PBR28 and mIns signals across pre- and post-lockdown cohorts. Then, to evaluate the possible clinical significance of our findings, we collected and analyzed a retrospective pandemic-specific questionnaire from the post-lockdown subjects. Finally, we investigated multivariate associations between the spatial pattern of [11C]PBR28 post-lockdown changes and constitutive brain gene expression in post-mortem brains (Allen Human Brain Atlas).
Results: Post-lockdown subjects showed increased [11C]PBR28 and mIns signals. These neuroinflammatory markers demonstrated elevated levels in cortical and subcortical regions including sensory, motor and higher order association areas, and white matter. Post-lockdown [11C]PBR28 signal elevations were correlated with individual physical/mental fatigue indicators. Further, the regional variability of increased [11C]PBR28 signal was spatially aligned with the constitutive expression of several genes highly expressed in glial/immune cells and/or involved in neuroimmune signaling.
Conclusion: Our results of markedly increased neuroimmune response associated with COVID-19-related disruptions highlight the impacts of the pandemic and suggest that brain alterations may have occurred in response to pandemic-associated stressors in non-infected individuals. This work lays the foundations for the investigation of possible long-term effects of pandemic-induced neuroinflammation on humans’ health.
Acknowledgements
This work was funded by R01-NS094306-01A1, R01-NS095937-01A1, R01-DA047088-01 and The Landreth Family Foundation.
References
HolmesEA, et al.Multidisciplinary research priorities for the COVID-19 pandemic: a call for action for mental health science. Lancet Psychiatr2020;
7: 547–560.AlbrechtDS, et al.The neuroinflammatory component of negative affect in patients with chronic pain.Mol Psychiatr2021;
26: 864–874.NakatomiY, et al.Neuroinflammation in patients with chronic fatigue syndrome/myalgic encephalomyelitis: an 11C-(R)-PK11195 PET study. J Nucl Med2014;
55: 945–950.WeerasekeraA, et al.Thalamic neurometabolite alterations in patients with knee osteoarthritis before and after total knee replacement. Pain2021;
162: 2014–2023.AlbrechtDSGranzieraCHookerJM, et al.In vivo imaging of human neuroinflammation. ACS Chem Neurosci2016;
7: 470–483.
2021-116
Imaging nociceptive opioid peptide (NOP) receptors in alcohol use disorders (AUD) with [11C]NOP-1A and PET: Findings from a second cohort (#350)
Savannah Tollefson1, Clara Stoughton1, Michael Himes1, Neale S. Mason1 and Rajesh Narendran1, 2
1University of Pittsburgh, School of Medicine, Department of Radiology, Pittsburgh Pennsylvania, USA
2University of Pittsburgh, School of Medicine, Department of Psychiatry, Pittsburgh Pennsylvania, USA
Abstract
Introduction: Nociceptin (N/OFQ), which binds to the NOP receptor, regulates anxiety, reward, and stress-resilience.1 Recent studies showing NOP antagonists ability to block alcohol reinstatement and relapse suggest a hyperactive N/OFQ-NOP receptor system in AUD.1,2 In a previous [11C]NOP-1A study, we found no significant differences in VT in non-treatment seeking AUD relative to controls (HC).3 Here, we used [11C]NOP-1A PET to measure VT in treatment-seeking AUD to relate to relapse.
Methods: [11C]NOP-1A VT was measured in 27 AUD and age, sex, and tobacco use matched HC. AUD were scanned following ∼10 days of abstinence confirmed with urine ethyl glucuronide (EtG). Recent heavy drinking (EtG greater than or equal to 30 pg/mg) prior to the scan was also quantified using hair samples. VT was measured with a two tissue-compartment kinetic analysis in eleven ROIs that regulate reward and stress behaviors. To document relapse, AUD subjects were followed for 12-weeks after PET using contingency management where they were received compensation to remain abstinent.
Results: [11C]NOP-1A VT was not significantly different between AUD and HC groups.
VT was significantly lower in hair EtG+AUD vs. EtG-AUD (Linear Mixed Model, EtG+, p = 0.049; region, p < 0.001; region X EtG, p < 0.001). Significant negative correlations between VT and the mean number of drinking days per week and drinks consumed per drinking day before PET were also present.
AUD subjects who relapsed (n = 8) and dropped out (n = 8) had significantly lower VT than AUD who abstained (n = 6) for 12-weeks (Linear Mixed Model, outcome, p = 0.012; region, p < 0.001; region X outcome, p < 0.001).
Conclusion: Lower NOP VT, which is related to heavy drinking in AUD, predicted relapse to alcohol during 12-week follow-up in which subjects received money to abstain. Assuming lower VT reflects increased N/OFQ levels, these data support the pursuit of NOP antagonists to reduce heavy drinking and promote abstinence in AUD.
Funding:
NIAAA RO1 AA025247
References
CiccocioppoR, et al.NOP-related mechanisms in substance use disorders. Handb Exp Pharmacol2019.PostA, et al.Proof-of-concept study to assess the nociceptin receptor antagonist LY2940094 as a new treatment for alcohol dependence. Alcohol Clin Exp Res2016;
40: 1935–1944.NarendranR, et al.Nociceptin receptors in alcohol use disorders: a positron emission tomography study using [(11)C]NOP-1A. Biol Psychiatr2018;
84: 708–714.
2021-117
In response to social acceptance, nucleus accumbens mu opioid receptor activation is associated with increased self-esteem and positive mood (#351)
Kathryn R. Hill1, Christine DeLorenzo1, Stephan F. Taylor2, R. Todd Ogden3, Ramin V. Parsey1 and David T. Hsu1,2
1Renaissance School of Medicine at Stony Brook University, Department of Psychiatry, Stony Brook New York, USA
2Michigan Medicine, University of Michigan, Department of Psychiatry, Ann Arbor Michigan, USA
3Columbia University Mailman School of Public Health, Department of Biostatistics, New York New York, USA
Abstract
Introduction: The mu opioid receptor (MOR) related biological underpinnings of rejection sensitivity (RS), a common feature of borderline personality disorder and major depressive disorder, have not yet been examined. We hypothesize that RS will be negatively associated with MOR activation during rejection and acceptance tasks. In an exploratory analysis, we assessed the relationships between MOR activation and changes in mood and self-esteem before and after the task.
Methods: RS was assessed with the Adult Rejection Sensitivity Questionnaire. Healthy participants, n = 75 (52% female), completed rejection and acceptance tasks during [11C]carfentanil positron emission tomography (PET) scans, with self-reported levels of self-esteem and mood collected during and after each task. Nondisplaceable binding potential (BPND) was calculated from bolus plus infusion delivery of tracer with occipital lobe reference-region based Logan graphical analysis. MOR activation (BPND neutral – BPND social task) in the regions of interest (amygdala, midline thalamus, anterior insula, and nucleus accumbens) were evaluated.
Results: RS was not related to regional MOR activation during acceptance or rejection tasks. MOR activation in the nucleus accumbens was positively correlated with increase in ratings of self-esteem (r = 0.35, uncorrected p = 0.0064) and also with the feelings “happy and accepted” (r = 0.31, uncorrected p = 0.0094) during the period between acceptance task administration and 5 minutes after the task completion. MOR activation was not associated with rejection task related self-esteem and negative mood.
Conclusion: Our previous publication1 demonstrates that this social rejection task induces MOR activation. We presently demonstrate that regional MOR activation during social rejection is not dependent on RS in healthy individuals, suggesting regional endogenous opioid response to social rejection is independent of this trait. MOR activation in the nucleus accumbens was associated with increased self-esteem and positive mood after experiencing social acceptance, warranting further investigation.
Reference
HsuDTSanfordBJMeyersKK, et al.Response of the µ-opioid system to social rejection and acceptance. Mol Psychiatr2013;
18: 1211–1217.
2021-118
[18F]Nifene binding to α4β2* nicotinic acetylcholinergic receptors is reduced in human hippocampus of postmortem Alzheimer’s disease brains (#352)
Jogeshwar Mukherjee, Christopher Liang, Anthony-David Campoy, Rommani Mondal, Reisha M. Ladwa and Oshini V. Keerthisinghe
University of California-Irvine, Preclinical Imaging, Radiological Sciences, Irvine California, USA
Abstract
Introduction: Nicotinic acetylcholinergic receptors (nAChRs), α4β2* may be adversely affected in the hippocampus (HP) in Alzheimer’s disease (AD).1 In our efforts to consider translational use of [18F]Nifene PET2 in AD, we have carried out quantitative autoradiographic evaluation of α4β2* nAChRs using HP (CA1/subiculum plus) from postmortem AD and compared to control (CN) brains.
Methods: Human post-mortem brain tissue sections (10 µm thick) consisting of HP CA1-subiculum regions, (AD, n = 16F and 13M and CN = 16F and 16M) were obtained from Banner Health, Sun City, Arizona. Brain slices were incubated [18F]Nifene (1 µCi/cc) in Tris/pH 7.4 buffer at 25°C for 1 hr. Nonspecific binding was measured using 300 µM nicotine. Adjacent sections were tested for Tau using [125I]IPPI (0.1 µCi/cc) in PBS pH 7.4 buffer at 25°C for 1 hr using our previously published protocols.3 Anti-tau and anti-Aβ immunostaining was carried out on adjacent slices. Using the Optiquant program, regions of interest were drawn and digital light units/mm2 (DLU/mm2) were used to quantify the percentage change in binding of [18F]Nifene and [125I]IPPI.
Results: All CN subjects exhibited significant [18F]Nifene binding (Figure 1(c)) in the HP CA1-Subiculum (SUB) regions consistent with our PET [18F]Nifene results in healthy human subjects.2 Significantly lower binding in AD (Figure 1(g)) resulted in a overall 40% reduction in [18F]Nifene AD HP CA1-SUB compared to CN accross all subjects (average [18F]Nifene ratio AD/CN = 0.58). Male-female [18F]Nifene HP CA1-SUB differences in the brain sections were not significant, both in CN and AD subjects. All AD subjects had higher [125I]IPPI Tau binding, Figure 1(f) compared to CN in Figure 1(b) (average [125I]IPPI ratio AD/CN = 7). As expected, [18F]Nifene binding in the AD subjects was inversely correlated to [125I]IPPI in HP CA1-SUB. Quantitative correlation of [18F]Nifene and [125I]IPPI is underway using recently reported methods.4 Additional correlations with anti-Aβ (Figure 1(h)) and [18F]flotaza5 for Aβ plaques are also underway.
Conclusion: [18F]Nifene binding is significantly reduced in HP CA1-SUB of the postmortem AD brain. [18F]Nifene PET may potentially have a complementary role along with ongoing NFT PET studies in AD subjects. Information on the status of the α4β2* nAChRs in AD may be useful in treatment planning using acetylcholinesterase inhibitors.
Acknowledgements
Research Support: NIH/NIA RF1 AG029479, UCI UROP (ADC, RM, RML), Banner Health Research Institute
References
ShimohamaSKawamataJ.Roles of nicotinic acetylcholine receptors in the pathology and treatment of Alzheimer’s and Parkinson’s diseases. In: AkaikeAShimohamaSMisuY (eds). Nicotinic acetylcholine receptor signaling in neuroprotection.
Singapore:
Springer, 2018, pp. 137–158.MukherjeeJLaoPBetthauserT, et al.Human brain imaging of nicotinic acetylcholine α4β2* receptors using [18F]Nifene: selectivity, functional activity, toxicity, aging effects, gender effects and extrathalamic pathways. J Comparative Neurol2018;
526: 80–95.MukherjeeJLiangCPatelKK, et al.Development and evaluation [125I]IPPI for tau imaging in post-mortem human Alzheimer’s disease brain. Synapse2021;
74: e22183.MondalRLadwaRMLiangL, et al. [125I]IPPI: A novel Tau radioligand for assessing neurofibrillary tangles in human hippocampus of postmortem Alzheimer’s disease subjects. In: Annual meeting of society of neuroscience, Chicago, Illinois, 13–17 November 2021, Abstract # 7939, Session # P941.KaurHFelixMRLiangC, et al.Development and evaluation [18F]Flotaza for Ab plaque imaging in post-mortem Alzheimer’s disease brain. Bioorg Med Chem Lett2021;
46: 128164.
2021-119
Imaging synaptic density in aging and frontotemporal dementia (#353)
Cécile Tissot1, 2, Gleb Bezgin1, João Pedro Ferrari-Souza2, 3, Pâmela C. Lukasewicz Ferreira2, Joseph Therriault1, Firoza Z. Lussier1, Mira Chamoun1, Yi-Ting Wang1, Jaime Fernandez-Arias1, Jenna Stevenson1, Nesrine Rahmouni1, Serge Gauthier1, Tharick A. Pascoal2 and Pedro Rosa-Neto1
1McGill University, Montreal, QC, Canada
2University of Pittsburgh, Pittsburgh Pennsylvania, USA
3Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil
Abstract
Introduction: Tissue loss has been observed in post-mortem data in aging and neurodegenerative conditions, such as frontotemporal dementia (FTD). In dementia, we can observe specific patterns of brain damage related to symptomatology. A hypothesis is that tissue degeneration is a consequence of synaptic loss. Our goal here is to assess synaptic density in vivo using the novel radiotracer [18F]SDM8 in aging and FTD.
Methods: We included cognitively unimpaired young (CUY) [21–22 years of age (yoa)], cognitively unimpaired elderlies (CUE) [60–81 yoa], and FTD [47–72 yoa] participants from the TRIAD cohort. They underwent an MRI, a neuropsychological evaluation, and a 90-minutes dynamic [18F]SDM8 PET-scan for synaptic density. We calculated the standardized uptake value ratio from 40–90 minutes and used the centrum semiovale as the reference region. Voxel-wise analyses were conducted to assess synaptic density differences between CUY, CUE, and FTD diagnostic groups. Analyses were always corrected for sex; when including only CUE and FTD, we also corrected for age.
Results: CUY individuals present high binding of [18F]SDM8 as compared to CUE individuals, especially in the precuneus, and the frontal and temporal lobes (Figure1(a)). Moreover, when comparing CUE and FTD individuals, we observed a greater [18F]SDM8 binding in the medial frontal lobe and the frontal pole (Figure 1(b)).
Conclusion: Synaptic loss seems to increase with age, and with dementia onset. Indeed, CUE show less synaptic density as compared to CUY, in regions vulnerable to the development of neurodegenerative conditions. Moreover, we also observed a strong difference between CUE and FTD individuals, in brain regions related to the disease symptomatology, i.e. the frontal lobe. Frontal synaptic loss seems to appear in aging around the orbitofrontal cortex. Conversely, in FTD, it is more located towards the dorsomedial prefrontal cortex. We hypothesize that synaptic loss is a process appearing in aging, which seems to be potentiated in individuals diagnosed with FTD.
Acknowledgements
The authors would like to acknowledge the help of the McGill Center for Studies in Aging staff, as well as the participants in the cohort.
Figure 1. Imaging synaptic density in aging and frontotemporal dementia.
2021-120
Associations between markers of synaptic dysfunction and tau accumulation, glial activation, and neurodegeneration in Alzheimer’s disease (#354)
Firoza Z. Lussier1, Stijn Servaes1, Cécile Tissot1, Joseph Therriault1, Min-Su Kang1, Gleb Bezgin1, Mira Chamoun1, Jenna Stevenson1, Nesrine Rahmouni1, Andréa L. Benedet2, Johanna Nilsson2, Nicholas J. Ashton2, Kaj Blennow2, Ann Brinkmalm2, Henrik Zetterberg2 and Pedro Rosa-Neto1
2Sahlgrenska Academy at the University of Gothenberg, Institute of Neuroscience and Physiology, Mölndal, Sweden
Abstract
Introduction: While Alzheimer’s disease (AD) is characterized by β-amyloid and tau, growing evidence suggests that synaptic dysfunction is a key feature of AD and may be a mechanism linking these pathologies to subsequent cognitive decline. The current aim was to investigate how levels of synapse proteins selected from a novel panel of synaptic biomarkers in cerebrospinal fluid (CSF) are associated with tau accumulation, glial activation, and neurodegeneration in an AD-enriched cohort.
Methods: Participants were selected from the TRIAD cohort (N = 105; 20 young, 44 cognitively normal, 41 cognitively impaired). Mass spectrometry methods were used to quantify pre- and post-synaptic proteins in CSF, including 14–3-3 zeta/delta, gamma-synuclein, neurogranin, and syntaxin and pentraxin proteins (Nilsson, 2021).1 Group comparisons and regression models with CSF levels of tau and glial activation markers were conducted, along with mediation analyses and correlations with AD-signature cortical thickness.
Results: Of the 12 quantified synaptic proteins, nine showed significant age effect and ten showed a significant difference in β-amyloid+/tau+ compared to β-amyloid-/tau-. Regressions with CSF p-tau231 and p-tau217 were significant for all synaptic proteins, as were regressions with CSF YKL-40 and GFAP. Mediation analysis with tau (predictor) and glial activation (mediator) showed that seven synaptic proteins’ associations with tau were significantly mediated by glial activation. These seven proteins matched those that were significantly correlated with AD-signature cortical thickness, with the exception of PEBP-1, which correlated with cortical thickness but did not have a significant mediation effect.
Conclusion: Our study of CSF synaptic dysfunction biomarkers in AD suggests that synaptic proteins whose relationship with tau is mediated by glial activation are more closely associated with reduced AD-signature neuronal degeneration assessed by cortical thickness.
Figure 1. Associations between synaptic dysfunction biomarkers and AD-signature cortical thickness
Reference
Nilsson J.
Cerebrospinal fluid biomarker panel for synaptic dysfunction in Alzheimer’s disease. Alzheimers Dement2021;
13: e12179.
2021-121
Detecting two-task dopamine release via residual analysis (#355)
Connor W.J. Bevington1, Jordan U. Hanania1, Giovanni Ferraresso2, Kevin (Ju-Chieh) Cheng1, 3, Alexandra Pavel3, A. Jon Stoessl3, 4 and Vesna Sossi1
1University of British Columbia, Physics and Astronomy, Vancouver British Columbia, Canada
2University of British Columbia, Department of Mechanical Engineering, Vancouver British Columbia, Canada
3University of British Columbia, Pacific Parkinson’s Research Centre, Vancouver British Columbia, Canada
4University of British Columbia, Faculty of Medicine, Division of Neurology, Vancouver British Columbia, Canada
Abstract
Introduction: Dopamine (DA) release can be detected in vivo using [11C]raclopride dynamic PET imaging: a voxel-level change in dopamine concentration alters the voxel-level PET temporal signal.1 Current detection methods suffer from low sensitivity (true positive rate), primarily due to image noise and relying on model comparison via an F-test which, at present, poorly separates true positives from true negatives.2 Additionally, the current methodology has only been investigated for single-task paradigms.3,4 Here we propose two alternative methods, one data-driven and the other task-informed, which aim to improve sensitivity and extend applicability to two-task paradigms. Sensitivity comparisons on simulated data are provided.
Methods: Denoised voxel-level signals are regressed on regional signals, producing structured residuals where DA release is present. The data-driven method then uses principal component analysis (PCA) to spatially localize the DA release, whereas the task-informed method uses known task timings to determine a predicted release timecourse, which is compared to the voxel-wise residuals via cross-correlation for spatial localization (Figure 1). We test these methods on two simulated datasets: (i) single-task: a single motor/cognitive task (∼200% basal DA increase in the L/R caudate and L putamen starting at 32 minutes); (ii) two-task: identical motor/cognitive task as above, followed by a reward task 17 minutes later (∼200% basal DA increase in the L ventral striatum). 20 noisy realizations are simulated in each case.
Results: Both methods increase true positive rate by ∼65% in the single-task simulation and ∼30% in the two-task simulation at equivalent false positive rate to the F-test method (Figure 2(a) and (b)). However, the task-informed method offers flexibility when release from multiple tasks are to be distinguished spatiotemporally, detecting the spatiotemporally distinct reward cluster with ∼2x higher sensitivity than the data-driven method (see white circles in Figure 2(c)).
Conclusion: Simulation results under varied task protocols show that our proposed data-driven and task-informed methods outperform the traditional F-test method in determining voxelwise patterns of DA release. These methods are general and can be adapted to signal detection in many domains. These methods are currently being applied to human data, comparing DA release behaviours of disease cohorts with healthy controls.
Acknowledgements
This work was supported by funding from the Natural Sciences and Engineering Research Council of Canada and the Pacific Parkinson’s Research Institute.
References
MorrisED, et al.ntPET: a new application of PET imaging for characterizing the kinetics of endogenous neurotransmitter release. Mol Imaging, 2005;
4:473–489.BevingtonCWJ, et al.A Monte Carlo approach for improving transient dopamine release detection sensitivity. JCBFM41: 116–131.KimSJ, et al.Voxelwise lp-ntPET for detecting localized, transient dopamine release of unknown timing: sensitivity analysis and application to cigarette smoking in the PET scanner. Hum Brain Mapp35: 4876–4891.WangS, et al.A framework for designing dynamic lp-ntPET studies to maximize the sensitivity to transient neurotransmitter responses to drugs: application to dopamine and smoking. Neuroimage146: 701–714.
2021-122
Using personalized longitudinal brain models to identify neurotransmitter receptor-mediated alterations in Alzheimer’s disease (#357)
Ahmed F. Khan2, 3, Quadri Adewale2, 3, Tobias R. Baumeister2, 3, Felix Carbonell4, Karl Zilles5, Nicola Palomero-Gallagher5, 6 and Yasser Iturria-Medina2, 3
1McGill University, Department of Neurology and Neurosurgery, Montreal Neurological Institute, Montreal, QC, Canada
2McGill University, McConnell Brain Imaging Center, Montreal Neurological Institute, Montreal, QC, Canada
3McGill University, Ludmer Centre for Neuroinformatics and Mental Health, Montreal, QC, Canada
4Biospective Inc., Montreal, QC, Canada
5Research Centre Jülich, Institute of Neuroscience and Medicine (INM-1), Jülich North Rhine-Westphalia, Germany
6Heinrich-Heine University, Cécile and Oskar Vogt Institute of Brain Research, Düsseldorf North Rhine-Westphalia, Germany
7RWTH Aachen, Department of Psychiatry, Psychotherapy, and Psychosomatics, Aachen North Rhine-Westphalia, Germany
8JARA, Translational Brain Medicine, Aachen North Rhine-Westphalia, Germany
9McGill University, Montreal Neurological Institute, Montreal, QC, Canada
Abstract
Introduction: Alzheimer’s disease (AD) involves degenerative alterations to several neurobiological processes,1 with inter-subject heterogeneity.2 While we do not currently understand its molecular basis, neurotransmitter receptors, as important signalling molecules affected in AD, are potential therapeutic targets. However, we currently lack integrative brain models characterizing how multiple biological processes and receptors interact to cause clinical deterioration. Here, we present a personalized, generative and whole-brain neurotransmitter receptor-enriched multifactorial causal model (re-MCM)3 of longitudinal neuroimaging changes. By interpreting model parameters as markers of receptor alterations, we are able to identify receptor-neuroimaging interactions associated with cognitive decline in AD.
Methods: We preprocessed longitudinal data from 6 neuroimaging modalities (structural, functional and arterial spin labeling MRI, and tau, amyloid-β and glucose PET) for a heterogeneous aged population from the ADNI dataset (N = 423). Using averaged (N = 3) receptor templates derived from post-mortem autoradiography for 15 neurotransmitter receptors from 6 key families (glutamatergic, GABAergic, cholinergic, adrenergic, serotonergic and dopaminergic receptors), we fit subject-specific regression models for each neuroimaging modality. Model parameters represent i) local interactions between neurobiological processes, ii) effects of local receptor densities, iii) functional alterations to receptor-based interactions, and iv) network propagation of neurobiological alterations. We then performed singular value decomposition (SVD) across AD patients with permutation testing to robustly identify parameters correlated with clinical measures of cognitive decline.
Results: Models with neuroimaging-receptor interactions were significantly more informative than neuroimaging-only models for 86.8% – 99.0% of subjects (F-test, P < 0.05), explaining 70% ( ± 20%) of the longitudinal variance in neuroimaging (across all subjects). We identified an axis of variability explaining up to 37% (P < 0.004, FWE-corrected) of inter-individual variability in AD cognitive deterioration, primarily affecting executive function, with functional alterations to glutamatergic interactions affecting tau accumulation and neural activity, GABAergic interactions concurrently affecting neural activity, amyloid and tau, and cholinergic effects on tau.
Conclusion: This project introduces the first data-driven framework for integrating neurotransmitter receptors, multi-modal neuroimaging and clinical data in an interpretable brain model. The estimation of receptor alterations despite the lack of in vivo data can allow the identification of candidate disease mechanisms and therapeutic targets, constituting a promising step towards personalized and precision neurotransmitter-based treatments.
Figure 1. Receptor-imaging interactions affecting cognitive decline in AD.
The angle of each sector is proportional to the contribution of the corresponding mechanism to explaining the variance in the rates of cognitive decline. The inner sectors represent the 6 neuroimaging modality models that together comprise each personalized re-MCM model. Within each modality, the intermediate sectors represent the neurotransmitter system involved, while the outer sector consists of the specific two-way receptor-neuroimaging interactions or direct predictor terms in the model. Together, the mechanisms robustly explain 37% of inter-subject variance in cognitive decline.
For each subject, longitudinal neuroimaging changes are decomposed into local effects from all (NROI = 88) brain regions due to i) the direct influence of all neuroimaging-quantified biological factors, ii) receptor density distributions, and iii) receptor-imaging interactions, and iv) global network-mediated intra-brain propagation. At a group level across AD subjects, receptor-based mechanisms robustly correlated to cognitive decline are identified using SVD.
References
FrancisPTMaríaJRLaiMK.
Neurochemical basis for symptomatic treatment of Alzheimer’s disease. Neuropharmacology2010;
59: 221–229.JackCRJrandCliffordR, et al.Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade. Lancet Neurol2010;
9: 119–128.YasserI-M, et al.Multifactorial causal model of brain (dis) organization and therapeutic intervention: application to Alzheimer’s disease. Neuroimage2017;
152: 60–77.
2021-123
Dosimetry and biodistribution of the novel PET radioligand (R)-[11C]Me-NB1 specific to the GluN2B subunit of the N-methyl-D-aspartate receptor (#358)
Lucas Rischka1, Verena Pichler2, 3, Chrysoula Vraka2, Ivo Rausch2, Dietmar Winkler1, Lukas Nics2, Sazan Rasul2, Leo Silberbauer1, Matej Murgaš1, Murray B. Reed1, Andreas Hahn1, Simon M. Ametamey4, Wolfgang Wadsak2, 5, Rupert Lanzenberger1 and Marcus Hacker2
1Medical University of Vienna, Department of Psychiatry and Psychotherapy, Vienna Wien, Austria
2Medical University of Vienna, Department of Biomedical Imaging and Image-guided Therapy, Division of Nuclear Medicine, Vienna Wien, Austria
3University of Vienna, Department of Pharmaceutical Sciences, Vienna Wien, Austria
4ETH Zurich, Centre for Radiopharmaceutical Sciences ETH-PSI-USZ, Institute of Pharmaceutical Sciences ETH, Vienna Zürich, Switzerland
5Center for Biomarker Research in Medicine (CBmed), Graz Wien, Austria
Abstract
Introduction: N-methyl-D-aspartate (NMDA) receptors are glutamate-gated ion channels involved in key physiological processes. However, upon activation, extrasynaptic, GluN2B-enriched NMDA receptors might cause excitotoxic processes 1. Since these processes could be related to neuropsychiatric disorders, the GluN2B subunit has become an interesting target for mapping pathological alterations and consequently, for drug development. Thus, the GluN2B-specific radioligand (R)-[11C]Me-NB1 was developed in rodents 2 and further successfully translated into humans 3. Here, we investigated the clinical applicability in terms of biodistribution and dosimetry.
Methods: Four healthy subjects (2 females, 2 males, mean age 25 ± 4 years) underwent a single whole-body measurement on a fully-integrated PET/MR. PET imaging was carried out using 9 passes of increasing duration in 6 bed positions (i.e. 0.5, 0.5, 0.5, 1, 1, 2, 3, 4, 7 min per bed position, head to mid-thigh). The radioligand (R)-[11C]Me-NB1 was administered via a cubital vein. The source organs for the biodistribution and dosimetry analysis are depicted in Figure 1. These organs were anatomically delineated utilizing the PET data. Time-integrated activity coefficients were computed for OLINDA/EXM 2.0 4 to estimate organ absorbed doses and total effective dose (ED). For visualization purposes, activity in percentage of injected dose was plotted over time for representative source organs and each subject separately (Figure 2).
Results: Highest uptake was observed in the urinary bladder, pancreas and spleen, and the lowest in the bone marrow. The urinary bladder wall was the critical organ with a mean absorbed organ dose of 1.5 ± 1.5 µSv/MBq. The total effective dose ranged from 4.4 to 7.1 µSv/MBq with an average ED of 6.0 ± 1.1 µSv/MBq.
Conclusion:(R)-[11C]Me-NB1 exhibited a similar low effective dose compared to other carbon-11 labelled radioligands 5. Considering a target dose of 6 MBq/kg as previously presented 3 and a person with 75 kg body weight would yield a reasonable effective dose of 2.7 mSv. In conclusion, longitudinal studies are feasible under commonly applied regulations, enabling drug development and interventional studies to further elucidate alterations in neuropsychiatric disorders.
Acknowledgements
This project was supported in part by the Swiss National Science Foundation grant numbers 310030E-160403/1 and 310030E-182872/1 to Prof. Simon M. Ametamey. Matej Murgaš is funded by the Austrian Science Fund FWF DOC 33-B27. Murray B. Reed and Leo R. Silberbauer are recipients of a DOC fellowship of the Austrian Academy of Sciences.
References
HardinghamGEBadingH.Synaptic versus extrasynaptic NMDA receptor signalling: implications for neurodegenerative disorders. Nat Rev Neurosci2010;
11: 682–696.HaiderAHerdeAMKrämerSD, et al.Preclinical evaluation of benzazepine-based PET radioligands (R)- and (S)-11C-Me-NB1 reveals distinct enantiomeric binding patterns and a tightrope walk between GluN2B- and σ1-receptor-targeted PET imaging. J Nucl Med2019;
60: 1167–1173.RischkaLVrakaCPichlerV, et al.First-in-human brain PET imaging of the GluN2B-containing N-methyl-D-aspartate receptor with (R)-11C-Me-NB1. J Nucl Med2021; in press.StabinMGSparksRBCroweE.OLINDA/EXM: the second-generation personal computer software for internal dose assessment in nuclear medicine. J Nucl Med2005;
46: 1023–1027.Zanotti-FregonaraPLammertsmaAAInnisRB.11C dosimetry scans should be abandoned. J Nucl Med2021;
62: 158–159.
2021-124
Longitudinal in vivo PET imaging of human ESC-derived dopamine neurons in minipig (#359)
Thea P. Lillethorup1, Deirdre B. Hoban2, Bengt Mattsson2, Dariusz Orlowski3, Jenny N. Wahlestedt2, Aage O. Alstrup1, Simone L. Bærentzen1, 4, Steen Jakobsen1, Andreas N. Glud3, Agnete Kirkeby5, David J. Brooks1, 6, Jens Christian H. Sørensen3, Malin J. Parmar2 and Anne M. Landau1, 4
1Aarhus University Hospital, Department of Nuclear Medicine and PET, Department of Clinical Medicine, Aarhus N, Denmark
2Lund University, Developmental and Regenerative Neurobiology, Wallenberg Neuroscience Center and Lund Stem Cell Centre, Department of Experimental Medical Science, Lund, Sweden
3Aarhus University Hospital, Department of Neurosurgery and Center for Experimental Neuroscience (CENSE), Department of Clinical Medicine, Aarhus N, Denmark
4Aarhus University, Translational Neuropsychiatry Unit, Department of Clinical Medicine, Aarhus N, Denmark
5University of Copenhagen, Novo Nordisk Foundation Center for Stem Cell Biology (DanStem), Copenhagen, Denmark
6University of Newcastle upon Tyne, Newcastle, UK
Abstract
Introduction: Human embryonic stem cells (hESC) can be robustly converted into authentic ventral midbrain dopaminergic neurons (hESC-DA) and have great potential as a dopamine-replacement therapy for Parkinson’s disease (PD). As part of the ongoing work to validate these cells for clinical use, it is important to verify that the cells survive, release dopamine, and innervate their target when transplanted into the living brain. The Gottingen minipig has a highly developed central nervous system with many physiological and anatomical similarities to humans. Their limited growth permits long-term longitudinal follow up studies with positron emission tomography (PET) and their brain size allows identification of sub-regional striatal distribution of pre- and post-synaptic dopaminergic tracers traditionally used in PD patients.
Methods: To obtain direct functional proof of the grafted cells integration into circuitry in vivo, we employed longitudinal PET imaging. Minipigs were immunosuppressed and imaged 2- and 7-months post-grafting of hESC-DA with [18F]-FDOPA PET for the uptake of dopamine precursors, and [18F]-PE2I PET, targeting the dopamine transporter (DAT).
Results: PET imaging revealed ipsilateral putamen increases in dopamine uptake already 2-months post-grafting, while DAT binding was first evident at 7-months post grafting, indicating a more mature graft. This effect was confirmed by post-mortem immunohistochemical analysis showing graft survival in 2/4 minipigs and dense DAT+ fiber staining including cell bodies after 7 months. Parallel studies using the same batch of hESC-DA neurons were conducted in athymic rats to ensure quality of the cells and compare longitudinal survival, volume and fiber outgrowth of the same cells between species.
Conclusion: This study confirms the translational potential of dopamine stem cell therapy and capability of innervating a large sized striatum, which can be visualized using PET imaging, supporting its use as a clinical treatment for PD patients.
Acknowledgements
This study was supported by The Lundbeck Foundation, Parkinsonforeningen, Jascha and the Danish Innovation Fund.
2021-125
Differential associations between neocortical tau pathology and blood flow with cognitive deficits in early-onset vs late-onset Alzheimer’s disease (#360)
Denise Visser1, Sander C.J. Verfaillie1, Emma E. Wolters1, 2, Emma M. Coomans1, Tessa Timmers1, 2, Hayel Tuncel1, Ronald Boellaard1, Sandeep S.V. Golla1, Albert D. Windhorst1, Philip Scheltens2, Wiesje M. van der Flier2, 4, Bart N.M. van Berckel1 and Rik Ossenkoppele2, 3
2Alzheimer Center Amsterdam, Department of Neurology, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, Netherlands
3Clinical Memory Research Unit, Lund University, Lund, Sweden
4Department of Epidemiology and Biostatistics, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, Netherlands
Abstract
Introduction: Early-onset Alzheimer’s disease (AD) and late-onset AD differ in neuropathological burden and type of cognitive deficits. Assessing tau pathology and relative cerebral blood flow (rCBF) measured with [18F]flortaucipir PET in relation to cognition may help explain these differences between early-onset and late-onset AD.
Methods: We included 79 amyloid positive individuals with MCI/AD dementia (early-onset AD: n = 35, age-at-PET = 59 ± 5, MMSE = 23 ± 4; late-onset AD: n = 44, age-at-PET = 71 ± 5, MMSE = 23 ± 4) who underwent a 130 minutes dynamic [18F]flortaucipir PET scan and extensive neuropsychological assessment. We extracted non-displaceable binding potentials (BPND) and R1 (proxy of rCBF) from parametric images using receptor parametric mapping, in medial and lateral temporal, parietal, occipital and frontal regions-of-interest and used continuous scores on nine neuropsychological tests covering memory, attention, language and executive functioning. We first examined differences between early-onset AD and late-onset AD in BPND or R1 using ANOVA (region-of-interests analysis) and voxel-wise contrasts. Next, we performed linear regression models to test for potential interaction effects between age-of-onset and BPND/R1 (for eachregion-of-interest) on cognition.
Results: Both region-of-interest and voxel-wise contrast showed a clear pattern of greater [18F]flortaucipir BPND across all neocortical regions in early-onset AD, while late-onset AD patients had subtle lower R1 values in medial temporal regions primarily (Figure 1). Furthermore, compared to late-onset AD, interaction analyses indicated that higher [18F]flortaucipir BPND and lower R1 values in lateral temporal, parietal and occipital regions were more strongly associated with more severe cognitive impairment in early-onset AD (Figure 2).
Conclusion: Our results show that early-onset AD is characterized by higher levels of tau pathology and stronger associations between lateral temporal, and occipito-parietal tau pathology or lower rCBF and cognitive impairment. These findings may have important implications for clinical trials, since effects of potential tau- or blood flow targeting therapeutic interventions might exert larger effects in early-onset compared to late-onset AD patients.
Acknowledgements
We kindly thank all participants for their contribution. Research of Amsterdam Alzheimer Center is part of the Neurodegeneration program of Amsterdam Neuroscience. The Amsterdam Alzheimer Center is supported by Alzheimer Nederland and Stichting VUmc funds. [18F]Flortaucipir PET scans were made possible by Avid Radiopharmaceuticals Inc.
2021-126
12 days of sucrose consumption reduces 3H-UCB-J binding in the caudate and prefrontal cortex of healthy Göttingen minipigs (#361)
Simone L. Bærentzen1, 2, Majken B. Thomsen1, 2, Aage O. Alstrup2, David J. Brooks2, 3, Gregers Wegener1, Michael Winterdahl1, 2 and Anne M. Landau1, 2
1Aarhus University, Translational Neuropsychiatry Unit, Department of Clinical Medicine, Aarhus C, Denmark
2Aarhus University, Department of Nuclear Medicine and PET, Department of Clinical Medicine, Aarhus N, Denmark
3University of Newcastle upon Tyne, Translational and Clinical Research Institute, Faculty of Medical Sciences, Newcastle, UK
Abstract
Introduction: Over the last decades, there has been an increase in the consumption of sugar worldwide, especially in the form of sugar-sweetened beverages. This change in diet has been proven to be a contributing factor to the development of cancers, cardiovascular diseases, diabetes, obesity, metabolic syndrome and mental ill-health. Additionally, prolonged periods of high sugar intake have been linked to a reduction in cognitive performance and altered neurotransmission in humans, minipigs and rodents, and may increase the risk of Alzheimer’s disease. Here we test the hypothesis that a diet high in sucrose leads to reduced synaptic vesicle glycoprotein 2A (SV2A), an indirect measure of synaptic density, in healthy Göttingen minipigs.
Methods: Seven minipigs had either a 12-day period of access to two liters of a 25% sucrose in water solution or to tap water ad libitum followed by euthanasia and brain extraction. Tritiated UCB-J, which binds to SV2A, was applied to brain sections from prefrontal cortex, nucleus accumbens, caudate, putamen, dorsal hippocampus, ventral hippocampus and thalamus, and autoradiography was performed in the presence and absence of levetiracetam, an SV2A blocker, in order to assess specific binding.
Results: [3H]UCB-J binding was significantly lower in prefrontal cortex and caudate of minipigs drinking sucrose solution compared to the water-drinking minipigs.
Conclusion: Our data suggest that sucrose consumption can lead to reduced synaptic SV2A protein density which could contribute to cognitive deficits.
2021-127
Longitudinal tau PET using [18F]flortaucipir: Comparison of (semi)quantitative parameters (#362)
Denise Visser1, Hayel Tuncel1, Rik Ossenkoppele2, 3, Maqsood Yaqub1, Emma E. Wolters1, 2, Tessa Timmers1, 2, Emma Weltings1, Emma M. Coomans1, Marijke E. den Hollander1, Wiesje M. van der Flier2, 4, Bart N.M. van Berckel1 and Sandeep S.V. Golla1
2Alzheimer Center Amsterdam, Department of Neurology, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, Netherlands
3Clinical Memory Research Unit, Lund University, Lund, Sweden
4Department of Epidemiology and Biostatistics, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, Netherlands
Abstract
Introduction: Semi-quantitative PET measures can be affected by changes in blood flow, whereas quantitative measures are not. The aim of the study was to compare semi-quantitative (SUVr) and quantitative (R1, BPND) parameters of longitudinal tau PET scans with [18F]flortaucipir, with respect to changes in blood flow.
Methods: Thirty-eight subjects with subjective cognitive decline (SCD; age 65 ± 7y, male/female 16/22, MMSE 29 ± 1, amyloid positive 12/38) and 24 Alzheimer’s disease (AD) patients (age 66 ± 7y, male/female 11/13, MMSE 24 ± 3, amyloid positive 24/24) underwent baseline (BL) and 2.1 ± 0.3 and 2.2 ± 0.3 year follow-up (FU) dynamic [18F]flortaucipir PET scans, respectively. BPND and R1 were estimated using RPM and SUVr(80–100min) was calculated (cerebellar gray as reference). For each region-of-interest ((trans)entorhinal, limbic and neocortical) and parameter, %change was calculated. Regional SUVrs were compared to corresponding DVR ( = BPND+1) using paired T-tests. Additionally, simulations were performed to model effects of flow changes on BPND and SUVr in different binding categories. Thereafter, %bias for SUVr with respect to underlying binding and flow were evaluated.
Results: In SCD, there was a difference between %change in the (trans)entorhinal ROI (DVR 2.56% vs SUVr 1.85%) only (Figure 1). In AD, a difference was found in the limbic ROI (DVR 6.61% vs SUVr 7.52%) only. R1 changes were small (+0.7% in SCD and -1.6% in AD). Simulations illustrated with increasing flow a decreased %bias for SUVr in low binding conditions, whereas a slightly increased bias was observed in high binding conditions (Figure 2).
Conclusion: SUVr provided an accurate estimate of specific binding for [18F]flortaucipir over a two-year follow-up. However, simulations showed that flow changes can affect [18F]flortaucipir SUVr, hence DVR/BPND should be preferred in more advanced disease stages and/or conditions that could induce significant flow changes like pharmacotherapeutic interventions.
Acknowledgements
We kindly thank all participants for their contribution. We thank Ronald Boellaard for sharing his knowledge and thoughts about the project. Research of Amsterdam Alzheimer Center is part of the Neurodegeneration program of Amsterdam Neuroscience. The Amsterdam Alzheimer Center is supported by Alzheimer Nederland and Stichting VUmc funds. [18F]Flortaucipir PET scans were made possible by Avid Radiopharmaceuticals Inc.
2021-128
From genes to networks: In-vivo CRISPR/Cas9-induced VMAT2 knockdown exerts functional connectome changes in the rat brain (#363)
Tudor M. Ionescu1, Sabina Marciano1, Ran Sing Saw1, Rachel Cheong2, Deniz Kirik2, Andreas Maurer1, Bernd J. Pichler1 and Kristina Herfert1
1University of Tuebingen, Werner Siemens Imaging Center, Department of Preclinical Imaging and Radiopharmacy, Tuebingen, Germany
2Lund University, Brain Repair and Imaging in Neural Systems, Department of Experimental Medical Science, Lund, Sweden
Abstract
Introduction: Recent advances in CRISPR/Cas9 technology have enabled its use as a powerful tool for in-vivo gene editing. Previously, we have validated this technique in-vivo by assessing vesicular monoamine transporter 2 (VMAT2) knockdown (KD) effects using PET, behavioral and ex-vivo experiments. However, its effects on whole-brain function have not been studied to date. Here, we targeted the Slc18a2 gene, encoding the VMAT2, using Cas9 from Staphylococcus aureus (SaCas9) in rats. We performed simultaneous PET/fMRI scans to elucidate the effects of gene editing on both local molecular and network-level functional connectivity (FC) parameters.
Methods: Baseline [11C]raclopride PET/fMRI scans were acquired in rats (n = 23) under medetomidine and isoflurane in air over 60min using a 7T small-animal MRI with a PET insert. AAV-SaCas9 and AAV-sgRNA-Slc18a2 were next injected into the right substantia nigra pars compacta (SNc) and DPBS into the left SNc. The rats were rescanned 8–14 weeks after using the same PET/fMRI protocol and additional PET scans were performed using [11C]DTBZ and [18F]FMZ. The PET data were assessed for molecular changes and fMRI was used to evaluate effects on default-mode (DMN) and sensorimotor networks (SMN).
Results: Based on the [11C]DTBZ scans we split the rats into low knockdown (KD < 20%,n = 13) and high KD (KD > 20%,n = 10) groups. Our data indicated both subtle FC increases and decreases (p < 0.05) at low KD. At high KD, strong increases were found both within the DMN and SMN, driven by the contralateral thalamus (30% change,p = 0.009), and between the two networks, indicating compensatory effects in network function following VMAT2 KD. Additionally, we found correlations between VMAT2 KD and D2R expressions (r = 0.46,p = 0.03), and robust left-right differences in regional GABAA expressions independently of KD magnitude in several regions (p < 0.001), hinting towards neurotransmitter cross-talk effects.
Conclusion: Our study is the first to assess CRISPR/Cas9 gene editing as an in-vivo method for the investigation of both local and network-level brain function. We could robustly separate the two cohorts, thus recommending the used study design for future research of both early and advanced disease phenotypes. We demonstrated the tremendous potential combining CRISPR/Cas9 and PET/fMRI imaging has to bridge the gap between targeted gene alterations and whole-brain functional changes.
Assessment of FC effects
(A) Mild KD and (B) Moderate KD: Correlation matrices indicate FC at baseline and after KD. Symbols indicate alterations (white: decrease, black: increase; +: p < 0.05, uncorr.; *: p < 0.05, FDR-corr.). The brain maps show changes (p < 0.05, uncorr.) in edge and node FC, their sizes being proportional to the change. (C) Comparison of functional effects at mild and moderate KD. The matrix compares the effects of mild and moderate KD. The scatter plots show correlations between the alterations both within and between DMN and SMN (+ p < 0.05, * p < 0.01, ** p < 0.001).
Molecular alterations induced by VMAT2 knockdown
(A) A significant decrease was observed in the ipsilateral striatal [11C]DTBZ confirming VMAT2 knockdown (left panel). Additionally, [11C]raclopride was increased on the ipsilateral side after VMAT2 KD (middle panel). Both effects correlated significantly (R2 = 0.52, p < 0.01, right panel). The analysis indicated a separation of the two groups at a knockdown above 20% based on [11C]DTBZ uptake. (B) [11C]Flumazenil revealed decreased ipsilateral GABA-A binding in two cortex areas in both cohorts. + p < 0.05; * p < 0.01, ** p < 0.001.
2021-129
Development of voxel-level EC50 Images for use in CNS Drug Development (#364)
Jocelyn Hoye, Bart de Laat, Heather Liu and Evan D. Morris
Yale University, Radiology and Biomedical Imaging, New Haven Connecticut, USA
Abstract
Introduction: We recently introduced voxel-level images of drug occupancy from PET via our “Lassen Plot Filter”.1 Occupancy images revealed clear dependence (locally) of 11C-flumazenil displacement on dose of GABAa inhibitor, CVL-865. We hypothesized that regions requiring higher drug concentrations to achieve desired occupancy would have higher EC50 values. We introduce novel “EC50 images” from human data and supporting simulations.
Methods: Five healthy subjects were scanned with the nonselective GABAa tracer, 11C-flumazenil, before and twice after administration of CVL-865. We created ten occupancy images (Figure 1) and applied an Emax model at the voxel-level to all images, combined, to create one EC50 image. We performed simulations (Figure 2) to confirm our observations of regional variation in EC50 and to assess the effects of plasma concentration sampling on EC50 variability. EC50 images were simulated using plasma concentrations from the human study (3–279 ng/mL) and a noise model consistent with observed concentration-response data. In a second simulation study the range of plasma concentrations was extended (3–1957 ng/mL), to allow all simulated EC50 regions to reach a minimum of 97% occupancy.
Results: As expected, the EC50 image revealed spatial variation in ‘apparent drug affinity’ (differential displacement for a given dose). High EC50 was found in areas of low occupancy, for a given drug dose (Figure 1). Simulations showed that it is possible to distinguish between regional differences in EC50 estimates (Figure 2). Simulations also demonstrated that sampling from a wider, more complete range of plasma drug concentrations could improve and regularize EC50 precision, spatially (Figure 2).
Conclusion: Our results argue for (a) confidence in the ability of the EC50 images to identify regional variation and (b) a need to tailor the range of drug doses and post-drug sampling times to ensure uniform precision of the EC50. The EC50 image could add value to early phase drug development by identifying regional variation in affinity that might impact therapy or safety, and by guiding dose selection for late-phase trials.
Reference
de Laat B and Morris ED.
A local-neighborhood Lassen plot filter for creating occupancy and non-displaceable binding images. J Cereb Blood Flow Metab2020;
41: 1379–1389.
2021-130
Agonist-induced µ-opioid receptor desensitization increases radiotracer binding: A positron emission tomography study in the mice brain (#365)
Chi-Hyeon Yoo, Sarah E. Reid and Hsiao-Ying Wey
Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown Massachusetts, USA
Abstract
Introduction: µ-opioid receptor (MOR) agonists are the most effective analgesics but with detrimental side-effects. The development of tolerance, which is associated with increased risks of opioid dependent and overdose, has been reported but the underlying mechanism remain elusive. Agonist-induced receptor desensitization was proposed as a key mechanism underlies opioid tolerance. Previous findings suggested that desensitized MORs have an increased affinity for ligand binding, consistent with our preliminary findings in nonhuman primates. The goal of this study was to validate the effect of agonist-induced desensitization on MOR ligand binding in the mice brain. Specifically, positron emission tomography (PET) experiments were performed using a MOR selective radiotracer, [11C]carfentanil. We measured MOR binding potential at baseline and after pretreatment of two MOR agonists (morphine and SR-17018) in wild-type (WT) mice and β-arrestin2 knockout (βarr2-KO) mice.
Methods: Bruker 4.7T MRI scanner with a PET insert were used to scan WT mice (n = 14). [11C]carfentanil (∼4.72 MBq; specific activity: > 39 GBq/µmol) was intravenously given as bolus. WT mice were scanned at baseline (n = 6), after 15-min pretreatment of morphine (1.0 mg/kg, i.v.; n = 4) or SR-17018 (1.0 mg/kg, i.v.; n = 4). PET data were binned into the 36 frames, and analyzed for binding potentials (BPND) using a simplified reference tissue model with cerebellum as the reference tissue. For statistical analysis, a one-way analysis of variance was used.
Results: High-level of specific bindings were identified in the striatum and thalamus (Figure 1) in all groups, consistent with known distribution of MOR in the brain. Based on ANOVA analysis in the WT mice, statistically higher BPND was observed in the thalamus and midbrain in morphine-pretreated mice than controls and SR17018-pretreated mice (p < 0.050; Figure 2).
Conclusion: The result of this study suggests that a potential increase in ligand-target affinity due to morphine-induced MOR desensitization, while a G-protein biased MOR agonist, SR-17018, does not trigger receptor trafficking. Experiments on βarr2-KO mice are completed, and data analysis is in progress. Additional data in βarr2-KO mice will verify whether the β-arrestin2 pathway is key to regulate receptor trafficking and will confirm the accessibility of PET for measuring changes in receptor affinity.
Acknowledgements
This research is support by NIH R00DA037928 and R21DA047133 (H.-Y. W.).
2021-131
Age and BMI associations with brain NET availability using [11C]MRB (#366)
Sheida Koohsari1, Brian Pittman1, Nabeel Nabulsi1, Yiyun H. Huang1, Richard E. Carson1, Christopher H. van Dyck2, Marc N. Potenza2 and David Matuskey1, 2
1Yale University, PET Center, Department of Radiology and Biomedical Imaging, New Haven Connecticut, USA
2Yale University, Department of Psychiatry, New Haven Connecticut, USA
3Yale University, Department of Neurology, New Haven Connecticut, USA
Abstract
Introduction: Understanding aging in the brain is crucial to understand disorders common in the elderly. Previous studies revealed the neuromodulatory effect of norepinephrine (NE) in conditions like dementia, Parkinson’s disease, and depression.1 (S,S)-[11C] O methylreboxetine ([11C]MRB) is a selective radiotracer for norepinephrine transporter (NET), the principal regulator of NE. Here we investigate aging, along with body mass index (BMI), on brain NET availability in the largest sample to date.
Methods: Forty-three healthy controls (20 females, 23 males; age range 18–49 years; BMI range: 21–39.4) were included from previous studies. All participants underwent an MRI scan and [11C]MRB PET scan. PET images of individuals were coregistered to their MRI scan. The multilinear reference tissue model 2 (MRTM2) was applied to quantify binding potential (BPND) values with the occipital cortex as a reference region. Norepinephrine-rich regions of the brain included the locus coeruleus (LC), pulvinar, raphe nuclei, red nucleus, thalamus, and hypothalamus. We investigated the correlations between BPND and age and BMI using a mixed linear model.
Results: Inverse associations between [11C]MRB BPND and age were observed in the LC (r = −0.3615; p = 0.0172) (Figure 1), raphe nuclei (r = −0.4619; p = 0.0018), and hypothalamus (r = −0.3074; p = 0.0449), with no gender effects noted. There were no statistically significant correlations between BMI and BPND.
Conclusion: We observed lower NET availability with aging consistent with a previous investigation.2 This work provides further evidence of NET decreases associated with aging, which could have implications in cognitive and attentional deficits of the elderly. Similar to a previous study,3 we could not replicate the correlation between BMI and NET availability reported previously.4 Further investigations with endocrine and behavioral measurements may provide additional insights.
References
BenarrochEE.
The locus ceruleus norepinephrine system: functional organization and potential clinical significance. Neurology2009;
73: 1699–1704.DingY-S, et al.PET imaging of the effects of age and cocaine on the norepinephrine transporter in the human brain using (S,S)-[(11)C]O-methylreboxetine and HRRT. Synapse2010;
64: 30–38.HesseS, et al.Central noradrenaline transporter availability in highly obese, non-depressed individuals. Eur J Nucl Med Mol Imaging2017;
44: 1056–1064.Li Chiang-shanR, et al.Decreased norepinephrine transporter availability in obesity: positron emission tomography imaging with (S,S)-[(11)C]O-methylreboxetine. NeuroImage2014;
86: 306–310.
2021-132
Measuring synaptic density and the dopamine transporter in Parkinson’s disease: a PET imaging study with11C-UCB-J and18F-FE-PE2I (#367)
David Matuskey1, Sule Tinaz1, Mark Dias1, Mika Naganawa1, Jean-Dominique Gallezot1, Sophie Holmes1, Shannan Henry1, Sheida Koohsari1, Jim Ropchan1, Robert A. Comley2, Nabeel Nabulsi1, Yiyun H. Huang1, Richard E. Carson1 and Sjoerd J. Finnema2
1Yale University, New Haven Connecticut, USA
2Abbvie, North Chicago Illinois, USA
Abstract
Introduction: Synaptic dysfunction and loss are associated with Parkinson’s disease (PD) pathology and have been investigated using positron emission tomography (PET) with the synaptic vesicle glycoprotein 2A (SV2A) radiotracer 11C‑UCB‑J.1–3 Here we present preliminary 11C‑UCB‑J and 18F-FE-PE2I data, both acquired within the same PD and healthy control (HC) subjects, to compare synaptic density with dopamine transporter (DAT) availability.
Methods: PD subjects (N = 16; 9M/7F; mean (SD) age 61(8); duration since diagnosis 7(5) years) were included along with matched HCs (N = 5; 2M/3F; mean age 56(8)). All subjects were in good physical health, with no signs of dementia (Montreal Cognitive Assessment score < 21) or severe psychiatric or medical conditions. All dopaminergic or psychotropic medications were withheld during PET imaging days to avoid potential confounds. Subjects underwent MRI, 11C‑UCB‑J PET and 18F-FE-PE2I PET; PET imaging took place on a high resolution research tomograph (HRRT). As in previous work, binding potential (BPND), calculated using the simplified reference tissue model 2 (SRTM2), was the primary outcome measure. Reference regions were centrum semiovale (11C‑UCB‑J) and cerebellum (18F-FE-PE2I). Primary regions of interest (ROI) were defined on MR and included important motor areas such as the caudate, putamen, substantia nigra (SN) and, for 11C‑UCB‑J only, cortical regions.
Results: In PD subjects compared with HCs,11C‑UCB‑J BPND was non-significantly lower in most brain regions, including the SN (-21%, p = 0.2). 18F-FE-PE2I BPND was significantly lower (-38% to -73%) with the greatest difference in the putamen (-73%, p < 0.001) (Figure 1).
Conclusion: In PD subjects, binding of 18F-FE-PE2I in the PD group was significantly lower, consistent with previous reports.3 11C‑UCB‑J was also lower throughout all motor brain regions examined, but was nonsignificant. Further work is ongoing to investigate the temporal relationship between SV2A and DAT binding in order to increase our understanding of degeneration in PD.
Acknowledgements
We thank AbbVie for financial support of this study, the Connecticut Advocates for Parkinson’s for their help with recruitment and UCB Pharma for providing precursor for 11C‑UCB‑J.
References
MatuskeyDTinazSWilcoxKC, et al.Synaptic changes in Parkinson Disease assessed with in vivo imaging. Ann Neurol2020;
87: 329–338.WilsonHPaganoGde NataleER, et al.Mitochondrial complex 1, sigma 1, and synaptic vesicle 2A in early drug-naive Parkinson’s disease. Mov Disord2020;
35: 1416–1427.DelvaAVan WeehaegheDKooleM, et al.Loss of presynaptic terminal integrity in the substantia nigra in early Parkinson’s disease. Mov Disord2020;
35: 1977–1986.
2021-133
Longitudinal test-retest reproducibility of 11C-UCB-J, a PET tracer for synaptic density imaging (#368)
Nikkita Khattar1, Mika Naganawa2, Jean-Dominique Gallezot2, Jayanta Mondal2, Yihuan Lu2, Samantha Rossano2, David Matuskey2, Yiyun H. Huang2, Richard E. Carson2 and Takuya Toyonaga2
1Yale University, School of Medicine, New Haven Connecticut, USA
2Yale University, PET Center, New Haven Connecticut, USA
Abstract
Introduction:11C-UCB-J, a radioligand for synaptic vesicle glycoprotein 2A (SV2A), is a potential in vivo biomarker for synaptic density.1,2 Excellent test-retest reproducibility of 11C-UCB-J was reported among same-day scans.3 To expand its applications, high longitudinal reliability is important. In this study, we evaluated same-day and longitudinal (short- to long-term) test-retest reliability of 11C-UCB-J in healthy subjects.
Methods: In total, thirty HS underwent two 60-min 11C-UCB-J scans. Ten subjects had two scans on the same day, while twenty subjects in the longitudinal group had scanned from 1 week to more than two years apart (median: 391 days, range: 7–1028 days). All subjects underwent arterial blood sampling and volume of distribution (VT) was estimated by one-tissue compartment modeling. Distribution volume ratio (DVR) was calculated using centrum semiovale (CS) or cerebellum as a reference region (DVR(CS) or DVR(CB)). CS is defined in AAL space and the other regions of interest (ROIs) were generated by FreeSurfer (Table 1). Reproducibility was examined using test-retest variability (TRV = 100*(test-retest)/{(test+retest)/2}). To evaluate the difference of variance between groups (same-day, longitudinal) and between measures (VT, DVR(CS), DVR(CB)), two-tailed F tests were applied. Longitudinal data were also analyzed with regression analysis.
Results: For VT, DVR(CS), and DVR(CB), variances were consistent across ROIs, except VT in CS (Table 1). In the same-day group, TRVs in gray matter (GM) regions (averaged over subjects) were 0.5 ± 6.7%, -1.4 ± 9.1%, and 0.3 ± 2.8% for VT, DVR(CS) and DVR(CB), respectively. In the longitudinal group, TRVs for VT, DVR(CS) and DVR(CB) were 0.8 ± 10.2%, 1.3 ± 15.9%, and -0.9 ± 3.8%, respectively. There was no variance difference between the same-day and longitudinal groups for any measure. BPND vs. DVR (same-day), BPND vs. DVR (longitudinal), and VT vs. DVR (longitudinal) showed significant variance differences with Bonferroni correction (Figure 1(a)). Regression analysis on longitudinal data showed no significant inter-scan interval effect on TRV (Figure 1(b)).
Conclusion: The reproducibility of 11C-UCB-J DVR (cerebellum) was excellent in both same day and longitudinal groups with BPND (CS) showing the poorest TRV. There was no inter-scan interval effect up to 1028 days. For appropriate cases, cerebellum could be a good pseudo reference region for future longitudinal clinical studies.
Table 1. Test-retest variability for each region.
References
NabulsiNBMercierJHoldenD, et al.Synthesis and preclinical evaluation of 11C-UCB-J as a PET tracer for imaging the synaptic vesicle glycoprotein 2A in the brain.J Nucl Med2016;
57: 777–784.FinnemaSJNabulsiNBEidT, et al.Imaging synaptic density in the living human brain. Sci Transl Med2016;
8: 348ra396.FinnemaSJNabulsiNBMercierJ, et al.Kinetic evaluation and test-retest reproducibility of [11C]UCB-J, a novel radioligand for positron emission tomography imaging of synaptic vesicle glycoprotein 2A in humans. J Cereb Blood Flow Metab2018;
38: 2041–2052.RossanoSToyonagaTFinnemaSJ, et al.Assessment of a white matter reference region for 11C-UCB-J PET quantification. J Cereb Blood Flow Metab2019; 271678X19879230.
2021-134
Simultaneous optogenetic [18F]FDG-fPET/fMRI to study brain circuits in rats (#369)
Sabrina Buss1, Tudor M. Ionescu1, Laura Kuebler1, Gerald Reischl1, 2, Bernd J. Pichler1, 2 and Kristina Herfert1
1Eberhard Karls University Tuebingen, Preclinical Imgaing and Radiopharmacy, Tuebingen Baden-Württemberg, Germany
2Eberhard Karls University Tuebingen, Cluster of Excellence iFIT (EXC 2180), Tuebingen Baden-Württemberg, Germany
Abstract
Introduction: Optogenetic functional magnetic resonance imaging (ofMRI) combines a precise neuronal stimulation technique, with fMRI as indirect readout of neuronal activation. This enables a cell-type specific mapping of the whole brain dynamic response to the activation or inhibition of neuronal circuits. In addition, functional positron emission tomography (fPET) enables the detection of metabolic changes with good temporal resolution on single subject level via a continuous [18F]FDG infusion. In this study, a simultaneous [18F]FDG-fPET/BOLD-fMRI protocol was applied in rats subjected to optogenetic stimulation.
Methods: Male rats were stereotactically injected with 2 µL of an adeno-associated virus vector overexpressing channelrhodopsin-2 (ChR2) (n = 17) or GFP (control) (n = 10) into the right substantia nigra compacta (rSNc). Eleven weeks post-surgery, an optical fiber connected to a 473 nm laser was implanted into the rSNc. [18F]FDG-fPET/BOLD-fMRI scans were performed on a 7T small-animal MRI equipped with an inhouse-built PET insertunder a constant infusion of α-chloralose and pancuronium bromide. [18F]FDG (142 ± 8 MBq) was infused using a bolus plus constant infusion protocol and fMRI data were simultaneously acquired using an EPI-BOLD sequence (TR: 2 s, TE: 18 ms) during a 20 Hz laser light stimulation. Data were preprocessed and analyzed using SPM12.
Results: Optogenetic stimulation of rSNc resulted in BOLD signal changes of 2.4% in the right caudate putamen (CPuR) of ChR2 injected rats, while no changes were detected on the contralateral side and in GFP control rats. BOLD signal changes highly correlated to the stimulation paradigm and mean t-scores of 4.9 ± 1.5 were observed in the dorsal, medial CPuR. A 7% higher [18F]FDG uptake in the CPuR of ChR2 injected rats was observed compared to the contralateral side and control rats in the last 10-minute time-interval of the PET acquisition (ChR2 > GFP CPuR mean t-scores: 3.6 ± 1.6). In contrast to fMRI, metabolic activation was more pronounced in the ventral, medial part of the CPuR.
Conclusion: We present for the first time simultaneous, dynamic [18F]FDG-fPET/BOLD-fMRI data during optogenetic stimulation in rats. Our data show a clear spatial and temporal mismatch between activity patterns observed with BOLD-fMRI and [18F]FDG-fPET, which may be related to the complementary readouts of both.
Study design (a). Toxic effects on dopaminergic neurons were excluded by [11C]MP PET (b,c). Confirmation of ChR2 expression in the rSTR/rSNc by fluorescence microscopy (d). [18F]FDG TACs and PET image show tracer uptake in the CPuR over time on single animal level (e, g)and group level (h) compared to the CPuL or GFP group. t-maps show CPuR activation in one ChR2 animal (f). Activation in the CPuR of ChR2 rats is confirmed by BOLD signal changes on group level (j). Overlay of fMRI and fPET reveals spatial matches and mismatches (i). Highest t -scores in ChR2 group in CPu, MB and VTA (k).
2021-135
Manifold component analysis and its use in evaluating stage-wise cortical tau propagation measured with 18F-MK-6240 PET (#370)
Gleb Bezgin1, Tharick A. Pascoal2, 1, Joseph Therriault1, Firoza Z. Lussier1, Min-Su Kang1, Cécile Tissot1, Stijn Servaes1, Yi-Ting Wang1, Peter Kunach1, Jenna Stevenson1, Serge Gauthier1 and Pedro Rosa-Neto1
1McGill Centre for Studies In Aging, Translational Neuroimaging Laboratory, Montreal, QC, Canada
2University of Pittsburgh, Department of Psychiatry, Pittsburgh Pennsylvania, USA
Abstract
Introduction: In systems neuroscience, it is often necessary to represent neuroimaging data in the context of ordered regions, wherein the order signifies particular stages of sensory/cognitive processing, or progression of a disease. One prominent example is the representation of six neuropathological stages of tau protein aggregation in AD patients, first systematically investigated in the seminal Braak & Braak study in 1991.1 With modern PET tracers, it is possible to track such progression in patients in vivo2; recent studies successfully untangled Braak stage manifestations using the whole brain tau PET/MRI representation, as well as reducing such representation into a discrete six-stage model, with six-region profiles using Braak stage regions.
Methods: We propose a novel approach called Manifold Component Analysis (MCA) for producing pseudo-continuous tau profiles, aiming to track the progression within Braak regions and across them, taking into account the uptake density and capacity. We achieve it in three steps: 1) define Braak regions anatomically on a cortical surface (Figure 1(a)); 2) use surface topology to make pseudo-continuous representation of Braak regions, ensuring gradual transition between neighbouring stage regions (Figure 1(b)); 3) sample 18F-MK-6240 PET tracer uptake across the resulting pseudo-continuous scheme (Figure 1(c)).
Results: Profiled 18F-MK-6240 accumulations across different MCA stages show an increase in follow-up compared to the baseline; such increase is predominant at the Braak region during the corresponding Braak stage, particularly prominent and significant at stages 4 and 5, and corresponding Braak regions IV and V (Figure 1(d) and (e)).
Conclusion: The Braak sub-stage profile representation using MCA helps identify changes within a given Braak stage and across the stages, both in terms of magnitude and spatial progression. It reduces surface manifold data into more interpretable spatial series, at the same time providing information on subtle between-stage transitions which represents a novel systematic way to quantify tau propagation patterns. Importantly, MCA can be used in a vast majority of other applications involving ordered regions, e.g. in the analyses of sensory processing hierarchies.
Acknowledgements
We thank Melissa Savard, Mira Chamoun, Andrea Lessa Benedet, Nesrine Rahmouni, Jaime Fernandez-Arias and Parissa Fereydouni for their help.
a: discrete Braak regions; b: pseudo-continuous Braak regions; c: 18F-MK-6240 tracer uptake profiles for subjects averaged across their corresponding pathophysiological stages; d: average profiles for subjects assessed longitudinally, involving baseline (blue) and follow-up (red) SUVRs; e: differences between follow-up and baseline for each pathophysiological stage and each Braak region.
References
BraakHBraakE.
Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol1991;
82: 239–259.PascoalTATherriaultJBenedetAL, et al.18F-MK-6240 PET for early and late detection of neurofibrillary tangles. Brain2020;
143: 2818–2830.
2021-136
Evaluation of brain structure and function in currently depressed adults with a history of childhood trauma (#371)
Joshua Jones1, Samantha J. Goldstein2, Junying Wang5, John Gardus2, Jie Yang3, Ramin V. Parsey2 and Christine DeLorenzo2, 4
1University of Rochester, Rochester New York, USA
2Stony Brook University, Department of Psychiatry and Behavioral Science, Stony Brook New York, USA
3Stony Brook University, Department of Family, Population & Preventive Medicine, Stony Brook New York, USA
4Stony Brook University, Department of Biomedical Engineering, Stony Brook New York, USA
5Stony Brook University, Department of Applied Mathematics and Statistics, Stony Brook New York, USA
Abstract
Introduction: Structural differences in the dorsolateral prefrontal cortex (DLPFC), anterior cingulate cortex (ACC), hippocampus, and amygdala were reported in adults who experienced childhood trauma; however, the functional consequences are unknown. This multimodal imaging study examined structural and functional consequences of childhood trauma in adults with major depressive disorder (MDD).
Methods: Adult participants ages 18–65 with MDD completed the Childhood Trauma Questionnaire (CTQ) and simultaneous positron emission tomography (PET)/magnetic resonance imaging (MRI). Structure (volume and cortical thickness, n = 83) was quantified from MRI using Freesurfer. Function (metabolic rate of glucose uptake) was quantified from dynamic 18F-fluorodeoxyglucose (FDG)-PET images (n = 77) using Patlak graphical analysis. A linear mixed model was utilized to examine the association between structural/functional variables and discrete or continuous childhood trauma measures while controlling for confounding factors.
Results: DLPFC metabolism and cortical thickness (Figure 1(c) and (f)) and ACC metabolism (Figure 1(a)) differed significantly between some discrete childhood trauma levels. Hippocampus and amygdala volumes were significantly inversely correlated with continuous CTQ scores (Figure 2(j) and (h)). Specifically, hippocampus and amygdala volumes were lower by 12.96 (95% Confidence Interval [CI]: -23.30, -2.61) and 7.95 mm3 (95% CI: -12.72, -3.18), respectively, per point increase in CTQ.
Conclusion: Examining childhood trauma within MDD as discrete and continuous measures uncovered unique relationships with thickness/metabolism and volume, respectively. This is the first study to report functional differences and provide estimates of volume differences based on childhood trauma within MDD. While longitudinal studies are required to establish causation, this study provides insight into potential consequences of, and therefore potential therapeutic targets for, childhood trauma in the prevention of MDD.
Acknowledgements
This study was funded by R01MH104512, Brain & Behavior Foundation, The Dana Foundation, and an NYS Faculty Development Grant. We acknowledge the biostatistical consultation and support provided by the Biostatistical Consulting Core at the School of Medicine, Stony Brook University.
References
BaekenCWuGRDe RaedtR.
Dorsomedial frontal cortical metabolic differences of comorbid generalized anxiety disorder in refractory major depression: a [(18)F] FDG PET brain imaging study. J Affect Disord2018;
227: 550–553.BernsteinDPFinkLHandelsmanL, et al.Initial reliability and validity of a new retrospective measure of child abuse and neglect. Am J Psychiatr1994;
151: 1132–1136.ParrLABoudreauMHechtE, et al.Early life stress affects cerebral glucose metabolism in adult rhesus monkeys (Macaca mulatta). Dev Cognit Neurosci2012;
2: 181–193.SuLCaiYXuY, et al.Cerebral metabolism in major depressive disorder: a voxel-based meta-analysis of positron emission tomography studies. BMC Psychiatr2014;
14: 321.VyasAMitraRShankaranarayana RaoBS, et al.Chronic stress induces contrasting patterns of dendritic remodeling in hippocampal and amygdaloid neurons. J Neurosci2002;
22: 6810–6818.
2021-137
Impact of direct-4D PET image reconstruction on within- and between-subject variance of [11C]UCB-J in Parkinson’s disease (#373)
Paul Gravel, Kathryn Fontaine, David Matuskey, Richard E. Carson and Jean-Dominique Gallezot
Yale University School of Medicine, Department of Radiology and Biomedical Imaging, New Haven Connecticut, USA
Abstract
Introduction: Several studies have demonstrated the benefits of direct-4D PET image reconstruction, which delivers lower-variance parametric images, over conventional indirect-3D reconstruction techniques. The goal of this work is to evaluate direct-4D reconstruction, applied on real patient PET scans, on within-subject and between-subject variance, compared to indirect-3D reconstruction.
Methods: Eight subjects diagnosed with Parkinson’s disease underwent PET imaging with [11C]UCB-J, a radioligand for measuring synaptic density (SV2A)1, on the Siemens HRRT2. During each scan, continuous head motion data were acquired using the Polaris Vicra, and arterial blood was collected for measurement of the input curve. Indirect-3D reconstruction was performed using the Motion-compensation OSEM List-mode Algorithm for Resolution-recovery (MOLAR)3 followed by estimation of the voxel-wise volume of distribution (VT = K1/k2) using the one-tissue compartment model. Direct-4D reconstruction was performed with Parametric-MOLAR for the one-tissue model (PMOLAR-1T)4,5, which directly delivers parametric images of VT. Both reconstruction methods were performed with up to 4 iterations and 30 subsets, including appropriate corrections, with event-by-event motion correction, and used 60 min. of data. In addition, each subject’s list-mode file was down-sampled to 10% of the total counts, to create a set of 10 noisy replicates. Both the voxel-level variance within-subjects (across replicates) and between-subjects were evaluated and compared between indirect-3D and direct-4D.
Results: As expected, the variability within-subject is considerably lower for direct-4D, compared to indirect-3D, (35 ± 7% reduction in SD across GM ROIs, Figures 1 and 2(a) and (c)). At the voxel level, direct‑4D also delivers lower between-subject variance albeit at a smaller degree (25 ± 6% Figure 2(b) and (d)). This is expected since the between-subject SD is composed of a biological and an intrinsic noise (within-subject) component, which direct-4D reduces.
Conclusion: These results show that direct-4D reconstruction, compared to indirect-3D, is considerably better for within-subject and between-subject studies. This approach should increase the sensitivity to detect longitudinal changes in SV2A/synaptic density due to disease progression and treatment effect.
References
FinnemaSJ, et al.Imaging synaptic density in the living human brain. Sci Transl Med. 2016;
8.MatuskeyD, et al. Synaptic changes in Parkinson’s disease assessed with in-vivo imaging. Ann Neurol2020.JohnsonCA, et al.Software architecture of the MOLAR-HRRT reconstruction engine. IEEE Nuclear Science Symp Conf Record2004; 3956–3960.YanJPlaneta-WilsonBCarsonRE.
Direct 4D PET list mode parametric reconstruction with a novel EM algorithm. IEEE Trans Med Imaging2012;
31: 2213–2223.GerminoMGallezotJDYanJ, et al.Direct reconstruction of parametric images for brain PET with event-by-event motion correction: evaluation in two tracers across count levels. Phys Med Biol2017;
62: 5344–5364.
2021-138
Initial validation of reconstruction parameters in [18F]FDG PET brain images aiming scan time reduction (#374)
Samara Pinto1, Paulo Caribe1, 2, Cristina S. Matushita3, Diego B. Pianta3, Lucas Narciso1, 4 and Ana Maria Marques da Silva1
1PUCRS, Medical Image Computing Laboratory (MEDICOM), Porto Alegre, Brazil
2Ghent University, Medical Imaging and Signal Processing (MEDISIP), Ghent, Belgium
3Brain Institute of Rio Grande do Sul, Porto Alegre, Brazil
4Western University, Department of Medical Biophysics, London, ON, Canada
Abstract
Introduction: Positron emission tomography (PET) imaging with [18F]FDG provides valuable information regarding the underlying pathological processes in neurodegenerative disorders, such as Alzheimer’s disease (AD)1,2. Therefore, image reconstruction optimization strategies to reduce acquisition time are essential for this population due to involuntary motion3. This study aims to evaluate retrospective [18F]FDG PET data of healthy and AD individuals reconstructed with optimized reconstruction parameters to verify the feasibility of acquisition time reduction while ensuring satisfactory image quality and minimal impact on quantification.
Methods: PET data were obtained from 17 healthy individuals and 18 AD patients on a PET/CT scanner (GE Discovery 600) after injecting 240 ± 50 MBq of [18F]FDG. Images were reconstructed with the standard clinical settings (OSEM, 8 iterations, 16 subsets, 3 mm smoothing filter, 8 min acquisition time) and optimized reconstruction parameters4 (OSEM, 4 iterations, 32 subsets, 4-mm smoothing filter, 5 min acquisition time). Contrast, standardized uptake value ratio (SUVR, pons was used as reference), coefficient of variation (COV), signal-to-noise ratio (SNR), and bias were measured. Additionally, both sets of images were randomly and blindly presented to two experienced physicians. Image quality was scored (1–5; score 3 represented clinical quality) in image noise, contrast, and overall image quality. Paired t-test was used to identify differences between metrics.
Results: Metrics in Figure 1(a) from images reconstructed with optimized and standard parameters are similar for both groups. Image quality assessment (Figure 1(b)) showed no differences between clinical and optimized reconstruction parameters (Figure 2 for visual comparison). The 5 min acquisition represents a 40% reduction in imaging time compared to the standard protocol (8 min), without substantial differences in both groups’ quantitative or qualitative analyses.
Conclusion: Optimized reconstruction parameters resulted in images with similar quantification and quality compared with the standard clinical protocol, reinforcing our previous assessment in a phantom study4. Shortening the acquisition time is therefore possible by optimizing image reconstruction parameters without a loss in image quality. The reduced imaging time with equivalent image quality is essential for patients with neurodegenerative diseases, both by increasing patient comfort and limiting image artifacts due to head movements.
References
LoweVJ., et al.
Neuroimaging correlates with neuropathologic schemes in neurodegenerative disease.Alzheimer’s Dement2019;
15: 927–939.NordbergARinneJOKadirA, et al.The use of PET in Alzheimer disease. Nat Rev Neurol2010;
6: 78–87.IllesJRosenAGreiciusM, et al.Prospects for prediction: ethics analysis of neuroimaging in Alzheimer’s disease. Ann N Y Acad Sci2007;
1097: 278–295.PintoSOCaribePRRVNarcisoL, et al.Optimization of reconstruction parameters in [18F]FDG PET brain images aiming scan time reduction. Rev Bras Física Médica2021;
15: 611.
2021-139
Investigating autism spectrum disorder with synaptic density PET imaging (#376)
David Matuskey, Mika Naganawa, Sheida Koohsari, Takuya Toyonaga, Kristen Torres, Lauren Pisani, Caroline Finn, Julie Wolf, Nabeel Nabulsi, Jim Ropchan, Yiyun H. Huang, Richard E. Carson, Adam Naples and James C. McPartland
Yale University, New Haven Connecticut, USA
Abstract
Introduction: The neuropathology of autism spectrum disorder (ASD) is poorly understood at the molecular level. Evidence from animal models, genetics, post-mortem studies, and single-gene disorders suggests possible involvement of synaptic changes. Here, we use positron emission tomography (PET) to assess the density of synapses with synaptic vesicle glycoprotein 2A (SV2A) in autistic adults using 11C‑UCB‑J.
Methods: Eleven individuals with ASD (mean age (SD) 25 (4); six males), and eleven demographically matched healthy comparison controls (HC; 27 (4); six males) participated in an 11C‑UCB‑J PET scan. Volume of distribution (VT: ratio of activity in tissue relative to blood) was the primary outcome measure and computed with an arterial input function. Gray matter masking (GMM) and partial volume correction (PVC) with both Muller-Gartner (MG) and Iterative Yang (IY) were applied to control for possible volumetric differences. T-tests were calculated for between group differences, and p values were uncorrected for multiple comparisons given the exploratory nature.
Results: With a GMM, there were no significant findings between ASD and HCs, with brain regions varying between -4%-4%. Using MG, a significant lower VT was found with ASD in the orbitofrontal cortex (-12%, p = 0.04) and partial lobe (-12%, p = 0.03), with trends in the ventromedial prefrontal cortex (-9%, p = 0.07), whole frontal lobe (-10%, p = 0.08) and posterior cingulate (-13%, p = 0.08). Using IY, regions varied between -4%-5% and were not significant.
Conclusion: To our knowledge, this is the first in vivo investigation with SV2A PET in ASD. Contrary to expectations, the results obtained by the two PVC methods were different. Analyses are ongoing to understand the factors that may influence the PVC methods, such as the definition of brain regions, differences in white matter and differences in volume.
2021-140
Centrum semiovale as reference region – An evaluation of methods to quantify [18F]FPEB PET binding data in man (#377)
Max Andersson, Katarina Varnäs, Christer Halldin, Jacqueline Borg, Lars Farde and Johan Lundberg
Karolinska Institutet, Center for Psychiatry Research, Department of Clinical Neuroscience, Stockholm, Sweden
Abstract
Introduction: Information have been reported on 3-[18F]Fluoro-5-[(pyridine-3-yl)ethynyl]benzonitrile ([18F]FPEB)1 binding to the metabotropic glutamate receptor 5, based on Positron Emission Tomography (PET) data alone2, or in combination with quantification of radioactivity in arterial blood.3 In the absence of arterial data, cerebellar white matter (cerWM) has been suggested as a possible reference region.4 However, no comparisons with other potential reference regions are published and blocking studies including cerWM are lacking.
Methods: Eight healthy male subjects (age 20–27) were recruited and examined twice (2–7 days apart at approximately the same time of day) using the High Resolution Research Tomograph (Siemens Molecular Imaging). A tracer dose of the radioligand [18F]FPEB was given as a bolus injection and dynamic PET data was collected for 93 minutes and divided into 38 time frames. Arterial blood radioactivity was measured both continuously for the first 10 minutes and with discrete samples to produce a metabolite corrected plasma radioactivity curve. Regions of interest (ROI) were automatically delineated using Freesurfer 6.0. ROIs used as reference regions were created by uniform and automated editing of the obtained cerebellar and cerebral white matter ROIs. Tracer kinetics was modeled for each region of interest using the two tissue compartment model (2TCM) as well as the multilinear reference tissue model 2 (MRTM2) to produce measures of distribution volume (VT) and binding potential (BPND). Binding potentials using both cerWM and Centrum semiovale (CSO) as reference region were calculated. To investigate specific binding in proposed reference regions, autoradiography using [18F]FPEB and blocking with 10µM MPEP was performed.
Results: Including seven of the eight examined subjects, VT was numerically lower in CSO than in cerWM. Using CSO as reference region produced higher binding potentials than cerWM as well as a lower coefficient of variation (COV) and lower absolute percentage difference (APD) between the examinations (Table 1).5 Results from autoradiography indicate low but existing specific binding in suggested reference regions and are consistent with the results from PET.
Conclusion: While still exhibiting specific binding, CSO could be considered as an alternative to cerWM as reference region in studies with [18F]FPEB in the absence of arterial data.
Table 1. Mean and standard deviation (SD) of distribution volume (VT) and binding potential (BPND) in frontal cortex (FC), temporal cortex (TC), anterior cingulate cortex (ACC), caudate (CAU), cerebellar white matter (cerWM) and centrum semovale (CSO). 2TCM using arterial input and MRTM2 using CSO and cerWM as reference regions. Absolute percetage difference (APD) and coefficient of variance (COV).
References
WongDWaterhouseR. ‘
18F-FPEB, a PET radiopharmaceutical for quantifying metabotropic glutamate 5 receptors: a first-in-human study of radiochemical safety, biokinetics, and radiation dosimetry. J Nucl Med2013;
54: 388–396.BrašićJNandiA.
Reduced expression of cerebral metabotropic glutamate receptor subtype 5 in men with fragile x syndrome. Brain Sci2020;
10: 1–17.Leurquin-SterkGCeccariniJ.
Lower limbic metabotropic glutamate receptor 5 availability in alcohol dependence. J Nucl Med2018;
59: 682–690.SullivanJLimK.
Kinetic analysis of the metabotropic glutamate subtype 5 tracer [(18)F]FPEB in bolus and bolus-plus-constant-infusion studies in humans. J Cereb Blood Flow Metab2013;
33: 532–541.Further statistical comparisons will be added with the inclusion of the 8th subject.
2021-141
Comparison of a simultaneous estimation method for quantifying [18F]FEPPA uptake to standard compartmental analysis (#378)
Praveen Dassanayake1, 2, Udunna Anazodo1, 2, Linshan Liu2, Pablo M. Rusjan4, 5, Elizabeth Finger2, 3 and Keith St Lawrence1, 2
1University of Western Ontario, Department of Medical Biophysics, London, ON, Canada
2Lawson Health Research Institute, Imaging, London, ON, Canada
3University of Western Ontario, Department of Clinical Neurological Sciences, London, ON, Canada
4Douglas Research Centre, Human Neuroscience Division, Montréal, QC, Canada
5McGill University, Department of Psychiatry, Montréal, QC, Canada
Abstract
Introduction: The accepted method for quantifying [18F]FEPPA, a translocator protein 18 kDa (TSPO) tracer, is by two-tissue compartment modelling (2TCM) to measure the total distribution volume (VT). However, this approach requires ≥ 2-h acquisition and arterial sampling.1 In this work, we investigated a minimally invasive alternative that combines simultaneous estimation (SIME)2,3 with an image-derived input function (IDIF), metabolite corrected using venous samples. The approach was evaluated on retrospective [18F]FEPPAdata to assess the feasibility of reducing the scan time to 90 min. Venous sampling was verified against arterial sampling in animal studies. The combination was used to measure the binding potential (BPND) in a separate set of [18F]FEPPA data.
Methods: Retrospective data from Twelve healthy volunteers (8 high affinity binders, HABs; 4 mixed affinity binders, MABs) consisting of 180 min of scanning, injection of 5 ± 0.5 mCi [18F]FEPPA, and arterial sampling.1 SIME and 2TCM were applied to compare VT estimates across six regions-of-interest (ROIs). Next, SIME was applied to 180, 120 and 90 min of data to determine if BPND was affected by duration. Animal study consisted of ten pigs in which metabolite correction was applied to simultaneously acquired arterial and venous samples. The last aim involved seven healthy volunteers, 90 min of imaging, and IDIFs extracted using caliPER.4
Results: SIME and 2TCM produced similar mean values for VT and VND; however, SIME reduced the inter-subject variability in VT by 45 ± 17% (Figure 1). Reducing imaging to 90 min did not significantly change BPND (Figure 1(c)), and BPND was significantly higher for HABs than for MABs (i.e. frontal lobe BPND = 5.4 ± 0.8 HABs, 3.4 ± 0.7 MABs). Animal experiments demonstrated good agreement between venous and arterial [18F]FEPPA fractions (Figure 2(a)). Finally, BPND measured using venous metabolite-corrected IDIFs were in agreement with BPND from the retrospective data (Figure 2(c)).
Conclusion: Regional VT and VND estimates from SIME were in good agreement with 2TCM. SIME worked well with 90-min acquisition and was sensitive to the difference between HABs and MABs. Figure 2 indicates that replacing arterial with venous samples for metabolite correction and using IDIFs are both feasible, showing the potential of the minimally invasive SIME approach.
References
RusjanPMWilsonAA.Quantitation of translocator protein binding in human brain with the novel radioligand [18F]-FEPPA and positron emission tomography. J Cereb Blood Flow Metabolism2011;
31: 1807–1816.SchainMZanderigoF.
Non-invasive estimation of C-11 PBR28 binding potential. Neuroimage2018;
169: 278–285.SchainMZanderigoF.
Estimation of the binding potential BPND without a reference region or blood samples for brain PET studies. Neuroimage2017;
146: 121–131.DassanayakePCuiL. CALIPER: a software for blood-free parametric Patlak mapping using PET/MRI input function. bioRxiv2021;.
2021-142
PET imaging of demyelination in traumatic brain injury with [18F]3F4AP in mice (#379)
Karla M. Ramos-Torres1, Emiri Mandeville2, Kazue Takahashi1, Eng Lo2 and Pedro Brugarolas1
1Massachusetts General Hospital, Gordon Center for Medical Imaging, Department of Radiology, Boston Massachusetts, USA
2Massachusetts General Hospital, Department of Radiology, Charlestown, USA
Abstract
Introduction: Traumatic brain injury (TBI) results from energy transmission to the brain, disrupting its normal function.1 While clinical TBI is markedly heterogenous, the primary injury is characterized by damage to neurons, glial cells, and blood vessels. A complex secondary injury can later be observed by metabolic, cellular, and molecular changes that lead to cell death, tissue damage and white matter degeneration involving demyelination.2
Assesing myelin changes via PET imaging can shed light on the role of demyelination in TBI, providing better understanding of the disease and improvement in diagnosis and treatment monitoring. To this end we postulated that [18F]3F4AP, a novel radiotracer based on a multiple sclerosis drug that binds to potassium channels on demyelinated axons,3 could be used to image demyelination after TBI. Indeed, on a prior [18F]3F4AP imaging study in a rhesus macaque, we observed a focal hotspot at a previously injured brain area that could not be explained by changes in perfusion or inflammation.4 This finding motivated us to further characterize [18F]3F4AP response following brain injury using a rodent model of TBI.
Methods: Controlled cortical impact (CCI), a well-established model of TBI, was used in mice in combination with microPET studies and ex-vivo examination of the brains. [18F]3F4AP was synthesized according to previous reports.3,5 Dynamic PET/CT imaging was performed for at least 30min after IV injection of ∼200µCi of [18F]3F4AP. Animals were imaged over a 4-week period following the injury. Postmortem analysis (gamma counting, immunoblotting and histochemical staining) was performed on a subset of animals.
Results: Longitudinal PET imaging in mice after CCI showed increased [18F]3F4AP uptake peaking at 7 days-post-injury (33 ± 8% increase in SUV0-30min in the injury site, n = 6 mice). Gamma counting of the left vs. right side of the brain showed a 40% increase in radiotracer concentration on the injured side compared to the contralateral side (n = 3), consistent with the PET findings. Histochemical staining showed reduced myelin staining (LFB) on the injured side consistent with demyelination.
Conclusion: This study shows that [18F]3F4AP is a potential tracer for imaging TBI and a prospective noninvasive method for diagnosing and monitoring demyelinating diseases.
FlygtJDjupsjoALenneF, et al.Myelin loss and oligodendrocyte pathology in white matter tracts following traumatic brain injury in the rat. Eur J Neurosci2013;
38: 2153–2165, 12179.ArmstrongRCMierzwaAJMarionCM, et al.White matter involvement after TBI: clues to axon and myelin repair capacity. Exp Neurol2016;
275: 328–333.BrugarolasP, et al.Development of a PET radioligand for potassium channels to image CNS demyelination. Sci Rep2018;
8: 607.GuehlNJRamos-TorresKMLinnmanC, et al. Evaluation of the potassium channel tracer [18F]3F4AP in rhesus macaques. J Cereb Blood Flow Metab2020; 271678X20963404.BasuliFZhangXBrugarolasP, et al.An efficient new method for the synthesis of 3-[18 F]fluoro-4-aminopyridine via Yamada-Curtius rearrangement. JLabelled Comp Radiopharm2017;
61: 112–117.
2021-143
Risk and resilience in a rodent model of posttraumatic stress disorder: an in vivo [18F]FPEB and positron emission tomography imaging study examining the role of metabotropic glutamate receptor 5 (#380)
Ruth H. Asch1, Santosh Pothula1, Takuya Toyonaga2, Krista Fowles3, Stephanie M. Groman4, Rolando Garcia-Milian5, TuKiet T. Lam6, 7 and Irina Esterlis1, 8
1Yale University School of Medicine, Department of Psychiatry, New Haven Connecticut, USA
2Yale University School of Medicine, Department of Radiology & Biomedical Imaging, New Haven Connecticut, USA
3Yale University, Yale Positron Emission Tomography Center, New Haven Connecticut, USA
4University of Minnesota Medical School, Department of Neuroscience, Minneapolis Minnesota, USA
5Yale University School of Medicine, Bioinformatics Support Program, Cushing/Whitney Medical Library, New Haven Connecticut, USA
6Yale University School of Medicine, Department of Molecular Biophysics and Biochemistry, New Haven Connecticut, USA
7Yale University School of Medicine, Keck MS & Proteomics Resource, New Haven Connecticut, USA
8VA Connecticut Healthcare System, US Department of Veterans Affairs National Center for Posttraumatic Stress Disorder, West Haven Connecticut, USA
Abstract
Introduction: Evidence suggests individual differences in developing posttraumatic stress disorder (PTSD). Clinical investigations implicate the metabotropic glutamate receptor 5 (mGluR5) and related signaling in PTSD. Here, we utilized a preclinical model to assess the utility of mGluR5 as a biomarker of traumatic-stress vulnerability.
Methods: Male (n = 16) and female (n = 12) rats were exposed to a stress-enhanced fear leaning (SEFL) paradigm- a model recapitulating aspects of PTSD symptomatology- and compared with control animals (n = 7 male; n = 4 female). SEFL rats were identified as vulnerable or resilient based on high vs low freezing behavior. Positron emission tomography at baseline and postSEFL was used to calculate [18F]FPEB non-displaceable binding potential (BPND) in four regions of interest (ROIs): amygdala, hippocampus, striatum, and prefrontal cortex (PFC). Baseline vs. postSEFL differences in BPND were calculated as ΔBPND. Additionally, we performed functional analyses of differentially expressed proteins in the PFC.
Results: There were no significant between group (control vs resilient vs vulnerable) differences in mGluR5 availability at baseline or postSEFL. However, there were significant time by sex (p = 0.021) and group by sex by time (p = 0.043) interactions across ROIs, with vulnerable rats showing trends towards increased mGluR5 availability, whereas resilient rats showed trends toward decreased mGluR5 availability postSEFL relative to baseline. Interestingly, vulnerable males exhibited greater postSEFL increases in mGluR5 availability as compared to vulnerable females. Within the SEFL groups, greater increases in mGluR5were associated with more freezing behavior on day 1 of the SEFL paradigm (fear expression/learning; r = 0.424–0.538, p = 0.039–0.007), and PFC ΔBPND was positively correlated with day 2 freezing behavior (contextual fear memory; r = 0.504, p = 0.012). Functional proteomic analyses revealed further sex differences, with SEFL males showing upregulation of synaptic long-term depression and glucocorticoid signaling, but vulnerable females exhibited changes in glutathione metabolism and acute phase response pathways.
Conclusion: Increased mGluR5 availability may be involved in the etiology of PTSD-related phenotypes. These results additionally highlight potentially important sex differences in responses to traumatic stress, including differential involvement of mGluR5 and related molecular networks.
2021-144
Progression of Tau and neuroinflammation PET are independently associated with structural network reorganization in Alzheimer’s disease (#381)
Julie Ottoy1, Min-Su Kang2, Jonah Isen1, Gleb Bezgin2, Jean-Paul Soucy3, Serge Gauthier2, Sandra E. Black1, 4, Pedro Rosa-Neto2, 3 and Maged Goubran1, 5
1University of Toronto, LC Campbell Cognitive Neurology Unit, Hurvitz Brain Sciences Program, Sunnybrook Research Institute, Toronto, ON, Canada
3McGill University, Montreal Neurological Institute, Montreal, QC, Canada
4University of Toronto, Department of Medicine (Division of Neurology), Toronto, ON, Canada
5University of Toronto, Department of Medical Biophysics, Toronto, ON, Canada
Abstract
Introduction: Alzheimer’s disease (AD) is characterized by alterations in the global and local topology of the brain’s structural network.1 No studies have yet investigated how network topology in AD relates to amyloid, tau, and neuroinflammation together.
Methods: We employed diffusion-weighted MRI with probabilistic fiber tractography and graph theory approaches to investigate the topological organization of the white matter tracts in 105 cognitively normal (CN) amyloid&tau-negative, 37 CN amyloid-positive, and 72 MCI-AD amyloid-positive [TRIAD cohort]. Graph metrics were quantified globally and locally including: 1) connectivity strength; 2) efficiency2; 3) eigenvector centrality,3 reflecting nodal reinforcement of network influence; and 4) global modularity,4 reflecting intra-network connectivity and inter-network separation. All metrics were relative to their corresponding random-based metrics. We investigated associations of global graph metrics with amyloid-PET (18F-NAV4694) in AD-signature, and with tau-PET (18F-MK6240) and neuroinflammation-PET (11C-PBR28, n = 93) in Braak-stages. All associations were corrected for age, sex, and APOE-ε4. Local associations (32 atlas-based regions) were additionally adjusted for local volume and co-pathology and multiple comparisons. Finally, each graph metric was associated with cognition (MMSE) while adjusting for global amyloid and BraakI-IV tau.
Results: Globally, we found that: 1) Increased tau was significantly associated with all graph metrics (Figure 1), which became stronger with Braak-stage progression; 2) Increased neuroinflammation in BraakIII-IV and BraakV-VI was significantly associated with reduced connectivity strength (P = 0.008 and P = 0.04), which remained significant in BraakIII-IV after tau adjustment; and 3) Increased amyloid was significantly associated with reduced connectivity strength, modularity, and centrality (all P < 0.05), which disappeared after tau adjustment. Reduced strength and centrality were associated with cognitive impairment (P = 0.001 and P = 0.026; after adjustment for all PET). Locally, tau inversely related to centrality in temporal and posterior/parietal regions (PFDR < 0.05), suggesting reduced communication between these regions and the remaining network (Figure 2). In contrast, tau related positively to centrality in frontal and motor-sensory regions.
Conclusion: Tau pathology and neuroinflammation are associated with structural network alterations in AD with weakened and strengthened central roles for temporo-parietal and frontal regions, respectively. This may indicate a hierarchal shift of the structural network in AD where anterior/frontal networks gain importance, possibly indicating a compensatory mechanism for posterior/temporal dissociation.
References
DaiZHeY.
Disrupted structural and functional brain connectomes in mild cognitive impairment and Alzheimer’s disease. Neurosci Bull2014;
30: 217–232.MårtenssonGPereiraJBMecocciP, et al.Stability of graph theoretical measures in structural brain networks in Alzheimer’s disease. Scient Rep2018;
8: 11592BonacichP.Factoring and weighting approaches to status scores and clique identification. J Math Sociol1972;
2: 113–120BlondelVDGuillaumeJLLambiotteR, et al.Fast unfolding of communities in large networks. J Stat Mechan: Theory Experiment2008; P10008
2021-145
A reference tissue forward model for improved PET accuracy using within-scan displacement studies (#382)
Joseph B. Mandeville1, 2, Michael A. Levine1, 2, Wim Vanduffel3, 1, Bruce R. Rosen1 and Christin Y. Sander1, 2
1Massachusetts General Hospital, A.A. Martinos Center for Biomedical Imaging, Boston Massachusetts, USA
2Harvard Medical School, Boston Massachusetts, USA
3KU Leuven, Laboratory for Neuro- and Psychophysiology, Leuven, Belgium
Abstract
Introduction: PET displacement experiments often use within-scan challenges as an alternative to sequential scans to decrease the temporal separation of neural states and to decrease costs or increase subjects. A one-tissue model is often used to analyze such data. In this study, we developed a novel reference region method and compared this to a 1-tissue analysis using a dataset of lateralized dopamine release in a baboon model of deep brain microstimulation, informed by simultaneous fMRI and previous literature.
Methods: A baboon was implanted with a unilateral microstimulation probe in ventral tegmentum area (VTA) to elicit dopamine release in the projection to nucleus accumbens (NAc) as performed by others.1,2 Twelve [11C]raclopride scans in eight sessions were acquired using within-scan stimulation designs. Data were analyzed by the MRTM method plus challenge method3,4 and by a novel alternative: a forward-model full reference tissue model (fFRTM) using a single value of k4 to stabilize results. The forward approach avoided problems associated with inverse solutions based upon noisy data regressors,3,5 including invalid cost functions and errors from matrix inversion in the presence of noise. The optimal value of k4 for each session was determined from goodness of fit across a series of automatically generated regions spanning binding potentials of interest. Model performance was assessed using indices of 1) test-retest reproducibility in BPND, 2) stability, which measures the rate of convergence of BPND versus time, and 3) sidedness of the stimulus-induced change ΔBPND.
Results: fFRTM identified a best value for 1/k4 of 16.9 + 3 min. Reproducibility for fFRTM and the MRTM2 method were not statistically different within striatum. However, fFRTM offered considerably better stability (Figure 1), a critical index using within-scan designs. Mapping of stimulus-induced DBPND was lateralized using fFRTM with localization similar to fMRI, whereas MRTM2-based mapping was not lateralized and was consistent with simulated predictions of model bias as a function of baseline BPND.5
Conclusion: A new fFRTM implementation outperformed a 1-tissue analysis based upon MRTM for analysis of within-scan challenges by providing faster convergence of BPND, lateralization of DBPND consistent with simultaneous fMRI, and no worse test-retest reproducibility of BPND.
Acknowledgements
This research was supported by grants from the National Institutes Health (R01NS112295, K99DA043629, R00DA043629, P41EB015896, S10RR022976, S10RR019933, S10RR017208); Leuven (C14/17/109); Fonds Wetenschappelijk Onderzoek-Vlaanderen (FWO-Flanders; G0B8617N, G0E0520N); and the European Union’s Horizon 2020 Framework Programme for Research and Innovation under Grant Agreement No 945539 (Human Brain Project SGA3).
References
ArsenaultJTRimaSStemmannH, et al.Role of the primate ventral tegmental area in reinforcement and motivation. Curr Biol2014;
24: 1347–1353.SchluterEWMitzARCheerJF, et al.Real-time dopamine measurement in awake monkeys. PLoS One2014;
9: e98692.IchiseMLiowJSLuJQ, et al.Linearized reference tissue parametric imaging methods: application to [11C]DASB positron emission tomography studies of the serotonin transporter in human brain. J Cereb Blood Flow Metab2003;
23: 1096–1112.AlpertNMBadgaiyanRDLivniE, et al.A novel method for noninvasive detection of neuromodulatory changes in specific neurotransmitter systems. Neuroimage2003;
19: 1049–1060.MandevilleJBSanderCYWeyHY, et al.A regularized full reference tissue model for PET neuroreceptor mapping. Neuroimage2016;
139: 405–414.
2021-146
First-in-human use of [11C]CPPC with PET for imaging the macrophage colony stimulating factor 1 receptor in healthy brain (#383)
Jennifer M. Coughlin1, 2, Yong Du2, Wojciech Lesniak2, Courtney Harrington1, Mary Katherine Brosnan2, Riley O’Toole1, Adeline Zandi2, Shannon Eileen Sweeney1, Yunkou Wu2, Daniel Holt2, Robert Dannals2, Andrew Horti2, Andrew Hall2 and Martin G. Pomper2, 1
1Johns Hopkins Medical Institutions, Department of Psychiatry and Behavioral Sciences, Baltimore, Maryland, USA
2Johns Hopkins Medical Institutions, Russell H. Morgan Department of Radiology and Radiological Science, Baltimore, Maryland\, USA
Abstract
Introduction: Imaging of microglia, the brain’s resident immune cells, and their contribution to brain injury and repair across neuropsychiatric conditions has been widely pursued using PET-based radiotracers. However, relevant targets such as 18 kDa translocator protein may be expressed on other cell types in human brain. [11C]CPPC was developed to image the macrophage colony stimulating factor 1 receptor (CSF1R) that is essentially only expressed by microglia in the brain. We recently demonstrated high, specific uptake of [11C]CPPC in murine and nonhuman primate models of neuroinflammation, and now present its pharmacokinetic behavior in the brains of healthy individuals.
Methods: Eight individuals within 18–65 years of age and in stable health participated. [11C]CPPC was injected via bolus intravenous injection at the initiation of dynamic PET neuroimaging on a High Resolution Research Tomograph scanner. Ninety minutes of dynamic emission data were collected after bolus injection of [11C]CPPC, with collection of arterial blood samples for generation of a metabolite-corrected arterial input function. PET images were reconstructed into 30 frames using the iterative ordinary-Poisson ordered-subset expectation-maximization algorithm. Time-activity curves (TACs) were generated for regions of interest using the co-registered magnetic resonance imaging data from each participant. One- and two-tissue compartmental models (1TCM and 2TCM), as well as Logan graphical analysis were then compared.
Results: Plasma activity peaked within 60 seconds post injection and decreased to < 5% of the peak by 10 minutes. HPLC isolated [11C]CPPC from its radiometabolites, which were more polar and well resolved from the parent compound. [11C]CPPC represented ∼80% of total plasma activity by 30 minutes and ∼45% by 90 minutes. [11C]CPPC rapidly entered the brain, and cortical and subcortical tissue TACs peaked by 30–35 minutes post injection and then declined. The 1TCM was preferred. Total distribution volume values computed from 1TCM aligned well with those from Logan graphical analysis (t* = 30), with VT values listed highest to lowest in thalamus > cortical regions > hippocampus.
Conclusion: For all healthy subjects, [11C]CPPC PET data yielded a pattern of brain uptake in vivo that fits well a 1TCM in key regions of interest, and supports study further of [11C]CPPC for imaging the CSF1R in the healthy human brain.
2021-147
Interaction between vascular risk and Alzheimer’s disease pathology boosts neurodegeneration and cognitive decline in cognitively unimpaired individuals (#384)
João Pedro Ferrari-Souza1, 2, Wagner S. Brum3, Lucas A. Hauschild1, Lucas U. Da Ros1, Pâmela C. Lukasewicz Ferreira1, 2, Bruna Bellaver1, 2, Douglas T. Leffa4, Andrei Bieger1, Cécile Tissot5, 2, Marco Antônio De Bastiani1, Andréa L. Benedet3, 5, Joseph Therriault5, Diogo O. Souza1, Pedro Rosa-Neto5, Thomas Karikari3, 2, Tharick A. Pascoal2, 5 and Eduardo R. Zimmer1, 6
1Universidade Federal do Rio Grande do Sul, Department of Biochemistry, Porto Alegre, Brazil
2University of Pittsburgh, Department of Psychiatry, Pittsburgh Pennsylvania, USA
3Sahlgrenska University Hospital, Clinical Neurochemistry Laboratory, Gothenburg, Sweden
4Hospital de Clínicas de Porto Alegre, ADHD Outpatient Program & Development Psychiatry Program, Porto Alegre, Brazil
5McGill University, McGill Centre for Studies in Aging, Montreal, QC, Canada
6Universidade Federal do Rio Grande do Sul, Department of Pharmacology, Porto Alegre, Brazil
7The Sahlgrenska Academy at the University of Gothenburg, Department of Psychiatry and Neurochemistry, Mölndal, Sweden
8McGill University, Department of Neurology and Neurosurgery, Montreal, QC, Canada
Abstract
Introduction: Vascular risk factors (VRFs) play an important role in the development of Alzheimer’s disease (AD). However, it is not fully understood how VRFs are associated with AD pathology to promote neurodegeneration and cognitive decline. In the present retrospective cohort study, our primary objective was to test whether VRF burden synergistically interacts with AD pathology to accelerate neurodegeneration and cognitive decline in cognitively unimpaired (CU) individuals. Secondarily, we aimed to assess whether VRF burden is related to changes in cerebrospinal fluid (CSF) AD biomarkers.
Methods: We assessed CU participants from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. Baseline VRF burden was calculated considering the history for cardiovascular disease, hypertension, diabetes mellitus, hyperlipidemia, stroke or transient ischemic attack, smoking, atrial fibrillation, and left ventricular hypertrophy. AD pathology was evaluated using CSF amyloid-β1-42 (Aβ1-42) and CSF tau phosphorylated at threonine 181 (p-tau181). Individuals were dichotomized as having an elevated VRF burden if ≥ 2 VRFs (V+) and as presenting biological AD if CSF p-tau181 ≥ 24 pg/mL and CSF Aβ1-42 ≤ 976.6 pg/mL [(AT)+]. Linear mixed-effects (LME) models were performed to evaluate the longitudinal trajectory of (i) plasma neurofilament light (NfL; n = 269), (ii) cognition indexed by modified version of Preclinical Alzheimer’s Cognitive Composite (mPACC; n = 503), and (iii) CSF AD biomarkers (Aβ1-42 and p-tau181; n = 284). Additionally, time-to-event analysis was carried to evaluate the risk of clinical progression to cognitive impairment (n = 503).
Results: LME models demonstrated that an elevated VRF burden interacted with AD pathology to promote higher rates of neurodegeneration (β = 5.68, p = .005; Figure 1(a)) and cognitive decline (β = −0.43, p = .019; Figure 1(b)). Nonetheless, VRF burden was not associated with CSF Aβ1-42 or p-tau181 changes over time. Survival analysis demonstrated that only individuals with the presence of both biological AD and elevated VRF burden [(AT)+V+] had a significantly greater risk of clinical progression to cognitive impairment (adjusted Hazard Ratio = 3.5, p < .001).
Conclusion: Our results suggest that VRF burden and Alzheimer’s disease pathology are independent processes, however, they synergistically lead to neurodegeneration and cognitive decline, favoring the onset of cognitive impairment. These findings support that the clinical evaluation of vascular risk factor burden might improve the clinical assessment especially of subjects at higher risk for developing cognitive impairment.
2021-148
APOE isoforms differentially modulate the associations between regional tau deposition and neuroinflammation in Alzheimer’s disease (#385)
Yi-Ting Wang1, 3, Andréa L. Benedet4, Cécile Tissot1, 3, Firoza Z. Lussier1, 3, Gleb Bezgin1, 3, Stijn Servaes1, 3, Joseph Therriault1, 3, Min-Su Kang1, 2, Jaime Fernandez Arias1, 3, Sulantha Mathotaarachchi1 and Pedro Rosa-Neto1, 2
1McGill University, The McGill University Research Centre for Studies in Aging, Montreal, QC, Canada
2McGill University, Department of Neurology and Neurosurgery, Montreal, QC, Canada
3McGill University, Douglas Research Centre, Montreal, QC, Canada
4University of Gothenburg, Institute of Neuroscience and Physiology, Gothenburg, Sweden
Abstract
Introduction: Apolipoprotein E (APOE) is a 34-kDa glycoprotein highly expressed in the CNS, primarily by astrocytes and microglia. Three allelic variants: APOE ε2, APOE ε3 and APOE ε4 exist, with this APOE polymorphisms being recognised as the most important risk factors for sporadic Alzheimer disease (AD). One hallmark of AD is the presence of neuroinflammation, which manifests as the presence of both reactive astrocytes and microglia. Mounting evidence suggests that APOE has immunomodulatory effects and it seems to modulate brain inflammatory responses in an isoform-dependent manner. The role of APOE ε4 in neuroinflammation was interrogated in previous studies. Experiments in APOE targeted-replacement mice showed that APOE ε4 mice had a greater inflammatory response to LPS injection than APOE ε3 mice. However, the relevance of these mechanisms to other isoform-dependent disease-relevant observations, such as differential tau deposition still warrants further investigation.
Methods: This was a cross-sectional study examining a total number of 162 subjects from the TRIAD cohort at McGill University Research Centre for Studies in Aging, Canada. Cerebral tau neurofibrillary tangles were assessed using positron emission tomography (PET) radiopharmaceuticals [18F]MK6240. Cerebrospinal fluid (CSF) neuroinflammation biomarkers including soluble triggering receptor expressed on myeloid cells 2 (sTREM2), YKL40 and glial fibrillary acidic protein (GFAP) were also measured. Voxelwise analyses were performed to evaluate the relationships between regional tau burden and neuroinflammation biomarkers.
Results: Voxelwise analyses revealed that subjects with different APOE variants showed different relationships between neuroinflammation biomarkers and tau burden. Noteworthy, in APOE ε4 carriers, a significant correlation between neuroinflammation biomarkers and tau burden was found in medial temporal regions, after corrected for age, sex and pathological status. These associations were absent in APOE ε2 or ε3 subjects, suggested APOE exerts immunomodulatory effects in an isoform-dependent way.
Conclusion:APOE isoforms differentially modulate the associations between regional tau deposition and neuroinflammation.
APOE modulates the relationships between neuroinflammation biomarkers and tau burden
The models were corrected for age, sex, pathological status and educational attainment. T-statistical parametric maps were corrected for multiple comparisons using a random field theory cluster threshold of P < 0.001, overlaid on the ADNI reference template.
2021-149
Employing simultaneous (EEG-)PET-MRI to map arousal-induced hemodynamic and metabolic dynamics (#387)
Jingyuan Chen1, 2, Ciprian Catana1, 2, Jonathan Polimeni1, 2, Kyle Droppa1, Nina Fultz1, Hsiao-Ying Wey1, 2, Julie C. Price1, 2, Bruce R. Rosen1, 2, Laura Lewis3 and Christin Y. Sander1, 2
1Massachusetts General Hospital, Radiology, Boston Massachusetts, USA
2Harvard Medical School, Radiology, Boston Massachusetts, USA
3Boston University, Biomedical Engineering, Boston Massachusetts, USA
Abstract
Introduction: Arousal-induced metabolic dynamics have previously been characterized by quantifying differential metabolism between two separate PET scans.1,2 In this study, we show that a recently developed technique functional PET (fPET)-FDG3,4 can track metabolic dynamics accompanying sleep-wake transitions within a single scan; and that by integrating fPET-FDG with simultaneous EEG-fMRI, one can jointly map global and local physiological and metabolic changes regulated by arousal.
Methods: Each subject underwent one night of partial sleep deprivation and took part in a 60∼120 min PET-MRI scan the next afternoon. Subjects’ arousal states were delineated according to concurrent EEG or behavioral measures. Data were collected on a 3T MR scanner with a BrainPET insert. FDG was administered using a 20% bolus plus constant infusion paradigm (injected activity up to 8 mCi).4 Reconstruction and attenuation correction of PET images (2.5 mm iso., 30-s temporal resolution) followed.3 BOLD-weighted EPI data (3 mm iso., TR = 2/2.4 s) were collected to track ongoing hemodynamic changes. Sleep-induced changes in hemodynamic signals vs. glucose metabolism: standard deviations of fMRI percent signal changes in each arousal state were computed and compared; the slopes of fPET signals (approximately proportional to glucose metabolism3) were fitted separately for each arousal state and compared.
Results: Our results demonstrated that fPET-FDG can track metabolic dynamics coupled to arousal (Figure 1), with decreased slopes (reduced glucose uptakes) at the transitions from wake to sleep. In line with previous studies, during sleep, the strongest changes in hemodynamic fluctuation amplitudes occurred in the sensory regions5; and the metabolic reductions were most salient in the association cortices and thalamus.2. We additionally noticed that the spatial distributions of sleep-induced hemodynamic increases and metabolic decreases were highly complementary to each other (Figure 2).
Conclusion: We have implemented a multi-modal framework that enables us to jointly map and link transient changes of hemodynamic and metabolic signals accompanying brain arousal state transitions, thus opening up the opportunity to model the interactions among various dynamic processes and achieve a holistic view on the mechanisms underlying arousal regulation in ensuing investigations.
Acknowledgements
This work was supported in part by the NIH (grants K99-NS118120, R01-MH111438, P41-EB015896), by the MGH/HST Athinoula A. Martinos Center for Biomedical Imaging; and was made possible by the resources provided by NIH Shared Instrumentation Grants S10OD010759, S10RR022976, S10RR019933 and S10RR017208.
References
NofzingerE, et al.Human regional cerebral glucose metabolism during non‐rapid eye movement sleep in relation to waking. Brain2002;
125: 1105–1115.BuchsbaumM, et al.Regional cerebral glucose metabolic rate in human sleep assessed by positron emission tomography.Life Sci1989;
45: 1349–1356.VillienM, et al.Dynamic functional imaging of brain glucose utilization using fPET-FDG. Neuroimage2014;
100: 192–199.HahnA, et al.Quantification of task-specific glucose metabolism with constant infusion of 18F-FDG. J Nucl Med2016;
57: 1933–1940.FukunagaM, et al.Large-amplitude, spatially correlated fluctuations in BOLD fMRI signals during extended rest and early sleep stages.Magn Reson Imaging2006;
24: 979–992.
2021-150
Biomarker modeling of Alzheimer’s disease using PET-based in vivo Braak staging (#388)
Joseph Therriault, Tharick A. Pascoal, Firoza Z. Lussier, Cécile Tissot, Mira Chamoun, Gleb Bezgin, Stijn Servaes, Andréa L. Benedet, Nicholas J. Ashton, Thomas Karikari, Juan Lantero-Rodriguez, Yi-Ting Wang, Jaime Fernazdez-Arias, Gassan Massarweh, Paolo Vitali, Jean-Paul Soucy, Paramita Saha-Chaudhuri, Kaj Blennow, Henrik Zetterberg, Serge Gauthier and Pedro Rosa-Neto
McGill University, Montreal, QC, Canada
Abstract
Introduction: Alzheimer’s disease (AD) is the leading cause of dementia worldwide and can be detected in vivo using biomarkers of amyloid-b and tau. Gold standard diagnostic methods for AD rely on staging systems to measure disease severity, which have not yet been incorporated into the in vivo biological research framework for AD. The topographical information conferred by tau-PET offers the potential to translate the gold standard histopathological Braak staging system to living individuals.
Methods: Using [18F]MK6240 tau-PET, we applied the Braak staging model to 345 individuals. We measured amyloid-PET, cerebrospinal fluid (CSF) and plasma phosphorylated tau (pTau) epitopes, neurodegeneration and neuropsychological function in relation to in vivo Braak stage. ANOVA assessed relationships between Braak stage with biomarker and clinical changes. Progression of in vivo Braak stage was assessed in 163 individuals with follow-up [18F]MK6240 scans.
Results: Advancing in vivo Braak stage was associated rising plasma pTau species, as well as rise and plateau of amyloid-PET & CSF pTau species. Early Braak stages were associated with isolated memory impairment, while later Braak stages were closely associated with the severity of dementia. Follow up tau-PET scans indicated sequential progression of in vivo Braak stages over time, with substantially higher rates of progression in amyloid-b+ individuals.
Conclusion:In vivo Braak stages had stage-specific associations with the severity of amyloid-b deposition, CSF measures of pTau and clinical function. In vivo Braak staging contributes to the understanding of the natural history of biological AD and provides a framework to measure AD severity in living humans.
2021-151
Neuro-immune signatures in chronic low back pain subtypes (#389)
Zeynab Alshelh1, Ludovica Brusaferri1, Atreyi Saha1, Erin J. Morrissey1, Paulina C. Knight1, Minhae Kim1, Yi Zhang2, Jacob M. Hooker1, Daniel Albrecht1, Angel Torrado-Carvajal1, 3, Michael S. Placzek1, Oluwaseun J. Akeju2, Julie C. Price1, Robert Edwards4, Jeungchan Lee1, Roberta Sclocco1, 5, Ciprian Catana1, Vitaly Napadow1, 4 and Marco L. Loggia1
1Massachusetts General Hospital/Harvard Medical School, Radiology, Boston Massachusetts, USA
2Massachusetts General Hospital/Harvard Medical School, Anaesthesia, Critical Care and Pain Medicine, Boston Massachusetts, USA
3Universidad Rey Juan Carlos, Medical Image Analysis an Biometry Laboratory, Madrid, Spain
4righam and Women’s Hospital/Harvard Medical School, Department of Anaesthesiology, Perioperative and Pain Medicine, Boston Massachusetts, USA
5Logan University, Radiology, Chesterfield Missouri, USA
Abstract
Introduction: We recently showed that patients with different chronic pain conditions demonstrated elevated brain levels of the glial marker 18kDa translocator protein, which suggests that neuroinflammation might be a pervasive phenomenon observable across multiple etiologically heterogeneous pain disorders. Interestingly, the spatial distribution of this neuroinflammatory signal appears to exhibit a degree of disease specificity, suggesting that different pain conditions may exhibit distinct “neuroinflammatory signatures”. To further explore this hypothesis, we tested whether neuroinflammatory signal can characterize putative etiological subtypes of chronic low back pain patients based on clinical presentation. Specifically, we explored neuroinflammation in patients whose chronic low back pain either did or did not radiate to the leg (i.e., “radicular” vs. “axial” back pain).
Methods: Fifty-four chronic low back pain patients, twenty-six with axial back pain (43.7 ± 16.6 y.o. [mean ± SD]) and twenty-eight with radicular back pain (48.3 ± 13.2 y.o.), underwent PET/MRI with [11C]PBR28, a second-generation radioligand for the 18kDa translocator protein. [11C]PBR28 signal was quantified using standardized uptake values ratio. Functional MRI data were collected simultaneously to the [11C]PBR28 data 1) to functionally localize the primary somatosensory cortex back and leg subregions and 2) to perform functional connectivity analyses (in order to investigate possible neurophysiological correlations of the neuroinflammatory signal). PET and functional MRI measures were compared across groups, cross-correlated with one another and with the severity of “fibromyalgianess”. Furthermore, statistical mediation models were employed to explore possible causal relationships between these three variables.
Results: For the primary somatosensory cortex representation of back/leg, [11C]PBR28 PET signal and functional connectivity to the thalamus were: 1) higher in radicular compared to axial back pain patients, 2) positively correlated with each other and 3) positively correlated with fibromyalgianess scores, across groups. Finally, 4) fibromyalgianess mediated the association between [11C]PBR28 PET signal and primary somatosensory cortex-thalamus connectivity across groups.
Conclusion: Our findings support the existence of “neuroinflammatory signatures” that are accompanied by neurophysiological changes, and correlate with clinical presentation in chronic pain patients. These signatures may contribute to the subtyping of distinct pain syndromes and also provide information about inter-individual variability in neuro-immune brain signals, within diagnostic groups, that could eventually serve as targets for mechanism-based precision medicine approaches.
Figure 1. [11C]PBR28 signal in S1 and S1-thalamus connectivity correlations with Fibromyalgia Survey Scores.
Top panel: Average SUVR was extracted from S1 (see Figure 4 caption) and plotted against Fibromyalgia Survey Scores (data have been adjusted for scanner and genotype). Bottom panel: S1 – thalamus connectivity values were extracted (see Figure 4 caption) and plotted against Fibromyalgia Survey Scores (data have been adjusted for scanner). Triangle denotes data from Protocol 1, circle denotes data from Protocol 2.
Figure 2. Voxel-wise group differences in [11C]PBR28 signal
a. Maps displaying areas with significantly elevated [11C]PBR28 SUVR in cLBPRAD compared to cLBPAX in a voxelwise analysis, adjusted for Protocol and genotype. b. Average ± standard deviation SUVR extracted from several clusters identified as statistically significant in the voxelwise SUVR analysis from A(adjusted for scanner and genotype). c. BOLD fMRI localizing the somatotopic representation of S1 area for the Back+Leg and the overlap between Back+Leg e-stim and [11C]PBR28 SUVR signal in cLBPRAD > cLBPAX.
2021-152
Tau accumulation using [18F]MK6240 PET is associated with increase in executive dysfunction in prodromal AD (#390)
Vanessa Pallen1, Firoza Z. Lussier2, Nina M. Poltronetti1, Joseph Therriault2, Cécile Tissot2, Gleb Bezgin2, Jaime F. Arias1, 5, Nesrine Rahmouni1, Jenna Stevenson1, Alyssa Stevenson1, Sulantha Mathotaarachchi2, Andréa L. Benedet3, Min-Su Kang2, Yi-Ting Wang2, Mira Chamoun1, Pedro Rosa-Neto1, 5, Serge Gauthier6, 4 and Tharick A. Pascoal2, 1
1McGill University Research Center for Studies in Aging, Verdun, QC, Canada
5Montreal Neurological Institute, Montreal, QC, Canada
6Douglas Hospital, Mental Health University Institute, Verdun, QC, Canada
Abstract
Introduction: Decline in executive function has been noted to precede other forms of cognitive decline in the prodromal stage of Alzheimer’s disease (Clark et al., 2011). The D-KEFS Color-Word Interference test (CWIT) is a cognitive test measuring executive function across several conditions. Few studies have explored the relationship between tau deposition and executive dysfunction. However, to date, no study has explored this relationship in prodromal AD. This study aimed at exploring such associations
Methods: We looked at cognitively unimpaired and impaired elderly individuals from the TRIAD cohort in Montreal, QC who underwent neuropsychological evaluations across 2 time points. In vivo tau accumulation was assessed at baseline with [18F]-MK6240 PET. Standardized uptake value ratio was calculated 90–110 minutes post-injection using cerebellar grey matter as the reference region. Voxel-wise linear regression models were applied with the rate of change of CWIT condition 3 and 4 completion times as predictor variables with tau tracer binding being the response variable. Age, sex, education, Geriatric Depression Scale score, APOEε4, diagnostic group, and amyloid-β status were included as covariates in regression models. Results were corrected for multiple comparisons using random field theory with a cluster threshold of p < .001.
Results: A total of 105 (CN = 65;MCI = 36;AD = 4) individuals were included in this analysis, with mean follow-up time of neuropsychological evaluation being 15.9 months. We found that brain [18F]-MK6240 uptake was correlated with change in D-KEFS CWIT condition 3 and 4 completion times, with a greater pattern of association in condition 3 (inhibition). For condition 3, associations were strongest in the precuneus, posterior cingulate, and primarily right dorsal frontal and parietal regions. For condition 4 (flexibility), we found a strong association in the left temporal pole and left dorsal frontal regions.
Conclusion: These results support previous claims that tau follows a distinct regional distribution similar to that of the dorsal attention network which is related to higher order cognitive function such an inhibition and switching (Hansson et al., 2017). Additionally, given that most individuals in this study were CN or MCI, our findings suggest that tau pathology is associated with a decrease in executive function even in prodromal stages of AD.
Figure 1. Tau & alternating task.
[18F]-MK6240 SUVR ∼ rate of change in inhibition/switching task completion time + age + education + GDS score + APOE4 + Dx group + amyloid-β status
Figure 2. Tau & inhibition task.
[18F]-MK6240 SUVR ∼ rate of change in inhibition task completion time + age + education + GDS score + APOE4 + Dx group + amyloid-β status
References
ClarkLRSchiehserDMWeissbergerGH, et al.Specific measures of executive function predict cognitive decline in older adults. J Int Neuropsychol Soc2011;
18: 118–127.HanssonOGrotheMJStrandbergTO, et al.Tau pathology distribution in Alzheimer’s disease corresponds differentially to cognition-relevant functional brain networks. Front Neurosci2017;
11. https://doi.org/10.3389/fnins.2017.00167
2021-153
Neuroimaging VMAT2 in Parkinson’s disease with rapid eye movement sleep behaviour disorder (#391)
Mikaeel Valli1, 2, Sang Soo Cho3, Carme Uribe1, Mario Masellis4, 2, Robert Chen5, 2, Alexander Mihaescu1, 2 and Antonio P. Strafella1, 5
1CAMH, Brain Health Imaging Centre, Toronto, ON, Canada
2University of Toronto, Institute of Medical Science, Toronto, ON, Canada
3Seoul National University, Seoul, South Korea
4Sunnybrook Health Sciences Centre, Toronto, ON, Canada
5Toronto Western Hospital, Toronto, ON, Canada
Abstract
Introduction: REM sleep behaviour disorder (RBD) is a common condition found in 50% of Parkinson’s disease (PD) patients. Molecular imaging evidence shows that PD with RBD (PD-RBD+) show lower dopamine transporter activity within the caudate and putamen compared to PD without RBD (PD-RBD–). However, the characterization of the vesicular monoamine transporter 2 (VMAT2), an index of nigrostriatal dopamine innervation, has been rarely explored in PD patients with RBD.
Methods: We enrolled 15 PD-RBD+, 15 PD-RBD– and 15 age matched healthy controls (HC) for the [11C]DTBZ PET imaging study. This technique measures VMAT2 availability within striatal regions of interest (ROI). Mixed effect model was used to compare the radioligand binding of VMAT2 between the three groups for each striatal ROI, while co-varying for sex, cognitive and depression scores. Multiple regressions were also computed to predict clinical measures from group condition and VMAT2 binding within all ROIs explored. Significant level was set at p < 0.05 (Bonferroni corrected).
Results: We observed significant main effect of group condition on VMAT2 availability within the caudate, putamen, ventral striatum, globus pallidus, substantia nigra, and subthalamus. Specifically, we observed that both PD-RBD+ and PD-RBD– group had lower VMAT2 availability compared to HC. PD-RBD– showed a negative relationship between motor severity and VMAT2 availability within the left caudate. This relationship was not found in the PD-RBD+ group.
Conclusion: Our findings reveal that both PD patient subgroups had reduced VMAT2 levels relative to HC—which reflects denervation within the nigrostriatal pathway. No significant interactions were detected between radioligand binding and clinical scores in PD-RBD+. Taken together, we found limited evidence that VMAT2 may contribute differently in PD-RBD+ relative to PD-RBD–. Future studies are encouraged to explore other underlying neural chemistry mechanisms contributing to RBD in PD patients.
Acknowledgements
The authors thank Alvina Ng, Laura Nguyen and Anusha Ravichandran for their technical assistance. The authors also thank Marcos Sanches for his invaluable assistance with the statistical analyses.
This work was supported by Canadian Institutes of Health Research (CIHR) (MOP 136778). APS was supported by the Canada Research Chair program. MV was supported by the CIHR’s Doctoral Award program.
2021-154
Olfactory impairment is related to tau pathology and neuroinflammation in Alzheimer’s disease (#392)
Julia I. Klein1, 3, Xinyu Yan2, Aubrey Johnson1, Zeljko Tomljanovic1, James Zou1, Krista Polly1, Lawrence S. Honig1, Adam M. Brickman1, Yaakov Stern1, 4, Davangere P. Devanand4, Seonjoo Lee2, 5 and William C. Kreisl1
1Columbia University Irving Medical Center, Taub Institute, New York New York, USA
2Columbia University Irving Medical Center, Mailman School of Public Health, New York New York, USA
3Weill Cornell Medicine, New York New York, USA
4Columbia University Irving Medical Center, Gertrude H. Sergievsky Center, New York New York, USA
5The Research Foundation for Mental Hygiene, Inc, New York New York, USA
Abstract
Introduction: Olfactory impairment is evident in Alzheimer’s disease (AD), however, its precise relationships with clinical biomarker measures of tau pathology and neuroinflammation are not well understood. To determine if odor identification performance measured with the University of Pennsylvania Smell Identification Test (UPSIT) is related to in vivo measures of tau pathology and neuroinflammation.
Methods: Cognitively normal and cognitively impaired participants were selected from an established research cohort of adults aged 50 and older who underwent neuropsychological testing, brain MRI, and amyloid PET. Fifty-four participants were administered the UPSIT. Forty-one underwent 18F-MK-6240 PET (measuring tau pathology) and fifty-three underwent 11C-PBR28 PET (measuring TSPO, present in activated microglia). Twenty-three participants had lumbar puncture to measure CSF concentrations of total tau (t-tau), phosphorylated tau (p-tau) and β-amyloid (Aβ42).
Results: Low UPSIT performance was associated with greater18F-MK-6240 binding in medial temporal cortex, hippocampus, middle/inferior temporal gyri, inferior parietal cortex and posterior cingulate cortex (p < 0.05). Similar relationships were seen for 11C-PBR28. These relationships were primarily driven by amyloid-positive participants. Lower UPSIT performance was associated with greater CSF concentrations of t-tau and p-tau (p < 0.05). Amyloid status and cognitive status exhibited independent effects on UPSIT performance (p < 0.01).
Conclusion: Olfactory identification deficits are related to extent of tau pathology and neuroinflammation, particularly in those with amyloid pathophysiology. The independent association of amyloid-positivity and cognitive impairment with odor identification suggests that low UPSIT performance may be a marker for AD pathophysiology in cognitive normal individuals, although impaired odor identification is associated with both AD and non-AD related neurodegeneration.
2021-155
Associations between neutrophils and amyloid deposition in the Alzheimer’s disease spectrum (#393)
Nesrine Rahmouni1, 2, Cecile Tissot1, 2, Gleb Bezgin1, 2, Jenna Stevenson1, 2, Alyssa Stevenson1, 2, Firoza Z. Lussier1, 2, Joseph Therriault1, 2, Mira Chamoun1, 2, Min-Su Kang1, 2, Dayna Baldo1, 2, Tharick A. Pascoal2, 3, Andréa L. Benedet2, 4 and Pedro Rosa-Neto1, 2
1McGill University Research Center for Studies in Aging, McGill, Montreal, QC, Canada
3University of Pittsburgh, Psychiatry, Pittsburgh Pennsylvania, USA
4University of Gothenburg, Gothenburg, Sweden
Abstract
Introduction: Neutrophils are key components of early innate immunity and contribute to uncontrolled systemic inflammation. Several studies have highlighted the link between systemic inflammation and the Alzheimer’s disease (AD) pathophysiology. In fact, experiments with animal models and studies with AD patients demonstrated the hyperactivation of neutrophils associated with AD pathology and cognitive decline. However, the comprehension of the inflammatory component in the AD spectrum is still uncomplete. We thus investigated whether the amount of systemic neutrophils is correlated with brain amyloid-β in the AD spectrum.
Methods: The present study was conducted in a population of 213 individuals (138 cognitively unimpaired (CU) and 75 cognitively impaired (CI; 35 MCI and 20 AD) from the Translational Biomarkers in Aging and Dementia TRIAD cohort. The neutrophil relative values were assessed using the Automated Beckman DXH hematology Analyzer. Amyloid (Aβ) deposition was assessed with [18F]AZD4694 PET and standardized uptake value ratio (SUVRs) were calculated between 40 to 70 min post-injection, using cerebellum grey matter as the reference region. A voxel-based regression model evaluated the relationship between neutrophil count and Aβ PET, adjusted for age, sex, years of education and diagnosis. RFT was used to account for multiple comparisons.
Results: In the present study, a positive association was found between systemic neutrophil counts and brain Aβ load, where the associated regions were the anterior cingulate, cuneus, and occipital pole areas (Figure 1). Also, CI individuals showed significantly higher neutrophil counts as compared to the CU group (Figure 2).
Conclusion: Our findings review the link between the peripheral immune system and the central nervous system amyloid deposition. Further studies should investigate the use of neutrophil counts as biomarkers of AD.
Figure 1. Correlation between neutrophil counts and AB global SUVR.
CI individuals showed significantly higher neutrophil counts as compared to the CU group.
Figure 2. Association between systemic neutrophil count and amyloid load assessed with [18F]AZD4694 PET.
A positive association was found between systemic neutrophil counts and brain Aβ load, where the associated regions were the anterior cingulate, cuneus, and occipital pole areas.
2021-156
Verbal Fluency associated with tau accumulation and not amyloid deposition in the Alzheimer’s disease spectrum (#395)
Nesrine Rahmouni1, 2, Alyssa Stevenson1, 2, Jenna Stevenson1, 2, Cecile Tissot1, 2, Firoza Z. Lussier1, 2, Gleb Bezgin1, 2, Joseph Therriault1, 2, Mira Chamoun1, 2, Nina M. Poltronetti1, 2, Vanessa Pallen1, 2, Min-Su Kang1, 2, Tharick A. Pascoal2, 3, Paolo Vitali1, 2, Andréa L. Benedet2, 4 and Pedro Rosa-Neto1, 2
1McGill University Research Center for Studies in Aging, McGill, Montreal, QC, Canada
3University of Pittsburgh, Department of Psychiatry, Pittsburgh Pennsylvania, USA
4University of Gothenburg, Gothenburg, Sweden
Abstract
Introduction: Recent studies have shown that pathological amyloid deposition and tau accumulation, hallmarks of the Alzheimer’s disease, are closely related to cognitive deficits in older individuals. The present study evaluates the associations between the Alzheimer’s disease hallmarks and verbal fluency and lexical speed access impairment.
Methods: The present study was conducted in a population of 262 individuals (162 cognitively unimpaired individuals (CU), 100 cognitively impaired individuals (CI; 61 MCI and 39 AD). The verbal Letter Fluency test was used to assess the vocabulary size and lexical speed access. Individuals underwent an MRI, a [18F]AZD4694 amyloid-PET scan and a [18F]MK6240 tau-PET scan. [18F]AZD4694 and [18F]MK6240 standardized uptake value ratio (SUVRs) were calculated between 40 to 70 min and 90 to 110 min post-injection, respectively, using cerebellum grey matter as the reference region. A voxel-based regression model evaluated the relationship between the cognitive scoring and the PET markers [18F]AZD4694 and [18F]MK6240, correcting for age, sex, education, APOE, diagnosis and RFT was used to account for multiple comparisons.
Results: In the present study, negative correlations were found between Letter Fluency scores and tau accumulation via PET-imaging in individuals where the associated regions were prefrontal cortex, which plays an important role in this neuropsychological task, as well as in the medial frontal, superior temporal, medial occipital lobes and precuneus area. No associations were found between the Letter Fluency scoring and amyloid deposition.
Conclusion: Our results suggest that tau accumulation is associated with lower vocabulary size and lexical speed access, whereas amyloid deposition does not influence this neuropsychological field.
Figure 1. Correlation between Letter Fluency scores and [18F]MK6240 Braak 3 SUVR.
Pearson’s correlation, p < 0.001, correlating Letter Fluency scores with the PET marker [18F]MK6240 in braak stage 3.
Figure 2. Association between Verbal Fluency scores and tau accumulation assessed with [18F]MK6240 PET.
A voxel-based regression model evaluated the relationship between the Letter Fluency score and the PET marker [18F]MK6240
2021-157
Reducing Model Bias in Measurement of Dopamine Response to Behavioral Challenge (#396)
Michael A. Levine1, Joseph B. Mandeville1, Finnegan Calabro2, Julie C. Price1, Beatriz Luna2 and Ciprian Catana1
1Massachusetts General Hospital, A. A. Martinos Center for Biomedical Imaging, Boston Massachusetts, USA
2University of Pittsburgh Medical Center, Psychiatry, Pittsburgh Pennsylvania, USA
Abstract
Introduction: [11C]Raclopride Positron Emission Tomography (PET) can be used to probe dopamine release in response to behavioral challenge, applying compartmental kinetic modeling and extracting macroparameters, including binding potential (BPND). However, behavioral challenges produce small changes in neurotransmitter release, and detection may be conflated with model-based sources of bias. Voxel maps of absolute change in binding potential (ΔBPND, i.e. BPND-POST – BPND-PRE) in humans were compared to simulated maps which describe the bias induced by the Multilinear Reference Tissue Model (MRTM2).1
Methods: 69 healthy adult participants underwent bolus plus constant infusion [11C]raclopride PET over 90 min with simultaneous functional Magnetic Resonance Imaging and a reward learning task (avg. start, end time: 40, 70 min).2 Bias was estimated from a forward simulation of the two-tissue compartment model (2TCM), using an analytic arterial input function,3 global K1’ and k4,4 a single k2’ (0.30/min), participant-averaged voxel maps of R1 and BPND, and no challenge. The acquired and simulated PET data were fit with an Extension of MRTM2 that includes a challenge term (unit step at 40 minutes), producing parametric maps of ΔBPND (E-MRTM2). Debiasing was performed by eliminating the contribution of the uptake period (0–27.5 min) to BPND estimation (DE-MRTM2).5
Results: In simulations, E-MRTM2 estimates of ΔBPND bias were -0.4% in striatum, 19.5% in white matter, and 16.1% in gray matter. DE-MRTM2 reduced ΔBPND bias to -0.3% in striatum, 4.1% in white matter, and 5.7% in gray matter. In human participants, E-MRTM2 estimated ΔBPND of -2.7% in striatum, 6.1% in white matter, and -4.3% in gray matter, while DE-MRTM2 estimated ΔBPND of -2.2% in striatum, 2.5% in white matter, and -6.6% in gray matter. Voxel maps of ΔBPND show that DE-MRTM2 decreases bias globally in simulations. In parametric human data, DE-MRTM2 reduces the ‘halo’ effect around the edges of the striatum and eliminates the spatial correlation of ΔBPND with gray and white matter (Figure 1).
Conclusion: The results suggest that model bias from fitting a single compartment model (E-MRTM2) to two-tissue data can be reduced by fitting a model that excludes the contribution of the uptake period to the estimation of BPND (DE-MRTM2).
Acknowledgements
This work was partly supported by National Institute of Biomedical Imaging and Bioengineering Grant 5R01EB014894-02, National Institute of Mental Health Grant Number R01MH080243, National Institute of Neurological Disorders and Stroke Grant Number R01NS112295, National Institute of General Medical Sciences Grant T32 GM008313, NIH Blueprint for Research Science Grant T90DA022759/R90DA023427, and NIH Shared Instrumentation Grant S10RR023043.
References
IchiseM, et al.Linearized reference tissue parametric imaging methods: application to [11C]DASB positron emission tomography studies of the serotonin transporter in human brain. J Cereb Blood Flow Metab2003;
23: 1096–1112.CalabroF, et al. Striatal dopamine supports reward reactivity and learning: a simultaneous PET/fMRI study.bioarxiv2020; .NormandinMDMorrisED.
Estimating neurotransmitter kinetics with ntPET: a simulation study of temporal precision and effects of biased data. Neuroimage2008;
39: 1162–1179.FardeL, et al.Kinetic analysis of central [llC]raclopride binding to D2-dopamine receptors studied by PET – a comparison to the equilibrium analysis. J Cereb Blood Flow Metab1989;
9: 696–798.LevineMA, et al. Characterization and mitigation of model bias in parametric mapping of dopamine response to behavioral challenge. medarxiv2021;
2021-158
Single dose of cocaine alters synaptic vesicle glycoprotein 2A density in adolescent rats (#397)
Rachele Rossi1, Simone Larsen Bærentzen1, 2, Majken Borup Thomsen1, 2, Caroline Cristiano Real1, Gregers Wegener1, Albert Gjedde1, 3 and Anne M. Landau1, 2
1University of Aarhus, Translational Neuropsychiatric Unit (TNU), Department of Clinical Medicine, Aarhus, Denmark
2University of Aarhus, Department of Nuclear Medicine and PET, Department of Clinical Medicine, Aarhus, Denmark
3McGill University, Department of Neurology and Neurosurgery, Montreal, QC, Canada
Abstract
Introduction: Synaptic vesicle glycoprotein 2A (SV2A) serves as a marker of synaptic density. As an anti-epileptic drug that binds to the SV2A protein, Levetiracetam is the source of radiolabelled tracers of SV2A. By binding to dopamine transporters, cocaine affects the function of synapses. Here, we tested the temporal and spatial properties of neuroimaging of SV2A after a single dose of cocaine by autoradiography of the specific binding of [3H]UCB-J to SV2A.
Methods: We injected male adolescent rats with cocaine (20 mg/kg; n = 16) or saline (n = 12). We tested the animals in open field trials for 30 minutes immediately after administration. We collected, froze, and sectioned the brains at 1 hour or at 7 days after injection, and generated autoradiograms of the binding of [3H]UCB-J to medial prefrontal cortex, striatum, nucleus accumbens, amygdala, and dorsal and ventral hippocampus in the presence or absence of Levetiracetam.
Results: Cocaine-injected rats had increased activity in the open field trials, unlike controls. [3H]UCB-J binding was slightly lower in the medial prefrontal cortex in cocaine vs. saline rats 1 hour after injection. In contrast, binding was higher in amygdala and hippocampus in cocaine vs. saline rats 7 days after injection. We detected no statistically significant differences in striatum or nucleus accumbens.
Conclusion: Changes of synaptic density in limbic brain regions indicate that effects of cocaine may regulate neurotransmission with differential effects in different areas over time. The findings add to the understanding of SV2A expression in physiological and pathological conditions that would improve the interpretation of imaging outcomes.
Acknowledgements
Rachele Rossi thanks the Erasmus+ Traineeship Program for the opportunity to do the research.
2021-159
Protective role of β-amyloid revealed by PET of [11C]PiB and [18F]FDG in Alzheimer’s disease (#398)
Elise Rischel1, Michael Gejl2, Birgitte Brock3, Jørgen Rungby4 and Albert Gjedde1, 2
1University of Copenhagen, Department of Neuroscience, Copenhagen N, Denmark
2Aarhus University, Department of Clinical Medicine, Aarhus C, Denmark
3Steno Diabetes Center Copenhagen (SDCC); Gentofte, Denmark, Gentofte, Denmark
4University of Copenhagen, Department of Endocrinology, Bispebjerg Hospital, Copenhagen N, Denmark
Abstract
Introduction: In Alzheimer’s disease (AD), evidence suggests that beta-amyloid accumulates in regions with the highest cerebral metabolic rate of glucose (CMRglc). However, in individual regions, the accumulation appears to be inversely related to the magnitude of CMRglc. This difference suggests a generally protective role of beta-amyloid in the maintenance of glucose metabolism that ultimately fails in AD. We tested the hypothesis by comparing the binding potentials (BP_ND) of tracer Pittsburgh Compound B ([11C]PiB), representing the beta-amyloid accumulation, and estimates of CMRglc in separate regions and individual subjects with AD.
Methods: We used positron emission tomography (PET) to examine the correlation between regionally averaged and individual regional values of the BP_ND of [11C]PiB, and regionally averaged and individual regional values of CMRglc measured with tracer [18F]fluorodeoxyglucose ([18 F]FDG) in specific brain regions of 29 patients with AD.
Results: We found significant direct correlation between the average values of the BP_ND of [11C]PiB and CMRglc of dfferent regions, and significant inverse correlations between individual values of PIB BP_ND and CMRglc values in precuneus, temporal lobe, occipital lobe, parietal lobe and whole cerebral cortex. Inverse correlations were not significant in frontal lobe, cingulate cortex, or cerebellum.
Conclusion: We found significant positive correlation between the two tracers in regions on average, but inverse correlations for individual subjects in separate regions. The results are consistent with β-amyloid acting as a protective mechanism that ultimately fails. The observation implies that the β-amyloid accumulates in separate regions in proportion to glucose metabolism, while in each separate region the metabolism declines in inverse proportion to the accumulation of the β-amyloid, suggesting a protective role of amyloid accumulation that ultimately fails.
References
GejlMBrockBEgefjordL, et al.Blood-brain glucose transfer in Alzheimer’s disease: effect of GLP-1 analog treatment. Scient Rep2017;
7: 1–10.RodellAAanerudJBraendgaardH, et al.Washout allometric reference method (WARM) for parametric analysis of [11C]PiB in human brains. Front Aging Neurosci2013;
5: 1–14.RodellABO’KeefeGRoweCC, et al.Cerebral blood flow and Abeta-amyloid estimates by WARM analysis of [11C]PiB uptake distinguish among and between neurodegenerative disorders and aging. Front Aging Neurosci2017;
8: 1–11.
2021-160
Using a support vector machine to identify signatures of different p-tau CSF species in incipient Alzheimer’s disease (#399)
Stijn Servaes1, Firoza Z. Lussier1, Joseph Therriault1, Gleb Bezgin1, Min-Su Kang1, Yi-Ting Wang1, Jenna Stevenson1, Cécile Tissot1, Jaime F. Arias1, Andréa L. Benedet1, 4, Mira Chamoun1, Tharick A. Pascoal1, 5, Serge Gauthier1, 2 and Pedro Rosa-Neto1, 2
1Translational Neuroimaging Laboratory, The McGill University Research Centre for Studies in Aging, Douglas Hospital, McGill University, Montreal, QC, Canada
2Department of Neurology and Neurosurgery, McGill University, Montreal, QC, Canada
3Montreal Neurological Institute, Montreal, QC, Canada
4Department of Psychiatry and Neurochemistry, Institute of Neuroscience and Physiology, The Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden
5Department of Psychiatry, University of Pittsburgh, Pittsburgh Pennsylvania, USA
Abstract
Introduction: The presence of p-tau in biofluids has previously been proposed to be a response to neurofibrillary tangle pathology.1 However, the increase of p-tau in cerebrospinal fluid (CSF) precedes detectable neurofibrillary tangle pathology, as indexed by tau PET, by up to a decade,2 suggesting that soluble tau could be an indication of early tau pathology. As Alzheimer’s disease (AD) is the only disorder that consistently shows an increase in CSF p-tau, whereas this biomarker is normal in other neurodegenerative disorders,3 research has focused on the potential of several p-tau species for early clinical diagnosis. With this study, we investigated the heterogeneity of p-tau species in CSF in order to assess the clinical status of participants of the TRIAD cohort.
Methods: Support vector machines were used to identify cutoff values of p-tau181, p-tau217, p-tau231 and p-tau235 in CSF, both individually and combined, to separate a group of AD patients (n = 25) and young controls (n = 28). Using these cutoff values, signatures were calculated on an individual level in a group of individuals with cognitive impairment (n = 62) and age-matched controls (n = 75). Additionally, [18F]MK6240 SUVR maps and memory composite scores were calculated and evaluated using the Logical Memory test, the Rey Auditory Verbal Learning Test, the Face Name Association Test and the Free and Cued Selective Reminding Test.
Results: When combining different CSF p-tau species, the largest contribution in order of importance came from p-tau181, followed by p-tau217, p-tau235 and p-tau231. Achieving the cutoff for multiple p-tau species was associated with higher Braak stages (Figure 1) and lower memory composite scores (Figure 2). In particular, achieving the cutoff value for p-tau217 was associated with later Braak stages.
Conclusion: Our findings suggest that heterogeneity in p-tau species carries predictive power in the identification of incipient Alzheimer’s Disease.
References
AshtonNJPascoalTAKarikariTK, et al.Plasma p-tau231: a new biomarker for incipient Alzheimer’s disease pathology. Acta Neuropathol2021;
141: 709–724.BatemanRJXiongCBenzingerTL, et al. Clinical and biomarker changes in dominantly inherited Alzheimer’s disease [published correction appears in N Engl J Med 2012 Aug 23;367:780]. N Engl J Med. 2012; 367: 795–804. doi:10.1056/NEJMoa1202753JackCRJrBennettDABlennowK, et al.NIA-AA research framework: toward a biological definition of Alzheimer’s disease. Alzheimers Dement2018;
14: 535–562.
2021-161
Ex-vivo analysis of metabotropic glutamate receptor type 5 hippocampal abnormalities in epileptogenic foci (#400)
Maria Zimmermann1, 2, Luciano Minuzzi2, 3, Arturo Aliaga2, 4, Marie-Christine Guiot5, Jeffrey A. Hall1, Jean-Paul Soucy1, 4, Gassan Massarweh4, Pedro Rosa-Neto1, 2 and Eliane Kobayashi1
1McGill University, Department of Neurology and Neurosurgery, Montréal Neurological Institute, Montréal, QC, Canada
2McGill University, Translational Neuroimaging Laboratory, Douglas Research Institute, Montréal, QC, Canada
3McMaster University, Department of Psychiatry and Behavioural Neurosciences, Hamilton, ON, Canada
4McGill University, PET Unit, McConnell Brain Imaging Centre, Montréal, QC, Canada
5McGill University, Department of Pathology, Montréal Neurological Institute, Montréal, QC, Canada
Abstract
Introduction: Hippocampal metabotropic glutamate receptor 5 (mGluR5) abnormalities in patients with intractable mesial temporal lobe epilepsy (MTLE) have been characterized ex-vivo via immunoreactivity studies of surgically-extracted specimens, and in-vivo via Positron Emission Tomography with [11C]ABP688, a negative allosteric modulator of mGluR5. Reduced [11C]ABP688 binding potentials (BPs) have been found within the involved hippocampus of refractory MTLE patients,1 contrasting with mGluR5 upregulation observed in MTLE immunohistochemistry studies.2,3 These reduced BPs may signify hyperexcitability-induced receptor conformation changes, the presence of excess glutamate, or decreased cell surface-expressed mGluR5. To further investigate mGluR5 alterations in MTLE, we performed [3H]ABP688 autoradiography on hippocampi surgically-extracted from MTLE patients compared to post-mortem healthy controls to evaluate ex-vivo receptor densities (Bmax) and dissociation constants (KD) in the absence of endogenous glutamate.
Methods: We assessed Bmax and KD in frozen hippocampal specimens (16 MTLE, 5 controls) using saturation autoradiography with six different concentrations of [3H]ABP688 (0.25–8nM). Non-specific binding was determined using 10mM 2-Methyl-6-(phenylethynyl)pyridine. Differences in Bmax and KD were evaluated using a two-tailed Welsh’s t-test and a two-tailed Mann-Whitney U Test, respectively. A linear regression model was used to assess Bmax changes with KD, age and sex as predictors.
Results: MTLE specimens showed a 46.8% reduction in Bmax compared to control hippocampi (p = 0.0046). There was no significant change in KD between the groups. A linear regression model revealed that the decrease in Bmax in the MTLE group was independent of age, sex, and KD.
Conclusion: The observed reductions in [3H]ABP688 binding and the nonsignificant difference in KD suggest that the decreased BPs found in vivo cannot be explained by the presence of excess glutamate or by reduced affinity of ABP688 for mGluR5. Our results are consistent with mGluR5 downregulation/internalization, or a conformational change resulting in complete occlusion of ABP688’s binding site in a proportion of mGluR5 receptors.
Acknowledgements
This study was supported by operating funds from the Savoy Foundation (www.savoy-foundation.ca;
pilot project grant to E.K. and P.R.-N and MSc studentship to M.Z.).
Figure 1. [3H]ABP688 binding in Hippocampus.
Autoradiographic saturation binding curves constructed from total and non-specific binding data using the one-site binding model (GraphPad Prism). Displayed are the mean curves for specific binding of mGluR5 by [3H]ABP688 in the resected hippocampus of mesial temporal lobe epilepsy patients and in healthy control hippocampal specimens. Error bars are SEM.
Figure 2. [3H]ABP688 Bmax and Kd values in the Hippocampus.
Receptor density (Bmax) and dissociation constant (KD) values in the surgically-extracted hippocampus of patients with intractable mesial temporal lobe epilepsy, and in hippocampal specimens from healthy controls. (A) Receptor density was reduced in patients compared to controls. Bmax values are represented as mean ± std. deviation. (B) Dissociation constants were unchanged in patients compared to controls. KD values are represented as median with interquartile range. ** indicates p < 0.005; ns indicates a non-significant comparison.
References
LamJDuBoisJMRowleyJ, et al.In vivo metabotropic glutamate receptor type 5 abnormalities localize the epileptogenic zone in mesial temporal lobe epilepsy. Ann Neurol2019;
85: 218–228.KandrataviciusLRosa-NetoPMonteiroMR, et al.Distinct increased metabotropic glutamate receptor type 5 (mGluR5) in temporal lobe epilepsy with and without hippocampal sclerosis. Hippocampus2013;
23: 1212–1230.NotenboomRGEHampsonDRJansenGH, et al.Up-regulation of hippocampal metabotropic glutamate receptor 5 in temporal lobe epilepsy patients. Brain2006;
129: 96–107.
2021-162
Covid-19 pandemic: Quantifying the effects of the first lockdown on behavioral and cognitive measures using TASIC (#401)
Jenna Stevenson1, Firoza Z. Lussier1, 2, Stijn Servaes1, 2, Mira Chamoun1, 2, Nesrine Rahmouni1, Cécile Tissot1, 2, Nina M. Poltronetti1, Tharick A. Pascoal2, 3, Andréa L. Benedet2, 4, Gleb Bezgin1, 2, Suzanne King5, Serge Gauthier1, 6 and Pedro Rosa-Neto1, 2
1The McGill University Research Centre for Studies in Aging, Montreal, Canada
3University of Pittsburgh, Department of Psychiatry, Pittsburgh, USA
4University of Gothenburg, Gothenburg, Sweden
5McGill University, Department of Psychiatry, Montreal, Canada
6Douglas Mental Health University Institute, Montreal, Canada
Abstract
Introduction: Following the rapid spread of the COVID-19 virus throughout Quebec, the TRIAD cohort, a longitudinal observational study, evaluated the effects of COVID-19 on it’s aging and vulnerable population and their caregivers. This study aims at investigating the behavioural and psychological effects of COVID-19 and social isolation on the aging population. The TRIAD Assessment of Social Isolation and Cognition (TASIC) was developed to assess these effects on participants of observational trials.
Methods: Pre-pandemic data including, demographical information, Clinical Dementia Rating (CDR), Mini-Mental State Examination (MMSE), Geriatric Depression Scale (GDS), Montreal Cognitive Assessment (MoCA) and a Social Support Questionnaire, were collected during in-person visits and take-home questionnaires. Following the onset of COVID-19, TASIC was created to include additional COVID-19 specific scales developed by Dr. Rosa-Neto and Dr. King that include Knowledge of COVID-19 scale, the Montreal Assessment of Stress related to COVID-19 (MASC), as well as the Impact of Events Scale (IESR), the Peritraumatic Distress Inventory (PDI), the Peritraumatic Dissociative Experiences Questionnaire (PDEQ) for COVID-19. Assessments were conducted via telephone interviews with TRIAD participants (n = 292) and their informants (n = 243) by eight trained research assistants from April through June 2020. All participants enrolled in previous TRIAD studies, deemed eligible through detailed screening criteria were contacted.
Results: As a result of this study, and previous work done by the cohort, a culmination of information is available, with 90% of TRIAD participants having completed the COVID-19 study also having plasma collected. 80% of participants from the COVID-19 study have PET imaging using [18F]AZD4694 and [18F]MK6240 and MRI sequencing.
Conclusion: The COVID-19 study conducted by TRIAD provides a unique opportunity to understand the effects of a global pandemic on our aging population as well as caregiver burden. This data, in conjunction with other measures available in the cohort, can make important strides in finding ways to help, and better understand those most impacted by the crisis.
2021-163
Amyloid-PET and free-water diffusion MRI of the white matter: a multi-center mixed cohort of small vessel disease and Alzheimer’s disease pathology (#402)
Julie Ottoy1, Miracle Ozzoude1, Katherine Zukotynski1, 2, Min-Su Kang3, Sabrina Adamo1, Christopher Scott1, Joel Ramirez1, Walter Swardfager4, Benjamin Lam5, Aparna Bhan1, Parisa Mojiri1, Alex Kiss6, Stephen Strother7, 8, Christian Bocti9, Michael Borrie10, Howard Chertkow11, Richard Frayne12, Robin Hsiung13, Robert J. Laforce14, Michael D. Noseworthy15, Frank S. Prato10, Demetrios J. Sahlas16, Eric E. Smith17, Philip H. Kuo18, Vesna Sossi13, Alexander Thiel11, Jean-Paul Soucy3, Jean-Claude Tardif19, Sandra E. Black1, 5 and Maged Goubran1, 7
1University of Toronto, LC Campbell Cognitive Neurology Unit, Hurvitz Brain Sciences Program, Sunnybrook Research Institute, Toronto, ON, Canada
2McMaster University, Departments of Medicine and Radiology, Hamilton, ON, Canada
3McGill University, Montreal Neurological Institute, Montreal, QC, Canada
4University of Toronto, Department of Pharmacology & Toxicology, Toronto, ON, Canada
5University of Toronto, Department of Medicine (Division of Neurology), Toronto, ON, Canada
6University of Toronto, Department of Research Design and Biostatistics, Sunnybrook Research Institute, Toronto, ON, Canada
7University of Toronto, Department of Medical Biophysics, Toronto, ON, Canada
8University of Toronto, The Rotman Research Institute Baycrest, Toronto, ON, Canada
9Université de Sherbrooke, Département de Médecine, Sherbrooke, QC, Canada
10Western University, Lawson Health Research Institute, London, ON, Canada
11McGill University, Jewish General Hospital and Department of Neurology and Neurosurgery, Montreal, QC, Canada
12University of Calgary, Departments of Radiology and Clinical Neuroscience, Calgary Alberta, Canada
13University of British Columbia, Physics and Astronomy Department and DM Center for Brain Health, Vancouver British Columbia, Canada
14Université Laval, Clinique Interdisciplinaire de Mémoire, Département des Sciences Neurologiques, Quebec, QC, Canada
15McMaster University, Department of Electrical and Computer Engineering, Hamilton, ON, Canada
16McMaster University, Department of Medicine, Hamilton, ON, Canada
17University of Calgary, Hotchkiss Brain Institute, Calgary Alberta, Canada
18University of Arizona, Department of Medical Imaging, Medicine, and Biomedical Engineering, Tucson Arizona, USA
19Université de Montréal, Montreal Heart Institute, Montreal, QC, Canada
Abstract
Introduction: To investigate the neurobiological underpinnings of amyloid-PET binding in the white matter (WM) using free-water diffusion MRI in a multi-centre mixed cohort of small vessel disease (SVD) and Alzheimer’s disease (AD) pathology.
Methods: We included sixty participants with moderate-to-severe white matter hyperintensity burden (WMH; median (IQR): 30.51(22.14)cm3) from dementia/stroke-prevention clinics (48% amyloid-positive) as part of the MITNEC-C6 project across seven sites in Canada. Additionally, we included sixty cognitively normal or early MCI with mild-to-moderate WMH burden (median (IQR): 5.82(9.29)cm3) from the ADNI-2 database (22% amyloid-positive). We applied a bi-tensor diffusion MRI model that differentiates between extracellular (free-water fraction) and tract-specific WM compartments (free-water adjusted fractional anisotropy or FAadjusted) (Figure 1). We tested associations of these diffusion metrics with amyloid standardized uptake value ratio (SUVR) in both WMH and normal-appearing WM, and with cognition. To further investigate how the diffusion metrics and the demographical variables including age, sex, education, WMH volume, and cortical amyloid SUVR covary with WM amyloid SUVR, we performed partial-least-square analysis using ten-fold cross-validation with five repeats.
Results: In WMH, amyloid SUVR was significantly lower compared to normal-appearing WM and associated strongly with higher free-water (β = −0.36 ± 0.13, P = 0.005). Partial-least-square analysis in the moderate-to-severe burden group showed that free-water as well as WMH volume were most strongly associated with amyloid SUVR in WMH (Figure 2-left; component-1 explaining 24% variance), while FAadjusted as well as cortical amyloid were most strongly associated with amyloid SUVR in normal-appearing WM (Figure 2-right; component-1 explaining 31% variance). Higher level of free-water was more closely related to cognitive impairment (MMSE: βWMH = −0.40 ± 0.13, P = 0.003; βnormal-appearing = −0.30 ± 0.11, P = 0.01) than FAadjusted (MMSE: βnormal-appearing = 0.21 ± 0.09, P = 0.02).
Conclusion: In mixed AD and SVD, representative of the more common AD population, amyloid-PET changes in WM lesions may largely reflect extracellular free-water. In contrast, normal-appearing WM changes may reflect tract-specific microstructural injury (possibly demyelination). Our study supports free-water as a potential [early] WM-related biomarker in AD.
Acknowledgements
For the Medical Imaging Trials Network of Canada (MITNEC) and the Alzheimer’s Disease Neuroimaging Initiative.
Figure 1. Bi-tensor model of the white matter.
Schematic representation of the two-compartment model per brain voxel (red color), including an extracellular (free-water; purple color) and tract-specific (FAadjusted for free-water; yellow color) compartment. Abbreviations: NAWM, normal-appearing white matter; WMH, white matter hyperintensities; FA, fractional anisotropy; FW, free water.
Figure 2. Amyloid SUVR in relation to diffusion MRI metrics and demographics.
PLS analysis showing the relationship of amyloid SUVR with DTI and demographical variables in WMH (left) and NAWM (right). Error bars represent 95%CI based on bootstrapping with 5,000 repetitions. Abbreviations: NAWM, normal-appearing white matter; WMH, white matter hyperintensities; FA, fractional anisotropy; FW, free water; MD, mean diffusivity.
2021-164
Evaluation of [18F]FR – a potential PET tracer for the diagnosis of cerebral amyloid angiopathy (#403)
Daniel Bleher1, Marilena Poxleitner1, Gregory D. Bowden1, 2, Sabrina Buss1, Ann-Kathrin Grotegerd1, Sonja Schembecker1, Bernd J. Pichler1, 2, Andreas Maurer1, 2 and Kristina Herfert1
1Eberhard Karls Universität, Werner Siemens Imaging Center, Department of Preclinical Imaging and Radiopharmacy, Tübingen Baden-Württemberg, Germany
2Eberhard Karls Universität, Cluster of Excellence iFIT (EXC 2180), Tübingen Baden-Württemberg, Germany
Abstract
Introduction: The β-amyloid peptide Aβ1-40 is the main component in cerebral amyloid angiopathy (CAA), a degenerative disorder of the brain vasculature with a prevalence of 50% in elderly people, but its diagnosis is still not possible without brain biopsy. A selective radiotracer would aid the clinical delineation of vascular from parenchymal amyloid depositions (PEA) present in Alzheimer’s Disease (AD) mainly composed of Aβ1-42. Herein, we evaluate the benzoxazinole derivative [18F]FR as PET tracer targeting CAA.
Methods: Specificity and selectivity of [3H]FR was tested in invitro binding assays using recombinant Aβ1-40 and Aβ1-42 fibrils. Two different AD mouse models, the APP23 model with, and the APPPS1 model without CAA pathology in the brain were used for further evaluation. In vitro autoradiography (AR) and immunofluorescence staining was performed on brain sections. ICTAD-1, a fluorescent dye able to distinguish between Aβ1-40 and Aβ1-42, was used as reference. After automated radiolabeling, purification, and formulation of [18F]FR, the pharmacokinetic profile was determined in mice. In vivo PET with [18F]FR was performed in wild type and transgenic mice.
Results: [3H]FR shows specific binding to both fibril entities with a selectivity towards Aβ1-40 (Kd = 5.7 nM) over Aβ1-42 (Kd = 122 nM). AR with [18F]FR demonstrated specific binding of the radiotracer towards CAA in mice brain sections and only moderate binding to PEA. Metabolite analysis revealed one radio-metabolite in plasma and brain with 91% of the parent compound remaining in the brain at 5 min. Preliminary in vivo PET data analysis revealed a high binding potential (BPnd) of [18F]FR in the cortex of APP23 mice with CAA (BPnd = 0.23 ± 0.04; n = 5) with negligible binding in APPPS1 mice (BPnd = 0.04 ± 0.04; n = 4) and wild type mice (BPnd = 0.02 ± 0.04; n = 9).
Conclusion: FR demonstrated selectivity towards CAAover PEAin vitro and in vivo. After completed invivo evaluation of FR, future studies will evaluate the compound in the CAA mouse model APPDutch and in pigs to investigate metabolite formation in higher species.
Acknowledgements
The authors want to thank Johannes Kinzler, Ramona Stremme, Linda Schramm and Christian Köder for support on radiosynthesis and animal handling.
2021-165
Cognitive reserve: Evaluating the relationship between WASI-II matrix reasoning and tau accumulation using [18F]MK6240 in monolingual and bilingual individuals (#404)
Alyssa Stevenson1, 2, Nesrine Rahmouni1, 2, Jenna Stevenson1, 2, Cécile Tissot1, 2, Joseph Therriault1, 2, Firoza Z. Lussier1, 2, Gleb Bezgin1, 2, Mira Chamoun1, 2, Nina M. Poltronetti1, 2, Vanessa Pallen1, 2, Min-Su Kang1, 2, Tharick A. Pascoal2, 3, Andréa L. Benedet2, 4 and Pedro Rosa-Neto1, 2
1McGill University Research Centre for Studies in Aging, McGill, Montreal, QC, Canada
3University of Pittsburgh, Department of Psychiatry, Pitsburgh Pennsylvania, USA
4University of Gothenburg, Gothenburg, Sweden
Abstract
Introduction: Cognitive reserve (CR), a protective mechanism, allows for sustained cognitive functioning in older adults with greater experiential resources. Bilingualism, an indicator of greater cognitive reserve, has been found to preserve cognitive performance in older adults despite significant disease pathology. While the association with amyloid has been well acknowledged, the effect of tau is still under investigation. In the present study, we aim to evaluate the effect of bilingualism on individuals’ cognition and tau load.
Methods: This study was conducted in 310 individuals (193 cognitively unimpaired (CU)) and 117 cognitively impaired (CI; 70 MCI and 47 AD) from the TRIAD cohort. 49.7% of the study participants were monolingual. Wechsler Abbreviated Scale of Intelligence – Second Edition (WASI-II) was used to measure cognitive ability and reasoning in individuals. Accumulation of tau was assessed with [18F]MK6240 PET and standardized uptake value ratio (SUVRs) were calculated between 90 to 110 min post-injection, using cerebellum grey matter as the reference region. A voxel-based regression model evaluated the relationship between the WASI-II score and the PET marker [18F]MK6240, correcting for age, sex, education, APOE, diagnosis and RFT was used to account for multiple comparisons.
Results: Our data showed a negative association between the cognitive score and tau accumulation in both monolingual and bilingual groupings. In the bilingual individuals, the associated regions were the anterior cingulate cortex and lateral frontal cortex. In the monolingual group, the associated regions were medial occipital, medial parietal and medial prefrontal cortex. In addition, bilingual individuals demonstrated significantly higher scores on the WASI-II test in comparison to monolingual individuals.
Conclusion: Our results demonstrate that bilingual individuals have fewer regions affected than their monolingual counterparts. This suggests that bilingual individuals are more resilient to the accumulation of tau; shown in higher scores in cognitive ability and reasoning.
References
CostumeroVMarin-MarinLCalabriaM, et al.A cross-sectional and longitudinal study on the protective effect of bilingualism against dementia using brain atrophy and cognitive measures. Alz Res Therapy2020;
12: 11.DuncanHDPhillipsNA. The contribution of bilingualism to cognitive reserve in healthy aging and dementia. In: NicoladisEMontanariS (eds) Bilingualism across the lifespan: Factors moderating language proficiency.
American Psychological Association, 2016, pp. 305–322.
2021-166
Amyloid beta deposition and cognitive decline in Parkinson’s disease: a study of the PPMI cohort (#405)
Alexander Mihaescu1, 3, Mario Masellis3, 5, Ariel Graff-Guerrero1, 3, Mikaeel Valli1, Carme Uribe1, 2, Maria Diez-Cirarda1, 7 and Antonio P. Strafella2, 3
1Centre for Addiction and Mental Health, Research Imaging Centre, Toronto, ON, Canada
2University Health Network, Krembil Research Institute, Toronto, ON, Canada
3University of Toronto, Institute of Medical Science, Toronto, ON, Canada
4Toronto Western Hospital, Morton and Gloria Shulman Movement Disorder Unit, Toronto, ON, Canada
5Sunnybrook Hospital, LC Campbell Cognitive Neurology Research Unit, Toronto, ON, Canada
6Sunnybrook Research Institute, Hurvitz Brain Sciences Program, Toronto, ON, Canada
7Biocruces Bizkaia Health Research Institute, Neurodegenerative Diseases Group, Barcelona, Spain
Abstract
Introduction: The accumulation of amyloid beta (Aβ) in the brain has a complex and poorly understood impact on the progression of Parkinson’s disease (PD) pathology.1,2 Increased Aβ burden has been associated with cognitive decline in PD, especially in PD with dementia.3 Some studies however have found no association between high Aβ burden and cognitive decline in PD patients,4,5 instead suggesting normal aging may be responsible for both Aβ accumulation and cognitive decline. To better understand the relationship between PD cognitive decline and Aβ deposition, our study focused on a homogenous group of idiopathic PD patients whose cognitive abilities were measured longitudinally for 2 years after their Aβ scan.
Methods: Our study used PD patient and healthy control (HC) data from the Parkinson’s Progression Marker Initiative (PPMI) cohort. 25 de novo idiopathic PD patients and 31 HC underwent a [18F]florbetaben (FBB) positron emission tomography (PET) scan which can measure the density of Aβ in vivo in the brain. [18F] FBB density was expressed as a standardized uptake value ratio (SUVR) of 20 bilateral cortical regions of interest (ROI) parcellated using an MNI-modified AAL template. Demographic information, clinical characteristics, and cognitive test scores were also collected. The data was analyzed using IBM SPSS version 27 software with linear regression modeling, Pearson correlations, and hierarchical cluster analysis.
Results: A stepwise linear regression of the PD group revealed a strong adjusted R2 of 0.495 in a model explaining their Montreal Cognitive Assessment (MoCA) score 1-year post-scan using the SUVR from 20 cortical ROIs as independent variables. The ROIs found in this stepwise model were the left rectus, the left anterior cingulate cortex, and the right parietal cortex. We found the PD group’s brain regions formed two clusters, with increased Aβ in cluster two correlating more strongly with a lower MoCA score.
Conclusion: The results suggest Aβ has a moderate association with cognitive decline in PD. We found Aβ accumulation in the brain had a patchwork effect on PD cognition depending on which ROIs are affected. More Aβ accumulation preferentially in cluster 2 ROIs may be involved in PD patients’ cognitive dysfunction.
Acknowledgements
PPMI – a public-private partnership – is funded in part by the Michael J. Fox Foundation for Parkinson’s Research.
References
IrwinDJ.
Parkinson’s disease dementia: convergence of α-synuclein, tau and amyloid-β pathologies. Nat Rev Neurosci2013;
14: 626–636.PetrouM.
Aβ-amyloid deposition in patients with Parkinson disease at risk for development of dementia. Neurology2012;
79: 1161–1167.IrwinDJ.
Neuropathological and genetic correlates of survival and dementia onset in synucleinopathies: a retrospective analysis. Lancet Neurol2017;
16: 55–65.MelzerTR.
Beta amyloid deposition is not associated with cognitive impairment in Parkinson’s disease. Front Neurol2019;
10: 391.KimJ.Network patterns of beta-amyloid deposition in Parkinson’s disease. Mol Neurobiol2019;
56: 7731–7740.
2021-167
[124I]IAZA: Development of a new radioiodinated PET imaging agent for human Alzheimer’s disease brain (#406)
Thrisha T. Reddy, Christopher Liang and Jogeshwar Mukherjee
UCI School of Medicine, Radiological Sciences, Irvine California, USA
Abstract
Introduction: Amyloid b (Ab) and Tau (neurofibrillary tangles) imaging Alzheimer’s disease (AD) is currently underway using various radiotracers. Using “azo” derivatives, we have reported Ab plaque binding properties of [11C]TAZA [1] and [18F]Flotaza2 in postmortem human brain AD. Here we report development and evaluation of a new iodine-124 “azo” analog, [124I]IAZA (4-[124I]iodo-4’-N,N-dimethylaminoazobenzene), as a potential PET imaging agent for AD.
Methods: The radiosynthesis of [124I]IAZA used electrophilic substitution of tributyltin by [124I]iodide (Figure 1) and purified on HPLC.3 Human AD post-mortem brain slices (10 µm) consisting of anterior cingulate (AC) and corpus callosum CC and temporal cortex were used for in vitro binding studies. Brain slices incubated with [124I]IAZA4 in 50% ethanol PBS buffer pH 7.4 (60 mL;0.5 µCi/mL) at 25°C for 1.25 hr. Drug effects of 10 mM 5-OH-BTA-0, curcumin, IAZA, IPPI and MK-6240 were measured. The slices were washed with PBS buffer, 50% and 90% ethanolic PBS buffer and cold water. Brain sections were dried, exposed on phosphor film and analyzed using Optiquant software and binding of [124/125I]IAZA was measured in DLU/mm2.
Results: The single step radiosynthesis of [124I]IAZA was very efficient. Slices from subjects were positively immunostained with anti-Ab and Anti-Tau. Selective binding of [124I]IAZA was observed in grey matter regions (Figure 2). The ratio of AC to CC was > 10. Similarly, temporal cortex exhibited significant binding of [124I]IAZA. Very little white matter binding was seen when 90% alcohol was used in the washing of the slices. IAZA displaced > 90% of [124I]IAZA, displacement by Ab drugs, 5-OH-BTA-0 and curcumin was low, similar to [125I]IAZA4 and Tau drugs, IPPI and MK-6240 displaced ∼40%. These findings suggest that [124/125I]IAZA may have unique binding sites in the AD brain.
Conclusion: [124I]IAZA exhibited high binding in postmortem human AD brains. Although it is a close analog of [11C]TAZA [1] and [18F]Flotaza,2 [124/125I]IAZA appears to be dissimilar. Along with our recently developed [125I]IPPI for Tau,5 it offers a useful tool for the study of postmortem AD brains. Studies are underway to evaluate the utility of [124I]IAZA in transgenic AD mice models, P301S for Tau and 5XFAD for Ab.
Acknowledgements
NIH/NIA RF1 AG029479 (JM), UCI UROP (TTR). Banner Health Research Institute for tissue samples and UCI Pathology for immunostaining.
References
PanMLMukherjeeMTPatelHH, et al.Evaluation of [11C]TAZA for amyloid Ab plaque imaging in postmortem Alzheimer’s disease brain region and whole body distribution in rodent PET/CT. Synapse2016;
70: 163–176.KaurHFelixMRLiangC, et al.Development and evaluation [18F]Flotaza for Ab plaque imaging in post-mortem Alzheimer’s disease brain. Bioorg Med Chem Lett2021;
46: 128164.PandeySKVenugopalAKantR, et al.124I-Epidepride: a high affinity and selective PET radiotracer with potential for extended imaging of dopamine D2/D3 receptors. Nucl Med Biol2014;
41: 426–431.ReddyTTLiangCMukherjeeJ. [125I]IAZA: a new imaging agent for human Alzheimer’s disease brain. To be presented at the World Molecular Imaging Congress, 6–9 October 2021 (Virtual Meeting), 2021.MukherjeeJLiangCPatelKK, et al.Development and evaluation [125I]IPPI for tau imaging in post-mortem human Alzheimer’s disease brain. Synapse2021;
74: e22183.
2021-168
Mapping the multivariate effects of amyloid, tau, and neuroinflammation on cortical thickness in AD (#408)
Min-Su Kang1, Julie Ottoy2, Gleb Bezgin4, Gassan Massarweh4, Jean-Paul Soucy4, Serge Gauthier3 and Pedro Rosa-Neto3
1McGill University, IPN, Montreal, QC, Canada
2Toronto University, Sunnybrook Research Institute, Toronto, ON, Canada
3McGill University, Neurology and Neurosurgery, Montreal, QC, Canada
4McGill University, Montreal Neurological Institute, Montreal, QC, Canada
Abstract
Introduction: Understanding the effects of amyloid, tau, and neuroinflammation on neurodegeneration is important for identifying an effective therapeutic target and combination therapy for AD. Here, we aimed to elucidate the amyloid, tau, and neuroinflammation effects on neurodegeneration and cognition using multimodal imaging biomarkers.
Methods: A total of 111 participants (57 A-T-, 12 A+T-, 42 A+T+) from the TRIAD cohort underwent T1, 3 PET, and MMSE procedures. Cortical thickness measurement was generated using the Freesurfer pipeline while static [18F]AZD4694, [11C]PBR28, and [18F]MK6240 SUVR images were generated using the PETsurfer pipeline. A multivariate PLS regression model was conducted to evaluate the effect of [18F]AZD4694, [11C]PBR28, and [18F]MK6240 on cortical thickness. Then, the significant latent variables (LVs) were evaluated to understand the pathologic effect on cognition based on a multiple regression between the MMSE and latent variable scores with age, sex, education, and APOEε4 as covariates.
Results: Our study revealed two significant LVs that explained 28% and 21.2% of the cortical thickness, respectively. Here, the first LV score showed a significant negative effect on MMSE and a significant increase in A+T+ compared to A-T- and A+T- and in A+T+ compared to A-T-. Upon further examination of the first LV, [11C]PBR28 in bilateral transverse temporal, hippocampal, and insular regions showed positive effects while all [18F]AZD4694 and [18F]MK6240 regions showed negative effects on cortical thickness.
Conclusion: This study highlights a complex pathologic mechanism, in which activated glial neuroinflammation showed protective effects while amyloid and tau showed degenerative effects on cortical thickness, leading to cognitive impairment in AD.
Figure 1. LV1 bootstrap ratio.
Figure 2. LV1 loadings.
2020 Abstract
2020-01
Anti-correlations between 18F-FDG PET and resting state dynamic functional connectivity: Insights into brain network variability (#3)
Tommaso Volpi1, Marco Aiello2, Valentin Riedl3, Maurizio Corbetta1, 4 and Alessandra Bertoldo1, 5
1Padova Neuroscience Center, University of Padova, Padova, Italy
2IRCCS SDN, Napoli, Italy
3TUM-Neuroimaging Center, Department of Neuroradiology, Technical University of Munich, Munich, Germany
4Department of Neurosciences, University of Padova, Padova, Italy
5Department of Information Engineering, University of Padova, Padova, Italy
Abstract
Introduction: There is increasing interest in understanding the relationship between the brain’s metabolic consumption, imaged by 18F-FDG PET, and its functional connectivity (FC) architecture emerging from resting-state fMRI studies, which is assumed to partly rely on local glucose metabolism. Correlations between 18F-FDG PET measures and FC in simultaneous recordings have been reported,1,2 but without investigating FC temporal variability.
Methods: Simultaneous 18F-FDG PET and resting state fMRI data were acquired in 28 healthy subjects (59.8 ± 10.8 yrs, 14 F) on a Siemens mMR Biograph 3T.1,2 Standard uptake value ratio (SUVR) to whole-brain average was computed from 18F-FDG data; rs-fMRI was acquired for 101 and 72 minutes (TE/TR = 30/2000 ms). Both PET and fMRI scans were registered to T1w images and sampled over the Gordon-Laumann functional atlas.3 Time-varying FC (Pearson’s correlation) was computed with a sliding windows approach (window size: 30 TRs, step: 1 TR). FC graph metrics were computed for each sliding window: degree (DEG), strength (STR), participation coefficient (PAR), clustering coefficient (CC), betweenness centrality (BC). Coefficients of variation CV% (graph metrics’ standard deviation across sliding windows divided by their mean) were computed for each node and averaged across subjects; Pearson correlation with SUVR was computed considering: all ROIs; only network hubs, identified on the average static FC matrix as 15% highest DEG nodes.
Results: We identified statistically significant anti-correlations between nodes’ SUVR and CVs% of DEG, STR, CC, BC (ranging from r = −0.30 to -0.50, p < 0.01), but not with CV% of PAR (Figure 1). Anti-correlations were stronger when considering only between-network links (r = −0.44 vs. -0.33). Network hubs, defined according to DEG, had CVs% that were less significantly anti-correlated with SUVR (around r = −0.10).
Conclusion: An unexpected pattern of anti-correlations between 18F-FDG SUVR and FC variability was identified, which was consistent across most graph metrics, implying that the least variable connections are the ones that require more energy. These results need further investigation, e.g. employing absolute quantification of PET data and alternative dynamic FC methods, as well as validation from a biological and/or computational standpoint.
References
RiedlVBienkowskaK.
Local activity determines functional connectivity in the resting human brain: a simultaneous FDG-PET/fMRI study. J Neurosci2014;
34: 6260–6266.AielloMSalvatoreE.
Relationship between simultaneously acquired resting-state regional cerebral glucose metabolism and functional MRI: a PET/MR hybrid scanner study. Neuroimage2015;
113: 111–121.GordonEMLaumannTO.
Generation and evaluation of a cortical area parcellation from resting-state correlations.Cereb Cortex2016;
26: 288–303.
2020-02
Does ebselen: A potential new lithium mimetic, decrease synaptic glutamate availability? (#9)
Nisha Singh1, 2, Filipa Mota1, Teresa Sementa1, Natasha Moses2, Carlotta Taddei1, Jayanta Borlodoi1, Stefan Hader1, Thomas Eykyn1, Diana Cash2, Mattia Veronese2 and Federico E Turkheimer2
1School of Biomedical Engineering & Imaging Sciences, King’s College London, London, UK
2Department of Neuroimaging, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London, UK
Abstract
Introduction: In bipolar disorder, magnetic resonance spectroscopy has shown elevations in brain glutamate. We recently showed that ebselen, a potential new lithium mimetic, decreased brain glutamate in healthy volunteers.1 However, if this change is reflected in glutamatergic neurotransmission, or is a consequence of metabolism, remains unknown. In this study, we used in vivo positron emission tomography to determine if ebselen reduced glutamate transmission using [18F]FPEB, an mGluR5 radiotracer. We hypothesised that ebselen, by virtue of inhibiting glutaminase,2 should decrease brain glutamate, and we would observe increased brain uptake of [18F]FPEB with ebselen treatment, compared to controls. Less glutamate at the receptor level implies more available sites for the tracer to bind to and consequently increased [18F]FPEB uptake.
Methods: Rats were divided into four groups (n = 4/group); ‘ebselen acute’, ‘vehicle acute’, ‘ebselen chronic’ (two weeks) and, ‘vehicle chronic’ groups. The acute treatment groups were administered 5 mg/kg ebselen or vehicle intravenously (IV) followed by an IV injection of [18F]FPEB 15 minutes later, and scanned for 45 minutes. The chronic treatment groups were at baseline, after one week, and two weeks of ebselen treatment (3 µg/mL in drinking water). The data was analysed using Logan graphical analysis with cerebellum as normative region and distribution volume ratio as the main parameter of interest.
Results: In the acute treatment group, there was an overall increase in [18F]FPEB brain uptake (p = 0.028), as hypothesized. In the chronic treatment group, there was an increased [18F]FPEB uptake (p = 0.004), however, the vehicle group also showed an increase (p < 0.001). On looking closer, there was a significant correlation between weight of the rat and brain uptake of [18F]FPEB (p = 0.03) and so the results were co-varied for weight. After corrected for weight, there was a significant increase in [18F]FPEB after 1 week of treatment, (p = 0.004) but not after two weeks.
Conclusion: Acute administration of ebselen decreased synaptic availability of glutamate. However, chronic dosing offered a more complicated picture, where synaptic glutamate levels decreased after one week of treatment but then normalised after two weeks.
Acknowledgements
BAP In Vivo training award, MRC.
Ebselen acutely reduces glutamate transmission
References
MasakiCSharpleyALGodlewskaBR, et al.Effects of the potential lithium-mimetic, ebselen, on brain neurochemistry: a magnetic resonance spectroscopy study at 7 tesla. Psychopharmacology (Berl)2016;
233: 1097–1104.ThomasAGRojasCTanegaC, et al.Kinetic characterization of ebselen, chelerythrine and apomorphine as glutaminase inhibitors. Biochem Biophys Res Commun2013;
438: 243–248.
2020-03
Examining the underpinnings of loudness dependence of auditory evoked potentials with positron emission tomography (#17)
Rajapillai LI Pillai1, Elizabeth A Bartlett2, Mala R Ananth1, Chencan Zhu3, Jie Yang4, Greg Hajcak5, Ramin V Parsey1 and Christine DeLorenzo1, 6
1Psychiatry, Stony Brook University, Stony Brook, NY, USA
2Molecular Imaging and Neuropathology, New York State Psychiatric Institute, New York, NY, USA
3Applied Mathematics and Statistics, Stony Brook University, Stony Brook, NY, USA
4Family, Population, and Preventive Medicine, Stony Brook University, Stony Brook, NY, USA
5Biomedical Sciences and Psychology, Florida State University, Tallahassee, FL, USA
6Biomedical Engineering, Stony Brook University, Stony Brook, NY, USA
Abstract
Introduction: Loudness dependence of auditory evoked potentials (LDAEP) has long been considered to reflect central basal serotonin transmission. However, the relationship between LDAEP and individual serotonin receptors and transporters has not been fully explored in humans, which limits its use as a potential biomarker. Establishing the correlation between LDAEP and PET-based estimates of the serotonin 1A receptor and serotonin transporter binding could increase the utility of this non-invasive modality.
Methods: To examine LDAEP’s relationship with the serotonin system, we performed PET using serotonin-1A (5-HT1A) imaging via [11C]CUMI-101 and serotonin transporter (5-HTT) imaging via [11C]DASB on a mixed sample of healthy controls (n = 4: 4 females, 0 males), patients with unipolar (MDD, n = 11: 4 females, 7 males) and bipolar depression (BD, n = 8: 4 females, 4 males). Full quantification was performed including arterial sampling or simultaneous estimation (SIME).1 On these same participants, we also performed electroencephalography (EEG) within a week of PET scanning, using 1000 Hz tones of varying intensity to evoke LDAEP. We then evaluated the relationship between LDAEP and 5-HT1Aor 5-HTT binding potential (BPFfor [11C]CUMI-101 and VT/fPfor [11C]DASB) in both the raphe2 (5-HT1A)/midbrain (5-HTT) areas and in the temporal cortex. In an exploratory analysis, we examined the relationship between LDAEP and 5-HT1A/5-HTT across 25 regions of the brain, as well as the effects of age and sex, as previous studies have found differences in PET and EEG measures for both.
Results: LDAEP was significantly correlated with 5-HT1ABPFpositively and with 5-HTT VT/fPnegatively in the temporal cortex (p < 0.05), but did not correlate with either in midbrain or raphe (Figure 1). In males only, exploratory analysis showed LDAEP significantly correlated with 5-HT1ABPFin temporal cortex, anterior cingulate cortex, medial prefrontal cortex, occipital cortex, orbitofrontal cortex, and parietal cortex; these correlations were not observed with 5-HTT (see Figure 2 for sample regions).
Conclusion: This multimodal study partially validates preclinical models of a serotonergic influence on LDAEP. Replication in larger samples is necessary to further clarify our understanding of the role of serotonin in perception of auditory tones.
Acknowledgements
We acknowledge the biostatistical computation and support provided by the Biostatistical Consulting Core at School of Medicine, Stony Brook University, and in particular Dr. Mengru Zhang, who provided key support for this study. This work was funded by NIMH grants F30MH109412 (PI: Pillai) and R01MH090276 (PI: Parsey). We would like to thank Emily Hale-Rude for EEG setup assistance and Dr. Craig Tenke for assistance in experimental design and for auditory stimuli.
References
BartlettEA, et al.Quantification of positron emission tomography data using simultaneous estimation of the input function: validation with venous blood and replication of clinical studies. Mol Imaging Biol2018;
21: 926–934.https://renaissance.stonybrookmedicine.edu/psychiatry/research/cubit/data (accessed 11 November 2021).
2020-04
Physiological brain activation by visual stimulation does not alter binding of the synaptic density tracer [11C]UCB-J (#26)
Kelly Smart, Heather Liu, David Matuskey, Ming-Kai Chen, Kristen Torres, Nabeel Nabulsi, David Labaree, Yiyun H Huang, Evan D Morris, Ansel T Hillmer and Richard E Carson.
Yale PET Center, Yale School of Medicine, New Haven, CT, USA
Abstract
Introduction: The radiotracer [11C]UCB-J provides a measure of synaptic vesicle density in vivo with clinical applications in neurological and psychiatric disorders. [11C]UCB-J binds to the synaptic vesicle glycoprotein 2A (SV2A), a ubiquitously expressed protein involved in regulation of neurotransmitter release through vesicle exocytosis. Changes in brain activity could potentially affect tracer binding if binding site availability is altered during the process of vesicle transport, fusion, or recycling. The objective of this study was to determine whether physiological brain activation induces measurable changes in [11C]UCB-J binding.
Methods: Seven subjects (four women) completed two 60-minute [11C]UCB-J PET scans (654 ± 99.0 MBq) on a high-resolution research tomograph (HRRT). In the baseline scan, a static black screen was displayed on video glasses worn by the subject. During the visual activation scan, an 8-Hz flashing radial checkerboard was presented in 3-minute intervals separated by 2-minute rest periods throughout the scan. Scans were performed on the same day with order counter-balanced across subjects. Tissue influx constant, K1, and volume of distribution, VT, were determined using a metabolite-corrected arterial input function and the 1-tissue compartment model. Regional values were determined in primary visual cortex (V1; Brodmann area 17), in temporal cortex as negative control, and in centrum semiovale (white matter) to assess nonspecific binding. Parametric whole-brain images of VT and K1 were compared using SPM.
Results: In V1, [11C]UCB-J K1 increased 25 ± 8.3% (p = 0.009; 95% CI 18–35%) during activation (Figure 1). No change was observed in temporal cortex (-3.1 ± 6.1%, p = 0.77) or centrum semiovale (-4.0 ± 6.1%, p = 0.64). There was no change in VT from baseline to activation in V1 (mean change -3.9 ± 8.8%, p = 0.34; 95% CI -12 to 4%), temporal cortex (-3.1 ± 9.1%, p = 0.61), or centrum semiovale (-1.1 ± 10%, p = 0.83). Representative parametric images during baseline and activation are shown in Figure 2. Whole-brain analyses confirmed an increase in K1 specific to visual cortex (cluster-level pFWE = 0.031), with no significant changes in VT across the brain.
Conclusion: Despite a robust increase in tracer influx to V1, [11C]UCB-J VT did not change during visual stimulation. [11C]UCB-J VT is therefore likely to be a stable measure of synaptic vesicle density regardless of brain state at the time of scan.
2020-05
Reliability assessment of neuronal activation using fPET, BOLD and ASL obtained with simultaneous PET/MR imaging (#38)
Lucas Rischka1, Godber M Godbersen1, Verena Pichler2, Paul Michenthaler1, Sebastian Klug1, Manfred Klöbl1, Vera Ritter1, Wolfgang Wadsak2, 3, Marcus Hacker2, Siegfried Kasper1, Rupert Lanzenberger1 and Andreas Hahn1
1Department of Psychiatry and Psychotherapy, Medical University of Vienna, Vienna, Austria
2Department of Biomedical Imaging and Image-guided Therapy, Division of Nuclear Medicine, Medical University of Vienna, Vienna, Austria
3Center for Biomarker Research in Medicine (CBmed), Graz Steiermark, Austria
Abstract
Introduction: Mapping of cognitive tasks on a neuronal level represents a crucial factor for our understanding of human brain function. Recently, the novel approach of functional PET (fPET) was introduced, enabling quantification of task-specific activation within a single measurement.1,2 Besides task-sensitivity, a valuable neuroimaging method also provides high reliability of activation. We therefore investigated the test-retest performance of task-specific glucose metabolism with fPET and [18F]FDG. Using simultaneous PET/MR imaging enabled an unbiased direct comparison to well-established fMRI methods measuring BOLD-derived activation (BOLD) and cerebral blood flow (CBF) with ASL.
Methods: Twenty healthy subjects underwent two measurements on a hybrid PET/MRI scanner. During the scan, subjects completed an adapted version of the video game Tetris® representing a complex visuo-spatial coordination task with varying task load (i.e. easy/hard condition). In order to track task-specific changes on a multimodal level, functional PET was acquired simultaneously with ASL and within the same session as BOLD (Figure 1). Data processing of CBF and BOLD was performed with standard procedures.2,3 Glucose metabolism (influx constant Ki) was quantified with a general linear model, including regressors for each task load, and the Gjedde-Patlak plot.2 Functional ROIs were defined across all three modalities by a conjunction analysis of significant activations during the hard condition (p < 0.05 FWE-corrected, Figure 2). Test-retest reliability was assessed for both task conditions using the within-subject Coefficient of Variation (CoV) and Intra-Class Correlation (ICC3,1).
Results: Regions active during the task are depicted in Figure 2. The median CoV of the hard/easy condition was 0.20/0.33 (glucose metabolism), 0.17/0.30 (CBF) and 0.21/0.27 (BOLD). The ICC showed moderate consistency for glucose metabolism (0.64/0.41) and CBF (0.60/0.56) but poor for BOLD (0.23/0.38).
Conclusion: We demonstrated matching activation across the three complementary imaging modalities. However, glucose metabolism and cerebral blood flow exhibited a higher consistency between the measurements than BOLD-derived activation. Hence, fPET shows at least similar reliability compared to well-established methods. Since fPET is still novel it has great potential for optimization, such as innovative task modeling or radioligand administration strategies. We propose to consider fPET as a complementary, reliable imaging approach to map task-specific changes in future studies.
Acknowledgements
This research was supported by a grant from the Austrian Science Fund to A. Hahn (FWF KLI 610). L. Rischka and M. Klöbl are recipients of DOC Fellowships of the Austrian Academy of Sciences at the Department of Psychiatry and Psychotherapy, Medical University of Vienna.
References
Hahn, et al.
Quantification of task-specific glucose metabolism with constant infusion of 18F-FDG.J Nucl Med2016;
57: 1933–1940.Rischka, et al.
Reduced task durations in functional PET imaging with [18F]FDG approaching that of functional MRI. NeuroImage2018;
181: 323–330.WangJ, et al.
Amplitude-modulated continuous arterial spin-labeling 3.0-T perfusion MR imaging with a single coil: feasibility study.Radiology2005;
235: 218–228.
2020-06
Assessment of model bias upon detection of dopamine response to challenge (#50)
Michael A Levine1, 2, Joseph B Mandeville1, Finnegan Calabro3, Julie C Price1, Beatriz Luna3 and Ciprian Catana1
1A.A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Boston, MA, USA
2Biophysics, Harvard University, Boston, MA, USA
3Psychiatry, University of Pittsburgh, Pittsburgh, PA, USA
Abstract
Introduction: [11C]Raclopride Positron Emission Tomography (PET) is used to probe dopamine release in response to a behavioral challenge by applying compartmental kinetic modeling, extracting macroparameters including binding potential (BPND). However, behavioral challenges produce small changes in neurotransmitter release, whose detection may be conflated with sources of systematic bias such as model selection. To investigate this effect, voxel maps of absolute change in binding potential (ΔBPND, i.e. BPND-POST – BPND-PRE) in humans were compared to simulated maps which describe the bias induced by the Multilinear Reference Tissue Model (MRTM,MRTM2).1
Methods: 69 healthy young adult participants underwent bolus plus constant infusion [11C]raclopride PET over 90 min with simultaneous functional Magnetic Resonance Imaging and a reward learning task (avg. start, end time: 40,70 min).2 A two-tissue compartment model (2TCM) was forward simulated using an analytic arterial input function,3 global K1’ and k4,4 a single k2’ (0.30/min), voxel maps of R1 and BPND created by averaging fit maps acquired across participants, and no challenge. The acquired and simulated PET data were fit voxel-wise with an extension of MRTM2 that includes a challenge term (unit step at 40 minutes) to produce parametric maps of ΔBPND. Real and simulated ΔBPND maps were compared for both MRTM and MRTM2 to evaluate the potential for bias.
Results: In human participants, MRTM and MRTM2 estimated average putamen BPND to be 3.3 ± SD 1.2 and 3.4 ± SD 0.3, respectively. In response to the behavioral challenge, an average change in putamen BPND of -0.17 ± SD 0.17 (-5.4% ± SD 5.1%) and -0.17 ± SD 0.09 (-5.2% ± SD 2.5%) were observed across participants for MRTM (Figure 1(a)) and MRTM2 (Figure 1(c)) respectively. Simulating a 2TCM without challenge and fitting the models produced a change in putamen BPND of 0.03 (0.9%) for MRTM (Figure 1(b)) and -0.04 (-1.2%) for MRTM2 (Figure 1(d)).
Conclusion: The results suggest a significant behavioral reward challenge response (ΔBPND), exceeding the level of model-induced bias observed in simulations. MRTM2 could lead to overestimation of dopamine responses to behavioral challenge. We are currently investigating approaches to overcome this bias including de-weighting the delivery period and regularizing the Full Reference Tissue Model.5
Acknowledgements
This work was supported by National Institute of Mental Health Grant Number 5R01MH080243-07 and National Institute of Neurological Disorders and Stroke Grant Number R01NS112295.
References
IchiseM, et al.
Linearized reference tissue parametric imaging methods: application to [11C]DASB positron emission tomography studies of the serotonin transporter in human brain.J Cereb Blood Flow Metab2003;
23: 1096–1112.LarsenB, et al.
Maturation of the human striatal dopamine system revealed by PET and quantitative MRI. Nature Commun2020;
11: 846.NormandinMDMorrisED.
Estimating neurotransmitter kinetics with ntPET: a simulation study of temporal precision and effects of biased data. Neuroimage2008;
39: 1162–1179.FardeL, et al.
Kinetic analysis of central [llC]raclopride binding to D2-dopamine receptors studied by PET – a comparison to the equilibrium analysis.J Cereb Blood Flow Metab1989;
9: 696–798.MandevilleJB, et al.
A regularized full reference tissue model for PET neuroreceptor mapping.Neuroimage2016;
139: 405–414.
2020-07
Non-displacable binding is a potential confounding factor in [11C]PBR28 PET studies (#53)
Gjertrud L Laurell1, Pontus Plavén-Sigray1, 2, Aurelija Jucaite2, 3, Andrea Varrone2, Kelly P Cosgrove4, 5, Claus Svarer1, Gitte M Knudsen1, 6, R Todd Ogden7, Francesca Zanderigo8, 9, Simon Cervenka2, Ansel T Hillmer4, 10 and Martin Schain1
1Neurobiology Research Unit, Copenhagen University Hospital Rigshospitalet, Copenhagen, Denmark
2Centre for Psychiatry Research, Department of Clinical Neuroscience, Karolinska Institutet & Stockholm Health Care Services, Stockholm, Sweden
3Precision Medicine and Genomics, PET Science Centre, AstraZeneca, Stockholm, Sweden
4PET Center, Department of Radiology and Biomedical Imaging, Yale University, New Haven, CT, USA
5Department of Psychiatry, Yale University, New Haven, CT, USA
6Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
7Department of Biostatistics, Columbia University, New York, NY, USA
8Department of Psychiatry, Columbia University College of Physicians and Surgeons, New York NY, USA
9Molecular Imaging and Neuropathology Area, New York State Psychiatric Institute, New York, NY, USA
10Department of Radiology and Biomedical Imaging, Yale University School of Medicine, New Haven, CT, USA
Abstract
Introduction: [11C]PBR28 is a frequently used PET radioligand that binds to the 18kDa translocator protein (TSPO), a biomarker sensitive to glial cell activation. In clinical TSPO PET studies, the radioligand total distribution volume, VT, is typically regarded as the preferred outcome measure. VT is the sum of the ligand specific (VS) and non-displaceable binding (VND). Thus, differences in brain-wide VND across groups are potentially a confounding factor because group differences in regional VT could be attributable to differences in VND, rather than differences in actual TSPO densities.
Methods: Here, we used a recently developed method for estimation of VND (simultaneous estimation, SIME1) to disentangle the components VND and VS in data from four previously published [11C]PBR28 PET studies: (i) before and after a lipopolysaccharide challenge2 (8 subjects); (ii) in alcohol use disorder3 (14 patients, 15 controls); (iii) in first-episode psychosis4 (16 patients, 16 controls); and (iv) in Parkinson’s disease5 (16 patients, 16 controls). In each dataset, regional VT estimates were obtained with a two-tissue compartment model, and brain-wide VND was estimated with SIME. Regional VS was calculated as VT – VND. Regional VT and VS, and brain-wide VND were then compared across groups.
Results: Lower VND values were found for individuals with alcohol use disorder (34%, p = 0.00084, Figure 1) and Parkinson’s disease (34%, p = 0.0032, Figure 2), when compared to their corresponding controls. We found no difference in VND between first-episode psychosis patients and their controls (p = 0.30), and the administration of lipopolysaccharide did not change VND (p = 0.38). We also observed a clear effect of genotype on VND; when pooling all controls, we observed that high-affinity binders not only displayed higher specific binding, but also had markedly higher VND than mixed affinity binders (p = 0.00016).
Conclusion: Our findings suggest that in TSPO PET studies, levels of non-displaceable binding may differ both between patient groups and between conditions, and should therefore be taken into account.
References
OgdenRTZanderigoFParseyRV.
Estimation of in vivo specific binding in positron emission tomography studies without requiring a reference region.Neuroimage2015;
108: 234–242.SandiegoCMGallezotJDPittmanB, et al.Imaging robust microglial activation after lipopolysaccharide administration in humans with PET.Proc Natl Acad Sci2015;
112: 12468–12473.HillmerATSandiegoCMHannestadJ, et al.In vivo imaging of translocator protein, a marker of activated microglia, in alcohol dependence.Mol Psychiatry2017;
22: 1759–1766.CollsteKPlavén-SigrayPFatouros-BargmanH.
Lower levels of the glial cell marker TSPO in drug-naive first-episode psychosis patients as measured using PET and [11C]PBR28.Mol Psychiatry2017;
22: 850–856.VarnäsKCselényiZJucaiteA, et al.PET imaging of [11C]PBR28 in Parkinson’s disease patients does not indicate increased binding to TSPO despite reduced dopamine transporter binding.Eur J Nucl Med Mol Imaging2019;
46: 367–375.
2020-08
Transcriptome-based human cerebral cortex parcellation (#55)
Matej Murgaš, Gregor Gryglewski, Manfred Klöbl, Murray B Reed and Rupert Lanzenberger.
Department of Psychiatry and Psychotherapy, Medical University of Vienna, Wien, Austria
Abstract
Introduction: Compartmentalization of the human cerebral cortex, in the terms of structural organization, e.g. Desikan-Killiany,1 cytoarchitecture, Brodmann areas,2 or functional specialization, Glasser atlas3 is an essential part of neuroimaging. However, mappings based on proteomic and transcriptomic data are still largely missing. Hence, the aim of this work was to find a specific delineation to the functionally specific areas using molecular information from the mRNA transcriptome.
Methods: The dataset of 18686 mRNA expression patterns mapped to the fsaverage space (FreeSurfer common space) consisting of 163842 vertices per hemisphere were used. mRNA maps were interpolated from 3702 microarray samples of six left hemispheres taken from the Allan Human Brain atlas.4 Using agglomerative hierarchical cluster analysis with average linkage and Pearson correlation as the distance metric, a structural representation of the data was created. The Bayesian Information Criterion (BIC) was used to optimize the number of clusters. Subsequently, the similarities with already existing atlases were quantified using the Dice coefficient, demonstrating the goodness of overlap between the parcellations.
Results: The hierarchical structure of the data is presented in Figure 1(a) as dendrogram. Based on the BIC, the dendrogram was cut so that the cerebral cortex was split into 33 regions (Figure 1(b) – middle). The transcriptome-based compartmentalization was compared to the Brodmann areas (10 of 42 regions with Dice > 0.5, mean 0.37), the Desikan-Killiany atlas (6 of 34 regions with Dice > 0.5 mean 0.35) and the Glasser atlas (42 of 180 regions with Dice > 0.5, mean 0.4). Although some regions show a certain overlap with all three parcellations, mostly low dice coefficients emphasize the difference of the atlases under comparison.
Conclusion: The presented work introduces a parcellation of the human cerebral cortex based on the whole-brain transcriptome. In future work, molecular-based mapping of the cortex could be used in the prediction of the activation and structural patterns associated with psychopharmacological challenges. Prospectively, molecular-based approach could help us to understand the interspecies homology on the level of functional organization comparing brains of various non-human primates.
Acknowledgements
M.M. is funded by the Austrian Science Fund FWF DOC 33-B27. M.K. and M.B.R. are recipients of a DOC-fellowship of the Austrian Academy of Sciences at the Department of Psychiatry and Psychotherapy, MUV.
References
DesikanRS.
An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest.Neuroimage2006;
31: 968–980.BrodmannK. Vergleichende Lokalisationslehre der Grosshirnrinde in ihren Prinzipien dargestellt auf Grund des Zellenbaues.
Leipzig:
Barth, 1909.GlasserMF.
A multi-modal parcellation of human cerebral cortex.Nature2016;
536: 171–178.GryglewskiG.
Spatial analysis and high resolution mapping of the human whole-brain transcriptome for integrative analysis in neuroimaging. NeuroImage2018;
176: 259–267.
2020-09
Neuroinflammation in autism spectrum disorder: A [18F]FEPPA PET Study. (#57)
Dominic Simpson1, Stephanie H Ameis2, 3, Avideh Gharehgazlou4, Romina Mizrahi2, 3 and Pablo M Rusjan1, 3
1Department of Pharmacology & Toxicology, University of Toronto, Toronto, ON, Canada
2Department of Psychiatry, University of Toronto, Toronto, ON, Canada
3Centre for Addiction and Mental Health, Toronto, ON, Canada
4Institute of Medical Science, University of Toronto, Toronto, ON, Canada
Abstract
Introduction: Converging evidence points to the significant involvement of the immune system and neuroinflammation in Autism Spectrum Disorders (ASD). For example, people diagnosed with ASD present with elevated levels of cytokines in peripheral blood and CSF. Monocytes derived from individuals on the autism spectrum show aberrant innate immune responses upon stimulation with lipopolysaccharide, with excessive production of proinflammatory cytokines. Translocator protein 18 kDa (TSPO) is a mitochondrial protein mostly localized in microglia, which can be quantified with PET, providing an in-vivo marker of neuroinflammation/microglial activation. This preliminary analysis aimed to explore whether markers of brain inflammation/microglial activation were altered in participants with ASD in-vivo using [18F]FEPPA PET imaging.
Methods: ASD participants were scanned for 2 hours with an HRRT following an injection of [18F]FEPPA. Clinical diagnosis of ASD was confirmed using the Autism Diagnostic Observation Schedule-2. All subjects had an IQ≥70. Subjects were classified based on their rs6971 TSPO polymorphism to high, medium, or low affinity binders (HAB, MAB, and LAB, respectively), and LABs were excluded. The 2-TCM was used to determine the total volume of distribution (VT) of 4 previously identified regions of interest (ROI): the prefrontal, temporal, cerebellar and anterior cingulate cortices. Data for typically developing volunteers (Controls) from previous studies were used for comparison. Linear mixed models were used with group and genotype as fixed factors, ROIs as a repeated within-subject factor, participants as a random effect (including intercept), and VT as the dependent variable. Posthoc testing was adjusted using Bonferroni.
Results: We scanned 13 individuals with ASD (5F, age 25 ± 5, 5 MAB). 13 Controls (9F, age 24 ± 5, 5 MAB) were chosen from previous studies to match age and genotype. The VT for the ACC of an ASD subject did not present identifiability. There were no significant differences in VT in participants with ASD versus controls (F = 1.7,p = 0.2), though VT was 15% lower in MABs, and 23% in HABs in ASD (Figure 1). Of note, a single subject (ASD, HAB) had VT > 2 SD of the mean.
Conclusion: While no clear evidence of altered inflammation/microglia activation was found in participants with ASD, this preliminary analysis was limited by a small sample size.
Acknowledgements
CIHR MOP-142376.
2020-10
Development of a carbon-11 positron emission tomography pro-radiotracer for imaging the astrocyte glutamate transporter 1 (#64)
Igor C Fontana1, 2, Salvatore Bongarzone1, Eduardo R Zimmer2, Diogo O Souza2 and Antony Gee1
1Division of Imaging Sciences and Biomedical Engineering, King’s College London, St Thomas’ Hospital, London, UK
2Graduation Program in Biological Sciences: Biochemistry, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil
Abstract
Introduction: The astrocyte glutamate transporter 1 (GLT-1) is essential for the clearance and recycling of glutamate in the brain. In fact, GLT-1 dysfunction is associated with neuronal death and linked to multiple brain disorders.1 Synthesizing inhibitors with high selectivity to GLT-1 may help understanding glutamatergic dysfunction in these conditions. Greenfield et al.2 described a GLT-1-selective inhibitor (Figure 1(a) – IC50in vitro = 0.7 µM), with suitable molecular structure for carbon-11 radiolabelling, but with low brain penetrance profile (typical for carboxylic acids). To overcome this limitation, developing an analogous ester to be hydrolysed to the active form inside the brain, seems a promising approach. Here, we aim to develop a carbon-11 labelled pro-radiotracer ([11C]IF1), which we hypothesize will be hydrolysed to the GLT-1-targeting radiotracer ([11C]IF2 – Figure 1(b)) after crossing the blood-brain barrier (BBB) (Figure 1(c)).
Methods: The [11C]IF1 radiosynthesis was performed using an Eckert & Ziegler Modular-Lab system. [11C]CO2 was converted to [11C]CO as previously reported by Taddei et al.3 The [11C]CO afforded [11C]IF0 via a palladium-mediated [11C]carbonylation reaction. [11C]IF1 was obtained after removal of the BOC protecting group with trifluoroacetic acid (Figure 1(d)). Positron emission tomography (PET) imaging experiments were conducted as shown in Figure 2(a).
Results: [11C]IF1 was obtained within 18 min after the end of bombardment with radiochemical yield of 79% (based on radio‐HPLC analysis of the crude product) and radiochemical purity of 99%. Preliminary results from PET imaging analysis were depicted in Figure 2(b). In brief, standardize uptake values (SUVs) were high in the intestine (SUV = 17) and liver (SUV = 6), but low in the brain (SUV = 0.8).
Conclusion: [11C]IF1, a potential GLT-1 pro-radiotracer, was successfully synthesized from [11C]CO. The low in vivo [11C]IF1 brain uptake suggests that the methyl ester is too rapidly metabolised to enable BBB penetration. Interestingly, astrocyte-like cells are highly abundant in the intestine and liver. Thus, it is likely that [11C]IF1 also binds to peripheral GLT-1. Work is in progress to further characterize [11C]IF1 peripheral binding and also to optimize this reaction, furnishing the pharmacophore with a more metabolically resistant ester function to allow greater BBB penetration.
Acknowledgements
[1] Universidade Federal do Rio Grande do Sul (UFRGS); [2] Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES); [3] King’s College London; [4] Wellcome EPSRC Centre For Medical Engineering.
Figure 1. [11C]IF1 PET imaging in healthy mice. a. [11C]IF1 and [11C]IF2 PET biodistribution scheme. Female mice were anesthetised with isoflurane 2.5% and 100 µL of [11C]IF1 or [11C]IF2 (saline with 10% ethanol) were administered into mouse tail vein. After a 60 min dynamic PET scan, CT was acquired and animals were culled. b. The pro-radiotracer [11C]IF1 was mostly taken up in the intestine, liver, kidneys and bladder. [11C]IF1 dynamic SUV. [11C]IF1 brain uptake was very low (SUV = 0.8), with highest accumulation in the liver (SUV = 6) and intestine (SUV = 17).
Figure 2. Development of a carbon-11 GLT-1 specific pro-radiotracer.
a. GLT-1 specific inhibitor developed by Greenfield et al. (2005). b. GLT-1-specific radiotracer – [11C]IF2. c. Hypothesis of work: the pro-radiotracer [11C]IF1 is hydrolysed, releasing inside the brain the EAAT2-specific radiotracer [11C]IF2. d. Radiosynthesis scheme to obtain the pro-radiotracer [11C]IF1
References
PetersonARBinderDK.
Post-translational regulation of GLT-1 in neurological diseases and its potential as an effective therapeutic target. Front Mol Neurosci2019;
12: 164.GreenfieldA, et al.Synthesis and biological activities of aryl-ether-, biaryl-, and fluorene-aspartic acid and diaminopropionic acid analogs as potent inhibitors of the high-affinity glutamate transporter EAAT-2. Bioorg Med Chem Lett2005;
15: 4985–4988.TaddeiC, et al.[(11)C]CO2 to [(11)C]CO conversion mediated by [(11)C]silanes: a novel route for [(11)C]carbonylation reactions. Chem Commun (Camb)2015;
51: 11795–11797.
2020-11
Amyloid (A), tau (T) and voxel-based morphometry (N) correlates of visual memory performance (#86)
Tharick A Pascoal1, 2, Andréa L Benedet1, 2, Joseph Therriault1, 2, Min-Su Kang1, 2, Melissa Savard1, 2, Sulantha Mathotaarachchi2, 3, Firoza Z Lussier1, 2, Cécile Tissot1, 2, Serge Gauthier1, 2, Pedro Rosa-Neto1, 2 and Jaime Fernandez Arias1, 2
1McGill University Research Center for Studies in Aging, McGill University, Montreal, Québec, Canada
Introduction: The Aggie Figures Learning Test (AFLT) is a memory test that was designed to be a visual analogue of the Rey Auditory Verbal Learning Test (RAVLT). Previous studies have found associations between tau and amyloid PET and brain volume and RAVLT scores. However, to date, no study has explored the associations between these markers and AFLT scores. We aimed at exploring such associations.
Methods: Structural MRI, amyloid PET ([18F]-NAV4694) and tau PET ([18F]-MK6240) were acquired for 161 individuals. We conducted analyses on two sub-samples. Demographic data is shown on Table 1. MRI were segmented into probabilistic grey (GM) and white (WM) maps, non-linearly registered to the ADNI template using Dartel and smoothed with an 8 mm FWHM gaussian kernel. Voxel-wise linear regression models were applied, using VoxelStats, with AFLT sub-scores as dependent variables and either tau and amyloid binding or Voxel-Based Morphometry as predictors. We corrected for sex, Apoe genotype, age and years of education.
Results: We found negative associations between tau binding and AFLT total (trials 1–5) and AFLT delayed recall (DR) scores in the MTL and temporo-occipital cortices, as well as in some of their WM tracts. These associations were stronger for the total scores and, in both cases, in the right hemisphere. For the amyloid biomarker, associations with AFLT total and AFLT DR scores were negative in right ventromedial PFC, right posterior thalamus and basal ganglia. In this case, strongest associations are reported between amyloid burden and AFLT DR scores. Finally, we found positive associations between AFLT total and DR scores and VBM in the fusiform gyrus bilaterally.1–3
Conclusion: Results resemble previously reported findings on associations between RAVLT and tau, amyloid and brain volume estimates. Our findings are also in line with initial reporting of patients with left hemisphere damage having more difficulties with RAVLT and patients with right hemisphere damage finding more trouble with AFLT tasks.4 In addition, relationships with tau tend to be more posterior, while relationships with amyloid were shown to be more anterior. Further analyses based on diagnostic categories are needed to explore how these associations change with pathology.5
Acknowledgements
Many thanks to the Translational Biomarkers of Aging and Dementia (TRIAD) cohort, the Weston Brain Institute, the Douglas Mental Health University Institute, Canadian Institutes of Health Research, and the Fundación San Rafael.
Demographic data
Imaging results
On the left, amyloid and tau correlates for AFLT Delayed Recall Scores; in the middle, amyloid and tau correlates for total AFLT scores; on the right, VBM correlates for both DR and total AFLT scores.
References
AltomareDde WildeAOssenkoppeleR, et al.Applying the ATN scheme in a memory clinic population: the ABIDE project.Neurology2019;
93: e1635–e1646.DigmaLAMadsenJRReasET, et al.;Alzheimer’s Disease Neuroimaging, I. Tau and atrophy: domain-specific relationships with cognition. Alzheimers Res Ther2019;
11: 65.FjellAMWalhovdKBAmlienI, et al.Morphometric changes in the episodic memory network and tau pathologic features correlate with memory performance in patients with mild cognitive impairment. AJNR Am J Neuroradiol2008;
29: 1183–1189MajdanASziklasVJones-GotmanM.Performance of healthy subjects and patients with resection from the anterior temporal lobe on matched tests of verbal and visuoperceptual learning. J Clin Exp Neuropsychol1996;
18: 416–430.SziklasVJones-GotmanM.RAVLT and nonverbal analog: French forms and clinical findings. Can J Neurol Sci2008;
35: 323–330.
2020-12
Neuroinflammation as a predictor of total knee arthroplasty recovery (#88)
Zeynab Alshelh1, Erin J Morrissey1, Angel Torrado-Carvajal1, 2, Atreyi Saha1, Minhae Kim1, Samantha Rotman1, Daniel Albrecht1, Oluwaseun J Akeju3, Yang Lin1, Kayla Florio3, Young-Min Kwon4, Hany Bedair4, Antonia Chen5, Zhongcong Xie3, Vitaly Napadow1, Robert Edwards3 and Marco L Loggia1
1Radiology, Massachusetts General Hospital/Harvard Medical School, Boston, MA, USA
2Medical Image Analysis and Biometry Laboratory, Universidad Rey Juan Carlos, Madrid, Spain
3Anesthesia, Massachusetts General Hospital, Boston, MA, USA
4Orthopaedics, Massachusetts General Hospital, Boston, MA, USA
5Orthopaedics, Brigham and Women’s Hospital, Boston, MA, USA
Abstract
Introduction: Knee-osteoarthritis (KOA) is one of the most prevalent causes of pain and disability. Identifying whether a patient will experience persistent post-surgical pain (PPSP) can be extremely useful in the clinical decision-making process. While the mechanisms underlying PPSP remain elusive, the decoupling between pain and peripheral pathology, suggests that alterations in the central nervous system might have a role1-2. A growing body of literature implicates a dysregulation in the activation of glial cells in preclinical pain models3-4. Here, we tested the hypothesis that neuroinflammation predicts PPSP in KOA patients scheduled to undergo primary-TKA. To this end, we used an integrated PET/MR imaging paired with [11C]PBR28, a radioligand that binds to the 18 kDa translocator protein, which is upregulated in activated glia5.
Methods: Eight subjects with KOA were recruited for the study. At the pre-TKA visit patients were assessed for pain, stiffness, functional limitation, physical function and pain interference, then were subsequently scanned. One-year post-TKA, patients were assessed again to determine changes from the pre-TKA visit. Correlation analyses were performed between the pain change scores and pre-TKA PET signal, in a voxel-wise analysis. [11C]PBR28 signal extracted from the regions significant in this analysis were assessed for correlations with change in stiffness, functional limitation, physical function and pain interference.
Results: In the voxel-wise analysis, subjects showed a negative correlation between [11C]PBR28 and change in pain intensity in regions encompassing parts of the anterior cingulate cortex, the dorsomedial prefrontal cortex and ventromedial prefrontal cortex, indicating that in these regions, higher pre-surgical [11C]PBR28 signal predicts poorer recovery one-year post-TKA. Within these regions, there was a negative correlation between [11C]PBR28 signal and change in functional limitation (r = −0.95, p = 0.004), physical function (r = −0.96, p = 0.003) and pain interference (r = −0.88, p = 0.021) and no significant correlation between [11C]PBR28 signal and change in stiffness (r = −0.70, p = 0.123).
Conclusion: Our results indicate that higher inflammatory signal in the brain of patients with KOA before undergoing primary TKA might be predictive of a poorer recovery 1-year post surgery. If these results are reproduced in a larger sample, then these findings could better inform our decision-making process on whether primary TKA should be performed in specific patients and also to individualize pre- and post-surgical treatment accordingly.
References
BakerAD. Abnormal magnetic-resonance scans of the lumbar spine in asymptomatic subjects. A prospective investigation.
Classic papers in orthopaedics:
Springer; 2014: 245–7.JensenMCBrant-ZawadzkiMNObuchowskiNModicMTMalkasianDRossJS.Magnetic resonance imaging of the lumbar spine in people without back pain. New England Journal of Medicine1994;
331(2): 69–73.HainsBCWaxmanSG.Activated microglia contribute to the maintenance of chronic pain after spinal cord injury. Journal of Neuroscience2006;
26(16): 4308–4317.TsudaMShigemoto-MogamiYKoizumiS, et al.P2X 4 receptors induced in spinal microglia gate tactile allodynia after nerve injury. Nature2003;
424(6950): 778–783.LoggiaMLChondeDBAkejuO, et al.Evidence for brain glial activation in chronic pain patients. Brain2015;
138(3): 604–615.
2020-13
Reduced metabotropic glutamate receptor subtype 5 in fragile X syndrome (#92)
James R Brasic1, Anil K Mathur1, Ayon Nandi1, Keith Slifer2, 3, Zabecca Brinson1, Matthew Ryan4, Rebecca Landa3, 5, Ebony Holliday5, Thomas W Sedlak1, 3, Emily Dillon5, Samuel D Martin1, Pankhuri Vyas6, Rohan Panaparambil1, Omar KAE El Mandouh1, Elizabeth M Berry-Kravis7, Ernest M Mahone3, 4, Dean F Wong1, 8 and Dejan B Budimirovic3, 9
1Russell H. Morgan Department of Radiology and Radiological Science, Division of Nuclear Medicine and Molecular Imaging, Section of High Resolution Brain Positron Emission Tomography Imaging, Johns Hopkins University, Baltimore, MD, USA
2Behavioral Psychology Department, Kennedy Krieger Institute, Johns Hopkins University, Baltimore, MD, USA
3Department of Psychiatry and Behavioral Sciences, Johns Hopkins University, Baltimore, MD, USA
4Neuropsychology Department, Kennedy Krieger Institute, Johns Hopkins University, Baltimore, MD, USA
5Center for Autism and Related Disorders, Kennedy Krieger Institute, Johns Hopkins University, Baltimore, MD, USA
6Otolaryngology-Head and Neck Surgery, Johns Hopkins University, Baltimore, MD, USA
7Rush University Medical Center, Departments of Pediatrics, Neurological Sciences, Biochemistry, Chicago, IL, USA
8Department of Radiology, Laboratory of Central Nervous System (CNS) Neuropsychopharmacology and Multimodal Imaging (CNAMI), Mallinckrodt Institute of Radiology, Washington University in Saint Louis, Saint Louis, MO, USA
9Psychiatry Department, Fragile X Clinic, Kennedy Krieger Institute, Johns Hopkins University, Baltimore, MD, USA
Abstract
Introduction: Fragile X Mental Retardation Protein (FMRP) deficits lead to fragile X syndrome (FXS), the leading known single-gene cause of intellectual disability and autism spectrum disorder (ASD). Converging evidence in animal models of FXS demonstrated reversal of overly active metabotropic glutamate receptor subtype 5 (mGluR5) by pharmacological interventions. Yet clinical trials of the mGluR5 antagonists have failed in adult or adolescent humans with FXS. While studies using positron emission tomography (PET) have shown that these drugs do engage mGluR5, their expression has not been measured in FXS.1 Here, we aim to show a feasibility to measure an expression of the mGluR5 receptors in humans with FXS.2
Methods: Positron emission tomography (PET) with a high resolution research tomograph (HRRT) was conducted for 90 minutes after the intravenous bolus injection of 185 megabequerels (MBq) (5 millicuries [mCi] 3-[18F]fluoro-5-(2-pyridinylethynyl)benzonitrile ([18F]FPEB), a potent, selective mGluR5 inhibitor3 to two men with the full mutation for fragile X syndrome (FXS) aged 24–27 (25.5 ± 2.12) years, six men with idiopahtic autism spectrum disoder (IASD) aged 18–23 (20 ± 2.10) years, and three men with typical development (TD) aged 26–31 (27 ± 3.61) years.4 The nondisplaceable binding potential (BPND) was calculated for relevant volumes of interest (VOIs) (Figure 1) on PET images coregistered with magnetic resonance imagery (MRI).3
Results: In contrast to both the IASD and TD groups, BPNDs for the men with FXS were reduced in all VOIs (Figure 2). Reductions in the men with FXS were particularly marked in the limbic system and the striatum (Figure 2).
Conclusion: This study supports the viability of PET to measure mGluR5 density in FXS. The marked reductions of mGluR5 density in the cortices and the limbic system support the finding that two distinct but interrelated social behavior abnormalities, one linked to impaired cognitive processes (delayed socialization) and the second one to disturbance in limbic circuits (avoidance), play a role in the development of ASD in boys with FXS.5 The protocol suggests that PET may be used to quantify a mGluR5 binding to confirm target engagement for clinical trials of novel glutamatergic agents for FXS and other subtypes of ASD.
Acknowledgements
This research is funded by a Radiology Bridge Funding Initiative to Stimulate and Advance Research (RAD BriteStar Bridge) Award, Johns Hopkins University, Baltimore, Maryland (JRB) and the Intellectual & Developmental Disabilities Research Center (U54 HD079123) at the Kennedy Krieger Institute of Johns Hopkins Medical Institutions in Baltimore, Maryland (JRB). Hiroto Kuwabara, MD, PhD, provided image analysis. The Positron Emission Tomography (PET) Radiotracer Service Center and the Research MRI Service Center at Johns Hopkins Medicine provided excellent performance of the scans of this study.
Figure 2. Nondisplaceable binding potentials for [18F]FPEB of men with FXS, ASD, and TD. Nondisplaceable binding potentials (BPNDs) after the intravenous bolus inject of approximately 185 megabequerels (MBq) (5 millicuries [mCi] 3-[18F]fluoro-5-(2-pyridinylethynyl)benzonitrile ([18F]FPEB), a radiotracer for metabotropic glutamate receptor subtype 5 (mGluR5) (3) were calculated for volumes of interests (Figure 1) of two men with fragile X syndrome (FXS), six men with autism spectrum disorder (ASD), and three men with typical development (TD) (4).
The densities of mGluR5s were markedly lower in the men with FXS than the men with ASD and TD.
References
BudimirovicDBBerry-KravisEEricksonCA, et al.Updated report on tools to measure outcomes of clinical trials in fragile X syndrome.J Neurodev Disorders2017;
9: 14. [PubMed: 28616097], [PubMedCentral: PMC5467057]BrašićJRMathurAKBudimirovicDB.
The urgent need for molecular imaging to confirm target engagement for clinical trials of fragile X syndrome and other subtypes of autism spectrum disorder.Arch Neurosci2019;
6: e91831.WongDFWaterhouseRKuwabaraH, et al.18F-FPEB, a PET radiopharmaceutical for quantifying metabotropic glutamate 5 receptors: a first-in-human study of radiochemical safety, biokinetics, and radiation dosimetry. J Nucl Med2013;
54: 388–396.FatemiSHWongDFBrašićJR, et al.Metabotropic glutamate receptor 5 tracer [18F]-FPEB displays increased binding potential in postcentral gyrus and cerebellum of male individuals with autism: a pilot PET study. Cerebellum Ataxias2018;
5: 3.BudimirovicDBBukelisICoxC, et al.Autism spectrum disorder in fragile X syndrome: differential contribution of adaptive socialization and social withdrawal. Am J Med Genet Part A2006;
140A: 1814–1826.
2020-14
Clinical validation of [18F]EKZ-001: A novel positron emission tomography ligand for quantifying HDAC6 in the human brain (#104)
Michel Koole1, Donatienne Van Weehaeghe1, 2, Guy Bormans3, Sofie Celen3, Kim Serdons2, Jan de Hoon4, Marissa Herbots4, Christopher Cawthorne1, Frederick A Schroeder5, Jacob M Hooker6, Janice E Kranz5, Koen Van Laere1, 2 and Tonya M Gilbert5
1Nuclear Medicine and Molecular Imaging, KU Leuven, Leuven, Belgium
2Nuclear Medicine and Molecular Imaging, University Hospitals Leuven, Leuven, Belgium
3Laboratory for Radiopharmaceutical Research, KU Leuven, Leuven, Belgium
4Center for Clinical Pharmacology, University Hospitals Leuven, Leuven, Belgium
5Eikonizo Therapeutics, Inc., Cambridge, MA, USA
6Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA
Abstract
Introduction: Histone deacetylase 6 (HDAC6) is a cytoplasmic enzyme that modulates intracellular transport1 and protein quality control.2 Inhibition of HDAC6 deacetylase activity has shown beneficial effects in disease models,3,4 including Alzheimer’s disease, amyotrophic lateral sclerosis and major depressive disorder. This first-in-human positron emission tomography (PET) study evaluated [18F]EKZ-001 ([18F]Bavarostat5), a radioligand selective for HDAC6, in healthy adult subjects.
Methods: [18F]EKZ-001 was administered by intravenous bolus injection (142 ± 32 MBq). Biodistribution and radiation dosimetry studies were performed in four healthy subjects (2M/2F, 23.5 ± 2.4 years) using sequential whole-body PET/CT (Figure 1). The most appropriate kinetic model to quantify brain uptake was determined in 12 healthy subjects (6M/6F, 57.6 ± 3.7 years) from 120-minute dynamic PET/MR using a metabolite-corrected arterial plasma input function. Four subjects underwent retest scans (2M/2F, 57.3 ± 5.6 years) with a one-day interscan interval to determine test-retest variability (TRV). Fittings of a one-tissue and two-tissue compartment model (1–2TCM) were compared using the Akaike information criterion (AIC) while regional distribution volumes (VT) were also calculated with Logan graphical analysis (LGA) and time stability of VT was assessed. VT differences between males and females were evaluated using a whole brain voxel-wise analysis.
Results: The effective dose was 39.1 ± 7.0 µSv/MBq. Based on AIC, 2TCM was the preferred model compared to 1TCM. Regional LGA VT were in line with 2TCM values, however with a lower absolute TRV of 7.7 ± 4.9%. Regional VT values were relatively homogeneous with highest values in hippocampus, entorhinal cortex, anterior cingulate, insula and cerebellum (Figure 2). Reduction of acquisition time was achieved by a coffee-break protocol with a 0- to 60-minute scan followed by a 90- to 120-minute scan resulting in regional LGA VT with TRV of 4.5 ± 7.9% and average bias < 1%. Males demonstrated significantly higher LGA VT than females in a majority of cortical and subcortical brain regions (Pheight < 0.001, kextent = 50). No relevant tracer related adverse events were reported.
Conclusion: [18F]EKZ-001 is safe and appropriate for quantifying HDAC6 expression in the human brain with the Logan plot as the preferred quantitative approach. Sex differences should be considered in future studies.
Acknowledgements
This study was supported by funding from the Alzheimer’s Drug Discovery Foundation, Inc. (ADDF) to Eikonizo (J.E.K.).
References
DompierreJP, et al.Histone deacetylase 6 inhibition compensates for the transport deficit in Huntington’s disease by increasing tubulin acetylation. J Neurosci2007;
27: 3571–3583.LeeJ-Y, et al.HDAC6 controls autophagosome maturation essential for ubiquitin-selective quality-control autophagy. EMBO J2010;
29: 969–980.ZhangL, et al.Tubastatin A/ACY-1215 improves cognition in Alzheimer’s disease transgenic mice. J Alzheimers Dis2014;
41: 1193–205.GuoW, et al.HDAC6 inhibition reverses axonal transport defects in motor neurons derived from FUS-ALS patients. Nat Commun2017;
8: 861.StreblMG, et al.HDAC6 brain mapping with [18F]bavarostat enabled by a Ru-mediated deoxyfluorination. ACS Cent Sci2017;
3: 1006–1014.
2020-15
Fetal imaging of synaptic vesicle glycoprotein 2A using 18F-SynVesT-1 PET on the primate miniEXPLORER (#105)
Samantha Rossano1, 2, Takuya Toyonaga1, Stephanie M Groman3, Songye Li1, Eric Berg4, Yiyun H Huang1, Alice F Tarantal5 and Richard E Carson1, 2
1Yale PET Center, Department of Radiology and Biomedical Imaging, Yale University, New Haven, CT, USA
2Department of Biomedical Engineering, Yale University, New Haven, CT, USA
3Department of Psychiatry, Yale University, New Haven, CT, USA
4Department of Biomedical Engineering, University of California, Davis, CA, USA
5Departments of Pediatrics and Cell Biology and Human Anatomy and California National Primate Research Center, University of California, Davis, CA, USA
Abstract
Introduction: The formation of synapses between neurons, or synaptogenesis, begins mid-gestation in primates and continues following birth into the neonatal period.1 Synaptic vesicle glycoprotein 2A (SV2A) PET imaging of gravid rhesus monkeys allows for longitudinal studies of synaptic density across gestation (term = 165 ± 10 days). The current study investigated changes in SV2A density in the fetal brain during the 2nd and 3rd trimester by collecting longitudinal SV2A PET images in four gravid rhesus monkeys using the primate miniEXPLORER.2
Methods: Four gravid rhesus were administered 18F-SynVesT-1 (SDM-8) twice between gestational days 100–140 (2nd and 3rd trimesters). PET images were acquired on the primate miniEXPLORER and reconstructed using TOF-OSEM (2 iterations/20 subsets). CT images were acquired for attenuation correction and anatomical delineation. Fetal brain uptake was quantified by distribution volume ratio (DVR) using the simplified reference tissue model (TMAX = 30mins) using the maternal brain as a reference region. Fetal brain images collected at gestation day 140 were registered to a 2-week-old neonatal rhesus template3 and DVRs were reported in ten regions: Frontal, parietal, temporal, occipital lobes; cerebellum, hippocampus, amygdala, caudate, putamen, thalamus/midbrain. Near term, tissues were harvested and samples from one animal were evaluated for SV2A using Western blots.4
Results: PET/CT images and corresponding time activity curves are shown for a representative subject in Figure 1. Across gestation, fetal brain volume and DVR increased such that the average ( ± SD, N = 4) fetal brain volume was 68 ± 9 mL (Figure 2(a)) and the average DVR was 0.32 ± 0.05 (Figure 2(b)) two weeks before tissue collection. The regional distribution of 18F-SynVesT-1 was heterogeneous at 140 days, with high uptake in subcortical regions compared to cortical regions (Figure 2(c) and (d)). Ex vivo SV2A measures significantly correlated with DVR (Figure 2(e)).
Conclusion: Using the primate miniEXPLORER to image synaptic density (SV2A), fetal/maternal brain DVR increased during the last two months of gestation, a pattern consistent with previous literature. However, the in vivo relative synaptic density substantially underestimates ex vivo data, which shows a fetal/adult ratio ³1.0 in the days before birth in cortical regions. These in vivo data are consistent with previous pilot data (Yale) using 11C‑UCB‑J. Ongoing work includes ex vivo validation of SV2A expression.
References
RakicPBourgeoisJP, et al.Concururent overproduction of synapses in diverse regions of the primate cerebral cortex. Science1986;
232: 232–235.BergEZhangX, et al.Development and evaluation of mini-EXPLORER: a long axial field-of-view PET scanner for nonhuman primate imaging. J Nucl Med2018;
59: 993–998.ShiYBudinF, et al.UNC-Emory infant atlases for macaque brain image analysis: postnatal brain development through 12 months. Front Neurosci2017;
10: 617.HarrisVM. Protein detection by simple Western(TM) analysis. In: KurienBScofieldR (eds) Western Blotting. Methods in Molecular Biology. vol
1312.
New York, NY:
Humana Press, 2015.
2020-16
Simultaneous multi-parameter multi-tracer estimation with dynamic neuro-PET data (#106)
Ayla Mansur1, 2, Gaia Rizzo1, Eugenii A Rabiner1, 3, Roger N Gunn1, 2 and for the MIND-MAPS Consortium.
1Invicro LLC, Boston, MA, USA
2Division of Brain Sciences, Imperial College London, London, UK
3Centre for Neuroimaging Sciences, King’s College London, London, UK
Abstract
Introduction: The MIND-MAPS program (https://lp.invicro.com/mind-maps) studies the mitochondria/endoplasmic reticulum/synaptic axis with 18F-BCPP-EF, 11C-SA-4503 and 11C-UCB-J scans acquired in each subject. This work takes advantage of this unique multi-tracer dataset to investigate whether simultaneous model fitting across tracers could provide further insight into the underlying physiology and improve the robustness of outcome measure estimation.
Methods: For each subject (n = 12), three sets (one per tracer) of arterial input functions and time-activity data from 13 grey matter regions were used. The one tissue compartment model is suitable for quantifying 11C-UCB-J,1 while the two tissue compartment model can be used to quantify 18F-BCPP-EF, 11C-SA-4503.2 For these models, the delivery K1 can be expanded as f(1-e-PS/f)3,4 where f is flow, PS is the permeability surface product and (1-e-PS/f) is extraction. f was shared across tracers and PS and the non-displaceable distribution volume (K1/K2) was shared across regions for each subject. Microparameters were estimated from a simultaneous nonlinear least squares estimation using the Levenberg-Marquardt optimisation algorithm which fitted all data together and enabled calculation of K1 and VT. Akaike information criteria5 were used to compare simultaneous and independent fitting methods. The inter-subject variability of parameters derived using the two methods was compared.
Results: The simultaneous and independent estimation approaches performed similarly with each providing good fits to the data (Figure 1). Mean flow was 73 ± 46 ml/100g/min (values with COV > 100% excluded). The average extraction for 11C-UCB-J, 18F-BCPP-EF and 11C-SA-4503 was 0.50 ± 0.17, 0.63 ± 0.17 and 0.53 ± 0.17, respectively. Simultaneous estimation was the most parsimonious model in 12/13 regions for 11C-UCB-J but only in 1/13 and 2/13 of the regions for 18F-BCPP-EF and 11C-SA-4503 respectively. The inter-subject variability of VT and K1 was not significantly different across methods (Figure 2). Outcome measures derived using the two methods were well correlated for all three tracers (18F-BCPP-EF R2VT = 0.99, R2K1 = 0.99; 11C-SA-4503 R2VT = 0.92, R2K1 = 0.90; 11C-UCB-J R2VT = 1.00, R2K1 = 0.97).
Conclusion: The simultaneous approach had little impact on improving the numerical identifiability of the main outcome measures. We showed that it is possible to derive the otherwise unidentifiable physiological parameters flow and extraction by taking advantage of multi-tracer data in the same subject
Acknowledgements
The authors thank the MIND-MAPS consortium members for their continued support.
Figure 2. Box plots showing distribution of (a) independtantly and simulatenoulsy derived VT estimates, and (b) independtantly and simulatenoulsy derived K1 estimates for three representative regions.
References
FinnemaSJNabulsiNBEidT.
Imaging synaptic density in the living human brain. Sci Transl Med2016;
8: 348ra96.MansurARabinerERComleyRA.
Characterization of 3 PET tracers for quantification of mitochondrial and synaptic function in healthy human brain: 18F-BCPP-EF, 11C-SA-4503, and 11C-UCB-J.J Nucl Med2020;
61: 96–103.RenkinEM.
Transport of potassium-42 from blood to tissue in isolated mammalian skeletal muscles. Am J Physiol1959;
197: 1205–1210.CroneC.
The permeability of capillaries in various organs as determined by use of the ‘indicator diffusion’ method.Acta Physiol Scand1963;
58: 292–305.AkaikeH.
Information theory and extension of the maximum likelihood principle.Int Symp Inf theory1973; 267–281.
2020-17
Thalamic distributions of α4β2* subtype nicotinic acetylcholine receptors in two independently collected positron emission tomography datasets (#107)
Evan Gallagher1, Robert K Doot2, Kelly P Cosgrove3, 4, Ansel T Hillmer3 and Jacob G Dubroff2
1Neuroscience Graduate Group, University of Pennsylvania, Philadelphia, PA, USA
2Department of Radiology, University of Pennsylvania, Philadelphia, PA, USA
3Department of Radiology and Biomedical Imaging, Yale University School of Medicine, New Haven, CT, USA
4Department of Psychiatry, Yale University School of Medicine, New Haven, CT, USA
Abstract
Introduction: Positron emission tomography (PET) brain imaging of the α4β2* subtype nicotinic acetylcholine receptor (nAChR) has improved our understanding of both nicotine dependence and neurodegeneration. These studies consistently show greatest receptor availability in the thalamus.1,2 Although postmortem studies reveal heterogeneous distribution of nAChRs within the thalamus,3,4 PET studies have not rigorously examined these thalamic substructures. This study seeks to compare thalamic nAChR distributions in two independently collected PET datasets using two different nAChR-targeting radiotracers, 2[18F]A and [18F]Flubatine.
Methods: T1-weighted magnetic resonance images and dynamic PET images with venous metabolite sampling were collected (1) from 24 smokers scanned with 2[18F]A,1 and (2) from 5 non-smokers scanned with [18F]Flubatine.2 For each subject, 17 thalamic volumes of interest (VOIs) were generated using a Freesurfer-based segmentation pipeline.5 These VOIs underwent erosion to < 50% of their original volume and were then overlaid onto the co-registered PET images (Figure 1). Normalized distribution volumes (VT) were then collected from all VOIs to establish an average α4β2* nAChR distribution for each dataset.
Results: The mediodorsal lateral nucleus and the anterior pulvinar demonstrated the highest nAChR availability, while the medial pulvinar and anteroventral nucleus showed lower availability. Results from other VOIs were less consistent; see Figure 2.
Conclusion: In agreement with postmortem studies, thalamic segmentation of both 2[18F]A and [18F]Flubatine PET imaging data revealed greatest α4β2* nAChR availability in the mediodorsal lateral and anterior pulvinar nuclei of the thalamus. Discrepancies between the two datasets could reflect differences in radiotracer specificity, characteristics of the study populations (smokers vs non-smokers), or PET instrumentation. These differences warrant further exploration. Furthermore, this analysis pipeline can readily be applied to other PET brain imaging studies using other radiotracers with substantial thalamic binding.
DubroffJGDootRKFalconeM, et al.Decreased nicotinic receptor availability in smokers with slow rates of nicotine metabolism. J Nucl Med 2015;
56: 1724–1729.HillmerATEsterlisIGallezotJD, et al.Imaging of cerebral α4β2* nicotinic acetylcholine receptors with (−)-[18F]Flubatine PET: implementation of bolus plus constant infusion and sensitivity to acetylcholine in human brain. NeuroImage2016;
141: 71–80.RubboliFCourtJASalaC, et al.Distribution of nicotinic receptors in the human hippocampus and thalamus. Eur J Neurosci1994;
6: 1596–1604.SpurdenDPCourtJALloydS, et al.Nicotinic receptor distribution in the human thalamus: autoradiographical localization of [3H]nicotine and [125I]α-bungarotoxin binding. J Chem Neuroanatomy1997;
13: 105–113.IglesiasJEInsaustiRLerma-UsabiagaG, et al.A probabilistic atlas of the human thalamic nuclei combining ex vivo MRI and histology. NeuroImage2018;
183: 314–326.
2020-18
Quantification and kinetic analysis of [11C]deschloroclozapine positron emission tomography imaging for designer receptors exclusively activated by designer drugs in monkey brain (#108)
Xuefeng Yan1, Rachel Dick1, Sanjay Telu1, Mark Eldridge2, Jeih-San Liow1, Paolo Zanotti-Fregonara1, Cheryl Morse1, Lester Manly1, Robert Gladding1, Yuji Nagai3, Takafumi Minamimoto3, Sami S Zoghbi1, Barry J Richmond2, Victor W Pike1 and Robert B Innis1
1Molecular Imaging Branch, National Institute of Mental Health, Magnuson Clinical Center, National Institutes of Health, Bethesda, MD, USA
2Laboratory of Neuropsychology, National Institute of Mental Health, Magnuson Clinical Center, National Institutes of Health, Bethesda, MD, USA
3Department of Functional Brain Imaging, National Institutes for Quantum and Radiological Science and Technology, National Institute of Radiological Sciences, Chiba, Japan
Abstract
Introduction: Designer Receptors Exclusively Activated by Designer Drugs (DREADD) is a powerful chemogenetic tool for manipulation of neuronal activity. Positron emission tomography (PET) imaging with a selective radioligand allows non-invasive visualization of DREADD. Although studies have shown that [11C]deschloroclozapine ([11C]DCZ) is a novel high-potency ligand for DREADD, no full quantitative analysis has been reported. Here we aimed to establish the gold standard for [11C]DCZ kinetic analysis and explore the suitability of reference tissue-based quantification methods.
Methods: Dynamic [11C]DCZ PET scans with arterial input function were acquired for 120 minutes at baseline and after intravenous administration of either clozapine-N-oxide (CNO, 10 mg/kg) or DCZ (1 or 10 mg/kg) in one monkey that received injection of virus expressing human M4 muscarinic DREADD (hM4Di) into the right amygdala. Goodness-of-fit for 1TCM and 2TCM were compared with F-tests, Akaike and MSC values. Regional binding potentials (BPND) derived from the total distribution volume (VT) were compared with those obtained from reference tissue models, with the cerebellum as the reference region.
Results: The baseline scans showed a high [11C]DCZ uptake in the hM4Di-DREADD region, which could be blocked by CNO and DCZ. Cerebellum uptake was lowest, but still almost 10% of the activity could be displaced by the blockers. 2TCM fitted the data better than 1TCM. Compared to 2TCM, BPND-hM4Di estimated with reference tissue models was about 50% lower, while the underestimation was only about 20% in all the other regions.
Conclusion: [11C]DCZ PET images can be quantified with a 2TCM. However, reference tissue methods show underestimation, which is more important in the regions with high BPND, such as the hM4Di region.
Acknowledgements
We would like to thank NIH veterinary staff for all of their help with this project.
Figure 1. Regional plot of BPND difference between reference tissue models and 2TCM.
Compared to 2TCM, BPND-hM4Di estimated with reference tissue models was about 50% lower, while the underestimation was only about 20% in all the other regions.
Representative Logan-derived VT parametric images
2020-19
Pretreatment brain metabolism imaging for prediction of major depressive disorder outcome (#109)
Kathryn R Hill, John Gardus, Ramin V Parsey and Christine DeLorenzo.
Department of Psychiatry, Stony Brook University, Stony Brook, NY, USA
Abstract
Introduction: Prediction of antidepressant treatment outcome could reduce burden of depression. Following our proposal (R01MH104512), we evaluated pretreatment metabolic rate of glucose uptake (MRGlu) in three regions, raphe nucleus (RN),1 right insula,2 and left ventral prefrontal cortex (vPFC), for prediction of major depressive disorder (MDD) treatment outcome in a placebo controlled randomized trial of selective serotonin reuptake inhibitors (SSRI).
Methods: All participants (n = 73) were diagnosed with MDD and scored > 21 on the Montgomery–Åsberg Depression Rating Scale. Participants were either medication naïve (n = 42) or medication free 3 weeks before study initiation (n = 31). The outcome measure was the Hamilton Rating Scale for Depression (HRDS17) after 8 weeks of escitalopram (SSRI) or placebo. Pretreatment MRGlu was calculated using 2-[18F]-fluorodeoxyglucose (FDG) acquired for 60 minutes on a Siemens Biograph mMR. Simultaneous Estimation (SimE)3 was used to estimate the arterial input function and MRGlu from a venous sample acquired 39–43.7 minutes post injection (4 subjects: 63–75 minutes, 2 subjects: arterial sample at 40 minutes). Subject motion was corrected for by rigid body registration to a reference frame4 with subsequent coregistration of MRI to average PET image. The RN, insula and vPFC were automatically delineated using atlases as previously described4 (RN atlas: https://renaissance.stonybrookmedicine.edu/psychiatry/research/cubit/data). Patlak modeling, correcting for blood glucose and lumped constant, was applied to the regional time activity curves for MRGlu estimation. Linear regression was used to examine the association between week-8 HRDS17 and pretreatment MRGlu of the 3 regions. Pretreatment HRDS17 (1 subject: week-1), treatment type (SSRI or placebo) and age were covariates (alpha = 0.05).
Results: Covariates of treatment type and age (average = 29.3 ± 13.4 years) were not significantly associated with week-8 HRDS17. However, pretreatment HRDS17 was significantly associated with week-8 HRDS17. Pretreatment MRGlu(mg/(min*100mL)) in all regions was not significantly associated with outcome (Figure 1).
Conclusion: In our cohort, regional pretreatment metabolism in the RN, right insula, and left vPFC does not predict antidepressant treatment outcome. Further, treatment group did not influence this relationship. Voxel-based analysis may reveal findings outside these regions. However, given the heterogeneity of MDD, prediction from pretreatment metabolism may be symptom dimension dependent, providing a future direction for investigation.
Acknowledgements
Funding source: R01MH104512.
References
KaufmanJSullivanGMYangJ, et al.Quantification of the serotonin 1A receptor using PET: identification of a potential biomarker of major depression in males.Neuropsychopharmacology2015;
40: 1692–1699.McGrathCLKelleyMEHoltzheimerPE, et al.Toward a neuroimaging treatment selection biomarker for major depressive disorder.JAMA Psychiatry2013;
70: 821–829.BartlettEAAnanthMRossanoS, et al.Quantification of positron emission tomography data using simultaneous estimation of the input function: validation with venous blood and replication of clinical studies.Mol Imaging Biol2019;
21: 926–934.PillaiRLIZhangMYangJ, et al.Will imaging individual raphe nuclei in males with major depressive disorder enhance diagnostic sensitivity and specificity?Depression Anxiety2018;
35: 411–420.PillaiRLZhangMYangJ, et al.Molecular connectivity disruptions in males with major depressive disorder.J Cereb Blood Flow Metab2019;
39: 1623–1634.
2020-20
Parametric mapping of [11C]PBR28 brain PET imaging using spectral analysis (#112)
Marcello Tuosto1, Tiago R Marques2, Oliver D Howes2, Paolo Zanotti-Fregonara3, Federico E Turkheimer4, Alessandra Bertoldo1 and Mattia Veronese4
1Department of information Engineering, Padova University, Padova, Italy
2Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London, UK
3Houston Methodist Hospital, Houston, TX, USA
4Department of Neuroimaging, King’s College London, London, UK
Abstract
Introduction: The aim of this study was to investigate the use of Spectral Analysis (SA) for voxelwise analysis of [11C]PBR28 brain PET imaging studies. As any other TSPO PET imaging ligand, [11C]PBR28 quantification is methodologically complicated by the heterogeneity of TSPO expression and its cell-dependent modulation during neuroinflammatory response.1 Compartmental models to account for this complexity exist, but they are inapplicable at the high noise of voxel data. On the contrary SA is noise-robust and provides useful information about tracer kinetics with a free-compartmental structure.
Methods: SA impulse response function calculated at 90 minutes after tracer injection (IRF90) was used as proxy of distribution volume (VT) [Fan et al, EJNMMI 2018]. To test IRF90-VT consistency, 21 dynamic [11C]PBR28 brain PET scans (age:38 ± 16 years, 16/5 males/females, 17/4 HABS/MABS, all healthy controls) from a previous study2 were reanalysed with SA and compartmental modelling (both standard 2TCM and 2TCM-1K with endothelial modelling). A second group of 8 patients with psychosis from a [11C]PBR28 PET blocking study with XBD173 [Veronese et al, JCBFM 2017] was reanalysed to test IRF90 specificity to TSPO availability.
Results: Regional averages of voxelwise IRF90 estimates were strongly associated with region of interest VT estimates provided by 2TCM-1K (r = 0.86 ± 0.11) but less with those obtained with standard 2TCM (r = 0.76 ± 0.32) (Figure 1). Distribution of spectral components were also more consistent with endothelial TSPO modelling in 89% of the ROIs as compared to 2TCM. SA-IRF90 parametric mapping proved to have great spatial quality and to be sensitive to both TSPO genotype (HABs vs MABs mean relative differences = 25%) and TSPO availability (mean signal displacement after 90 mg oral administration of XBD173 = 39%) (Figure 2).
Conclusion: SA-IRF90 can be used for voxelwise quantification of [11C]PBR28 PET data as it provides high quality parametric maps, is sensitive to TSPO availability and individual TSPO genotype, and returns information of tissue tracer kinetics comparable to compartmental modelling analysis. Specifically, the inclusion of a slow irreversible component seems to be a frequent solution (both at ROI and voxel level), in agreement with the modelling provided by 2TCM-1K.
Acknowledgements
NIHR Maudsley BRC.
References
Turkheimer FE, Rizzo G, Bloomfield PS, et al. The methodology of TSPO imaging with positron emission tomography. Biochem Soc Trans 2015; 43(4): 586–592.Bloomfield PS, Selvaraj S, Veronese M, et al, Microglial Activity in People at Ultra High Risk of Psychosis and in Schizophrenia: An [(11)C]PBR28 PET Brain Imaging Study. Am J Psychiatry 2016; 173(1): 44–52.
2020-21
Test-retest and α2-receptor challenge with [11C]yohimbine with search for a suitable reference region (#113)
Bénédicte Ballanger1, Chloé Laurencin1, 2, Sophie Lancelot1, 3, Florent Gobert2, Thibault Iecker3, François Liger3, Frédéric Bonnefoi3, Zacharie Irace34, Inés Mérida3, Jérôme Redouté3, Justine Debatisse4, 5, Didier Le Bars3 and Nicolas Costes3
1Lyon Neuroscience Research Center, CRNL, INSERM, CNRS, Université Claude Bernard, Lyon, France
2Hospices Civils de Lyon, Lyon University Hospital, Lyon, France
3CERMEP – Life Imaging, Lyon, France
4Siemens-Healthcare, SAS, Saint-Denis, France
5CarMeN Laboratory, INSERM, INRA, INSA Lyon, Université Claude Bernard, Lyon, France
Abstract
Introduction: Previous work1 introduced the [11C]yohimbine as a suitable ligand for positron emission tomography (PET) of central a2-adrenoreceptors (a2-Ars). The corpus callosum (CC) was an acceptable reference region. However, reproducibility of non-displaceable binding potential (BPnd) in healthy humans estimated with a simplified modelling method with reference region, as well as displaceability with an a2-ARs agonist was not evaluated. In this work, we performed a human investigation with dynamic PET including test-retest reliability study, a a2-ARs challenge, and search for a convenient reference regions.
Methods: Twelve healthy humans underwent two [11C]yohimbine investigations in a 90-minute dynamic acquisition performed on a PET-MRI scanner. Seven had arterial blood sampling with metabolite assessment and plasmatic Yohimbine free fraction evaluation at the first scan. The second scan was a simple retest for 5 subjects, and for 7 subjects, a scan performed 30-min after oral administration of clonidine for occupancy evaluation.2 With the arterial input function (AIF), distribution volumes (DV) were estimated with 1TC, 2TC direct fitting, and with Logan plot (LP) and SRTM methods in various cortical and subcortical brain regions, and in white matter (WM) area. Several regions: cerebellum (CER), vermis, cerebellum WM (CERWM), frontal WM (FLWM), caudate nucleus and CC) were considered as reference region for DV ratio (DVR) and BPnd estimation with LP and SRTM. Test-retest reliability parameters (bias and ICC) and occupancy parameters were evaluated.
Results: Bias of DVR estimation with LP and SRTM, relative to reference AIF-1TC modeling, showed that CERWM had the lowest biais (<7%). Test-retest showed that the LP with CC is highly reproducible (mean ICC > 0.87) but with a slight bias (11%), whereas CERWM has lower bias (7%), and an acceptable ICC (O.6) in cortical and subcortical regions. Similar conclusion but with lower performance were found with SRTM. 61% occupancy was found in cortical regions with CERWM as reference, whereas CC showed less accurate.
Conclusion: Test-retest reliability study confirmed that the CC is an good reference region, reproducible and able to evidence moderate occupancy with concurrent drug on a2-ARs. Alternatively, cerebellum white matter is a convenient reference region when measurements in the CC would not be reliable for structural or lesional reasons.
Acknowledgements
This work was supported by the french national ‘invest for the futur’ programs (LILI – Lyon Integrated Life Imaging: hybrid MR-PET ANR-11-EQPX-0026) and funded by the French "Agence Nationale pour la Recherche" (ANR -JCJC-2017-2021). ZI and JD are partially supported by Siemens-Heathcare SAS.
References
NahimiAJakobsenSMunkOL, et al.Mapping α2 adrenoceptors of the human brain with 11C-yohimbine. J Nucl Med2015;
56: 392–398.TakanoAVarroneAGulyásB, et al.Guidelines to PET measurements of the target occupancy in the brain for drug development. Eur J Nucl Med Mol Imaging2016;
43: 2255–2262.
2020-22
Design of a clinically applicable bolus-plus-constant-infusion PET imaging scheme for gold standard quantification of amyloid-beta in Alzheimer’s disease (#133)
Julie Ottoy1, Min-Su Kang2, Jeroen Verhaeghe1, Sigrid Stroobants1, Jenna Stevenson2, Nesrine Rahmouni2, Jean-Paul Soucy3, Serge Gauthier2, 3, Daniel Chartrand3, Pedro Rosa-Neto2, 3 and Steven Staelens1
1Molecular Imaging Center Antwerp, University of Antwerp, Antwerpen, Belgium
2The McGill University Research Centre for Studies in Aging, McGill University, Montréal, Québec, Canada
3MNI, McGill University, Montréal, Québec, Canada
Abstract
Introduction: The standardized uptake value ratio (SUVR) has been widely applied in anti-Aβ trials of Alzheimer’s disease. However, the method is prone to bias from variation in plasma clearance and cerebral blood flow. For more accurate Aβ quantification, the distribution volume ratio (DVR) could be applied, but requires long dynamic scanning. In addition, both SUVR and DVR require selection of a reference region. Here, we suggest an alternative tissue-to-plasma ratio method based on bolus-plus-constant-infusion (B/I) PET.
Methods: Five PSEN1 carriers (MMSE 26.4 ± 3) underwent dynamic 90-min [18F]-AV45-PET with bolus followed by continuous tracer infusion. 90-min continuous sampling as well as 10 arterial and 6 venous samples were taken to determine individual metabolite-corrected plasma inputs and calculation of the ‘gold standard’ VT,2TCM and graphical VT,Logan. Plasma free fraction (Fp) was determind in arterial and venous blood. Additionally, VT,B/I was determined as the equilibrium ratio (55–75min) of tracer concentration in tissue to the average of 3 metabolite-corrected arterial or venous plasma samples (Ctissue/Cplasma = VT,B/I). Last, DVR was calculated using cerebellar grey (CB-DVRB/I), subcortical white matter (WM-DVRB/I), or pons (pons-DVRB/I) as the reference and compared to DVR2TCM.
Results: After 55min, the B/I method demonstrated true equilibrium with low rate of [18F]-AV45 concentration change (e.g., Cparietal 2.0 ± 1.4%, CCB 3.1 ± 3.4%, and Cplasma 5.6 ± 3.5% change/20min). The arterial parent fraction stabilized at 31 ± 2% and VT,B/I_arterial corresponded well to the full-kinetic VT,2TCM and VT,Logan (average cortical difference: 2.9 ± 0.8 and 3.3 ± 0.7%, respectively) (Figure 1). Tracer concentractions remained 32 ± 10% lower in venous compared to arterial plasma. Fp (9 ± 2%) did not differ between arterial and venous blood. When using tissue reference, CB-DVRB/I underestimated true Aβ load possibly due to the presence of cerebellar Aβ (Figure 2). In contrast, WM-DVR and pons-DVR were positively correlated to the full-kinetic VT (Figure 2). The WM-DVRB/I showed +6.2 ± 1.3 and the pons-DVRB/I +6.9 ± 1.2% difference compared to their DVR2TCM.
Conclusion: 20-min static B/I-PET in combination with few arterial samples is sufficient to reproduce the full-kinetic VT. Further investigation is needed to understand arterial-venous [18F]-AV45 concentration differences to replace arterial by venous samples. Additionally, in the presence of a suitable reference region, B/I-PET can be applied to calculate DVR using a static scan and without the need for blood data.
2020-23
Validation of longitudinal [18F]FEPPA-PET in rat subcortical stroke revealed that TSPO misses chronic remote white matter inflammation (#134)
Nassir Al-Khishman1, 2, Austyn Roseborough3, Matthew S Fox2, Qi Qi1, 2, Alexander Levit3, Vladimir Hachinski4, Whitehead N Shawn2 and Jonathan D Thiessen1, 2
1Medical Biophysics, University of Western Ontario, London, ON, Canada
2Lawson Health Research Institute, London, ON, Canada
3Anatomy and Cell Biology, University of Western Ontario, London, ON, Canada
4Clinical Neurological Sciences, University of Western Ontario, London, ON, Canada
Abstract
Introduction: Soon after their first stroke, 1 in 10 people will develop dementia.1 Cognitive deficits develop alongside inflammation that subsides in the infarct but persists in remote white matter after ischemic subcortical stroke according to [11C]PK11195.2 Our group showed that in rats, ischemic subcortical stroke leads to proinflammatory microglia in remote white matter by day-28 and cognitive decline by month-3.3,4 In this study, our objectives were to (i) spatiotemporally quantify remote WM inflammation using [18F]FEPPA TSPO-PET and MRI and (ii) validate the cerebellum as a reference region for quantifying [18F]FEPPA uptake.
Methods: To induce ischemic subcortical stroke, the right dorsal striatum was injected with the vasoconstrictor endothelin-1 (n = 10) or saline (n = 13). Out of these, 5 stroke and 6 saline rats underwent in vivo imaging at baseline then post-stroke at day-7 and day-28. Dynamic [18F]FEPPA PET was acquired to quantify inflammation (Siemens Inveon). Rats were transferred to a 3T MRI (Siemens Biograph mMR) with a dedicated rat brain RF coil (Cubresa) to acquire anatomical MRI. T2-weighted MRI was used to delineate the infarct, corpus collosum (CC), and reference regions used to calculate uptake ratios (UR): cerebellum, contralateral region to the infarct, and periaqueductal gray. The in vivo imaging rats were sacrificed at day-28 and the remaining rats at day-7 for immunohistochemistry to quantify TSPO, activated astrocytes using GFAP, and proinflammatory microglia using iNOS and OX6.
Results: UR trends were similar between reference regions, but using the cerebellum correlated best with TSPO immunohistochemistry analysis. Infarct [18F]FEPPA UR in the stroke group peaked at day-7 and slightly dropped by day-28 (Figure 1). OX6, iNOS, and GFAP increased from day-7 to day-28. As for remote WM, the contralateral CC showed day-28 stroke-induced inflammation only in terms of OX6 (Figure 2).
Conclusion: We validated the cerebellum as a reference region for [18F]FEPPA PET. The opposing trends of activated astrocytes and proinflammatory microglia suggest that their contribution to stroke-induced TSPO is small. Interestingly, OX6 detected a unique type of chronic microglial activation in remote white matter. Overall, this study suggests that a whole-brain spatiotemporal quantification of inflammation requires the development of in vivo protocols that can quantify multiple inflammatory markers.
Acknowledgements
We would like to thank Lynn Wang, Lise Desjardins, Jennifer Hadway, Kevin Nishimura, and Alec Pencz for their technical help.
Immunohistochemistry of TSPO shows only an increase in TSPO at day-7 post stroke of the ipsilateral corpus collosum (CC). Another protein, OX6, detects day-28 inflammation in remote regions such as the contralateral CC.
Uptake ratio maps (to cerebellum) of [18F]FEPPA show a response in stroke rats only. The response is focal to the infarct, peaking post-stroke at day-7 then slightly dropping by day-28.
References
PendleburySTRothwellPM.
Prevalence, incidence, and factors associated with pre-stroke and post-stroke dementia: a systematic review and meta-analysis. Lancet Neurol2009;
8: 1006–1018.ThielARadlinskaBAPaquetteC, et al.The temporal dynamics of poststroke neuroinflammation: a longitudinal diffusion tensor imaging-guided PET study with 11C-PK11195 in acute subcortical stroke. J Nucl Med Off Publ Soc Nucl Med2010;
51: 1404–1412.WeishauptNZhangADezielRA, et al.Prefrontal ischemia in the rat leads to secondary damage and inflammation in remote gray and white matter regions. Front Neurosci2016;
10: 81.LevitARegisAMGarabonJR, et al.Behavioural inflexibility in a comorbid rat model of striatal ischemic injury and mutant hAPP overexpression. Behav Brain Res2017;
333: 267–275.
2020-24
Strength in numbers: Multilevel modelling of time activity curves (#140)
Granville J Matheson1, 2, Yakuan Chen3, 4, Francesca Zanderigo1, 5 and R Todd Ogden1, 3
1Molecular Imaging and Neuropathology Division, New York State Psychiatric Department, Columbia University, New York, NY, USA
2Department of Clinical Neuroscience, Centre for Psychiatry Research, Karolinska Institutet, Stockholm, Sweden
3Department of Biostatistics, Mailman School of Public Health, Columbia University, New York, NY, USA
4AT&T Services, Inc., Middletown, NJ, USA
5Department of Psychiatry, Columbia University, New York, NY, USA
Abstract
Introduction: The predominant approach for quantifying PET radiotracer influx or binding involves application of pharmacokinetic models to time activity curve (TAC) data from each subject separately, to derive estimates of binding that are subsequently compared between individuals or groups. However, this approach is both inefficient and imprecise, as all insights learned by the model when applied to a TAC from one subject are effectively “forgotten” when presented with the next. Multilevel modelling, in contrast, exploits similarities between data from different individuals, leading to improved estimates and predictions.1
Methods: We have developed and applied a multilevel two-tissue compartmental model, in which both estimation of kinetic parameters and group-level statistical comparisons are performed simultaneously within one large model. This approach leverages similarities between individuals to adaptively regularise parameter fitting and thereby improve individual estimates. Simulations were performed by generating model data and adding noise.2 We have also extended the work of Chen et al.,2 which used maximum likelihood estimation, to a Bayesian model using Hamiltonian Markov Chain Monte Carlo. We demonstrate the performance of this approach using a [11C]WAY100635 dataset of 97 individuals, using TACs derived from the midbrain.
Results: Based on our simulations, we found that a multilevel approach was more accurate than the conventional approach, providing equal statistical power for effect sizes of as little as a quarter.2 All approaches produced virtually identical fits; however the variance of estimated parameters was greatly reduced using multilevel approaches (Figure 1), and standardised effect sizes for patient-control comparisons were increased by between 1.7 and 4.3 times for different outcome measures (Figure 2).
Conclusion: Multilevel modelling improves the accuracy and statistical sensitivity of PET studies for investigating clinical questions. By making use of existing compartmental models, this approach should theoretically be applicable, and provide more accurate analysis, for any tracer for which compartmental modelling is possible. While a Bayesian approach to fitting these models is more computationally intensive, it allows us, in ongoing work, to flexibly extend this framework to additionally model multiple regions across multiple individuals simultaneously, to allow fitting of more complex compartmental models, and to incorporate prior knowledge effectively to improve estimation further.
Acknowledgements
This research was supported by Grant 1R01EB024526 from the National Institute of Biomedical Imaging and Bioengineering, by Karolinska Institutet Travel Grant, by Psykiatrifonden (4–498/2019) and by the Paul Janssen Fellowship in Translational Neuroscience.
References
GelmanAHillJ.Data analysis using regression and multilevel hierarchical models.
New York, NY, USA:
Cambridge University Press, 2007.ChenYGoldsmithJOgdenRT.
Nonlinear mixed-effects models for PET data. IEEE Trans Biomed Eng2018;
66: 881–891.
2020-25
Decreased binding of [C-11]UCB-J PET in cognitively impaired (#142)
Alexandra DiFilippo1, Tyler Tullis1, Matthew Zammit1, Todd Barnhart2, Jonathan Engle2, Dhanabalan Murali1, 2, Grace McKinney2, Nancy Davenport2, Tobey Betthauser2, Sterling Johnson2, Barbara Bendlin2 and Bradley Christian1, 2
1Waisman Center, University of Wisconsin-Madison, Madison, WI, USA
2School of Medicine and Public Health, University of Wisconsin-Madison, Madison, WI, USA
Abstract
Introduction: [11C]UCB-J is a PET radioligand which specifically binds to synaptic vesicle protein SV2A. As synaptic loss is a hallmark of neurodegenerative diseases, PET imaging can provide for an in vivo analysis of synaptic changes and the corresponding cognitive decline in Alzheimer’s Disease (AD). The purpose of this work is to investigate changes in [11C]UCB-J uptake in subjects presenting with cognitive impairment.
Methods: [11C]UCB-J dynamic PET imaging was conducted over 70 minutes in fifteen participants recruited from the UW ADRC. Amyloid (A+/-) status was determined using [11C]PiB PET scans. The participants included eleven cognitively unimpaired (CU) (A-), one CU A+ subject, three cognitively impaired (CI) (one MCI, two AD-dementia) A- subjects. [11C]UCB-J binding was quantified using DVR values calculated using SRTM2. ROIs were taken using FreeSurfer templates acquired using individual T1w MR images.
Results: Relative to the CU (A-) subjects, decreased [11C]UCB-J DVR measures were present in all four non-control subjects. However, the CI (A-) subjects exhibited less binding than the CU (A+) subject. The largest decreases were seen in the hippocampus, where the CI (A-) and CU (A+) cases had DVRs of 0.48 ± 0.10 and 0.71, respectively compared to 0.88 ± 0.13 for the control CU (A-) subjects. The parahippocampus showed decreases in the CI (A-) cases (0.59 ± 0.14 vs 0.90 ± 0.14) but not in the CU (A+) subject (0.74).
Conclusion: For all three subjects in cognitive decline [11C]UCB-J binding shows the most reduction in the hippocampus and parahippocampus when compared to the cognitively unimpaired A- subjects. Less significant decreases in binding were also seen in these regions in the CU (A+) subject. Ongoing, longitudinal studies with larger sample sizes will investigate associations between cognitive, amyloid, and tau status with [11C]UCB-J measured synaptic density.
2020-26
Evaluation of [18F]MNI-1188: A reversible monoacylglycerol lipase (MAGL) PET radiotracer in non-human primates (#143)
Cristian Constantinescu1, Alexandra Gouasmat1, Cedric Tresse1, Roger N Gunn2, Jan Passchier2 and Vincent Carroll1
1Invicro, New Haven, CT, USA
2Invicro, London, UK
Abstract
Introduction: MAGL is a key enzyme regulating the level of 2-arachidonoylglycerol (2-AG), an endocannabinoid ligand. Inhibition of this enzyme mobilizes 2-AG and induces non-opioid stress-induced analgesia. Furthermore, inhibition of MAGL leads to selective activation of presynaptic CB1 receptors, which play a role in inhibition of epileptic seizures. A reversible MAGL PET tracer could support development of new drug candidates against this target in vivo. Herein, we report the evaluation of [18F]MNI-1188, as the first reversible MAGL-specific radioligand.
Methods: Two 180 min scans, at baseline and following homologous competition (1.0 mg/kg, 5 min iv bolus at 10 min before tracer) were conducted with the active enantiomer, [18F]MNI-1188, in rhesus monkey. The same subject received an additional scan with the inactive enantiomer, [18F]MNI-1190. All scans were performed on a microPET Focus 220 scanner with associated arterial blood and metabolite assays. Data were modeled with 1 and 2-tissue compartmental models (1TC and 2TC), and multilinear analysis (MA1) to estimate total distribution volume, VT. Occupancy and the non-displaceable volume of distribution (VND) were estimated from Lassen plot analysis.
Results: [18F]MNI-1188 entered the brain with a peak uptake of ∼3% ID at ∼ 7 min post injection (p.i.) accompanied by a clear washout and a heterogenous signal consistent with the known distribution of MAGL (Figure 1). VT ranged from ∼10 ml.cm−3 in brainstem to ∼20 ml.cm−3 in cortex. Tracer kinetics were best described with 2TC and MA1 models. Time stability analysis showed that a scan as short as 40 min produced good parameter estimates (VT within 5% of the value at 180 min). [18F]MNI-1188 underwent fast declining metabolism with ∼20% parent remaining at 30 min p.i. Plasma free fraction was ∼26%. Plasma-to-blood partition is initially low (∼0.4) and monotonically increases towards 1.0 between 0 and 90 min p.i., suggesting peripheral binding to blood cells. Homologous competition yielded ∼92% occupancy in the brain (VND∼1.2) and blocked the peripheral binding sites. The inactive enantiomer [18F]MNI-1190 quickly entered the brain and washed out rapidly with no regional selectivity (VT∼1.5).
Conclusion: [18F]MNI-1188 is a promising reversible PET radiotracer with favorable properties for imaging and quantitation of MAGL enzyme in the brain.
Acknowledgements
Authors would like to acknowledge Amy Amenta (Invicro) and the staff at Yale University PET Center, in particular Daniel Holden and Krista Fowles for their support of imaging operations.
2020-27
Early stopping in clinical PET studies: Save money and mSv! (#147)
Jonas E Svensson1, Martin Schain2, Gitte M Knudsen2, 3, R Todd Ogden4, 5 and Pontus Plavén-Sigray1, 2
1Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden
2Neurobiology Research Unit, Copenhagen University Hospital, Copenhagen, Denmark
3Institute of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark
4Department of Biostatistics, Mailman School of Public Health, Columbia University, New York, USA
5Molecular Imaging and Neuropathology Area, New York State Psychiatric Institute, New York, USA
Abstract
Introduction: Two major issues in clinical PET studies are the high costs of the examinations, and the radiation exposure to people. Clearly, it is in the interest of the PET researcher to keep the number of included subjects to a minimum, while still being able to collect enough data to make sound scientific inference. Here we evaluated a statistical framework with the aim to stop the recruitment of subjects in a clinical PET study as soon as enough data have been collected to make a conclusion, so called sequential testing.
Methods: The Bayes Factor (BF) metric, which captures the relative likelihood of one hypothesis over another, was calculated after the collection of every data point in a paired-sample design (e.g., pre- and post-scan). When the BF reached three times more support for the null-hypothesis (H0) over the alternative (H1), or vice versa, the study was stopped. Using simulations over a range of effects based on data from real clinical PET studies, we evaluated the 1) average sample size needed to reach an early stopping decision and 2) the long-running error rates of false negative and false positive evidence, i.e. declaring support for the wrong hypothesis.
Results: Compared to employing the classical null-hypothesis significance testing (NHST) procedure, i.e., the use of a fixed sample size and p-value, BF sequential testing allowed for stopping a study early (Table 1). By using “default” specifications of the one-sided BF t-test, the procedure can save on average one third of subjects and resources compared a fixed sample size design. For most effects, the false positive and negative evidence rates were both kept below 10%. However, at true small effects, the method showed a high rate of false negative evidence.
Conclusion: BF sequential testing allows researcher to save money and injected radioactivity, while still keeping error rates at acceptable levels at a range of effect sizes relevant for PET research. Importantly, studies can also be stopped early when the effect is zero or negligible. The method shows potential for saving resources that can then be spend on examining other neurobiological effects of interest instead.
Table 1. Simulation results of a pre-intervention-post study using [11C]Raclopride BPND in striatum as example. A set of Planned N was defined in order detect a range of changes in BPND, using an alpha = 0.05 and 80% power. 1000 simulations were then run for each setting, in which samples were drawn from a population having the specified effect sizes. We also used the same Planned N but simulated data where there was no change in BPND. The BF was calculated sequentially after the collection of each data point. At a BF > 3 in favour of either an effect or no effect, the study was stopped (Actual N).
2020-28
Meta-analysis of the glial marker TSPO in psychosis revisited: Reconciling inconclusive findings of patient-control differences (#148)
Pontus Plavén-Sigray1, 2, Granville J Matheson1, Jennifer M Coughlin3, 4, Sina Hafizi5, Heikki Laurikainen6, Julie Ottoy7, Livia de’Picker8, Pablo M Rusjan5, Jarmo Hietala6, Oliver D Howes9, 10, Romina Mizrahi5, Manuel Morrens8, Martin G Pomper3, 4 and Simon Cervenka1
1Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden
2Neurobiology Research Unit, Copenhagen University Hospital, Copenhagen, Denmark
3Department of Psychiatry and Behavioral Sciences, Johns Hopkins Medical Institutions, Baltimore, MD, USA
4Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins Medical Institutions, Baltimore, MD, USA
5Department of Psychiatry, University of Toronto, Toronto, Canada
6Department of Psychiatry, University of Turku and Neuropsychiatric Imaging group, Turku University Hospital, Turku, Finland
7Molecular Imaging Center Antwerp, University of Antwerp, Antwerp, Belgium
8Collaborative Antwerpen Psychiatric Research Institute (CAPRI), University of Antwerp, Antwerp, Belgium
9IoPPN, King’s College London, London, UK
10MRC London Institute of Medical Sciences, Hammersmith Hospital, London, UK
Abstract
Introduction: Positron emission tomography (PET) studies examining the glial marker translocator protein (TSPO) in schizophrenia have been inconclusive. We previously demonstrated lower TSPO in psychosis patients in an individual participant data (IPD) meta-analysis of studies using second generation TSPO radioligands.1 Subsequently, a summary-statistics meta-analysis, including one newly published study, showed no difference.2 Here, the aim was to repeat the IPD analysis including two new samples unavailable at the time of our previous analysis to reevaluate the conclusions. The primary objective was to examine the hypotheses of 1) higher or 2) lower or 3) no difference in radioligand binding between patients and healthy control subjects.
Methods: Individual participant data were obtained from PET studies that used a second generation TSPO radioligand, reported distribution volume (VT) values in brain in patients with psychosis as compared to healthy controls, and reported TSPO affinity type of all participants. The outcome measure was VT in frontal cortex (FC), temporal cortex (TC) and hippocampus (HIP). We applied linear mixed modelling and Bayes factors (BF) to examine the relative support for higher, lower, or no-change of TSPO levels in patients compared to healthy controls.
Results: Individual participant data from seven studies were included, amounting to 99 patients with first-episode psychosis or schizophrenia and 109 healthy control subjects. In all regions investigated, BF showed moderate to strong support (BF > 5) for lower VT in patients as compared to no difference, and strong support (BF > 10) for lower VT compared to higher VT in patients. Mean patient-control differences in standardized VT values are shown in Figure 1.
Conclusion: In this updated meta-analysis, we found moderate to strong support for lower TSPO in psychosis patients compared to control subjects. In vitro data has shown a lack of correspondence between TSPO and pro-inflammatory activation, also recently confirmed in a post-mortem study in schizophrenia. Hence, based on the present results no firm conclusions can be made regarding the pro-inflammatory versus anti-inflammatory status of glial cells in psychosis patients. Additional work is needed to understand the biological relevance of the observed lower TSPO in patients.
References
Plavén-SigrayPMathesonGJCollsteK, et al.Positron emission tomography studies of the glial cell marker translocator protein in patients with psychosis: a meta-analysis using individual participant data. Biol Psychiatry2018;
84: 433–442.MarquesTRAshokAHPillingerT, et al.Neuroinflammation in schizophrenia: meta-analysis of in vivo microglial imaging studies. Psychol Med2018;
49: 2186–2196.
2020-29
Bayesian partial volume correction for image derived input function (#153)
Zacharie Irace1, 2, Inés Merida1, Florent Gobert3, 4, Frédéric Dailler4 and Nicolas Costes1
1CERMEP-Life Imaging, Lyon, France
2Siemens Healthcare SAS, Saint Denis, France
3Lyon Neuroscience Research Center, Lyon, France
4Hospices Civils de Lyon, Lyon, France
Abstract
Introduction: In PET kinetic modeling, the collection of the input function is necessary for absolute quantification. The growing popularity of hybrid PET-MR devices offers new expectations towards Image Derived Input Function (IDIF) methods, to avoid invasive arterial sampling. Direct quantification with IDIF extracted from MRI-mask is impractical because of strong Partial Volume Effects (PVE). To correct for PVE, Geometric Transfer Matrix (GTM)-based methods1 consider the observed Time Activity Curves (TACs) as a weighted mixture of the unknown true TACs. The GTM method requires the inversion of a low rank matrix, which makes the method sensitive to noise and biased on small regions such as carotids. We propose a Bayesian approach for a more robust resolution of the GTM model, with the aim of estimating an IDIF.
Methods: A mask of the carotids was defined by applying an adaptive threshold to a smoothed arterial MR-angiography with connexity constraints. The volume defined by 5 mm around the carotids was considered as the background mask. The IDIF was parametrized as the first three coordinates of a Principal Components Analysis, and the background TAC with 3 parameters of an irreversible 2-tissue compartment model. The probability distribution of these 6 parameters was jointly estimated with Markov Chain Monte Carlo (MCMC) in a Bayesian context.2 The Maximum-A-Posteriori of these densities lead to an estimate of IDIF. The method was applied to 22 patients with lesioned brain that had 90-min dynamic PET [18F]FDG scan. Their Cerebral Metabolic Rate of Glucose (CMRGlu) values were obtained in 78 regions from the estimated IDIF, and from arterial sampling for comparison.
Results: Better fits of IDIF with the arterial curve were obtained with the proposed Bayesian GTM method compared to the original GTM. The average bias on CMRGlu was 25.9 ( ± 103)% for GTM and 0.5( ± 47)% for Bayesian GTM. Regression on CMRGlu also showed slope and determination coefficients closer to 1, and intercept closer to zero.
Conclusion: Bayesian framework allowed to integrate prior knowledge from kinetics modeling to guide the GTM resolution. This, combined with a MCMC sampling approach, lead to more robust and accurate IDIF estimates than with the original GTM method.
Acknowledgements
This work was supported by the french national ‘invest for the futur’ programs (LILI – Lyon Integrated Life Imaging: hybrid MR-PET ANR-11-EQPX-0026) and the Hospital University Institut CESAME (Brain and Mental Health ANR-10-IBHU-0003). ZI is partially supported by Siemens-Heathcare SAS.
Example for a specific subject
Results on all subjects
References
RoussetOGMaYEvansAC.
Correction for partial volume effects in PET: principle and validation.J Nucl Med1998;
39: 904–911.RobertCPCasellaG.Monte Carlo statistical methods.
Berlin:
Springer-Verlag, 1999.
2020-30
In vivo relationship between brain arachidonic acid incorporation and cerebral metabolic rate of glucose in the healthy brain and in bipolar disorder: A pilot study (#164)
Francesca Zanderigo1, 2, R Todd Ogden2, 3, Shankar Vallabhajosula4, Paresh Kothari4, Anastasia Nikolopoulou4, Dileep Kumar2, Stanley Rapoport5, Maria Oquendo6, John Mann1, 7 and Elizabeth Sublette1, 2
1Psychiatry, Columbia University, New York, NY, USA
2Molecular Imaging and Neuropathology Area, New York State Psychiatric Institute, New York, NY, USA
3Biostatistics, Mailman School of Public Health, Columbia University, New York, NY, USA
4Radiology, Weill Cornell Medicine, New York, NY, USA
5National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, Bethesda, MD, USA
6Psychiatry, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA
7Radiology, Columbia University, New York, NY, USA
Abstract
Introduction: Polyunsaturated fatty acids (PUFA) are major determinants of brain development and functioning, with roles in balancing pro- and anti-inflammatory cascades, regulation of cell-cell signaling, and binding to peroxisome proliferator-activated receptors (PPARs), which affect genes involved in lipid storage, beta-oxidation and glucose metabolism. We performed a pilot PET study using 1-[11C]arachidonic acid ([11C]AA), an omega-6 PUFA, and [18F]FDG to directly assess how brain arachidonic acid (AA) incorporation coefficient (K*) relates to cerebral metabolic rate of glucose (CMRglu) in vivo.
Methods: Six healthy volunteers (HVs) and 2 participants with bipolar disorder (BD) were imaged using [11C]AA and then [18F]FDG three hours later. K* and CMRglu were quantified in 26 bilateral regions of interest (ROIs). Pearson’s linear correlation coefficient (r) between CMRglu and corresponding K* values was calculated in each ROI.
Results: Representative within-subject maps of CMRglu and K* are shown (Figure 1). In HVs, CMRglu correlated inversely with K* across all ROIs, with r values ranging from -0.068 to -0.643 (in medial prefrontal cortex, mPFC). CMRglu values in the 2 BD participants were within the range observed in HVs, although their values were in the lower end of the range, and K* values were higher. Considering all subjects together, CMRglu correlated inversely with K* across all ROIs, with r values ranging from -0.445 to -0.752 (mPFC). Figure 2 shows the scatter plots in 7 ROIs where we previously observed a significant positive correlation between brain [18F]FDG activity and plasma levels of AA (esterified to the phosphotriglycerides),1 considering the HVs together with the 2 BD participants.
Conclusion: Consistent with previous animal studies concerning PUFAs and bioenergetics,2,3 our preliminary findings indicate an inverse correlation between K* and CMRglu. Possible mediators of this relationship are PPARs, which regulate genes involved in metabolic adaptation between glucose and fatty acid utilization.4,5 Although sample size is small, this is the first report of direct brain measurement of both K* and CMRglu within the same subjects. Larger studies are needed to confirm the relationship and compare bipolar and healthy groups.
Acknowledgements
The National Institute of Mental Health provided funding for this study (MH096255, PI: M. Elizabeth Sublette MD, PhD).
References
SubletteME, et al.Plasma polyunsaturated fatty acids and regional cerebral glucose metabolism in major depression. Prostaglandins Leukot Essent Fatty Acids2009;
80: 57–64.PifferiF, et al.Long-chain n-3 PUFAs from fish oil enhance resting state brain glucose utilization and reduce anxiety in an adult nonhuman primate, the grey mouse lemur. J Lipid Res2015;
56: 1511–1518.Ximenes da SilvaA, et al.Glucose transport and utilization are altered in the brain of rats deficient in n-3 polyunsaturated fatty acids. J Neurochem2002;
81: 1328–1337.NakamuraMTYudellBELoorJJ.Regulation of energy metabolism by long-chain fatty acids. Prog Lipid Res2014;
53: 124–144.HenekaMTLandrethGE.PPARs in the brain. Biochim Biophys Acta2007;
1771: 1031–1045.
2020-31
Quantification of vesicular monoamine transporter type 2 (VMAT2) occupancy with [18F]AV‑133 in non-human primate (#170)
Christine Sandiego1, Eugenii A. Rabiner2, Roger N Gunn2, Richard Wong3, Gordon Loewen3, Dietrich Haubenberger3 and Ryan Terry-Lorenzo3
1Invicro, New Haven, CT, USA
2Invicro, London, UK
3Neurocrine Biosciences, San Diego, CA, USA
Abstract
Introduction: Vesicular monoamine transporter 2 (VMAT2) has been implicated in a broad range of CNS disorders. [18F]AV-133 (aka [18F]FP-DTBZ) is widely used to study VMAT2 in humans, with a suitable imaging range between 90–120 min post injection for analysis with SUVr.1 NBI compound (DD-139) is a selective VMAT2 inhibitor. The aim of this study was to examine the relationship of DD-139 plasma concentration with VMAT2 occupancy, measured with [18F]AV‑133, and to compare dynamic with static quantification in non-human primate (NHP) brain.
Methods: Three cynomolgus macaques were imaged over 120 min with [18F]AV-133 (3.3 ± 0.5 mCi) in a Focus-220 PET camera at baseline and 60 min post intravenous infusion of DD-139 (0.026, 0.084, 0.26, 0.52 or 0.78 mg/kg) that continued throughout the scan. During each DD-139 scan, blood samples were collected to measure plasma concentration. [18F]AV-133 images were normalized into cynomolgus MR atlas space and time-activity curves were generated for brain regions: caudate, putamen, and cerebellum (reference region). Binding potential (BPND) was estimated with non-invasive Logan graphical analysis (t* = 35 min) using the full kinetic data. Using 90–120 min, target to reference tissue activity concentration, CT/CREF-1, was computed. Occupancy of DD-139 was calculated in the caudate and putamen (i.e., striatum) using BPND and CT/CREF-1. The relationship between DD-139 occupancy (averaged between caudate and putamen) and plasma concentration (averaged between 0–120 min), EC50, was determined with the Emax model (Emax = 100, slope = 1).
Results: DD-139 demonstrated dose-dependent occupancy of the VMAT2 in the striatum, as measured with [18F]AV-133 (Figure 1). BPND and CT/CREF-1 were related by BPND = 1.12*(CT/CREF-1) – 0.23, R2 = 0.97. Occupancy using BPND and CT/CREF-1 predicted a similar EC50, 11.0 ng/mL and 12.6 ng/mL, respectively (Figure 2).
Conclusion: The relationship of VMAT2 occupancy and DD-139 plasma concentration were measured with [18F]AV-133 in NHP brain. Estimation of occupancy and EC50 were very similar between dynamic or static quantification. This may be useful for studying therapeutic agents in patients who cannot withstand 120 min of imaging.
Acknowledgements
We would like to thank the staff at the Yale PET Center for carrying out the NHP studies and the Discovery Team at Invicro, New Haven, for providing radiochemistry and study coordination.
Reference
LinKJLinWYHsiehCJ, et al.Optimal scanning time window for 18F-FP-(+)-DTBZ (18F-AV-133) summed uptake measurements. Nucl Med Biol2011;
38: 1149–1155.
2020-32
Full quantification of brain glucose metabolism using a portable positron emission tomography (PET) camera: A preliminary report (#171)
Elizabeth A Bartlett1, 2, Mohammad Lesanpezeshki1, Ragy Girgis2, David Beylin3, Morgan Cambareri1, R Todd Ogden1, 4, John Mann1, 5, Jeffrey Miller1, 2 and Francesca Zanderigo1, 2
1Molecular Imaging and Neuropathology Area, New York State Psychiatric Institute, New York, NY, USA
2Department of Psychiatry, Columbia University Medical Center, New York, NY, USA
3Brain Biosciencies, Inc., Rockville, MD, USA
4Department of Biostatistics, Mailman School of Public Health, Columbia University, New York, NY, USA
5Department of Radiology, Columbia University Medical Center, New York, NY, USA
Abstract
Introduction: PET quantifies components of neurotransmitter systems, and incorporation and metabolism of specific compounds, in the living brain. Current PET scanners, however, are expensive and non-portable.1 New portable, less expensive PET cameras,2,3 including CerePET™ by Brain Biosciences, Inc. (Rockville, MD),4,5 can radically expand the realm of possible PET applications. We are in Year 2 of an NIH-funded study that is gathering PET and arterial blood data in 20 healthy controls (HCs) imaged with [18F]FDG using both the standard Siemens Biograph™ and portable CerePET™ devices. The data will allow for the first within-subject comparison of images acquired both with a traditional and a portable scanner. Here we report a preliminary comparison.
Methods: Thus far, 8 HCs have undergone dynamic [18F]FDG imaging with both scanners on 2 separate days (range of days between imaging sessions: 1 to 154) and concurrent arterial blood sampling. The net influx rate (Ki) and corresponding cerebral metabolic rate of glucose (CMRglu) were quantified in 24 bilateral regions of interest (ROIs) across the brain and at the voxel level, using the two-tissue irreversible compartment model (2TCIRR) and Patlak approach. Within-subject ROI-level estimates of Kiand CMRglu were compared across scanners using Pearson’s linear correlation coefficient (r) and regression analysis. Results are summarized across ROIs for each subject.
Results: Within-subject correlation of CerePET™ estimates with corresponding Biograph™ estimates is high regardless of outcome measure (CMRglu or Ki) and quantification approach (r range: 0.781 to 0.983). Regression slopes between scanners range from 0.337 to 0.781 (CMRglu, 2TCIRR), 0.301 to 0.802 (CMRglu, Patlak; Figure 1), 0.310 to 0.651 (Ki, 2TCIRR) and 0.278 to 0.665 (Ki, Patlak). The largest subject-level between-scanner bias is observed in the individual with the highest number of days between imaging sessions. CMRglu maps for a representative subject are reported in Figure 2.
Conclusion: Our preliminary comparison indicates good correspondence between brain glucose metabolism measurements obtained via portable CerePET™ and traditional Biograph™ scanning. We anticipate the agreement will increase as we harmonize reconstruction, pre-processing (e.g., motion correction), and attenuation-correction across scanners. The data will also be used for validation of approaches for noninvasive quantification of the Kiof tracers with irreversible kinetics.
Acknowledgements
The National Institute of Biomedical Imaging and Bioengineering provided funding for this study (R01-EB026481, PI: Francesca Zanderigo, PhD).
Comparison of cerebral metabolic rate of glucose (CMRglu) maps across the CerePET and Biograph
Voxel-wise glucose cerebral metabolic rate (CMRglu) maps shown in Montreal Neurological Institute (MNI) space for a representative subject for the non-portable, Biograph™ scanner (TOP) and the portable, CerePET™ scanner (BOTTOM).
Comparison of [18F]FDG glucose cerebral metabolic rate across the CerePET and Biograph
Glucose cerebral metabolic rate (CMRglu) is plotted for the portable CerePET™ scanner (y-axis) and for the non-portable Biograph™ scanner (x-axis). CMRglu was estimated via the Patlak approach. The regression lines are plotted for each subject with dotted lines and identity is shown with a solid black line.
References
JonesTTownsendD.
History and future technical innovation in positron emission tomography. J Med Imaging2017;
4: 011013.MelroyS, et al.Development and design of next-generation head-mounted ambulatory microdose positron-emission tomography (AM-PET) system.Sensors2017;
17: 1164.SchulzD, et al.Simultaneous assessment of rodent behavior and neurochemistry using a miniature positron emission tomograph. Nat Methods2011;
8: 347–352.SprietM, et al.(18) F-sodium fluoride positron emission tomography of the equine distal limb: exploratory study in three horses.Equine Vet2018;
50: 125–132.SprietM, et al.Positron emission tomography of the equine distal limb: exploratory study.Vet Radiol Ultrasound2016;
57: 630–638.
2020-33
Simultaneous assessment of α4β2 nicotinic acetylcholine receptor (nAChR) availability and neuronal response to rewarding food-cues in human obesity using PET-MRI (#172)
Swen Hesse1, 2, Michael Rullmann1, 2, Georg A Becker1, Julia Luthardt1, Eva Schweickert-de Palma2, Tilman Guennewig2, Franziska Zientek1, 2, Thies Jochimsen1, Matthias Blüher3, 4, Anja Landsmann2, Sarah Martin2, Philipp M Meyer1, Marianne Patt1, Anja Hilbert2, 5, Jane Neumann6, 7, Peter Brust8 and Osama Sabri1
1Department of Nuclear Medicine, University of Leipzig, Leipzig, Saxony, Germany
2Integrated Research and Treatment Center (IFB) Adiposity Diseases, University of Leipzig, Leipzig, Saxony, Germany
3Helmholtz Zentrum München, Helmholtz Institute for Metabolic, Obesity and Vascular Research (HI-MAG), Leipzig, Germany
4Department of Internal Medicine, University of Leipzig, Leipzig, Saxony, Germany
5Department of Psychosomatic Medicine and Psychotherapy, University of Leipzig, Leipzig, Saxony, Germany
6Department of Medical Engineering and Biotechnology, Ernst-Abbe-Hochschule, University of Applied Sciences, Jena Thuringia, Germany
7Department of Neurology, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Saxony, Germany
8Department of Neuroradiopharmaceuticals, Helmholtz-Zentrum Dresden-Rossendorf, Leipzig, Saxony, Germany
Abstract
Introduction: Cholinergic modulation of brain reward and attentional networks appears to play a crucial role in information processing about salience as a key biological mechanism in obesity (OB). We investigated the α4β2-nAChRs availability and neuronal network activity applying PET-MRI and (-)-[18F]flubatine in individuals with OB and normal-weight controls (NW) at rest and in response to salient food cues with focus on the ventral tegmental area (VTA) and the thalamus (THAL).
Methods: Twenty-six individuals with OB (n = 11; 7♀; age 42 ± 15 yrs; BMI 37 ± 3 kg/m2) and NW (n = 15; 12♀; 28 ± 7 yrs; BMI 22 ± 2 kg/m2) underwent PET-MRI with (-)-[18F]flubatine twice on separate days (rest and stim) using a bolus-infusion protocol (298 ± 6 MBq) with list-mode acquisition 0–60 min and 120–165 min p.i. paralleled by MPRAGE and EPI sequence. Total distribution volumes VT (mL/cm3) were estimated as the ratio between mean (−)-[18F]flubatine activity in tissue between 120 and 165 min and free parent (−)-[18F]flubatine in plasma obtained from venous blood. Food pictures (stim) were shown 120–135 min p.i.
Results: Under rest, VT VTA tend lower in OB vs NW (16.1 ± 2.1 vs. 17.5 ± 2.2; p = 0.1). During stim, VT VTA as well as VT THAL tend higher in OB (17.0 ± 1.7, p = 0.3; 28.3 ± 3.5 vs. rest 25.9 ± 3.2, p = 0.2) but not in NW (17.7 ± 2.6, p = 0.9; 26.3 ± 3.4 vs. 26.2 ± 2.5, p = 0.9). Eigenvector centrality mapping revealed higher centrality (puncorr < 0.001) of the insula as a core node of the salience network and the nucleus accumbens in NW but not in OB, in fronto-parietal attentional/executive networks in both (NW > OB) under stim vs rest. Applying VT VTA as covariate in functional connectivity analyses using VTA as a seed markedly weakened the strong connectivity between VTA and the orbito-frontal cortex (OFC) (see Figure 1). Using VT THAL as seed, strongest connectivity was found with visual input areas of the temporal cortex.
Conclusion: Compared with NW, individuals with OB displayed different neuronal presentation in the control over attention to high-incentive, salient cues apparently modulated by changes of the mesolimbic and cortico-thalamic cholinergic system. Most interestingly, the number of available α4β2-nAChR binding sites in the VTA has an impact on the function of the OFC which is the most important region for providing a value judgement of food.
Acknowledgements
The work was supported by the Federal Ministry of Education and Research, Germany, FKZ: 01E01001.
Figure 1. Functional connectivity analysis. Seed-based (SPM) analysis indicating the effect of VT VTA on functional connectivity between rest and stim (food cues) study (OFC, orbito-frontal cortex; VTA, ventral tegmental area).
2020-34
Determination of the 5-HT2C receptor fraction in the human hippocampus in vivo: A [11C]Cimbi-36 PET study (#173)
Gaia Rizzo1, 2, Graham E Searle1, Jan Passchier1, Yvonne Lewis1, David Erritzoe2, Roger N Gunn1, 2, Gitte M Knudsen3, John D Beaver4 and Eugenii A Rabiner1, 5
1Invicro LLC, London, UK
2Imperial College London, London, UK
3University of Copenhagen, Copenhagen, Denmark
4Biogen, Cambridge, MA, USA
5King’s College London, London, UK
Abstract
Introduction: Evaluation of drug binding at the brain 5HT2C receptors (5HT2CR) in vivo has been hampered by the lack of suitably selective PET ligands. [11C]CIMBI-36 has similar affinity for the 5HT2A receptor and 5HT2CR in vitro.1 Non-human primate PET studies estimated that 5HT2CR represents 30% to 70% of [11C]Cimbi-36 specific binding in the hippocampus.2 In the human brain 5-HT2C mRNA expression is high in the choroid plexus and hippocampus.3 We evaluated the 5HT2CR fraction in the human hippocampus, by quantifying the 5-HT2A and 5-HT2C components of the [11C]Cimbi-36 PET signal after the administration of risperidone, an antipsychotic with 50–100 fold higher affinity for the 5-HT2A than for the 5HT2CR in vitro.
Methods: Two 90-min dynamic [11C]Cimbi-36 PET scans, at baseline and 2 hours post-administration of a single oral dose of risperidone (0.5, 1 or 2 mg), were administered to seven healthy males. All image analysis was performed using MIAKAT™. Regional volumes of distribution were derived using MA1 (t* = 20 min) and binding potential was calculated using the grey-masked ventrolateral cerebellum as a reference region.
Under the assumption that the specific binding in the cortex constitutes almost entirely of 5-HT2A binding and that risperidone does not bind to 5HT2CR, cortical occupancy (including frontal, parietal and occipital cortices, OccCORT) and risperidone concentration in plasma (CRISP) were used to estimate E50-HT2ARISP (risperidone concentration that leads to 50% occupancy of the 5-HT2A) as:
OCCCORT = CRISP/(CRISP+E50-5HT2ARISP)
Occupancy values in the hippocampus (OccHIPP) were used to estimate the f2C_HIPP (fraction of [11C]Cimbi-36 signal in the hippocampus attributable to the 5-HT2C) as:
OCCHIPP = (1-f2C_HIPP) CRISP/(CRISP+E50-5HT2ARISP)Amygdala (OccAMY) was used as a control region, similar in size and anatomical location to the hippocampus, but with low expected 5-HT2C expression.
Results: OccCORT was significantly higher than OccHIPP (p < 0.01, paired Student’s t-test, Figure 1), but not OccAMY. The estimated f2C_HIPP was 0.50 (95% CI, 0.24–0.76), while the estimated E50-HT2ARISP was 2.51 ng/ml (95% CI, 1.40–3.63 ng/ml) (Figure 2).
Conclusion: 1. The 5HT2CR represents between 25% and 75% of the total [11C]Cimbi-36 hippocampus BPND
2. [11C]Cimbi-36 may be useful in evaluating drug interaction with the 5HT2CR in the human brain.
References
EttrupAHansenMSantiniMA, et al.Radiosynthesis and in vivo evaluation of a series of substituted 11C-phenethylamines as 5-HT (2A) agonist PET tracers.Eur J Nucl Med Mol Imaging2011;
38: 681–693.FinnemaSJStepanovVEttrupA, et al.Characterization of [(11)C]Cimbi-36 as an agonist PET radioligand for the 5-HT(2A) and 5-HT(2C) receptors in the nonhuman primate brain. Neuroimage2014;
84: 342–353.PandeyGNDwivediYRenX, et al.Regional distribution and relative abundance of serotonin(2c) receptors in human brain: effect of suicide.Neurochem Res2006;
31: 167–176.
2020-35
Magnitude of translocator protein binding and its association with C-reactive protein in depression: An 11C-PK11195 PET study (#174)
Julia J Schubert1, Mattia Veronese1, Tim D Fryer2, 3, Roido Manavaki4, Manfred G Kitzbichler5, Edward T Bullmore5, 6, Federico E Turkheimer1 and NIMA Consortium (https://www.neuroimmunology.org.uk/)
1Neuroimaging, King’s College London, London, UK
2Clinical Neuroscience, University of Cambridge, Cambridge, UK
3Wolfson Brain Imaging Centre, University of Cambridge, Cambridge, UK
4Radiology, University of Cambridge, Cambridge, UK
5Psychiatry, University of Cambridge, Cambridge, UK
6Cambridgeshire and Peterborough NHS Foundation Trust, Cambridge, UK
Abstract
Introduction: Inflammation has previously been implicated in the pathophysiology of major depressive disorder.1,2 A relationship between peripheral and central inflammation has been proposed, in which peripheral cytokines directly induce central immune activation.3,4 The aims of the current work are to use 11C-PK11195 positron emission tomography (PET) to investigate whether translocator protein (TSPO) binding, as a measure for neuroinflammation, in anterior cingulate (ACC), prefrontal (PFC) and insular (INS) cortical regions is 1) significantly increased in depression and 2) associated with C-reactive protein (CRP), as a biomarker for peripheral inflammation.
Methods: 51 depressed cases (DCs) and 25 healthy controls (HCs) underwent 60-minute dynamic 11C-PK11195 PET. A venous blood sample was collected to assess CRP. DC was stratified into high CRP (>3mg/L; N = 20) and low CRP (<3mg/L; N = 31) subgroups. Binding potentials (BPND) for ACC, PFC, and INS regions of interest (ROIs) were calculated using a simplified reference tissue model with a supervised clustering reference region approach.5
Results: Analysis of variance revealed significant BPND case-control differences across ACC, PFC, and INS ROIs (F(1,72) = 5.59, P = 0.02). Significantly higher BPND was observed in ACC of DC compared to HC (d = 0.49; t(74) = 2.00, P = 0.03), which remained significant when corrected for age and sex (F(1,72) = 6.60, P = 0.01) (Figure 1(a)). Significantly higher BPND was observed in ACC of low CRP DC compared to HC (d = 0.53; t(54) = 1.96, P = 0.03) (Figure 1(b)). No other significant case-control differences were observed. No significant correlations were observed between BPND and CRP in ACC, PFC, or INS ROIs.
Conclusion: We successfully replicated previous results of increased TSPO binding in DC compared to HC and provide further support for a relationship between neuroinflammation and depression, although the effect size of case-control differences in the TSPO PET signal is small (Figure 2). TSPO PET signal is not correlated with peripheral CRP concentrations in this cohort. These results encourage future, more mechanistic studies that will require novel PET radioligands with specific brain immune targets.
Acknowledgements
We acknowledge the Cambridgeshire and Peterborough NHS Foundation Trust, University of Cambridge, Wellcome Trust, National Institute of Health Research (NIHR) Clinical Research Network: Kent, Surrey and Sussex & Eastern, NIHR Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London, and NIHR Cambridge Biomedical Research Centre. We gratefully thank all study participants, research teams and laboratory staff.
References
HolmesSEHinzRConenS, et al.Elevated translocator protein in anterior cingulate in major depression and a role for inflammation in suicidal thinking: a positron emission tomography study. Biol Psychiatry2017;
83: 61–69.Köhler-ForsbergOButtenschønHNTanseyKE, et al.Association between C-reactive protein (CRP) with depression symptom severity and specific depressive symptoms in major depression. Brain Behav Immun2017;
62: 344–350.DantzerRConnorJCOFreundGG, et al.From inflammation to sickness and depression: when the immune system subjugates the brain.Nat Rev Neurosci2008;
9: 45–56.MyintAMKimYK.
Cytokine-serotonin interaction through IDO: A neurodegeneration hypothesis of depression. Med Hypotheses2003;
61: 519–525.TurkheimerFEEdisonPPaveseN, et al.Reference and target region modeling of [11C]-(R)-PK11195 brain studies. J Nucl Med2007;
48: 158–167.
2020-36
Novel software for computer-aided differential diagnosis of Parkinsonism using positron emission tomography (#175)
Elin Lindström1, 2, Charles Widström2, Torsten Danfors2, Mark Lubberink1, 2 and My Jonasson1, 2
1Uppsala University, Uppsala, Sweden
2Uppsala University Hospital, Uppsala, Sweden
Abstract
Introduction: The PET tracer [11C]PE2I (N-(3-iodoprop-2E-enyl)-2β-carbomethoxy-3β-(4-methylphenyl)nortropane) is a selective dopamine transporter (DAT) compound that has been successfully introduced for differential diagnosis of Parkinsonism.1 Using tracer kinetic analysis of a dynamic [11C]PE2I-PET scan, DAT availability and relative cerebral blood flow (rCBF) can be measured at the voxel level. The aim was to develop an automated workflow from dynamic PET-data to statistical comparison of the patient’s regional rCBF and DAT availability values to a normal control database, allowing for computer-aided differential diagnosis of Parkinsonism.
Methods: The automated workflow was based on previously validated and published methodology.2,3 In brief, the dynamic PET images were realigned to correct for inter-frame patient motion. A volume of interest (VOI) template in MNI space was projected onto an early sum image. rCBF images were generated from 40-min dynamic PET-data using a basis function implementation of the simplified reference tissue model and SUVR-1 was estimated for the 30–40 min interval, with cerebellar gray matter as reference region. Regionally average voxel values were extracted in conjunction with z-scores from comparison to a normal database (n = 30). Surface projection maps of rCBF and rCBF z-score were computed after translation to MNI space. The software was written in Qt and C++ and implemented within the Hermes environment. Results in 22 subjects were compared to a previous analysis using in-house written software in Matlab.
Results: Quantitative results of the automated software agreed well with previous analysis (R2 correlation of SUVR-1 in putamen of 0.83 and rCBF in the limbic area of 0.77) (Figure 1). So far, the software has been used for differential diagnosis of about 300 patients at Uppsala University Hospital. Total processing time of approximately 5 min/patient.
Conclusion: [11C]PE2I enables a differential diagnosis of Parkinsonism based on both DAT availability and general brain function (rCBF) using a single scan, replacing a dual scan approach using DATscan-SPECT and FDG-PET. The software-aided approach improves logistics and may be helpful for experienced [11C]PE2I-PET readers analyzing challenging cases. The single scan approach and software have been successfully introduced as a routine clinical application at Uppsala University Hospital and the software is freely available for other interested centers.
References
AppelLJonassonMDanforsT, et al.Use of 11C-PE2I PET in differential diagnosis of Parkinsonian disorders. J Nucl Med2015;
56: 234–242.JonassonMAppelLEngmanJ, et al.Validation of parametric methods for [11C]PE2I positron emission tomography. NeuroImage2013;
74: 172–178.JonassonMAppelLDanforsT, et al.Development of a clinically feasible [11C]PE2I PET method for differential diagnosis of parkinsonism using reduced scan duration and automated reference region extraction. Am J Nucl Med Mol Imaging2017;
7: 263–274.
2020-37
Comparison of centiloids and amyloid load for evaluation of amyloid change in Down syndrome (#176)
Matthew Zammit1, Charles Laymon2, Dana Tudorascu2, Ann Cohen2, Davneet Minhas2, Shahid Zaman3, Beau Ances4, Chester A Mathis2, Sterling Johnson1, William E Klunk2, Benjamin Handen2 and Bradley Christian1
1University of Wisconsin-Madison, Madison, WI, USA
2University of Pittsburgh, Pittsburgh, PA, USA
3University of Cambridge, Cambridge, UK
4Washington University in St. Louis, St. Louis, MO, USA
Abstract
Introduction: Individuals with Down syndrome (DS) carry a triplicate copy of the gene encoding APP production, leading to an early presence of Aβ plaques in the brain. As a population, a majority of DS adults will develop clinical dementia by their late 60’s. The recent development of both Centiloids and Amyloid Load provide methodologies for standardizing the quantification of Aβ using PET. This work aims to compare these two measures in the DS population.
Methods: N = 169 adults with DS (age = 39.3 ± 8.4 years) were evaluated for Aβ using [C-11]PiB PET. A subset (n = 68) underwent longitudinal evaluation (3.0 ± 0.7 scans; 2.4 ± 0.6 years apart). All images were spatially normalized to MNI152 space, and 50–70 minute SUVr images were generated using a cerebellum reference region. Mean SUVr values were extracted from a global composite ROI and converted to units of Centiloids (CL).1 The Amyloid Load index (AβL) was calculated from each SUVr image, given inputs of DS-specific image templates of specific and nonspecific PiB binding.2,3 Aβ(+) was defined as having AβL > 20.0.3 Change scores between baseline and follow-up scans were evaluated for both CL and AβL across Aβ(-), Aβ converter (converted from Aβ(-) to Aβ(+)), and Aβ(+) groups. Effect sizes (Cohen’s d) were determined for CL and AβL change scores across each group.
Results: Change scores (presented as change/year) for CL (Aβ(-): 1.37 ± 2.36; Aβ converter: 6.25 ± 3.00; Aβ(+): 7.59 ± 4.10) and AβL (Aβ(-): 0.60 ± 1.07; Aβ converter: 3.20 ± 0.77; Aβ(+): 3.35 ± 1.69) reveal a lower variance in the AβL measure (Figure 1). Effect sizes for each measure are displayed in Table 1. For the Aβ(-) and Aβ(+) groups, CL and AβL show similar effect sizes. For the Aβ converter group, the lower variance of AβL results in a greater effect size compared to CL.
Conclusion: These results show that both Centiloids and Amyloid Load have similar performance to detect longitudinal increases in Aβ. Compared to Centiloids, the larger effect size for Amyloid Load in the Aβ converter group suggests this measure provides lower variability when evaluating Aβ change during the earliest stages of Aβ accumulation.
Table 1. Effect sizes (Cohen’s d with 95% CI’s) between baseline and follow-up scans for Centiloids (CL) and Amyloid Load (AβL) in Down syndrome.
References
KlunkWEKoeppeRAPriceJC, et al.The Centiloid Project: standardizing quantitative amyloid plaque estimation by PET. Alzheimer’s Dementia2015;
11: 1–15.WhittingtonAGunnRN.
Amyloid load: a more sensitive biomarker for amyloid imaging. J Nucl Med2019;
60: 536–540.ZammitMDLaymonCMBetthauserTJ, et al.Amyloid accumulation in Down syndrome measured with amyloid load. Alzheimer’s Dementia2020;
12: e12020.
2020-38
First-in-human evaluations of [11C]PS13 for imaging cyclooxygenase-1 and [11C]MC1 for imaging cyclooxygenase-2 (#180)
Min-Jeong Kim1, Fernanda Juarez Anaya1, Jae-Hoon Lee1, Jinsoo Hong1, William Miller1, Sanjay Telu1, Cheryl Morse1, Prachi Singh1, Michelle Y Cortes-Salva1, Katharine Henry1, Yanira Ruiz-Perdomo2, Jose A Montero Santamaria1, Jeih-San Liow1, Sami S Zoghbi1, Masahiro Fujita1, James D Katz2, Victor W Pike1 and Robert B Innis1
1Molecular Imaging Branch, National Institute of Mental Health, Bethesda, MD, USA
2National Institute of Arthritis and Musculoskeletal and Skin Diseases, Bethesda, MD, USA
Abstract
Introduction: The cyclooxygenase (COX) system comprises two isoforms—COX-1 and COX-2—that are important targets for neuroinflammatory biomarkers.1 We developed two PET radioligands: [11C]PS13 for COX-1 and [11C]MC1 for COX-2.2–4 Previous monkey and human whole blood assays as well as PET imaging studies in monkey demonstrated the excellent in vitro and in vivo selectivity of these ligands,5 justifying the extension of studies into human individuals. This study sought to assess the distribution of binding and in vivo selectivity of [11C]PS13 in healthy individuals and those of [11C]MC1 in patients with rheumatoid arthritis.
Methods: Injection of radioligand was followed by dynamic PET scans. For [11C]PS13, test-retest brain scans in 10 healthy individuals and whole-body scans in 24 healthy individuals were obtained. For [11C]MC1, whole-body scans were obtained in two patients with rheumatoid arthritis and two healthy controls. Whole-body scans were repeated after administration of blocking drugs that preferentially target COX-1 (aspirin and ketoprofen) or COX-2 (celecoxib). Concurrent blood samples were obtained to measure concentrations of parent radioligand and radiometabolites.
Results: In baseline, substantial uptake of [11C]PS13 was observed in most major organs, including the spleen, gastrointestinal tract, kidneys, and brain. This uptake was blocked by aspirin or ketoprofen but minimally blocked by celecoxib. In brain, [11C]PS13 showed high uptake in the hippocampus and occipital cortices as well as pericentral cortices. When total distribution volume of [11C]PS13 was calculated in brain, the overall test-retest variability was 6.0–8.5%. With [11C]MC1, significantly higher uptake was observed in arthritic joints of the patients with rheumatoid arthritis compared to the corresponding joints of healthy controls. The increased uptake was partially blocked by celecoxib.
Conclusion: Our results suggest that COX-1 is constitutively expressed in major organs and that COX-2 is expressed in inflamed joints with rheumatoid arthritis. The in vivo selectivity of [11C]PS13 and [11C]MC1 was well demonstrated by pharmacological blockade. The test-retest reliability of brain [11C]PS13 binding was also validated. Both [11C]PS13 and [11C]MC1 are potential probes for measuring brain and systemic inflammation in various conditions as well as target engagement by therapeutic drugs.
References
SimmonsDLBottingRMHlaT.Cyclooxygenase isozymes: the biology of prostaglandin synthesis and inhibition. Pharmacol Rev2004;
56: 387–437.SinghPShresthaSCortes-SalvaMY, et al.3-Substituted 1,5-diaryl-1 H-1,2,4-triazoles as prospective PET radioligands for imaging brain COX-1 in monkey. Part 1: synthesis and pharmacology. ACS Chem Neurosci2018;
9: 2610–2619.ShresthaSSinghPCortes-SalvaMY, et al.3-Substituted 1,5-diaryl-1 H-1,2,4-triazoles as prospective PET radioligands for imaging brain COX-1 in monkey. Part 2: selection and evaluation of [(11)C]PS13 for quantitative imagingACS Chem Neurosci2018;
9: 2620–2627.CortesMShresthaSSinghP, et al.2-(4-Methylsulfonylphenyl)pyrimidines as prospective radioligands for imaging cyclooxygenase-2 with PET – synthesis, triage, and radiolabeling. Molecules2018;
23: pii E2850.KimMJShresthaSSCortesM, et al.Evaluation of two potent and selective PET radioligands to image COX-1 and COX-2 in rhesus monkeys. J Nucl Med2018;
59: 1907–1912.
2020-39
Using hybrid PET/MRI to determine if perfusion MRI has comparable sensitivity to 15O-water PET for detecting dementia-related hypoperfusion (#181)
Tracy Ssali1, 2, Lucas Narciso1, 2, Justin W Hicks1, 2, Matthais Günther3, Frank S Prato1, 2, Udunna Anazodo1, 2, Elizabeth Finger4 and Keith St Lawrence1, 2
1Medical Imaging, Lawson Health Research Institute, London, Canada
2Medical Biophysics, Western University, London, Canada
3Fraunhofer Institute for Medical Image Computing MEVIS, Bremen, Germany
4Department of Clinical Neurological Sciences, Western University, London, Canada
Abstract
Introduction: The ability of arterial spin labeling (ASL) MRI to detect perfusion abnormalities in clinical populations can be limited by poor signal to noise and transit-time artifacts. This has been evident in studies involving frontotemporal dementia (FTD) patients, where reports on the diagnostic value of ASL have been inconsistent.1,2 Recent advances in ASL protocols including optimized labeling parameters should enable detection of more subtle perfusion abnormalities.3 This study presents a head-to-head comparison of regional hypoperfusion detected by ASL and PET with radiolabeled water (15O-water) – the gold standard for measuring cerebral blood flow (CBF) in humans.
Methods: Data were acquired from 9 controls (age: 63 ± 10, sex: 3 female/6 male) and 7 FTD patients across FTD subtypes (age: 69 ± 9, sex: 4 female/3 male) on the Siemens biograph mMR. Five minutes of list mode data were acquired after a bolus injection of approximately 800Mbq of 15O-water. Perfusion was quantified using a double-integration method.4 ASL data were collected immediately following PET acquisition and CBF images generated using a one compartment model.3 Statistical maps of regional hypoperfusion were determined using a case-control design using absolute and relative CBF (normalized to whole-brain blood flow).
Results: Global CBF measured by ASL was: 55.9 ± 9.6 and 49.9 ± 11.4 ml/100g/min (ns) and by 15O-water was: 51.1 ± 8.2 and 45 ± 9.9 ml/100g/min (ns) in controls and patients, respectively. Control CBF maps are shown in Figure 1. Both ASL MRI and 15O-water PET images showed hypoperfusion in regions commonly associated with each FTD subtype (Figure 2).
Conclusion: This work highlights the potential of ASL for identifying regional hypoperfusion in participants with FTD. Preliminary data showed that ASL was capable of detecting disease-specific perfusion abnormalities and the pattern were very similar to those detected by 15O-water PET. However, the hypoperfusion regions identified by PET were typically greater in number and larger in area than those identified by ASL. Further work is required to determine the impact of these differences on the diagnostic accuracy of ASL.
Acknowledgements
This work is supported by: Frederick Banting and Charles Best Canada Graduate Doctoral Award, Canadian Institutes of Health Research Grant and Alzheimer’s Drug Discovery Foundation Grant.
Regional hypoperfusion detected by ASL and 15O-water
T-maps generated using absolute (top) and relative (bottom) CBF measured by ASL (red) and 15O-water (blue) in 4 representative frontotemporal dementia FTD subtypes. Areas in red/blue indicate regions of significant hypoperfusion in the patient participant compared to the control group.
Normalized perfusion maps
Perfusion maps measured by 15O-water and ASL. Data shown has been averaged across healthy controls (n = 9) and normalized to mean global flow.
References
AnazodoUC, et al.Using simultaneous PET/MRI to compare the accuracy of diagnosing frontotemporal dementia by arterial spin labelling MRI and FDG-PET. NeuroImage Clin2018;
17: 405–414.BronEE, et al.Diagnostic classification of arterial spin labeling and structural MRI in presenile early stage dementia. Hum Brain Mapp2014;
35: 4916–4931.AlsopDC, et al.Recommended implementation of arterial spin-labeled perfusion MRI for clinical applications: a consensus of the ISMRM perfusion study group and the European consortium for ASL in dementia. Magn Reson Med2015;
73: 102–116.SsaliTAnazodoUCThiessenJD, et al.A non-invasive method for quantifying cerebral blood flow by hybrid PET/MR. J Nucl Med2018;
59: 1329–1334.
2020-40
Olfactory impairment is related to tau pathology and neuroinflammation in Alzheimer’s disease (#187)
Julia I Klein1, 2, Xinyu Yan3, Aubrey Johnson1, Zeljko Tomljanovic1, James Zou1, Krista Polly1, Estrella Morenas-Rodriguez4, 5, Christian Haass4, 6, Lawrence S Honig1, Adam M Brickman1, Yaakov Stern1, Seonjoo Lee3, 7 and William C Kreisl1
1Columbia University Irving Medical Center, Taub Institute, New York, NY, USA
2Weill Cornell Medical College, New York, NY, USA
3Mailman School of Public Health, Columbia University Irving Medical Center, New York, NY, USA
4German Center for Neuroegenerative Disease, Munich, Germany
5Institut d’Investigacion Biomèdiques, Universitat Autònoma de Barcelona, Barcelona, Spain
6Metabolic Biochemistry Biomedical Center, Ludwig-Maximilians-Universität München, Munich, Germany
7The Research Foundation for Mental Hygiene, New York, NY, USA
Abstract
Introduction: Olfactory impairment is observed early in Alzheimer’s disease (AD).1–4 and can be quantified using the University of Pennsylvania Smell Identification Test (UPSIT). The relationships between odor identification and in vivo measures of known pathological features of Alzheimer’s disease, including tau pathology and neuroinflammation, are less established. We sought to determine if low UPSIT performance was associated with increased neuroinflammation, measured by 11C-PBR28 PET and CSF sTREM2 and YKL-40. We further evaluated the relationship between odor identification and tau pathology using PET imaging with 18F-MK-6240, an improved tau radioligand, and CSF concentrations of total tau and phosphorylated tau.
Methods: Participants were selected from an established research cohort of adults aged 50 and older who underwent neuropsychological testing, brain MRI, and amyloid PET. Fifty-four participants were administered the UPSIT. Forty-one of these underwent 18F-MK-6240 PET to measure tau pathology and fifty-three underwent 11C-PBR28 PET to measure the 18 kDa translocator protein (a biomarker of inflammation). Twenty-three participants had lumbar puncture to measure CSF concentrations of total tau (t-tau), phosphorylated tau (p-tau), and inflammatory biomarkers sTREM2 and YKL-40.
Results: Negative correlations between UPSIT performance and 18F-MK-6240 binding were seen in medial temporal cortex, hippocampus, middle/inferior temporal gyri, inferior parietal cortex and posterior cingulate cortex (p < 0.05). Similar inverse relationships were seen for 11C-PBR28. UPSIT performance negatively correlated with CSF concentrations of t-tau, p-tau, and YKL-40 (p < .05). Moreover, UPSIT scores correlated with hippocampal volume and cognitive performance. Analysis of variance, controlling for age and sex showed that amyloid status and cognitive status exhibit independent effects on UPSIT performance (p < 0.01).
Conclusion: Olfactory identification declines with increasing tau pathology and neuroinflammation along the Alzheimer’s continuum. Amyloid-positivity and cognitive impairment exhibit independent effects on odor identification, suggesting that 1) low UPSIT performance may signify risk of AD in cognitively normal individuals, and 2) impaired odor identification is associated with both AD- and non-AD-related neurodegeneration.
Acknowledgements
18F-Florbetaben was supplied by Life Molecular Imaging. 18F-MK-6240 was supplied by Cerveau Technologies. TSPO genotyping was performed by Regina Santella, PhD and the Columbia University Biomarkers Shared Resource. We wish to acknowledge the contributions of the staff and faculty of the Irving Institute for Clinical Research, the MRI Center and the David A. Gardner PET Imaging Center at the Columbia University Irving Medical Center.
References
DevanandDPMichaels-MarstonKSLiuX, et al.Olfactory deficits in patients with mild cognitive impairment predict Alzheimer’s disease at follow-up. Am J Psychiatry2000;
157: 1399–1405.DjordjevicJJones-GotmanMDe SousaK, et al.Olfaction in patients with mild cognitive impairment and Alzheimer’s disease. Neurobiol Aging2008;
29: 693–706.DotyRLReyesPFGregorT.Presence of both odor identification and detection deficits in Alzheimer’s disease. Brain Res Bull1987;
18: 597–600.MurphyCGilmoreMMSeeryCS, et al.Olfactory thresholds are associated with degree of dementia in Alzheimer’s disease. Neurobiol Aging1990;
11: 465–469.
2020-41
Automated data quality control in [18F]FDOPA brain PET imaging using deep learning (#190)
Antonella D Pontoriero1, Barbara Santangelo1, 2, Sameer Jahuar2, Ilaria Bonoldi2, Maria Rogdaki2, Federico E Turkheimer1, Oliver D Howes2 and Mattia Veronese1
1Department of Neuroimaging, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London, UK
2Department of Psychosis Studies, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London, UK
Abstract
Introduction: With biomedical imaging research racing for larger datasets, it becomes critical to find new methods to guarantee that the quality of data collected is not compromised before using them for further analysis or clinical assessment. Some attempts to use Artificial Intelligence have already been applied to Magnetic Resonance Imaging (MRI), to perform automated quality control (QC) both for single-site and multi-site datasets.1–3 No automated QC methods for PET imaging exist today. The aim of this study is to develop and validate an automated QC pipeline for brain [18F]FDOPA PET imaging as biomarker for dopamine system in mental health applications.
Methods: This project uses Convolutional Neural Networks (CNNs), a deep-learning approach which provides better outcomes when compared to other learning models aimed at classification of images.4 Here, 3 different CNNs are combined to assess 1) the spatial alignment to standard MNI brain template, 2) the spatial distribution of [18F]FDOPA tracer uptake and 3) signal-to-noise ratio (SNR) of the image. A dataset of 200 manually QC [18F]FDOPA PET images (injected dose: 150 ± 12MBq) from three different PET/CT scanners is combined with 200 scans in which misalignment, altered tracer uptake and/or low SNR are simulated. A cross-validation is performed; 80% of the data is used for training and 20% for validation while training. Categorical cross-entropy (target value 0) and categorical accuracy (target value 1) are used performances indexes. The same models are trained on the original 3D images, 2D images of a representative single slice of the 3D image and on 1D data obtained by tracing a representative line of the 2D image (Figure 1).
Results: With 3D datasets, the best performances are obtained for the spatial distribution of [18F]FDOPA tracer uptake (accuracy = 0.76, cross-entropy = 0.53) while the SNR network is the poorest. When training on low-dimensional datasets, the time required for training is extremely reduced while the performances remain comparable (Table 1).
Conclusion: This feasibility study shows that it is possible to perform QC of [18F]FDOPA imaging with CNNs. However, it must be investigated how generalisable the approach is when using a completely new dataset not included in the training.
Table 1. Performance and training time for [18F]FDOPA CNNs for 1D, 2D and 3D datasets.
References
PriceRR, et al.Quality assurance methods and phantoms for magnetic resonance imaging: report of AAPM nuclear magnetic resonance Task Group No. Med Phys1990;
17: 287–295.PizarroRA, et al.
Automated quality assessment of structural magnetic resonance brain images based on a supervised machine learning algorithm. Front Neuroinfor. 2016;
10: 52.EstebanO, et al.MRIQC: advancing the automatic prediction of image quality in MRI from unseen sites. PLoS One2017;
12: e0184661.DuttaS, et al.A comparative study of deep learning models for medical image classification. IOP Conf Series: Mater Sci Eng2017;
263: 042097.
2020-42
Human blocking study to assess selectivity of [18F]FTP PET for dopamine D3 receptors (#193)
Robert K Doot1, Anthony J Young1, Tiffany L Dominguez1, Christopher G Ward2, Shihong Li1, Zeinab Helili1, Regan Sheffer1, Hsiaoju Lee1, Erin K Schubert1, Robert H Mach1 and Jacob G Dubroff1
1Dept. of Radiology, Perelman School of Medicine, Univ. of Pennsylvania, Philadelphia, PA, USA
2Dept. of Anesthesiology and Critical Care Medicine, The Children’s Hospital of Philadelphia, Philadelphia, PA, USA
Abstract
Introduction: Dopamine D3 receptors (D3R) in the D2 family of receptors are thought to be important to neurobiological reward pathways and implicated in drug addiction, Parkinson’s disease, and schizophrenia. [18F]Fluortriopride ([18F]FTP) was developed as a D3R-selective radiotracer1 and based on nonhuman primate imaging1 was anticipated to be more selective for D3R than current human D2 family radiotracers.2 First-in-human [18F]FTP images revealed unexpected uptake outside of the dopamine system. This distribution may be due to [18F]FTP binding to serotonin 5-HT1A receptors (HT1AR), which have high affinity for N-phenyl piperazine analogs like [18F]FTP.3 The goal of this blocking study goal was to assess whether the unexpected [18F]FTP uptake was due to nonselective binding or selective binding to HT1AR.
Methods: Search of the NIMH Psychoactive Drug Screening Program (PDSP) Ki database for drugs that could selectively block [18F]FTP from either HT1AR or D3R suggested using either HT1AR-blocking Lysergic acid diethylamide (LSD) or D3R-blocking perphenazine. LSD was not safe to use at levels required for blocking. 8-mg PO of perphenazine was selected based on a safe record in over 45,000 humans4 and a report estimating a maximum D2-family receptor occupancy of 98% in a [11C]raclopride human study.5 Five healthy volunteers (3 female, 2 male) underwent two 60-minute dynamic brain scans with arterial blood sampling following bolus injection of [18F]FTP (230 ± 29MBq (mean ± SD, n = 10) both without perphenazine and starting 1-hr following perphenazine administration. Total distribution volumes (VT) were estimated (Pmod v3.7) in D3R-rich (caudate nucleus, globus pallidus, putamen, and thalamus) and HT1AR-rich regions (frontal lobe, insula, parahippocampus, hippocampus, temporal lobe, and entorhinal area). Two-tailed paired t-tests were performed to test for significant differences (SPSS 25, p < 0.05).
Results: Representative [18F]FTP brain images of a healthy control before and after perphenazine challenge are in Figure 1. VT results are shown in Figure 2 with no significant differences observed for D3R-rich regions (p = 0.38) or for HT1AR-rich regions (p = 0.33).
Conclusion: [18F]FTP uptake in D3R-rich regions was not significantly changed following perphenazine challenge (p = 0.38) suggesting nonselective binding of [18F]FTP throughout human brains and future efforts to image D3R should focus on alternative D3R radiotracers including those under development in preclinical studies.
Acknowledgements
Authors appreciate research support provided by the United States of America’s National Institute on Drug Abuse of the National Institutes of Health under Award Numbers K01DA040023 (RD), K23DA038726 (JD) and R01DA029840-06A1 (RM).
References
MachRHTuZXuJ, et al.Endogenous dopamine (DA) competes with the binding of a radiolabeled d3 receptor partial agonist in vivo: a positron emission tomography study.Synapse2011;
65: 724–732.DootRKDubroffJGLabbanKJ, et al.Selectivity of probes for PET imaging of dopamine D3 receptors. Neurosci Lett2019;
691: 18–25.ChuWTuZMcElveenE, et al.Synthesis and in vitro binding of N-phenyl piperazine analogs as potential dopamine D3 receptor ligands.Bioorg Med Chem2005;
13: 77–87.HenaoJPPeperzakKALichvarAB, et al.Extrapyramidal symptoms following administration of oral perphenazine 4 or8mg. Eur J Anaesthesiol2014;
31: 231–235.AppelLGeffenYHeurlingK, et al.BL-1020, a novel antipsychotic candidate with GABA-enhancing effects: D2 receptor occupancy study in humans.Eur Neuropsychopharmacol2009;
19: 841–850 (Table 3, pg 848).
2020-43
ApoE4 packs a punch in women: Sex-specific vulnerability for tau (#194)
Yi-Ting Wang1, 2, Min-Su Kang1, 2, Joseph Therriault1, 2, Tharick A Pascoal1, 2, Melissa Savard1, 2, Firoza Z Lussier1, 2, Andréa L Benedet1, 2, Cécile Tissot1, 2, Jaime Fernandez Arias1, 2, Serge Gauthier1, 2 and Pedro Rosa-Neto1, 2
1Research Centre for Studies in Aging, McGill University, Montreal, Québec, Canada
Introduction: Apolipoprotein E (APOE) is the strongest genetic risk factor for sporadic Alzheimer’s disease (AD), with the ε4 allele conferring increased risk. APOE has been implicated in various neuropathological cascades relevant to AD, including amyloidosis, tau tangle pathology,1 microglial activation and neurodegeneration. Research leveraging in vivo tau biomarkers indicates that APOEε4 is associated with higher levels of tau in the cerebrospinal fluid2 (CSF). A meta-analysis published in 2017 states that the effect of APOEε4 is more robust among women compared with men, indicating the presence of a sex difference.3 The role that sex plays in AD has long been the subject of intense investigation. Previous works reported mixed results about the sex-specific effect of APOEε4 on tau pathology. Some studies suggest that women show a more robust association between APOEε4 and tau, while other works reveal no sex difference. The overarching goal of this study is to investigate the APOE effect on tau pathology in a sex-specific manner.
Methods: This was a cross-sectional study in subjects with AD (n = 44), MCI (n = 54), cognitively normal controls (n = 154) and young healthy controls (n = 35) from the TRIAD study at McGill University Research Centre for Studies in Aging, Canada. A total of 287 subjects were studied, among which 108 were APOEε4 carriers and 179 were non-APOEε4 carriers. Cerebral amyloid load and tau level were assessed using positron emission tomography (PET) radiopharmaceuticals [18F]AZD4694 ([18F]NAV4694) and [18F]MK6240 respectively.
Results: Male and female APOE ε4/ε4carriers showed a 13.2% and 65.1% higher [18F]MK6240 SUVR repectively when compared with their ε3/ε3counterparts. Female APOEε4 carriers showed significant higher tau burden in hippocampus (P < 0.01), entorhinal cortex (P < 0.05) and parahippocampal cortex (P < 0.05) compared to male ε4 carriers. A sex-specific vulnerability for tau was observed. More specifically, in amyloid-positive female APOEε4carriers, higher tau burden was found to be strongly associated with worse cognitive function assessed by Mini-Mental State Examination (MMSE).
Conclusion: We provide strong evidence that women are more vulnerable for the effects of APOEε4 on tau, suggesting APOE may modulate AD pathology in a sex-specific manner, particularly in the presence of amyloidosis.
References
FarfelJMYuLDe JagerPL, et al.Association of APOE with tau-tangle pathology with and without β-amyloid. Neurobiol Aging2016;
37: 19–25.HohmanTJDumitrescuLBarnesLL, et al.Sex-specific association of apolipoprotein E with cerebrospinal fluid levels of tau. JAMA Neurol2018;
75: 989–998. doi:10.1001/jamaneurol.2018.0821.NeuSCPaJKukullW, et al.Apolipoprotein E genotype and sex risk factors for Alzheimer disease: a meta-analysis. JAMA Neurol2017;
74: 1178–1189.
2020-44
Measurement of HDAC6 target occupancy in macaque using [18F]EKZ-001 positron emission tomography (#201)
Tonya M Gilbert1, Sofie Celen2, Michel Koole3, Isabeau Vermeulen2, Kim Serdons4, Frederick A Schroeder1, Florence Wagner5, Tom Bleeser6, Wim Vanduffel7, Koen Van Laere3, 4, Janice E Kranz1, Jacob M Hooker8, Guy Bormans2 and Christopher Cawthorne3
1Eikonizo Therapeutics, Inc., Cambridge, MA, USA
2Laboratory for Radiopharmaceutical Research, KU Leuven, Leuven, Belgium
3Nuclear Medicine and Molecular Imaging, Department of Imaging and Pathology, KU Leuven, Leuven, Belgium
4Nuclear Medicine and Molecular Imaging, University Hospitals Leuven, Leuven, Belgium
5Center for the Development of Therapeutics, Stanley Center for Psychiatric Research, The Broad Institute of MIT and Harvard, Cambridge, MA, USA
6Anesthesiology and Algology, KU Leuven, Leuven, Belgium
7Laboratory for Neuro- and Psychophysiology, KU Leuven, Leuven, Belgium
8Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA
Abstract
Introduction: Histone deacetylase 6 (HDAC6) is a multifunctional enzyme involved in diverse cellular processes such as intracellular transport1 and protein quality control.2 Inhibition of HDAC6 can alleviate defects in cell and rodent models of neurodegenerative disorders, including Alzheimer’s disease3 and amyotrophic lateral sclerosis.4 However, while HDAC6 represents a potentially powerful therapeutic target, development of effective brain-penetrant HDAC6 inhibitors remains challenging. Here [18F]EKZ-001 ([18F]Bavarostat5) positron emission tomography (PET) was applied to measure dose-occupancy relationships of HDAC6 inhibitors in the non-human primate brain.
Methods: [18F]EKZ-001 was administered by intravenous bolus injection (∼185 MBq). The most appropriate kinetic model for brain imaging was determined in rhesus macaque (N = 3 males) from 120-minute dynamic PET/MR using radiometabolite-corrected arterial plasma as an input function. The total volume of distribution (VT) was calculated using one-tissue and two-tissue compartment models (1TCM and 2TCM) and fits were evaluated using the Akaike information criterion (AIC). VT was also calculated with Logan graphical analysis (LGA) using a starting time of 40 minutes post-injection. Heterologous-blocking studies were performed in rhesus macaque (N = 1 male) with the novel candidate HDAC6 inhibitor, EKZ-317, and an existing HDAC6 inhibitor tool compound, ACY-775, at 0.1 and 2 mg/kg, using an intravenous pre-treatment paradigm 5 minutes before [18F]EKZ-001 injection. Target occupancy was estimated using the Lassen plot.
Results: Based on AIC, 2TCM was the preferred model compared to 1TCM. Regional LGA VTsignificantly correlated with 2TCM valuesin baseline and blocking conditions. Lassen plots using VTderived from either 2TCM or LGA gave comparable target occupancy estimates. Non-displaceable distribution volume (VND) was consistent between blocking conditions and quantitation methods (approximately 10%), indicating a high degree of specific binding. EKZ-317 achieved full HDAC6 occupancy at 2 mg/kg, and greater than 90% occupancy at 0.1 mg/kg, demonstrating a higher level of HDAC6 target engagement than ACY-775 in brain (Figure 1).
Conclusion: [18F]EKZ-001 PET showed favorable kinetic properties in non-human primate for calculation of VT in brain. This work supports the translation of [18F]EKZ-001 PET for HDAC6 target occupancy studies in humans and may facilitate the development of brain penetrant HDAC6 inhibitor therapeutics.
Acknowledgements
This study was supported by funding from Eikonizo Therapeutics, Inc.
References
DompierreJP, et al.Histone deacetylase 6 inhibition compensates for the transport deficit in Huntington’s disease by increasing tubulin acetylation. J Neurosci2007;
27: 3571–3583.LeeJ-Y, et al.HDAC6 controls autophagosome maturation essential for ubiquitin-selective quality-control autophagy. EMBO J2010;
29: 969–980.ZhangL, et al.Tubastatin A/ACY-1215 improves cognition in Alzheimer’s disease transgenic mice. J Alzheimers Dis2014;
41: 1193–1205.GuoW, et al.HDAC6 inhibition reverses axonal transport defects in motor neurons derived from FUS-ALS patients. Nat Commun2017;
8: 861.StreblMG, et al.HDAC6 brain mapping with [18F]bavarostat enabled by a Ru-mediated deoxyfluorination. ACS Cent Sci2017;
3: 1006–1014.
2020-45
Extra-striatal D2/3 receptor availability in youth at risk for addiction (#202)
Natalia Jaworska1, Sylvia Cox2, Maria Tippler2, Natalie Castellanos-Ryan3, Chawki Benkelfat2, Sophie Parent3, Alain Dagher4, Frank Vitaro3, Michel Boivin5, Robert Pihl6, Sylvana Côté7, Richard Tremblay8, Jean Séguin8 and Marco Leyton2
1Institute of Mental Health Research, University of Ottawa, Ottawa, ON, Canada
2Department of Psychiatry, McGill University, Montreal, Québec, Canada
3School of Psychoeducation, Université de Montréal, Montreal, Québec, Canada
4Montreal Neurological Institute, McGill University, Montreal, Québec, Canada
5Department of Psychology, Université Laval, Quebec, Canada
6Department of Psychology, McGill University, Montreal, Québec, Canada
7Department of Social and Preventative Medicine, Université de Montréal, Montreal, Québec, Canada
8Department of Pediatrics & Psychology, Université de Montréal, Montreal, Québec, Canada
9CHU Ste-Justine Research Centre, Université de Montréal, Montreal, Québec, Canada
Abstract
Introduction: The neurobiological traits that confer risk for addictions remain poorly understood. However, dopaminergic function throughout the prefrontal cortex, limbic system, and upper brainstem has been implicated in behavioral features that influence addiction vulnerability, including poor impulse control and altered sensitivity to rewards and punishments; i.e., externalizing features. To test these associations in humans, we measured type-2/3 dopamine receptor (DA2/3R) availability in youth at high vs. low risk for substance use disorders.
Methods: Fifty-eight participants (18.5 ± 0.6yr) were recruited from cohorts that have been followed since birth: half had high (high EXT, N = 27, 16F/11M) and half had low externalizing traits (low EXT, N = 31, 20F/11M). All underwent a 90-min positron emission tomography [18F]fallypride scan and completed the Barratt Impulsiveness Scale (BIS-11), Substance Use Risk Profile Scale (SURPS), and Sensitivity to Punishment (SP) & Sensitivity to Reward (SR) Questionnaire.
Results: The high vs. low EXT participants reported elevated substance use, BIS-11, SR and SURPS Impulsivity scores, had a greater prevalence of psychiatric disorders, and exhibited higher [18F]fallypride binding potential (BPND) values in prefrontal, limbic and paralimbic regions, even when controlling for substance use. Group differences were not evident in the midbrain dopamine cell body region, but, across all participants, low midbrain BPND values were associated with low SP scores.
Conclusion: Altered DA2/3R availability in extra-striatal terminal and dopamine cell body regions might constitute biological vulnerability traits, generating an EXT trajectory for addictions with and without co-occurring alterations in punishment sensitivity; i.e., an internalizing feature.
2020-46
Evaluation of [18F]APN-1607 to image tau protein in patients with Alzheimer’s disease and progressive supranuclear palsy: Test-retest and cross-sectional analysis (#204)
Christine Sandiego1, Roger N Gunn2, David S Russell2, 3, Ken Marek3, Richard Margolin4, 5, PJ Chen4, Paul Tempest4 and Ming-Kuei Jang4
1Invicro, New Haven, CT, USA
2Invicro, London, UK
3Institute for Neurodegenerative Disorders, New Haven, CT, USA
4Aprinoia Therapeutics, Taipei, Taiwan
5CNS Research Solutions LLC, Cambridge, MA, USA
Abstract
Introduction: [18F]APN-1607 is a novel PET tracer for imaging fibrillar tau, a key characteristic of several neurodegenerative diseases known as tauopathies. These conditions include Alzheimer’s disease (AD) and progressive supranuclear palsy (PSP). This study aimed to assess [18F]APN-1607 binding in patients with AD and PSP, compared with healthy volunteers (HV) and to evaluate test-retest variability (TRTV) of blood- and reference region-based outcome measures.
Methods: [18F]APN-1607 PET was conducted over 180 min in 5 HV, 7 AD and 8 PSP subjects; 2 HV, 4 AD and 3 PSP subjects returned for a retest scan. In 7 subjects who completed test and retest scans, arterial blood samples were collected throughout the imaging sessions, as required for quantitative analysis. Distribution volume ratio (DVR) was estimated with invasive and non-invasive Logan graphical analysis (LGA). SUVr was averaged between 60–90 and 120–150 min, using the cerebellar cortex as reference region. TRTV of DVR and SUVr were evaluated across cortical (temporal, parietal and frontal lobes) and subcortical (thalamus, pallidum, and substantia nigra) regions. DVR(invasive) was considered the gold standard and was compared via linear regression analysis with DVR(non-invasive) and SUVr. SUVr was compared between HV, PSP and AD subjects.
Results: TRTV of [18F]APN-1607 in HV, PSP and AD subjects was less than 10% for DVR and SUVr, averaged across regions. SUVr(120–150 min) was in best agreement (R2 = 0.97, slope = 1.04) with invasive DVR, compared with non-invasive DVR(R2 = 0.97, slope = 0.85) and SUVr(60–90 min)(R2 = 0.90, slope = 0.87). Images of [18F]APN-1607 SUVr(120–150 min) in representative HV, PSP and AD subjects are shown in Figure 1. Regional cross-sectional analysis of [18F]APN-1607 SUVr(120–150) is shown in Figure 2. In cortical regions SUVr(120–150 min) (mean ± SD) was 0.97 ± 0.13 in HV, 0.99 ± 0.24 in PSP and 1.76 ± 0.53 in AD subjects. In subcortical regions, SUVr(120–150 min) was 1.19 ± 0.17 in HV, 1.37 ± 0.21 in PSP and 1.20 ± 0.23 in AD subjects.
Conclusion: [18F]APN-1607 showed excellent test-retest reliability across outcome measures and is a promising PET radioligand for imaging tau in specific neurodegenerative diseases. Cross-sectional analysis demonstrated substantially higher binding in cortical regions for AD compared to HV and PSP, and moderately higher binding in subcortical regions for PSP, compared to HV and AD.
Acknowledgements
For this work, would like to acknowledge the chemistry and clinic teams at Invicro, New Haven.
2020-47
Validation of a non-invasive hybrid PET/MRI method for imaging the cerebral metabolic rate of oxygen (#209)
Lucas Narciso1, 2, Tracy Ssali1, 2, Linshan Liu1, Heather Biernaski1, John Butler1, Laura Morrison1, Jennifer Hadway1, Justin W Hicks1, 2, Michael C Langham3, Felix W Wehrli3, Hidehiro Iida4 and Keith St Lawrence1, 2
1Lawson Health Research Institute, London, ON, Canada
2Department of Medical Biophysics, Western University, London, ON, Canada
3Department of Radiology, University of Pennsylvania Medical Centre, Philadelphia, PA, USA
4Turku PET Centre, University of Turku, Turku, Finland
Abstract
Introduction: PET is the gold standard for imaging the cerebral metabolic rate of oxygen (CMRO2) in humans; however, the procedure requires multiple 15O-tracers and arterial blood sampling. Hybrid PET/MR offers a means of simplifying the procedure by using MRI-based measurements of whole-brain (WB) CMRO2 as a reference to calibrate dynamic 15O-oxygen-PET data.1 This hybrid approach eliminates the need for invasive arterial sampling, only requires PET images of 15O-oxygen, and reduces the total duration to 5 min. It is also predicted to be insensitive to errors related to blood-borne activity and recirculating water because they do not affect MRI CMRO2 measurements.2 In this study, we present initial validation of the approach conducted in a large animal model that enabled arterial sampling for measurement of CMRO2 by a previously validated PET method.
Methods: PET and MRI data were obtained from juvenile pigs (n = 9, 18.9 ± 2.1 kg) under two metabolic conditions on a 3T Siemens Biograph mMR system. MR imaging included arterial spin labelling (ASL) and Oxflow to measure regional cerebral blood flow (CBF) and WB CMRO2, respectively.3 Concurrent PET imaging involved 5-min list-mode acquisitions after injecting 500 MBq of 15O-water, followed by inhaling 2200 MBq of 15O-oxygen. Arterial sampling was obtained using an MR-compatible system (Swisstrace). CT images were acquired post-mortem for attenuation correction. Dynamic PET images were reconstructed into 48 time-frames (30 × 3s, 6 × 5s, 6 × 10s and 6 × 20s). CMRO2 images were generated from the PET data alone4 and by the hybrid PET/MR procedure.
Results: Results from the hybrid PET/MR approach (n = 6, Figure 1) presented mean CMRO2 within the expected range, for both baseline (1.88 ± 0.24 mL/100 g/min) and lower metabolic conditions (1.20 ± 0.33 mL/100 g/min; 36% reduction, p < 0.01).
Conclusion: Initial WB CMRO2 results obtained with the hybrid PET/MR approach were within the expected range, further reinforcing our previous assessments.1 These results suggest that quantitative measurements of CMRO2 can be obtained without the need for arterial blood sampling. The complete analysis of our experiments can be found in Narciso et al.2
Acknowledgements
This work is supported by the Canadian Institutes of Health Research.
References
NarcisoLSsaliTIidaH, et al.A non-invasive reference-based method for imaging the cerebral metabolic rate of oxygen by PET/MR: theory and error analysis. Phys Med Biol2021;
66: 065009.NarcisoL, et al.A noninvasive method for quantifying cerebral metabolic rate of oxygen by hybrid PET/MRI: validation in a porcine model. J Nucl Med2021;
120: 260521.WehrliFW, et al.Time-resolved MRI oximetry for quantifying CMRO2 and vascular reactivity. Acad Radiol2014;
21: 207–214.KudomiN, et al.Rapid quantitative CBF and CMRO2 measurements from a single PET scan with sequential administration of dual 15O-labeled tracers. J Cereb Blood Flow Metab2013;
33: 440–448.
2020-48
Non-invasive quantification of [11C]PBR28 binding in non-human primates (#211)
Lucero G Aceves-Serrano1, Jacob Goddard2, Vesna Sossi2 and Doris J Doudet1
1Department of Medicine/Neurology, University of British Columbia, Vancouver, BC, Canada
2Department of Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada
Abstract
Introduction: Quantification of neuroinflammation tracers (i.e. TSPO tracers) can be troublesome as it requires arterial cannulation. We assessed the usefulness of four different quantitative and semi-quantitative blood-free measurements of [11C]-PBR28 binding and compared them to the standard blood dependent measurements.
Methods: Thirteen [11C]-PBR28 scans from non-human primates (NHP) were acquired along with metabolite and plasma activity to create arterial input functions (AIFs). BPND was quantified with AIF (BPNDAIF)1 using activity in white matter as VND. Additionally, four alternate measurements were computed: BPND with a population-derived input function (BPNDPBIF), standard uptake value (SUV), SUV ratio to white matter (SUVr) and BPND‐‐‐ using white matter as non-specific input (BPNDWM).2 Four animals were scanned twice providing test-retest reproducibility data.
Results: Free fraction was stable across animals (0.093 ± 0.012). Parent and metabolite fractions were similar in arterial (N = 9) and venous samples (N = 3) (Figure 1(a)). There was good reproducibility between the test and retest scans for BPNDAIF (ICC = 0.721) BPNDPBIF (ICC = 0.699) SUVr (ICC = 0.855) and BPNDWM (ICC = 0.868) but not for SUV(ICC = 0.225). There was a good correlation between the BPNDAIF and BPNDPBIF within scans (mean r across animals = 0.944 ± 0.15) but only moderate correlation for SUV, SUVr and BPNDWM (mean r across animals ranging between 0.64 ± 0.09 and 0.71 ± 0.09). The shape of the late portion of the non-corrected plasma curve greatly resembled that of the white matter activity confirming that at least some [11C]-PBR28 metabolite(s) cross the blood-brain barrier (Figure 1(b)).
Conclusion: In NHP, the stability of peripheral metabolism permits the use of a PBIF to measure BPND. The high correlation between BPNDAIF and BPNDPBIF and the good test-retest value suggest that a population curve may be a good alternative to individual sampling to study the effect of intervention within subjects. Additionally, arterial and venous plasma appears to be equally useful to provide an estimate of peripheral metabolism. Further work on this data set will test the performance of SIME.
Activity curves
Parent and metabolite fraction
References
LoganJFowlerJSVolkowND, et al.Graphical analysis of reversible radioligand binding from time-activity measurements applied to [N-11C-methyl]-cocaine PET studies in human subjects. J Cereb Blood Flow Metab1990;
10: 740–747.LoganJFowlerJSVolkowD, et al.Distribution volume ratios without blood sampling from graphical analysis of PET data. J Cereb Blood Flow Metab1996;
16: 843–840.
2020-49
Type 5 metabotropic glutamate receptor availability in youth at risk for addictions: Effects of vulnerability traits and cannabis use (#217)
Sylvia Cox1, Maria Tippler1, Natalia Jaworska2, Kelly Smart3, Natalie Castellanos-Ryan4, France Durand1, Dominique Allard1, Chawki Benkelfat1, Sophie Parent4, Alain Dagher5, Frank Vitaro4, Michel Boivin6, Robert Pihl7, Sylvana Côté8, Richard Tremblay9, Jean Séguin10 and Marco Leyton1, 5
1Department of Psychiatry, McGill University, Montreal, Québec, Canada
2Institute of Mental Health Research, University of Ottawa, Ottawa, ON, Canada
3Yale PET Center, Yale University, New Haven, CT, USA
4School of Psychoeductation, Université de Montréal, Montreal, Québec, Canada
5Department of Neurology and Neurosurgery, McGill University, Montreal, Québec, Canada
6Department of Psychology, Université Laval, Québec, Canada
7Department of Psychology, McGill University, Montreal, Québec, Canada
8Department of Social and Preventative Medicine, Université de Montréal, Montreal, Québec, Canada
9Department of Pediatrics and Psychology, Université de Montréal, Montreal, Québec, Canada
10Department of Psychiatry and Addictology, Université de Montréal, Montreal, Québec, Canada
Abstract
Introduction: The excitatory neurotransmitter glutamate has been implicated in experience-dependent neuroplasticity and drug-seeking behaviors. Type 5 metabotropic glutamate (mGlu5) receptors might be particularly important. In humans, mGlu5 receptor availability has been reported to be lower in people with alcohol, tobacco, and cocaine use disorders. Since the reductions could reflect effects of drug use or pre-existing traits, we used positron emission tomography (PET) to measure mGlu5 receptor availability in young adults at elevated risk for addictions.
Methods: Fifty-nine participants (18–20 y.o.) were recruited from a longitudinal study that has been followed since birth. Based on diverse externalizing behaviors that predict future substance use problems, half of the participants were at low risk, half were at high risk. Cannabis use histories varied markedly, and participants were divided into three groups: zero, low and high use. Participants were scanned using high-resolution research tomograph PET with tracer 3-(6-methyl-pyridin-2-ylethynyl)-cyclohex-2-enone-O-11C-methyl-oxime ([11C]ABP688) and 3T magnetic resonance imaging for anatomical co-registration.
Results: Compared to low risk volunteers, those at elevated risk for substance use disorders had lower [11C]ABP688 binding potential (BPND) values in cortico-limbic regions including the striatum, amygdala, insula, and orbitofrontal cortex (OFC). Cannabis use by risk group interactions were observed in the striatum and OFC. In these regions, high cannabis use was associated with low [11C]ABP688 binding in the high risk group only. When the high risk, high cannabis using individuals (n = 9) were compared to all other participants (n = 50), [11C]ABP688 BPND values were significantly lower in the striatum, OFC, and insula. These effects were evident when controlling for the use of other substances (Figure 1).
Conclusion: The results indicate that mGlu5 receptor availability is low in youth at elevated risk for addictions, particularly those who are frequently using cannabis.
2020-50
Morphine administration increases translocator protein availability in humans (#218)
Eric A Woodcock1, Gustavo A Angarita1, David Matuskey1, 2, Yiyun H Huang2, Ansel T Hillmer1, 2, Richard E Carson2, 3 and Kelly P Cosgrove1, 2
1Psychiatry, Yale University, New Haven, CT, USA
2PET Center, Yale University, New Haven, CT, USA
3Radiology and Biomedical Imaging, Yale University, New Haven, CT, USA
Abstract
Introduction: There is tremendous need to investigate novel treatment targets for opioid use disorder, such as the neuroimmune system. In preclinical studies, opioid administration reliably evokes pro-inflammatory responses in the periphery and brain. These pro-inflammatory responses have been shown to influence appetitive (e.g., opioid-seeking) and dysphoric (e.g., pain and withdrawal) addiction processes which perpetuate opioid use. In this paradigm development study, we investigated the neuroimmune effects of acute opioid administration using Positron Emission Tomography (PET) imaging with [11C]PBR28, a radiotracer that binds to the 18kDa translocator protein (TSPO), a marker of microglia.
Methods: Healthy individuals with prior medical opioid exposure (n = 4; 3M; 2 ‘high-affinity’ binders; Age = 30 years; BMI = 26.5 [range = 24-30]) completed two 120-minute [11C]PBR28 PET scans in one day: before and 2-hours after intramuscular morphine (0.07mg/kg). Arterial blood was acquired to measure the metabolite-corrected input function. Volumes of distribution (VT), i.e., TSPO availability, were calculated in 10 regions of interest (ROIs) using multilinear analysis–1 (t* = 30). Regional [11C]PBR28 VT values were evaluated using a repeated-measures analysis of variance with rs6971 genotype as a fixed factor (‘high’ vs. ‘mixed affinity’ TSPO binders). Subjective and behavioral effects were also assayed before and after morphine.
Results: Morphine increased TSPO availability by 28%-39% across ROIs, F(1,2) = 9.56, p = .09, partial η2 = 0.83, ‘very large’ effect (Figure 1). Morphine increased withdrawal latency on the Cold Pressor Task, i.e., pain tolerance, F(1,2) = 3.98, p = .18, partial η2 = 0.66, ‘very large’ effect. Morphine decreased systolic and diastolic blood pressure by 11 mmHg and 7 mmHg, respectively (ps < .09; Figure 2(a)). Subjective drug effects were modest (Figure 2(b)).
Conclusion: These preliminary findings suggest that morphine evokes a neuroimmune response in people. The magnitude of TSPO increase observed herein is comparable to prior research in non-human primates. Notably, TSPO availability increased in all brain regions, which suggests that morphine’s neuroimmune effects are not limited to regions with µ opioid receptor binding, consistent with non-human primate findings. The morphine dose administered (<6mg) is comparable to a standard-of-care post-operative analgesic dose. Indeed, preliminary data suggest morphine enhanced thermal pain tolerance. Future studies could use this paradigm to investigate the role of neuroimmune signaling in addiction processes among individuals with opioid use disorder.
Acknowledgements
The authors acknowledge the staff at the Yale University PET center, Jon Mikael Anderson, Aleksandra Rusowicz, Brittany LeVasseur, Stephen Baldassarri, Olivia Wilson, Nicole DellaGioia, Ryan Cool, Sarah DeBonee, and Elizabeth Yanac.
Funding was generously provided by the National Institute on Drug Abuse (K99 DA048125; EAW), the National Institute of Mental Health (R01 MH110674; KPC), and the Veterans Affairs National Center for PTSD (KPC).
2020-51
Evaluation of PET quantitation methods using a 3D-printed anatomically accurate brain phantom (#234)
Davneet Minhas1, Anish Ghodadra1, Zheming Yu1, Sarah K Royse1, Howard Aizenstein2, Ann Cohen2, Dana Tudorascu3, Brian Lopresti1, Chester A Mathis1, William E Klunk2 and Charles Laymon1
1Department of Radiology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
2Department of Psychiatry, University of Pittsburgh, Pittsburgh, PA, USA
3Department of Medicine, University of Pittsburgh, Pittsburgh, PA, USA
Abstract
Introduction: A variety of reconstruction- and image-based methods have been developed to improve the resolution of PET. However, there remains a lack of consensus on their use due to a lack of ground truth-based evaluation. In this work, we assess techniques using a human-based PET phantom with physiologically relevant radiotracer distributions.
Methods: A lateral temporal phantom with six fillable, thin-walled (1.0 mm) chambers corresponding to inferior, middle, and superior temporal gray matter (GM) and white matter (WM) was constructed via 3D-printing (Formlabs Form 3) using the FreeSurfer v5.3 segmented sample subject “bert” as a pattern. Inferior and superior temporal GM regions were filled with a 2.2 kBq/cc concentration of F-18 solution and WM regions with 1.2 kBq/cc at time of scan, simulating a high amyloid-burden subject with a GM-to-WM [11C]PiB radioactivity ratio of 1.8. The phantom was scanned (Siemens Biograph mCT) for 180 minutes. Data were reconstructed using three techniques – FBP; OSEM (4 and 5 iterations); and TrueX (4 and 5 iterations), an iterative method incorporating a spatially variant point-spread function – into 7 frames with each containing the counts expected in a 20-minute [11C]PiB scan. Image-based geometric transfer matrix (GTM) partial-volume correction (PVC) was applied to the FBP and OSEM images. Average regional radioactivity concentrations across frames, uncorrected and PVC, were compared to known ground truth. Method-associated variability was assessed using coefficients of variation (CoV) across frames.
Results: CT (empty phantom), FBP, OSEM, and TrueX images are presented in Figure 1. Average regional radioactivity for each of the reconstruction methods, uncorrected and PVC, are presented in Figure 2.
Conclusion: GTM PVC resulted in overestimated but more accurate regional quantification relative to uncorrected data, regardless of reconstruction method. GTM PVC also resulted in increased variability with CoVs 1.2-to-1.9 times that of uncorrected CoVs across methods and regions. TrueX improved apparent image quality but not quantitative accuracy relative to uncorrected FBP and OSEM. This study adhered to one fundamental assumption of the GTM technique: radioactivity uniformity within a region. This assumption may be violated in human imaging. Future studies will address this and will examine the effect of varying levels of cortical atrophy.
Acknowledgements
This study was funded by United States National Institutes of Health (NIH) grant number P01 AG025204-13S.
2020-52
Alterations in cerebral blood flow in patients with secondary progressive and stem cells treated multiple sclerosis – A perfusion study with 15O-water-PET (#237)
Lieuwe Appel1, 2, Andreas Tolf3, 4, Torsten Danfors1, 2, Elna-Marie Larsson1, 2, Anne-Marie Landtblom3, 4, Gunnar Antoni2, 5, Jens Sörensen1, 2, Joachim Burman3, 4 and Mark Lubberink1, 6
1Surgical Sciences, Uppsala University, Uppsala, Sweden
2Medical Imaging Centre, Uppsala University Hospital, Uppsala, Sweden
3Neuroscience, Uppsala University, Uppsala, Sweden
4Neurology, Uppsala University Hospital, Uppsala, Sweden
5Medicinal Chemistry, Uppsala University, Uppsala, Sweden
6Medical Physics, Uppsala University Hospital, Uppsala, Sweden
Abstract
Introduction: Multiple sclerosis (MS) is an autoimmune and degenerative disorder where the immune system malfunctions attack the central nervous system, leading to demyelination and neuronal/axonal loss. Impaired cerebral blood flow (CBF) has been associated with disease progression in neurodegenerative disorders. Chemotherapy combined with haematopoietic stem cell transplantation (HSCT) was introduced as an alternative treatment in 2004 at our hospital. Nowadays there is a growing evidence that HSCT is a successful treatment, but the question remains whether it is an ultimate cure for MS. This study aimed to investigate differences in CBF between three cohorts: patients with secondary progressive MS (SPMS), treated patients (HSCT) and healthy controls (HC).
Methods:15O-water-PET scans with arterial blood sampling were acquired at rest for 10 SPMS, 9 HSCT (10 years after treatment), and 10 HC (age and gender matched). Arterial input functions were corrected for delay and dispersion. Hence, CBF and distribution volume (VT) values were generated using a standard single-tissue compartment model including a fitted blood volume parameter. T1-weighted MR images were used for segmentation and automated definition of volumes of interest. Nonparametric tests were used to detect significant differences between cohorts (p < 0.05).
Results: Global CBF, including gray and white matter, was significantly lower in SPMS compared to HSCT, whereas no significant differences were found between HSCT and HC. Regional CBF revealed similar patterns for SPMS, and significant differences with HC and HSCT were especially found in subcortical regions – Figure 1. Further, VT was significantly higher in HSCT compared to SPMS, both global and regional – Figure 2.
Conclusion: These preliminary results revealed an impaired CBF and decreased VT in SPMS. Similar levels of CBF and VT for HSCT and HC indicates that HSCT treatment might restore overall brain functionality. In part, the current findings also confirm the outcome of clinical investigations/observations.1,2 Specific CBF and VT alterations in gray matter suggest a relationship with various disabilities in MS patients.
Acknowledgements
Thanks to the MS-patients and healthy volunteers for their study participation, anesthetists for a careful setup of arterial lines and the imaging staff for a skillfull data acquisition.
References
BurmanJ, et al.Autologous haematopoietic stem cell transplantation for aggressive multiple sclerosis: the Swedish experience. J Neurol Neurosurg Psychiatry2014;
85: 1116.TolfA, et al.Sustained remission in multiple sclerosis after hematopoietic stem cell transplantation. Acta Neurol Scand2019;
40: 320–327.
2020-53
Development of novel PET imaging agents for neurodegeneration (#239)
Christopher Liang, Rommani Mondal, Krystal Patel and Jogeshwar Mukherjee
Preclinical Imaging, Radiological Sciences, University of California, Irvine, CA, USA
Abstract
Introduction: Proteinopathies, including α-synuclein aggregates, a hallmark Parkinson’s disease (PD) and tau aggregates in Alzheimer’s disease related disorders (ADRD) are a target for PET radiotracer development. Coexistence of these aggregates in neurodegeneration has been suggested,1,2 but recent flortaucipir PET findings suggest lack of tau aggregates in PD patients.3 In an effort to develop PET agents for ADRD and PD, we examined molecules containing overlapping Aβ-amyloid, dopamine receptor or tau binding structures. Here we report a new fluoroalkylazaindole ([18F]FAZIN3) as a potential PET imaging agent for neurodegeneration.
Methods: AZIN3-Tosyl was radiolabeled with 10–20 mCi of [18F]fluoride in acetonitrile containing Kryptofix/K2CO3. [18F]FAZIN3 (0.5–1 mCi) was purified by HPLC. Male C57BL/6 mice were used for in vivo Inveon PET/CT studies. Human post-mortem brain tissues consisting of anterior cingulate (AC) and corpus callosum (CC) were used (controls (CN), n = 6; PD, n = 6; AD, n = 6). Brain slices were incubated [18F]FAZIN3 (1 µCi/cc) in PBS (pH 7.4) buffer at 25°C for 60 min. Competition with clorgyline, deprenyl and MK-6240 were also carried out.
Results: PET/CT scans showed clearance of [18F]FAZIN3 from the brain with low retention in any brain region (Figure 1). No radioactivity was observed in the bones suggesting that [18F]FAZIN3 was resistant to defluorination (Figure 1). All CN subjects exhibited lower [18F]FAZIN3 binding in the AC compared to the PD and AD subjects (Figure 2). In the case of PD subjects, AC exhibited a more than 2-fold increase in binding of [18F]FAZIN3 (PD/CN = 2.12) for AC. Similarly, AD subjects showed a two-fold increase in [18F]FAZIN3 binding (AD/CN = 2.04). Binding of [18F]FAZIN3 was reduced > 90% in both AD and PD brain slices using clorgyline, a monoamine oxidase A (MAO-A) inhibitor. Deprenyl (MAO-B inhibitor) and MK-6240 had little effect on [18F]FAZIN3 binding.
Conclusion: We have developed a novel PET imaging agent, [18F]FAZIN3. The exquisite binding of [18F]FAZIN3 in the PD and AD brain suggests unique upregulation of MAO-A. Thus, PET imaging studies using [18F]FAZIN will be complimentary to studies reported with flortaucipir3 and [18F]MK-62404,5 in understanding the role of MAO-A in neurotransmitter depletion, neuronal damage from reactive oxygen species and inflammation in AD and PD.
Acknowledgements
NIH/NIA RF1 AG029479 (JM).
[18F]FAZIN3 Mouse PET/CT Images
Postmortem Human Brain Autoradiographs
References
Castillo-CarranzaDL, et al.
a-Synuclein oligomers induce a unique toxic tau strain. Biol Psychiatry2018;
84: 499–508.MoussaudS.
Alpha-synuclein and tau: teammates in neurodegeneration?Mol Neurodegener2014;
9: 43.HansenAK, et al.Tau tangles in Parkinson’s disease: a 2-year follow-up flortaucipir PET study. J Parkinson’s Dis2020;
10: 161–171.AgueroC, et al.Autoradiography validation of novel tau PET tracer [F-18]-MK-6240 on human postmortem brain tissue. Acta Neuropathologica Comm2019;
7:37, 1-15.WaljiAM, et al.Discovery of 6-(Fluoro-18F)-3-(1H-pyrrolo[2,3-c]pyridin-1-yl)isoquinolin-5-amine ([18F]-MK-6240): A positron emission tomography (PET) imaging agent for quantification of neurofibrillary tangles (NFTs). J Med Chem2016;
59: 4778–4789.
2020-54
Evaluation of neuroinflammation in HIV patients and “elite controllers” using [11C]-PBR28 PET (#241)
Hasan Sari1, Yang Lin1, Andrew Salvatore1, Zeynab Alshelh1, Angel Torrado-Carvajal1, 2, Shibani Mukerji3, Eva-Maria Ratai1, Jacqueline Chu4, Rajesh Gandhi4, Oluwaseun J Akeju5, Atreyi Saha1, Vitaly Napadow1, Robert Edwards5, Julie C Price1, Marco L Loggia1 and Cristina Granziera6
1MGH/HST Martinos Center for Biomedical Imaging, Charlestown, MA, USA
2Laboratory of Medical Image Analysis, Universidad Rey Juan Carlos, Madrid, Spain
3Department of Neurology, Massachusetts General Hospital, Boston, MA, USA
4Department of Infectious Diseases, Massachusetts General Hospital, Boston, MA, USA
5Department of Anaesthesia, Massachusetts General Hospital, Boston, MA, USA
6Department of Biomedical Engineering, University of Basel, Allschwil Basel-Land, Switzerland
Abstract
Introduction: [11C]-PBR28 is a PET radioligand with a high affinity for the translocator protein (TSPO), which can mark the presence of neuroinflammation. In recent studies, an increase in TSPO expression was shown in people with HIV (PWH) compared to controls.1,2 In this work, we evaluate [11C]-PBR28 PET standardized-uptake value ratios (SUVR) in PWH including a group of “elite controllers (EC)”, a group of participants who are able to maintain low viral loads without treatment.
Methods: Thirteen PWHs (age: 55 ± 5.9 years, sex:12M/1F, genotype: 10HAB/3MAB), 18 healthy controls (HC) (age: 48 ± 13.5 years, sex:10M/8F, genotype: 11HAB/5MAB) and 6 ECs (age: 59 ± 7.3 years, sex 5M/1F, genotype: 4HAB/2MAB) were scanned using an integrated PET/MR scanner. All PWHs were aviremic. Standardized-uptake value (SUV) images were generated (60–90min post-injection) and transformed to Montreal Neurological Institute (MNI) space using FSL.3 To generate a pseudo-reference region, an unpaired t-test was performed to identify voxels with no statistically difference between HCs and PWHs. Given the large preponderance of males in our sample, only datasets from male participants were used in this comparison. The resultant z-maps were thresholded to include voxels between -0.2 and 0.2 (p > 0.84).4 The resulting region was used to normalize the region-of-interests (ROIs) and compute SUVRs. Regions with significant difference between PWH and HCs were identified and Dunnett’s test was used to test if ECs were different from remaining groups.
Results: Regional SUVR values for each group are shown in Table 1. After Bonferroni correction, statistically significant group differences in [11C]PBR28 SUVRs were observed between HCs and PWHs in 4 of the 9 ROIs. The most significant differences were seen in thalamus, putamen and parietal operculum cortex (p < 0.01, corrected). Figure 1 shows box-plots of SUVRs in these regions. ECs demonstrated significantly less inflammation (p < 0.05) compared to the PWHs in thalamus, putamen, and parietal operculum cortex.
Conclusion: The results of this work demonstrate significantly greater [11C]-PBR28 signal in PWHs compared to HCs, suggestive of neuroinflammation. ECs demonstrated lower [11C]PBR28 signal than PWHs with higher viral loads, suggesting that these patients’ ability to naturally control their viral load might be also accompanied by lower levels of neuroinflammation.
Acknowledgements
This work was supported NIH grant R01DA047088-01 (ML and EMR).
Table 1. Regional 11C-PBR28 SUVR values for healthy control and HIV positive patients, grouped for each genotype. 2-way ANOVA test was used to compute p-values using genotypes as the fixed factor. Regions with significant group differences after Bonferroni correction are marked with *.
References
VeraJHGuoQ.
Neuroinflammation in treated HIV-positive individuals: a TSPO PET study. Neurology2016;
86: 1425–1432.CoughlinJMWangY.
Regional brain distribution of translocator protein using [11C]DPA-713 PET in individuals infected with HIV. J NeuroVirol2014;
20: 219–232.WoolrichMWJbabdiS.Bayesian analysis of neuroimaging data in FSL. NeuroImage2009;
45: S173–S86.AlbrechtDSMaineroC.Imaging of neuroinflammation in migraine with aura: a [11C]PBR28 PET/MRI study. Neurology2019;
92: 2038–2050.
2020-55
Extension of pseudo-CT method for attenuation correction of simultaneous PET/MR brain imaging in rhesus macaques (#244)
Jonathan M DuBois1, Lu Wang2, William Zhou1, Steven Liang3, David Izquierdo-Garcia1 and Hsiao-Ying Wey1
1Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Harvard Medical School, Massachusetts General Hospital, Charlestown, MA, USA
2Department of Nuclear Medicine and PET/CT-MRI Center, The First Affiliated Hospital of Jinan University & Institute of Molecular and Functional Imaging, Jinan University, Guangzhou, China
3Division of Nuclear Medicine and Molecular Imaging, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
Abstract
Introduction: Non-human primates (NHP) are a valuable model for neuroimaging research, allowing neurobiological investigation of brain function as well as facilitating the development and translation of new biomarkers and therapeutics. Unfortunately, many of the methodological advances of human brain imaging have not yet been applied to NHPs. While MR-based attenuation correction (AC) methods have been developed for human imaging, AC remains a major challenge to accurate quantification of PET data acquired on a PET/MR scanner for NHPs. To address this, we have extended our well validated human pseudo-CT method1,2 to develop MR and CT population templates for the rhesus macaque. This method combines a 6 tissue-class segmentation and diffeomorphic non-rigid deformation from a single MRI to generate a pseudo-CT image.1 The pseudo-CT attenuation (µ) maps from MR data can be used in PET/MR scanners to improve PET quantification for NHP studies.
Methods: Ten NHPs (rhesus macaque) were imaged on a PET/CT (GE Discovery 690) and on a 3T MRI (GE Discovery 750). The T1-weigthed MR images and CT images were segmented to obtain 6 tissue-classes: WM, GM, CSF, bone, soft-tissue, and air. The tissue classes from all animals were non-rigidly co-registered using a diffeomorphic approach to generate tissue class templates and a CT template (using SPM8). CT-like (pseudo-CT) images can be obtained by applying the inverse transformation, which were converted to linear attenuation coefficients to be used for AC of PET data.3 Bland-Altmant plots were calculated to test accuracy using a leave-one-out cross-validation (LOOCV) analysis across all 10 subjects.
Results: Qualitatively, the pseudo-CT template closely resembles the real CT image in individual subjects (Figure 1), which is consistent with previous results shown in the human brain. Voxel-wise LOOCV of the 10 NHPs showed only small errors in the brain, with an absolute relative change of 1.6% when comparing the pseudo-CT to the real CT.
Conclusion: We present an approach for deriving the head pseudo-CT µ map for rhesus macaques, which will allow for robust AC in integrated PET/MR scanners. We are currently working on comparing the accuracy of this technique against PET-based AC methods in NHPs to evaluate the impact on PET quantification.
References
Izquierdo-GarciaD, et al.An SPM8-based approach for attenuation correction combining segmentation and nonrigid template formation: application to simultaneous PET/MR brain imaging. J Nucl Med2014;
55: 1825–1830.LadefogedCN, et al.A multi-centre evaluation of eleven clinically feasible brain PET/MRI attenuation correction techniques using a large cohort of patients. NeuroImage2016;
147: 346–359.BurgerC, et al.PET attenuation coefficients from CT images: experimental evaluation of the transformation of CT into PET 511-keV attenuation coefficients. Eur J Nucl Med Mol Imaging2002;
29: 922–927.
2020-56
Validation of kinfitr: An open-source tool for reproducible PET kinetic modelling (#246)
Jonathan Tjerkaski, Simon Cervenka, Lars Farde and Granville J Matheson
Department of Clinical Neuroscience, Centre for Psychiatry Research, Karolinska Institutet, Stockholm, Sweden
Abstract
Introduction: In PET imaging, binding is typically assessed by fitting pharmacokinetic models to time-activity curves (TACs). However, there are wide variety of different models, and a multitude of other analytical decisions which must be made before these models can be applied. As such, there is a need for both analytical flexibility and transparency. The recently developed kinfitr1 is an open-source tool designed to provide this functionality, allowing reproducible PET modelling using the R programming language. Here, we validate the performance of kinfitr against the widely-used PMOD software.2
Methods: We used previously-collected TAC data from test-retest studies of four different radioligands, using six kinetic models. Two of these tracers ([11C]SCH23390, n = 15, and [11C]AZ10419369, n = 8) were modelled using non-invasive models, and two ([11C]PBR28, n = 12, and (R)- [11C]PK11195, n = 6) were modelled using invasive models with metabolite-corrected arterial input function. We assessed the agreement of estimated binding outcomes and microparameters, as well as test-retest performance for each tool, and agreement between models within each tool.
Results: We obtained high correlations between kinfitr and PMOD for both non-invasive and invasive models (median Pearson’s r = 0.99, range: 0.95–1.00). Linearised methods exhibited lower agreement due to differences in the selection of t* values between tools. No substantial differences were observed in between- or within-individual variation. We obtained high correlations between the microparameters of the simplified reference tissue model (mean r > 0.99) and of the two-tissue compartment model (mean r: K1 > 0.99, k2 0.81, k3 0.80, k4 0.88). Strong agreement was also observed between different models using each tool independently, for both non-invasive (median r: kinfitr 0.99, PMOD 0.99) and invasive models (median r: kinfitr 0.99, PMOD 0.79).
Conclusion: There was excellent agreement between kinfitr and PMOD. We therefore conclude that kinfitr is a valid and reliable tool for kinetic modelling of PET data. kinfitr is operated using R code, thereby ensuring computational reproducibility3: all analysis steps are transparently recorded and can be reproduced by re-running the code. In this way, re-analysis using different parameters or models only requires modification of the relevant lines of code, rather than repeating all steps of the analysis anew.
Acknowledgements
This research was supported by the Swedish Society of Medicine (Svenska Läkaresällskapet).
References
MathesonGJ. kinfitr: reproducible PET pharmacokinetic modelling in R. bioRxiv2019; 755751MikolajczykKSzabatinMRudnickiP, et al.A JAVA environment for medical image data analysis: initial application for brain PET quantitation. Med Inform (Lond)1998;
23: 207–214SandveGKNekrutenkoATaylorJ, et al.Ten simple rules for reproducible computational research. PLoS Comput Biol2013;
9: e1003285.
2020-57
Lower thalamic dopamine D2-receptor binding and connectivity in drug-naive patients with psychosis – A combined [11C]FLB 457 PET and DTI study (#248)
Pauliina Victorsson1, Pontus Plavén-Sigray1, 2, Alexander Santillo1, 3, Granville J Matheson1, Maria Lee1, Karin Collste1, Helena Fatouros-Bergman1, Ingrid Agartz1, 4, Christer Halldin1, Lars Farde1 and Simon Cervenka1
1Department of Clinical Neuroscience, Centre for Psychiatry Research, Karolinska Institutet, Stockholm, Sweden
2Neurobiology Research Unit, Copenhagen University Hospital, Copenhagen, Denmark
3Clinical Memory Research Unit, Lund University, Lund, Sweden
4Norwegian Centre for Mental Disorders Research (NORMENT), Institute of Clinical Medicine, University of Oslo, Oslo, Norway
5Department of Psychiatric Research, Diakonhjemmet Hospital, Oslo, Norway
Abstract
Introduction: Previous PET studies have demonstrated lower levels of the dopamine D2-receptor (D2-R) in thalamus in patients with schizophrenia.1 However, some studies were performed using radioligands with suboptimal affinity or included patients on antipsychotic treatment. Moreover, the relationship to cortical connectivity has not been investigated. Here we examined antipsychotic-naïve first-episode psychosis patients using PET and the high-affinity D2-R radioligand [11C]FLB457, aiming to a) replicate findings of lower D2-R in whole thalamus in patients, and b) investigate D2-R differences in connectivity-based thalamic subregions. Moreover, thalamocortical connectivity was examined using diffusion tensor imaging (DTI) in a subset of individuals.
Methods: Nineteen patients and 19 age- and sex matched healthy comparison subjects were examined using a High Resolution Research Tomograph (HRRT). Whole thalamus was defined using the Harvard Oxford Subcortical Atlas, whereas thalamic subregions were based on the Oxford Thalamic Connectivity Atlas. Binding potential (BPND) was calculated using the Logan graphical analysis with cerebellum as reference region. Statistical analyses (pre-registered at https://osf.io/nhr3w/) were performed using frequentist and Bayesian paired-samples t-tests. In 11 patients and 15 healthy subjects the anterior and inferior thalamic radiation were manually extracted using DTI whole brain tractography.
Results: Patients had significantly lower binding than control subjects in whole thalamus (Cohen’s D = −0.479, p = 0.026). The Bayes Factor indicated approximately 5 times more support for the hypothesis of lower BPND in patients, compared to the null hypothesis. Among subregions, the ROI corresponding to frontal thalamic connectivity showed the largest effect (Cohen’s D = −0.527, p = 0.017, Bayes Factor > 6). Patients showed significantly lower fractional anisotropy values compared with controls (Cohen’s D = −0.692, p = 0.036) in the inferior thalamic radiation, which predominantly projects to and from the orbitofrontal, temporal and insular cortex.
Conclusion: This study replicates previous findings of lower thalamic D2-R availability in patients. The strongest effect was observed in the subregion dominated by connections to prefrontal cortex, and aberrations in thalamocortical connectivity was confirmed in a subsample. The findings may reflect a dysregulation of the thalamic dopamine system in schizophrenia, which in turn could affect functional thalamocortical connectivity.
Acknowledgements
The contributions from the staff at the Karolinska PET Centre, the staff at the participating psychiatric units in Stockholm (Prima Vuxenpsykiatri, Psykiatri Nordväst and Norra Stockholms Psykiatri), as well as the members of Karolinska Schizophrenia Project are gratefully acknowledged.
Reference
KambeitzJAbi-DarghamAKapurS, et al.Alterations in cortical and extrastriatal subcortical dopamine function in schizophrenia: systematic review and meta-analysis of imaging studies.Br J Psychiatry2014;
204: 420–429.
2020-58
Tau deposition assessed by [18F]MK6240 PET is associated with longitudinal decrease in grey matter density across the spectrum of Alzheimer’s disease (#250)
Firoza Z Lussier1, Tharick A Pascoal1, Joseph Therriault1, Andréa L Benedet1, Cécile Tissot1, Yi-Ting Wang1, Jaime F Arias1, Melissa Savard1, Serge Gauthier2 and Pedro Rosa-Neto1
1Translational Neuroimaging Laboratory, McGill Research Centre for Studies in Aging, Verdun Québec, Canada
2Alzheimer’s Disease Research Unit, McGill Research Centre for Studies in Aging, Verdun Québec, Canada
Abstract
Introduction: Reduction in grey matter (GM) is a well-established neuroimaging finding in Alzheimer’s disease (AD). Here, we sought to determine whether baseline tau-PET using the high-affinity tracer [18F]MK6240 is predictive of longitudinal changes in GM across the AD spectrum.
Methods: Baseline [18F]MK6240 PET data and longitudinal structural MRI was acquired for 79 participants in the TRIAD cohort (47 CN, 20 MCI, 12 AD). [18F]MK6240 standardized uptake value ratio (SUVR) were calculated 90–110 minutes post-injection using cerebellar GM as the reference region. T1-weighted MR images were segmented into probabilistic GM and WM maps, which were non-linearly registered to the ADNI template using DARTEL and smoothed with an 8 mm FWHM Gaussian kernel. Voxel-based morphometry (VBM) was run on GM and WM maps. Longitudinal changes in GM density were indexed by voxel-wise percentage change in GM VBM. Voxel-based regression analyses were conducted to examine associations between baseline [18F]MK6240 SUVR and change in GM density, with age, gender, years of education, diagnosis, and time interval between MRI acquisitions employed as covariates.
Results: Baseline [18F]MK6240 SUVRs in Braak I/II, III/IV, and V/VI were significantly correlated with longitudinal decrease in GM density in the lateral and medial temporal lobe, including hippocampal regions. All results survived correction for multiple comparisons using random field theory at p < 0.001.
Conclusion: Our results suggest that tau pathology at baseline is associated with the progression of GM atrophy over time in brain regions vulnerable to AD pathological changes, providing evidence that tau-PET may identify individuals who are more susceptible to atrophy.
Tau assessed by [18F]MK6240 is associated longitudinal decreases in grey matter density
Results from voxel-based linear regression analyses between [18F]MK6240 SUVR in (A) Braak I/II, (B) Braak III/IV, (C) Braak V/VI and longitudinal change in GM density. Positive associations were found bilaterally in the lateral and medial temporal lobe. (Greater change in GM density represented decrease in density). Analyses have been corrected for multiple comparisons using random field theory at p < 0.001.
2020-59
Pre-clinical evaluation of novel COX-2 PET radiotracers for imaging neuroimmune dysregulation (#255)
Michael S Placzek1, 2, Michel Weiwer3, Misha M Riley1, Daniel K Wilton2, 4, Florence Wagner3, Beth Stevens2, 4 and Jacob M Hooker1, 2
1Department of Radiology, Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, USA
2Harvard Medical School, Boston, MA, USA
3Broad Institute of MIT and Harvard, Stanley Center for Psychiatric Research, Cambridge, MA, USA
4Department of Neurology, F. M. Kirby Neurobiology Center, Boston Children’s Hospital, Boston, MA, USA
Abstract
Introduction: In the brain, COX-2 is specifically upregulated in response to inflammatory events, including those that occur in neurodegenerative diseases.1 COX-2-targeting radiotracers have been pursued by several groups, but translation to human imaging has yet to succeed.2 With a focus on high COX-2 binding affinity and fast on-rate, we discovered BRD1158, with improved potency and faster on-rate for COX-2 compared to rofecoxib, celecoxib, and MC1. In vitro and in vivo evaluation in rodents and nonhuman primates demonstrated [11C]BRD1158 has high brain uptake and specific binding for COX-2, warranting further evaluation.
Methods: [11C]BRD1158 was radiolabeled at the methylsulfone by treating the corresponding thioester precursor with [11C]CH3I. [11C]BRD1158 in vitro autoradiography with slide-mounted brain tissue slices (rat and human) was performed by incubating tissue with either radioligand+vehicle or radioligand+blocking (10 µM self-block). Huntington’s Disease post-mortem brain tissue (globus pallidus) and age matched healthy control was used for autoradiography. To evaluate [11C]BRD1158 brain uptake we measured whole brain SUV in rats and baboons with PET-MR. Regional brain time activity curves for both rat and baboon were generated with PMOD using PxRat atlas and b2k baboon atlas.
Results: Using a COX-2 enzymatic assay (2 min pre-incubation), BRD1158 was a potent COX-2 inhibitor (IC50 = 20 nM), compared to rofecoxib (600 nM), and MC1 (43 nM). In vitro autoradiography (Figure 1) with rat brain slices demonstrated appreciable specific binding (22 ± 11%; n = 6 slices). In a pilot experiment with human HD tissue (1 subject), we observed robust [11C]BRD1158 specific binding in globus pallidus tissue (23 ± 5%, n = 6 slices). In vivo, [11C]BRD1158 had high brain uptake in rat (whole brain SUV Cmax = 1.9) and baboon (whole brain SUV Cmax = 3.3), and favorable washout kinetics for a brain PET radiotracer (rat t1/2 = 7.8 min; baboon t1/2 = 17.2 min) (Figure 2).
Conclusion: We have identified BRD1158 as a potent and selective COX-2 inhibitor that has high brain uptake in both rodent and baboon. [11C]BRD1158 also showed moderate specific binding in both rat and human tissue, and further evaluation is underway to prepare for eIND and first-in-human COX-2 imaging.
Acknowledgements
The authors would like to acknowledge support from the Stanley Center for Psychiatric Research at Broad Institute, Huntington’s Disease Society of America, and NIH-NINDS 1R01NS111168-01A1.
References
BartelsALeendersK.Cyclooxygenase and neuroinflammation in Parkinson’s disease neurodegeneration. Curr Neuropharmacol2010;
8: 62–68.JanssenBVugtsDWindhorstA, et al.PET imaging of microglial activation – beyond targeting TSPO. Molecules2018;
23: 607.
Author Index
A
Abe, J.
2021-18 39
2021-21 42
2021-92 119
Abrahamsson, N.
2021-24 45
Aceves-Serrano, L.G.
2020-48 250
2021-110 139
Adamo, S.
2021-163 193
Adewale, Q.
2021-122 151
Aeschbach, D.
2021-28 49
2021-30 50
Agalliu, D.
2021-77 102
Agartz, I.
2020-57 260
Aguirre-Aranda, I.
2021-33 53
Ahmed, H.
2021-22 43
Aiello, M.
2020 Abstract 199
2021-65 89
Aizenstein, H.
2020-51 253
Akeju, O.J.
2020-12 211
2020-54 256
2021-151 181
Albrecht, D.
2020-12 211
2021-115 144
2021-151 181
Alfaifi, B.Q.
2021-85 110
Aliaga, A.
2021-11 31
2021-161 190
2021-70 94
Al-Khishman, N.
2020-23 223
Allard, D.
2020-49 251
Almby, K.
2021-24 45
Alotaibi, M.
2021-82 106
Alshelh, Z.
2020-12 211
2020-54 256
2021-115 144
2021-151 181
Alstrup, A.O.
2021-124 154
2021-126 155
2021-47 67
2021-49 70
2021-51 72
Althubaity, N.S.
2021-62 86
Alves, I.L.
2021-86 111
Ameis, S.H.
2020-09 207
Ametamey, S.M.
2021-123 152
2021-22 43
Ananth, M.R.
2020-03 201
Anazodo, U.
2020-39 240
2021-105 133
2021-141 170
2021-89 115
Ances, B.
2020-37 238
Anderson, J.M.
2021-79 103
Andersson, M.
2021-140 169
Angarita, G.A.
2020-50 252
2021-39 59
2021-57 81
Antoni, G.
2020-52 254
Antonov, A.
2021-84 109
Appel, L.
2020-52 254
Appelhoff, S.
2021-91 117
Arakawa, R.
2021-81 105
Araneta, M.F.
2021-111 140
Arcego, D.M.
2021-96 123
Arias, J.F.
2020-11 210
2020-43 245
2020-58 261
2021-107 135
2021-148 177
2021-152 182
2021-160 189
Arias, J.Fernández.
2021-75 99
Asch, R.H.
2021-143 172
Ashok, A.H.
2021-113 142
2021-40 60
Ashton, N.J.
2021-112 141
2021-120 149
2021-150 180
Asselin, M.-C.
2021-82 106
Atassi, N.
2021-17 38
Audrain, H.
2021-47 67
2021-51 72
Aumont, E.
2021-55 78
Auvity, S.
2021-66 90
Avendaño-Estrada, A.
2021-33 53
Ávila-Rodríguez, M.Á.
2021-33 53
Åhs, F.
2021-25 46
B
Bærentzen, S.L.
2021-124 154
2021-126 155
2021-158 188
Babu, S.
2021-17 38
Baldo, D.
2021-155 185
Ballanger, B.
2020-21 221
Bard, J.
2021-01 21
Barkhof, F.
2021-101 129
2021-102 130
2021-27 47
2021-86 111
Barletta, V.
2021-06 26
Barnhart, T.
2020-25 225
Barret, O.
2021-10 30
2021-29 49
2021-56 79
Bars, D.L.
2020-21 221
Barth, B.
2021-96 123
Bartlett, E.A.
2020-03 201
2020-32 232
2021-02 22
Bastiani, M.A.D.
2021-147 176
Bauer, A.
2021-28 49
2021-30 50
Baumann, M.
2021-52 74
Baumeister, T.R.
2021-122 151
Baur, D.M.
2021-28 49
2021-30 50
Bdair, H.
2021-11 31
Beaver, J.D.
2020-34 234
Becker, A.
2021-59 83
Becker, G.A.
2020-33 233
2021-84 109
Bedair, H.
2020-12 211
Bedard, M.-A.
2021-55 78
Beer, S.
2021-28 49
2021-30 50
Beliveau, V.
2021-04 24
Bellaver, B.
2021-112 141
2021-147 176
Belov, V.
2021-109 138
Bendlin, B.
2020-25 225
Benedet, A.L.
2020-11 210
2020-43 245
2020-58 261
2021-107 135
2021-112 141
2021-120 149
2021-147 176
2021-148 177
2021-150 180
2021-152 182
2021-155 185
2021-156 186
2021-160 189
2021-162 192
2021-165 195
Benkelfat, C.
2020-45 247
2020-49 251
2021-11 31
Berg, E.
2020-15 215
Bernard-Gauthier, V.
2021-70 94
Berry-Kravis, E.M.
2020-13 212
2021-56 79
Bertoglio, D.
2021-01 21
Bertoldo, A.
2020 Abstract 199
2020-20 220
2021-105 133
2021-114 143
2021-38 58
2021-65 89
Best, L.M.
2021-95 122
Betthauser, T.
2020-25 225
2021-74 98
2021-91 117
Bevington, C.W.J.
2021-121 149
Beylin, D.
2020-32 232
Bezgin, G.
2021-107 135
2021-119 148
2021-120 149
2021-135 164
2021-144 173
2021-148 177
2021-150 180
2021-152 182
2021-155 185
2021-156 186
2021-160 189
2021-162 192
2021-165 195
2021-168 198
2021-55 78
2021-75 99
Bhan, A.
2021-163 193
Bieger, A.
2021-147 176
Biernaski, H.
2020-47 249
Björkstrand, J.
2021-25 46
Blüher, M.
2020-33 233
Black, S.E.
2021-144 173
2021-163 193
Blair, R.W.
2021-91 117
Bleeser, T.
2020-44 246
Bleher, D.
2021-164 194
2021-93 120
Blennow, K.
2021-112 141
2021-120 149
2021-150 180
Bloomfield, P.
2021-16 37
Bocti, C.
2021-163 193
Boctor, E.M.
2021-52 74
Boehm, M.
2021-23 44
Boellaard, R.
2021-101 129
2021-125 154
2021-27 47
Boileau, I.
2021-36 56
2021-76 101
2021-95 122
Boivin, M.
2020-45 247
2020-49 251
2021-99 127
Bonanno, F.
2021-93 120
Bonaventura, J.
2021-23 44
Bongarzone, S.
2020-10 208
Boniface, H.
2021-91 117
Bonnefoi, F.
2020-21 221
Bonoldi, I.
2020-41 243
Borg, J.
2021-140 169
Borlodoi, J.
2020-02 200
Bormans, G.
2020-14 213
2020-44 246
Borrie, M.
2021-163 193
Bowden, G.D.
2021-164 194
2021-93 120
Boyle, A.J.
2021-09 29
2021-16 37
Bradberry, C.
2021-23 44
Brasic, J.R.
2020-13 212
2021-56 79
Brendgen, M.
2021-99 127
Breuil, L.
2021-100 127
2021-66 90
2021-78 103
Brickman, A.M.
2020-40 242
2021-154 184
Brinkmalm, A.
2021-120 149
Brinson, Z.
2020-13 212
Brock, B.
2021-159 188
Brooks, A.F.
2021-11 31
Brooks, D.J.
2021-124 154
2021-126 155
Brosnan, M.K.
2021-146 176
Brouillette, K.N.
2021-95 122
Brown, T.
2021-10 30
Brugarolas, P.
2021-142 171
Brum, W.S.
2021-112 141
2021-147 176
Brusaferri, L.
2021-115 144
2021-151 181
Brust, P.
2020-33 233
2021-104 132
2021-84 109
Budimirovic, D.B.
2020-13 212
2021-56 79
Bullmore, E.T.
2020-35 236
2021-62 86
Burman, J.
2020-52 254
Buskirk, M.V.
2021-111 140
Buss, S.
2021-134 163
2021-164 194
2021-93 120
Butler, J.
2020-47 249
C
Côté, S.
2020-45 247
2020-49 251
Cai, L.
2021-37 57
Caillé, F.
2021-66 90
2021-78 103
Calabro, F.
2020-06 204
2021-157 187
Cambareri, M.
2020-32 232
Campoy, A.-D.
2021-118 146
Carbonell, F.
2021-122 151
Caribe, P.
2021-138 168
Carlin, A.R.
2021-82 106
Carroll, V.
2020-26 226
2021-10 30
Carson., R.E.
2020-04 202
Carson, R.E.
2020-15 215
2020-50 252
2021-131 160
2021-132 161
2021-133 162
2021-137 167
2021-139 169
2021-22 43
2021-35 55
2021-39 59
2021-55 78
2021-57 81
2021-91 117
Cash, D.
2020-02 200
Castellanos-Ryan, N.
2020-45 247
2020-49 251
2021-99 127
Castro-Blanco, K.A.
2021-115 144
Catana, C.
2020-06 204
2021-149 178
2021-151 181
2021-157 187
Catanese, M.C.
2021-06 26
Cawthorne, C.
2020-14 213
2020-44 246
Celen, S.
2020-14 213
2020-44 246
Cernasov, P.M.
2021-17 38
Cervenka, S.
2020-07 205
2020-28 228
2020-56 258
2020-57 260
Chamoun, M.
2021-112 141
2021-119 148
2021-120 149
2021-150 180
2021-152 182
2021-155 185
2021-156 186
2021-160 189
2021-162 192
2021-165 195
2021-75 99
Chan, J.
2021-17 38
Chan, P.
2021-29 49
Chappie, T.A.
2021-81 105
Chartrand, D.
2020-22 222
Chen, A.
2020-12 211
Chen, D.Y.-T.
2021-44 64
Chen, J.
2021-149 178
Chen, M.-K.
2020-04 202
Chen, P.
2020-46 248
Chen, R.
2021-153 183
Chen, Y.
2020-24 224
2021-82 106
Cheng, K.(Ju-Chieh).
2021-121 149
Cheong, R.
2021-128 157
Chertkow, H.
2021-163 193
Chew, S.
2021-17 38
Cho, S.S.
2021-153 183
Christian, B.
2020-25 225
2020-37 238
2021-74 98
Chu, J.
2020-54 256
Cohen, A.
2020-37 238
2020-51 253
Collij, L.E.
2021-102 130
2021-86 111
Collste, K.
2020-57 260
Comley, R.A.
2021-132 161
2021-67 92
Constantinescu, C.
2020-26 226
2021-10 30
Coomans, E.M.
2021-102 130
2021-125 154
2021-127 156
2021-27 47
Coope, D.
2021-82 106
2021-85 110
Corbetta, M.
2020 Abstract 199
2021-65 89
Cortes-Salva, M.Y.
2020-38 239
Cosgrove, K.P.
2020-07 205
2020-17 217
2020-50 252
2021-12 32
2021-57 81
2021-79 103
Costes, N.
2020-21 221
2020-29 229
Cote, S.
2021-99 127
Coughlin, J.M.
2020-28 228
2021-146 176
Cousins, O.
2021-63 87
Cox, S.
2020-45 247
2020-49 251
2021-55 78
2021-99 127
Cutforth, T.
2021-77 102
Cybulska, K.
2021-01 21
D
Díaz-Ruíz, A.
2021-33 53
d’Ambrosio, E.
2021-31 51
Dagher, A.
2020-45 247
2020-49 251
2021-54 76
2021-55 78
2021-99 127
Dahoun, T.
2021-38 58
Dailler, F.
2020-29 229
Dalenberg, J.R.
2021-91 117
Danfors, T.
2020-36 237
2020-52 254
Dannals, R.
2021-146 176
Dassanayake, P.
2021-105 133
2021-141 170
Dauba, A.
2021-100 127
Davenport, N.
2020-25 225
Davies, R.
2021-37 57
Davis, M.T.
2021-39 59
de Geus5, E.J.C.
2021-102 130
de’Picker, L.
2020-28 228
Debatisse, J.
2020-21 221
Debonee, S.
2021-39 59
DeLorenzo., C.
2020-19 219
DeLorenzo, C.
2020-03 201
2021-117 146
2021-136 165
Demjaha, A.
2021-31 51
den Braber, A.
2021-102 130
den Hollander, M.E.
2021-102 130
Deng, H.P.
2021-34 54
Deuther-Conrad, W.
2021-104 132
2021-84 109
Devanand, D.P.
2021-154 184
Dhaynaut, M.
2021-109 138
Dias, M.
2021-132 161
Dick, R.
2020-18 218
Diez-Cirarda, M.
2021-166 196
DiFilippo, A.
2020-25 225
2021-74 98
Dillon, E.
2020-13 212
Dima, D.
2021-62 86
Dipasquale, O.
2021-38 58
Djoukhadar, I.K.
2021-82 106
Dominguez, C.
2021-01 21
Dominguez, T.L.
2020-42 244
2021-19 40
Donovan, L.L.
2021-14 35
Doot, R.K.
2020-17 217
2020-42 244
2021-19 40
Doran, S.D.
2021-81 105
Doubrovin, M.
2021-77 102
Doudet, D.J.
2020-48 250
2021-110 139
2021-47 67
2021-51 72
Droppa, K.
2021-149 178
Du, Y.
2021-146 176
DuBois, J.M.
2020-55 257
Dubroff, J.G.
2020-17 217
2020-42 244
2021-19 40
Duffy, M.F.
2021-15 36
Dukić-Stefanović, S.
2021-104 132
Durand, F.
2020-49 251
Duvvuri, S.
2021-109 138
E
Easmin, R.
2021-31 51
Ebenau, J.
2021-101 129
Edison, P.
2021-45 66
Edwards, R.
2020-12 211
2020-54 256
2021-151 181
Eldridge, M.
2020-18 218
Eldridge, M.A.G.
2021-83 108
Elmenhorst, D.
2021-28 49
2021-30 50
Elmenhorst, E.-M.
2021-28 49
2021-30 50
Elmore, J.S.
2021-52 74
Engle, J.
2020-25 225
Erfani, M.
2021-50 71
Eriksson, J.
2021-24 45
2021-32 52
2021-90 116
Erritzoe, D.
2020-34 234
Esterlis, I.
2021-143 172
2021-35 55
2021-39 59
Evans, A.C.
2021-108 136
2021-48 69
Eykyn, T.
2020-02 200
F
Fahlstrom, M.
2021-24 45
Fakhri, G.E.
2021-109 138
2021-59 83
Fan, A.P.
2021-44 64
Farajdokht, F.
2021-50 71
Farde, L.
2020-56 258
2020-57 260
2021-140 169
Farrell, M.
2021-59 83
Fatouros-Bergman, H.
2020-57 260
Feigen, E.
2021-37 57
Feingold, F.
2021-91 117
Felchner, Z.
2021-79 103
Fernandez-Arias, J.
2021-119 148
Fernazdez-Arias, J.
2021-150 180
Ferraresso, G.
2021-121 149
Ferrari-Souza, J.P.
2021-112 141
2021-119 148
2021-147 176
Ferreira, P.C.L.
2021-112 141
2021-119 148
2021-147 176
Finger, E.
2020-39 240
2021-141 170
2021-89 115
Finn, C.
2021-139 169
Finnema, S.J.
2021-132 161
Florio, K.
2020-12 211
Fontaine, K.
2021-137 167
Fontana, I.C.
2020-10 208
Fowles, K.
2021-143 172
Fox, M.S.
2020-23 223
Frayne, R.
2021-163 193
Fredriksson, M.
2021-25 46
Fregonara, P.Z.
2021-81 105
Frick, A.
2021-25 46
Fronczek-Poncelet, J.
2021-28 49
Fryer, T.D.
2020-35 236
Fu, J.F.
2021-03 23
2021-13 33
2021-59 83
2021-88 114
Fujita, M.
2020-38 239
Fultz, N.
2021-149 178
Funck, T.
2021-108 136
2021-55 78
2021-91 117
G
Gündel, D.
2021-104 132
Günther, M.
2020-39 240
Gallagher, E.
2020-17 217
2021-106 134
Gallezot, J.-D.
2021-132 161
2021-133 162
2021-137 167
2021-35 55
2021-91 117
Gandhi, R.
2020-54 256
Ganz, M.
2021-04 24
2021-08 28
2021-91 117
Gao, H.
2021-12 32
2021-22 43
García, D.Vállez.
2021-86 111
Garcia-Milian, R.
2021-143 172
Gardus, J.
2020-19 219
2021-136 165
Garibotto, V.
2021-86 111
Gau, R.
2021-91 117
Gaudette, E.
2021-36 56
2021-95 122
Gauthier, S.
2020-11 210
2020-22 222
2020-43 245
2020-58 261
2021-107 135
2021-112 141
2021-119 148
2021-135 164
2021-144 173
2021-150 180
2021-152 182
2021-160 189
2021-162 192
2021-168 198
2021-75 99
Gebara, A.
2021-43 63
Gee, A.
2020-10 208
Gejl, M.
2021-159 188
Gharbawie, O.
2021-23 44
Gharehgazlou, A.
2020-09 207
Ghodadra, A.
2020-51 253
Giacomel, A.
2021-38 58
2021-91 117
Giese, A.
2021-93 120
Gilbert, T.M.
2020-14 213
2020-44 246
Gill, T.
2021-36 56
Gillman, A.
2021-91 117
Gingnell, M.
2021-24 45
Girgis, R.
2020-32 232
Gjedde, A.
2021-158 188
2021-159 188
2021-46 66
2021-47 67
2021-49 70
2021-50 71
2021-51 72
2021-52 74
Gladding, R.
2020-18 218
Glud, A.N.
2021-124 154
Gobbi, G.
2021-11 31
Gobert, F.
2020-21 221
2020-29 229
Godbersen, G.M.
2020-05 203
2021-71 95
2021-72 96
Goddard, J.
2020-48 250
Goghbi, S.
2021-09 29
Goislard, M.
2021-78 103
Goldstein, S.J.
2021-136 165
Golla, S.S.V.
2021-102 130
2021-125 154
2021-127 156
2021-27 47
Gomez, J.
2021-23 44
Gonçalves, A.J.
2021-82 106
Goodman, J.A.
2021-56 79
Gorgolewski, K.J.
2021-91 117
Gouasmat, A.
2020-26 226
Goubran, M.
2021-144 173
2021-163 193
Goutal, S.
2021-100 127
2021-66 90
2021-78 103
Gouw, A.
2021-27 47
Grace, A.A.
2021-52 74
Graff-Guerrero, A.
2021-166 196
Granziera, C.
2020-54 256
Gravel, P.
2021-137 167
Gray, K.R.
2021-86 111
Green, D.
2021-95 122
Greenwood, R.J.
2021-63 87
Greve, D.N.
2021-04 24
2021-08 28
2021-41 61
Griesinger, C.
2021-93 120
Groman, S.M.
2020-15 215
2021-143 172
Grotegerd, A.-K.
2021-164 194
2021-93 120
Gryglewski, G.
2020-08 206
2021-68 92
Gu, Z.-Q.
2021-29 49
Guehl, N.J.
2021-109 138
2021-59 83
Guennewig, T.
2020-33 233
Guimond, S.
2021-55 78
Guiot, M.-C.
2021-161 190
Gunn, R.N.
2020-16 216
2020-26 226
2020-31 231
2020-34 234
2020-46 248
2021-10 30
Gustavsson, T.
2021-32 52
H
Haass, C.
2020-40 242
Hachinski, V.
2020-23 223
Hacker, M.
2020-05 203
2021-123 152
2021-71 95
2021-72 96
Hader, S.
2020-02 200
Hadjikhani, N.
2021-115 144
Hadway, J.
2020-47 249
Hafizi, S.
2020-28 228
Hahn, A.
2020-05 203
2021-123 152
2021-71 95
2021-72 96
2021-96 123
Hajcak, G.
2020-03 201
Halbert, K.
2021-44 64
Hall, A.
2021-146 176
Hall, J.A.
2021-161 190
Halldin, C.
2020-57 260
2021-140 169
2021-81 105
Haller, S.
2021-24 45
Hammers, A.
2021-114 143
Hampson, M.
2021-73 97
Hanania, J.U.
2021-121 149
Handen, B.
2020-37 238
Hansen, H.D.
2021-14 35
2021-34 54
2021-91 117
Hansen, J.Y.
2021-54 76
2021-55 78
Harraz, M.M.
2021-52 74
Harrington, C.
2021-146 176
Harrison, N.A.
2021-62 86
Hattori, S.
2021-92 119
Haubenberger, D.
2020-31 231
Hauschild, L.A.
2021-147 176
Heeman, F.
2021-86 111
Helili, Z.
2020-42 244
2021-19 40
Hendriks, J.
2021-86 111
Hennecke, E.
2021-30 50
Henry, K.
2020-38 239
Henry, S.
2021-132 161
Herbots, M.
2020-14 213
Herfert, K.
2021-128 157
2021-134 163
2021-164 194
2021-93 120
Hesse, S.
2020-33 233
2021-84 109
Hicks, J.W.
2020-39 240
2020-47 249
2021-89 115
Hietala, J.
2020-28 228
2021-55 78
Hightower, B.G.
2021-06 26
2021-17 38
2021-69 93
Hilbert, A.
2020-33 233
Hill, K.R.
2020-19 219
2021-117 146
Hillebrand, A.
2021-27 47
Hillmer, A.T.
2020-04 202
2020-07 205
2020-17 217
2020-50 252
2021-39 59
2021-57 81
2021-79 103
Himes, M.
2021-116 145
2021-61 85
Hinz, R.
2021-85 110
Hoban, D.B.
2021-124 154
Hollander, M.E. den.
2021-127 156
Holley, D.
2021-44 64
Holliday, E.
2020-13 212
Holmes, S.
2021-132 161
2021-35 55
Holt, D.
2021-146 176
Hong, J.
2020-38 239
Honig, L.S.
2020-40 242
2021-154 184
Hooker, J.M.
2020-14 213
2020-44 246
2020-59 262
2021-06 26
2021-115 144
2021-151 181
2021-17 38
2021-69 93
Hoon, J.de.
2020-14 213
Hopewell, R.
2021-70 94
Horti, A.
2021-146 176
Houle, S.
2021-36 56
Housman, H.
2021-115 144
Howes, O.D.
2020-20 220
2020-28 228
2020-41 243
2021-113 142
2021-31 51
2021-38 58
2021-60 84
2021-67 92
Hoye, J.
2021-129 158
2021-73 97
Hsiung, R.
2021-163 193
Hsu, D.T.
2021-117 146
Huajie, J.
2021-31 51
Huang, Y.H.
2020-04 202
2020-15 215
2020-50 252
2021-12 32
2021-131 160
2021-132 161
2021-133 162
2021-139 169
2021-22 43
2021-57 81
2021-79 103
Hughes, Z.A.
2021-81 105
Hugon, G.
2021-100 127
Hutchison, M.G.
2021-36 56
I
Ichise, M.
2021-18 39
2021-21 42
2021-92 119
Iecker, T.
2020-21 221
Iida, H.
2020-47 249
Ikawa, M.
2021-20 41
Ikenuma, H.
2021-18 39
2021-21 42
2021-92 119
Imamura, S.
2021-92 119
Innis, R.B.
2020-18 218
2020-38 239
2021-09 29
2021-111 140
2021-37 57
2021-81 105
2021-83 108
2021-91 117
Ionescu, T.M.
2021-128 157
2021-134 163
Iqbal, M.
2021-99 127
Irace, Z.
2020-29 229
Irace3, Z.
2020-21 221
Iredale, P.
2021-109 138
Isen, J.
2021-144 173
Ito, K.
2021-18 39
2021-21 42
2021-92 119
Iturria-Medina, Y.
2021-122 151
Izquierdo-Garcia, D.
2020-55 257
J
Jackson, A.
2021-82 106
2021-85 110
Jahuar, S.
2020-41 243
Jakobsen, S.
2021-124 154
2021-49 70
Jang, M.-K.
2020-46 248
Jani, I.
2021-43 63
Jauhar, S.
2021-31 51
Jaworska, N.
2020-45 247
2020-49 251
2021-99 127
Jedema, H.
2021-23 44
Jenkins, M.
2021-111 140
Jennings, D.
2021-56 79
Jensen, P.S.
2021-04 24
Jesso, S.
2021-89 115
Jetly, R.
2021-36 56
2021-95 122
Jochimsen, T.
2020-33 233
Johnson, A.
2020-40 242
2021-154 184
2021-77 102
Johnson, K.
2021-59 83
Johnson, S.
2020-25 225
2020-37 238
2021-74 98
Johnson-Akeju, O.
2021-115 144
Jonasson, M.
2020-36 237
Jones, J.
2021-136 165
Joutsa, J.
2021-43 63
Juarez Anaya, F.
2020-38 239
Jucaite, A.
2020-07 205
K
Kadam, S.D.
2021-52 74
Kang, J.
2021-52 74
Kang, J.U.
2021-52 74
Kang, M.-S.
2020-11 210
2020-22 222
2020-43 245
2021-11 31
2021-120 149
2021-135 164
2021-144 173
2021-148 177
2021-152 182
2021-155 185
2021-156 186
2021-160 189
2021-163 193
2021-165 195
2021-168 198
2021-75 99
Karikari, T.
2021-112 141
2021-147 176
2021-150 180
Karjalainen, T.
2021-41 61
Karlsson, A.
2021-24 45
Kasper, S.
2020-05 203
Kato, T.
2021-18 39
2021-21 42
2021-92 119
Katz, J.D.
2020-38 239
Kaur, H.
2021-87 112
Kaur, T.
2021-11 31
Keerthisinghe, O.V.
2021-118 146
Keller, S.H.
2021-04 24
Kemp, A.F.
2021-46 66
Kemp, C.J.
2021-15 36
Kenou, B.
2021-111 140
Khalighi, M.M.
2021-44 64
Khan, A.F.
2021-122 151
Khattar, N.
2021-133 162
Khetarpal, V.
2021-01 21
Kim, M.
2020-12 211
2021-115 144
2021-151 181
Kim, M.-J.
2020-38 239
2021-111 140
Kimura, Y.
2021-18 39
2021-21 42
2021-92 119
King, S.
2021-162 192
Kinnerup, M.B.
2021-46 66
Kirik, D.
2021-128 157
Kirkeby, A.
2021-124 154
Kish, S.
2021-36 56
2021-95 122
Kiss, A.
2021-163 193
Kitzbichler, M.G.
2020-35 236
Kiyono, Y.
2021-20 41
2021-26 46
Klöbl, M.
2020-05 203
2020-08 206
Klein, J.I.
2020-40 242
2021-154 184
Klug, S.
2020-05 203
2021-72 96
Klunk, W.E.
2020-37 238
2020-51 253
Klyuzhin, I.S.
2021-03 23
2021-13 33
2021-88 114
Knight, A.C.
2021-16 37
Knight, P.C.
2021-115 144
2021-151 181
Knudsen, G.M.
2020-07 205
2020-27 227
2020-34 234
2021-04 24
2021-08 28
2021-14 35
2021-91 117
Kobayashi, E.
2021-161 190
Koohsari, S.
2021-131 160
2021-132 161
2021-139 169
Koole, M.
2020-14 213
2020-44 246
Korat, Špela.
2021-01 21
Kosaka, H.
2021-20 41
2021-26 46
Kostikov, A.
2021-11 31
Kothari, P.
2020-30 230
Koyama, H.
2021-18 39
2021-21 42
Kozinski, K.
2021-61 85
Kranz, J.E.
2020-14 213
2020-44 246
Kranzler, H.R.
2021-19 40
Kreisl, W.C.
2020-40 242
2021-154 184
2021-77 102
Krokos, G.
2021-82 106
Kroll, T.
2021-28 49
2021-30 50
Kuebler, L.
2021-134 163
2021-93 120
Kuhnast, B.
2021-78 103
Kumar, D.
2020-30 230
Kunach, P.
2021-135 164
Kuo, P.H.
2021-163 193
Kvernby, S.
2021-24 45
Kwon, Y.-M.
2020-12 211
L
López, J.Domingo G.
2021-86 111
Laat, B.de.
2021-129 158
Labaree, D.
2020-04 202
Ladwa, R.M.
2021-118 146
2021-64 88
2021-80 104
2021-97 124
Laforce, R.J.
2021-163 193
Lai, T.H.
2021-104 132
Lam, B.
2021-163 193
Lam, T.T.
2021-143 172
Lammertsma, A.A.
2021-86 111
Lancelot, S.
2020-21 221
Landa, R.
2020-13 212
Landau, A.M.
2021-124 154
2021-126 155
2021-158 188
2021-47 67
2021-49 70
2021-51 72
Landolt, H.-P.
2021-28 49
2021-30 50
Landsmann, A.
2020-33 233
Landtblom, A.-M.
2020-52 254
Lange, D.
2021-28 49
2021-30 50
Langer, O.
2021-66 90
Langham, M.C.
2020-47 249
Lantero-Rodriguez, J.
2021-150 180
Lanzenberger., R.
2020-08 206
Lanzenberger, R.
2020-05 203
2021-123 152
2021-68 92
2021-71 95
2021-72 96
2021-96 123
Larrat, B.
2021-100 127
Larsson, E.-M.
2020-52 254
Laurell, G.L.
2020-07 205
Laurencin, C.
2020-21 221
Laurikainen, H.
2020-28 228
Lawrence, K.S.
2020-39 240
2020-47 249
2021-105 133
2021-141 170
2021-89 115
Laymon, C.
2020-37 238
2020-51 253
Lee, E.
2021-58 82
Lee, H.
2020-42 244
2021-19 40
Lee, J.
2021-151 181
Lee, J.-H.
2020-38 239
Lee, M.
2020-57 260
Lee, S.
2020-40 242
2021-154 184
Leffa, D.T.
2021-147 176
Lehel, S.
2021-14 35
Leitzke, M.
2021-84 109
Leonov, A.
2021-93 120
Lepage, C.
2021-108 136
Lerchner, W.
2021-83 108
Lesanpezeshki, M.
2020-32 232
Lesniak, W.
2021-146 176
Levine, M.A.
2020-06 204
2021-145 174
2021-157 187
Levit, A.
2020-23 223
Lewis, D.
2021-85 110
Lewis, L.
2021-149 178
Lewis, Y.
2020-34 234
Leyton, M.
2020-45 247
2020-49 251
2021-11 31
2021-55 78
2021-99 127
Li, B.
2021-83 108
Li, S.
2020-15 215
2020-42 244
Liang, C.
2020-53 255
2021-118 146
2021-167 197
2021-58 82
2021-64 88
2021-80 104
2021-87 112
2021-97 124
2021-98 125
Liang, S.
2020-55 257
Liger, F.
2020-21 221
Lillethorup, T.P.
2021-124 154
Lin, Y.
2020-12 211
2020-54 256
Lindberg, A.
2021-70 94
Lindström, E.
2020-36 237
Liow, J.-S.
2020-18 218
2020-38 239
2021-111 140
2021-37 57
2021-81 105
2021-83 108
Liu, H.
2020-04 202
2021-05 25
2021-12 32
2021-129 158
Liu, L.
2020-47 249
2021-01 21
2021-105 133
2021-141 170
2021-89 115
Liu, S.-Y.
2021-29 49
Lo, E.
2021-142 171
Loew, L.M.
2021-52 74
Loewen, G.
2020-31 231
Loggia, M.L.
2020-12 211
2020-54 256
2021-115 144
2021-151 181
Lois, C.
2021-59 83
Lopresti, B.
2020-51 253
2021-61 85
Lora, S.J.
2021-76 101
Lu, Y.
2021-133 162
Lubberink, M.
2020-36 237
2020-52 254
2021-24 45
2021-25 46
Ludwig, F.-A.
2021-104 132
Luk, K.C.
2021-106 134
2021-15 36
Luna, B.
2020-06 204
2021-157 187
Lundberg, J.
2021-140 169
Lussier, F.Z.
2020-11 210
2020-43 245
2020-58 261
2021-107 135
2021-112 141
2021-119 148
2021-120 149
2021-135 164
2021-148 177
2021-150 180
2021-152 182
2021-155 185
2021-156 186
2021-160 189
2021-162 192
2021-165 195
2021-75 99
Luthardt, J.
2020-33 233
M
Mérida, I.
2020-21 221
Møller, A.
2021-47 67
Mach, R.H.
2020-42 244
2021-106 134
2021-19 40
Mahmoudi, J.
2021-50 71
Mahone, E.M.
2020-13 212
Mainero, C.
2021-06 26
Majdi, A.
2021-50 71
Makary, M.M.
2021-17 38
Makino, A.
2021-20 41
Malla, A.P.
2021-52 74
Manavaki, R.
2020-35 236
Manber, R.
2021-86 111
Mandeville, E.
2021-142 171
Mandeville, J.B.
2020-06 204
2021-145 174
2021-157 187
2021-34 54
Mandouh, O.KAE E.
2020-13 212
Manly, L.
2020-18 218
2021-111 140
Mann, J.
2020-30 230
2020-32 232
2021-02 22
Mannheim, J.G.
2021-13 33
2021-88 114
Mansur, A.
2020-16 216
Marciano, S.
2021-128 157
Marcus, R.E.
2021-69 93
Marek, K.
2020-46 248
Maresca, K.P.
2021-81 105
Margolin, R.
2020-46 248
Maria Marques da Silva, A.
2021-138 168
Marie, S.
2021-66 90
Markello, R.D.
2021-54 76
2021-55 78
Markiewicz, C.J.
2021-91 117
Marques, T.R.
2020-20 220
2021-113 142
Martin, S.
2020-33 233
Martin, S.D.
2020-13 212
2021-56 79
Martins, D.
2021-115 144
2021-62 86
Masellis, M.
2021-153 183
2021-166 196
Mason, N.S.
2021-116 145
2021-61 85
Massarweh, G.
2021-150 180
2021-161 190
2021-168 198
2021-70 94
Matheson, G.J.
2020-24 224
2020-28 228
2020-56 258
2020-57 260
2021-103 131
2021-91 117
2021-94 121
Mathis, C.A.
2020-37 238
2020-51 253
Mathotaarachchi, S.
2020-11 210
2021-107 135
2021-148 177
2021-152 182
Mathur, A.K.
2020-13 212
2021-56 79
Matsumoto, H.
2021-26 46
Mattsson, B.
2021-124 154
Matushita, C.S.
2021-138 168
Matuskey, D.
2020-04 202
2020-50 252
2021-131 160
2021-132 161
2021-133 162
2021-137 167
2021-139 169
2021-39 59
2021-57 81
2021-79 103
Matys, T.
2021-40 60
Maurer, A.
2021-128 157
2021-164 194
2021-93 120
McCluskey, T.
2021-36 56
2021-76 101
2021-95 122
McCrone, P.
2021-31 51
McCutcheon, R.
2021-38 58
2021-60 84
McDougle, C.J.
2021-69 93
McKeown, M.
2021-03 23
2021-13 33
McKinney, G.
2020-25 225
McLachlan, M.
2021-74 98
McManus, M.
2021-106 134
McPartland, J.C.
2021-139 169
McQuade, P.
2021-10 30
McVea, A.
2021-74 98
Meaney, M.J.
2021-96 123
Mehndiratta, A.
2021-06 26
Mercaldo, N.D.
2021-115 144
Merida, I.
2020-29 229
Meyer, J.
2021-36 56
Meyer, P.M.
2020-33 233
2021-84 109
Michaelides, M.
2021-23 44
2021-83 108
Michenthaler, P.
2020-05 203
2021-68 92
Mihaescu, A.
2021-153 183
2021-166 196
Miller, J.
2020-32 232
Miller, K.M.
2021-15 36
Miller, W.
2020-38 239
2021-111 140
Minamimoto, T.
2020-18 218
Minhas, D.
2020-37 238
2020-51 253
Minuzzi, L.
2021-161 190
Minyoung, J.
2021-20 41
Miranda, A.
2021-01 21
Misic, B.
2021-54 76
2021-55 78
Mita, K.
2021-26 46
Mizrahi, R.
2020-09 207
2020-28 228
2021-36 56
Mizuno, T.
2021-26 46
Mojiri, P.
2021-163 193
Moldovan, R.-P.
2021-104 132
Mondal, J.
2021-133 162
Mondal, R.
2020-53 255
2021-118 146
2021-64 88
2021-80 104
2021-97 124
Mondelli, V.
2021-62 86
Montero Santamaria, J.A.
2020-38 239
Moon, S.-H.
2021-109 138
Morales, J.
2021-33 53
Moran, T.R.
2021-97 124
Morenas-Rodriguez, E.
2020-40 242
Mori, T.
2021-20 41
Morrens, M.
2020-28 228
Morris, E.D.
2020-04 202
2021-05 25
2021-12 32
2021-129 158
2021-73 97
Morrisey, E.J.
2021-115 144
Morrison, L.
2020-47 249
Morrissey, E.J.
2020-12 211
2021-151 181
Morse, C.
2020-18 218
2020-38 239
2021-111 140
2021-37 57
2021-81 105
Moses, N.
2020-02 200
Mossine, A.V.
2021-70 94
Mota, F.
2020-02 200
Mrzljak, L.
2021-01 21
Muelas, E.H.
2021-06 26
Mukerji, S.
2020-54 256
Mukherjee, J.
2020-53 255
2021-118 146
2021-167 197
2021-58 82
2021-64 88
2021-80 104
2021-87 112
2021-97 124
2021-98 125
Mukherjee, T.
2021-98 125
Mullett, J.E.
2021-69 93
Munoz-Sanjuan, I.
2021-01 21
Murali, D.
2020-25 225
Murgaš, M.
2020-08 206
2021-123 152
2021-68 92
Murrell, E.
2021-07 27
Myers, J.
2021-113 142
N
Nørgaard, M.
2021-04 24
2021-08 28
Nabulsi, N.
2020-04 202
2021-12 32
2021-131 160
2021-132 161
2021-139 169
2021-39 59
2021-57 81
2021-79 103
Nagai, Y.
2020-18 218
Naganawa, M.
2021-132 161
2021-133 162
2021-139 169
2021-35 55
Nahimi, A.
2021-46 66
Najafzadeh, S.
2021-12 32
Nandi, A.
2020-13 212
2021-56 79
Napadow, V.
2020-12 211
2020-54 256
2021-151 181
Naples, A.
2021-139 169
Narayanaswami, V.
2021-16 37
Narciso, L.
2020-39 240
2020-47 249
2021-105 133
2021-138 168
2021-89 115
Narendran, R.
2021-116 145
2021-61 85
Nasrallah, I.M.
2021-19 40
Nasser, A.
2021-14 35
Nettis, M.A.
2021-62 86
Neumaier, B.
2021-28 49
2021-30 50
Neumann, J.
2020-33 233
Nguyen, A.N.
2021-58 82
2021-87 112
Nics, L.
2021-123 152
2021-71 95
Nikolopoulou, A.
2020-30 230
Nilsson, J.
2021-120 149
Niso, G.
2021-91 117
Noer, O.
2021-47 67
2021-49 70
2021-51 72
Nordio, G.
2021-31 51
2021-60 84
Norgaard, M.
2021-91 117
Normandin, M.
2021-59 83
Normandin, M.D.
2021-109 138
Noseworthy, M.D.
2021-163 193
Nour, M.
2021-38 58
Novell, A.
2021-100 127
Nummenmaa, L.
2021-41 61
O
O’Toole, R.
2021-146 176
Ogata, A.
2021-18 39
2021-21 42
2021-92 119
Ogden, R T.
2020-07 205
2020-24 224
2020-27 227
2020-30 230
2020-32 232
Ogden, R.T.
2021-02 22
2021-103 131
2021-117 146
2021-94 121
Okazawa, H.
2021-20 41
2021-26 46
Olayo, R.
2021-33 53
Omata, N.
2021-26 46
Omidyeganeh, M.
2021-108 136
Ooms, M.
2021-81 105
Oostenveld, R.
2021-91 117
Oquendo, M.
2020-30 230
Orlowski, D.
2021-124 154
2021-49 70
Ossenkoppele, R.
2021-102 130
2021-125 154
2021-127 156
2021-27 47
Ottoy, J.
2020-22 222
2020-28 228
2021-144 173
2021-163 193
2021-168 198
Ozzoude, M.
2021-163 193
P
Paganoni, S.
2021-17 38
Pallen, V.
2021-107 135
2021-152 182
2021-156 186
2021-165 195
Palma, E.Schweickert-de.
2020-33 233
Palomero-Gallagher, N.
2021-108 136
2021-122 151
2021-54 76
2021-55 78
Panaparambil, R.
2020-13 212
Parent, S.
2020-45 247
2020-49 251
2021-99 127
Pariante, C.
2021-62 86
Park, J.-H.
2021-44 64
Parmar, A.J.
2021-06 26
2021-69 93
Parmar, M.J.
2021-124 154
Parsey, R.V.
2020-03 201
2020-19 219
2021-117 146
2021-136 165
Pascoal, T.A.
2020-11 210
2020-43 245
2020-58 261
2021-107 135
2021-112 141
2021-119 148
2021-135 164
2021-147 176
2021-150 180
2021-152 182
2021-155 185
2021-156 186
2021-160 189
2021-162 192
2021-165 195
2021-53 75
2021-75 99
Passchier, J.
2020-26 226
2020-34 234
Patel, K.
2020-53 255
2021-98 125
Patel, S.
2021-96 123
Patt, M.
2020-33 233
2021-84 109
Patterson, J.R.
2021-15 36
Pavel, A.
2021-121 149
Peraldi, E.
2021-96 123
Pernet, C.
2021-91 117
Phillips, C.
2021-91 117
Pianta, D.B.
2021-138 168
Pichler, B.J.
2021-128 157
2021-134 163
2021-164 194
2021-93 120
Pichler, V.
2020-05 203
2021-123 152
2021-71 95
2021-72 96
Pierling, A.L.
2021-28 49
2021-30 50
Pietrzak, R.H.
2021-35 55
Pihl, R.
2020-45 247
2020-49 251
2021-99 127
Pijanowski, O.R.
2021-17 38
Pike, V.W.
2020-18 218
2020-38 239
2021-09 29
2021-111 140
2021-37 57
2021-81 105
Pillai, R.LI.
2020-03 201
Pinborg, L.H.
2021-04 24
Pinto, S.
2021-138 168
Pisani, L.
2021-139 169
Pittman, B.
2021-131 160
Placzek, M.S.
2020-59 262
2021-151 181
Platt, M.
2021-77 102
Plavén-Sigray, P.
2020-07 205
2020-27 227
2020-28 228
2020-57 260
Pokhvisneva, I.
2021-96 123
Polimeni, J.
2021-149 178
Polly, K.
2020-40 242
2021-154 184
Poltronetti, N.M.
2021-107 135
2021-152 182
2021-156 186
2021-162 192
2021-165 195
2021-75 99
Pomper, M.G.
2020-28 228
2021-146 176
Pontoriero, A.D.
2020-41 243
Potenza, M.N.
2021-131 160
Pothula, S.
2021-143 172
Poxleitner, M.
2021-164 194
2021-93 120
Prato, F.S.
2020-39 240
2021-163 193
Price, J.C.
2020-06 204
2020-54 256
2021-149 178
2021-151 181
2021-157 187
2021-59 83
Prins, N.
2021-101 129
Q
Qi, C.-Y.
2021-29 49
Qi, Q.
2020-23 223
Qiao, H.-W.
2021-29 49
R
Rabiner, E.A.
2020-16 216
2020-31 231
2020-34 234
2021-113 142
2021-67 92
Rahmim, A.
2021-52 74
Rahmouni, N.
2020-22 222
2021-107 135
2021-112 141
2021-119 148
2021-120 149
2021-152 182
2021-155 185
2021-156 186
2021-162 192
2021-165 195
2021-75 99
Ramirez, J.
2021-163 193
Ramlackhansingh, A.F.
2021-63 87
Ramos-Torres, K.M.
2021-142 171
Rapoport, S.
2020-30 230
Rasul, S.
2021-123 152
Ratai, E.-M.
2020-54 256
2021-115 144
Rausch, I.
2021-123 152
Raval, N.R.
2021-14 35
Real, C.C.
2021-158 188
Reddy, T.T.
2021-167 197
Redouté, J.
2020-21 221
Reed, M.B.
2020-08 206
2021-123 152
2021-71 95
Reichwein, S.M.
2021-19 40
Reid, S.E.
2021-130 159
Reischl, G.
2021-134 163
Renger, J.
2021-109 138
Reynolds, B.V.
2021-17 38
Rhind, S.
2021-36 56
2021-76 101
2021-95 122
Rice, P.A.
2021-109 138
Richardson, D.
2021-36 56
2021-95 122
Richmond, B.J.
2020-18 218
2021-83 108
Riedl, V.
2020 Abstract 199
Riley, M.M.
2020-59 262
Rios, C.
2021-33 53
Rischel, E.
2021-159 188
Rischka, L.
2020-05 203
2021-123 152
2021-71 95
2021-72 96
Ritchie, C.
2021-86 111
Ritter, V.
2020-05 203
Rizzo, G.
2020-16 216
2020-34 234
2021-67 92
2021-91 117
Rogdaki, M.
2020-41 243
Rogeau, A.
2021-60 84
Rokka, J.
2021-32 52
2021-90 116
Ropchan, J.
2021-12 32
2021-132 161
2021-139 169
2021-22 43
2021-57 81
2021-79 103
Ros, L.U.D.
2021-147 176
Rosa-Neto, P.
2020-11 210
2020-22 222
2020-43 245
2020-58 261
2021-107 135
2021-11 31
2021-112 141
2021-119 148
2021-120 149
2021-135 164
2021-144 173
2021-147 176
2021-148 177
2021-150 180
2021-152 182
2021-155 185
2021-156 186
2021-160 189
2021-161 190
2021-162 192
2021-165 195
2021-168 198
2021-53 75
2021-55 78
2021-70 94
2021-75 99
Roseborough, A.
2020-23 223
Rosen, B.R.
2021-115 144
2021-145 174
2021-149 178
2021-34 54
Roshanbin, S.
2021-90 116
Ross, K.
2021-11 31
Rossano, S.
2020-15 215
2021-133 162
Rossi, R.
2021-158 188
Rotman, S.
2020-12 211
Routier, A.
2021-91 117
Royse, S.K.
2020-51 253
Rubinstein, Z.
2021-59 83
Rubovits, S.
2021-111 140
Ruf, V.
2021-93 120
Ruiz-Perdomo, Y.
2020-38 239
Rullmann, M.
2020-33 233
2021-84 109
Rungby, J.
2021-159 188
Rusjan, P.M.
2020-09 207
2020-28 228
2021-141 170
2021-36 56
Russell, D.S.
2020-46 248
2021-10 30
2021-56 79
Ryan, M.
2020-13 212
Ryan, M.J.
2021-27 47
Ryazanov, S.
2021-93 120
S
Séguin, J.
2020-45 247
2020-49 251
Sörensen, J.
2020-52 254
Sørensen, J.Christian H.
2021-124 154
Saba, W.
2021-66 90
2021-78 103
Sabri, O.
2020-33 233
2021-104 132
2021-84 109
Sadigh-Eteghad, S.
2021-50 71
Saha, A.
2020-12 211
2020-54 256
2021-151 181
Saha-Chaudhuri, P.
2021-150 180
Sahlas, D.J.
2021-163 193
Sakai, T.
2021-92 119
Salimbeni, H.
2021-31 51
Salo, T.
2021-91 117
Salvatore, A.
2020-54 256
2021-59 83
Sanchez, J.
2021-59 83
Sander, C.Y.
2021-145 174
2021-149 178
2021-34 54
Sandiego, C.
2020-31 231
2020-46 248
2021-10 30
Santamaria, J.A.M.
2021-111 140
Santangelo, B.
2020-41 243
2021-31 51
Santillo, A.
2020-57 260
Sari, H.
2020-54 256
2021-59 83
Sattler, B.
2021-104 132
2021-84 109
Sattler, T.
2021-84 109
Savard, M.
2020-11 210
2020-43 245
2020-58 261
2021-53 75
Saw, R.S.
2021-128 157
2021-93 120
Scala, S.
2021-55 78
Schacht, A.M.
2021-49 70
Schain, M.
2020-07 205
2020-27 227
2021-08 28
2021-14 35
2021-34 54
Scheltens, P.
2021-101 129
2021-102 130
2021-125 154
2021-27 47
Schembecker, S.
2021-164 194
Scheper, W.
2021-27 47
Scheunemann, M.
2021-84 109
Schibli, R.
2021-22 43
Schifani, C.
2021-16 37
Schimpf, S.
2021-84 109
Schlein, E.
2021-32 52
Schmidt, C.J.
2021-81 105
Schmidt, F.
2021-93 120
Schober, P.
2021-27 47
Schoonhoven, D.
2021-27 47
Schröder, S.
2021-104 132
Schroeder, F.A.
2020-14 213
2020-44 246
Schubert, E.K.
2020-42 244
2021-19 40
Schubert, J.J.
2020-35 236
2021-62 86
2021-63 87
Schuit, R.C.
2021-27 47
Schwarz, A.J.
2021-10 30
Sclocco, R.
2021-151 181
Scott, C.
2021-163 193
Scott, G.
2021-63 87
Scott, P.J.H.
2021-11 31
2021-70 94
Searle, G.
2021-91 117
Searle, G.E.
2020-34 234
Sedlak, T.W.
2020-13 212
2021-56 79
Seguin, J.
2021-99 127
Sehlin, D.
2021-32 52
2021-90 116
Seibyl, J.P.
2021-56 79
Sementa, T.
2020-02 200
Serdons, K.
2020-14 213
2020-44 246
Servaes, S.
2021-112 141
2021-120 149
2021-135 164
2021-148 177
2021-150 180
2021-160 189
2021-162 192
2021-55 78
2021-75 99
Shafiei, G.
2021-55 78
Shao, X.
2021-70 94
Sharp, D.J.
2021-63 87
Shatalina, E.
2021-113 142
2021-67 92
Shawn, W.N.
2020-23 223
Sheffer, R.
2020-42 244
2021-19 40
Shen, B.
2021-44 64
Shin, D.D.
2021-44 64
Shine, J.M.
2021-55 78
Shvetz, C.
2021-43 63
Siderowf, A.D.
2021-19 40
Silberbauer, L.
2021-123 152
Silveira, P.P.
2021-96 123
Silvestri, E.
2021-114 143
2021-65 89
Simpson, D.
2020-09 207
Singh, N.
2020-02 200
Singh, P.
2020-38 239
Singleton, T.A.
2021-11 31
Skinbjerg, M.
2021-01 21
Skosnik, P.D.
2021-79 103
Slifer, K.
2020-13 212
2021-56 79
Smart, K.
2020-04 202
2020-49 251
2021-22 43
2021-55 78
2021-79 103
Smith, E.E.
2021-163 193
Soddu, A.
2021-105 133
Sortwell, C.E.
2021-15 36
Sossi, V.
2020-48 250
2021-03 23
2021-110 139
2021-121 149
2021-13 33
2021-15 36
2021-163 193
2021-88 114
Soucy, J.-P.
2020-22 222
2021-11 31
2021-144 173
2021-150 180
2021-161 190
2021-163 193
2021-168 198
2021-55 78
Souza, D.O.
2020-10 208
2021-147 176
Spreng, R.N.
2021-55 78
Ssali, T.
2020-39 240
2020-47 249
2021-89 115
Staelens, S.
2020-22 222
2021-01 21
Stein, E.
2021-23 44
Steinberg, G.K.
2021-44 64
Stenkrona, P.
2021-81 105
Stephenson, N.
2021-07 27
Stern, Y.
2020-40 242
2021-154 184
Stevens, B.
2020-59 262
Stevenson, A.
2021-152 182
2021-155 185
2021-156 186
2021-165 195
Stevenson, J.
2020-22 222
2021-107 135
2021-112 141
2021-119 148
2021-120 149
2021-135 164
2021-152 182
2021-155 185
2021-156 186
2021-160 189
2021-162 192
2021-165 195
2021-75 99
Stoessl, A.J.
2021-03 23
2021-121 149
2021-13 33
2021-88 114
Stoll, A.C.
2021-15 36
Stoughton, C.
2021-116 145
Strafella, A.P.
2021-153 183
2021-166 196
Stroobants, S.
2020-22 222
2021-01 21
Strother, S.
2021-163 193
Sublette, E.
2020-30 230
Sullivan, B.J.
2021-52 74
Sundbom, M.
2021-24 45
Suzuki, M.
2021-18 39
2021-21 42
2021-92 119
Svarer, C.
2020-07 205
2021-04 24
2021-08 28
2021-14 35
Svensson, J.E.
2020-27 227
Swardfager, W.
2021-163 193
Sweeney, S.E.
2021-146 176
Sweeney, S.P.
2021-27 47
Syed, A.U.
2021-98 125
Syvänen, S.
2021-32 52
2021-90 116
T
Taddei, C.
2020-02 200
Tagare, H.
2021-42 62
Taha, A.
2021-105 133
Takahashi, K.
2021-142 171
Takano, A.
2021-10 30
2021-81 105
Talebi, M.
2021-50 71
Tamagnan, G.
2021-22 43
2021-29 49
Tarantal, A.F.
2020-15 215
Tardif, J.-C.
2021-163 193
Tartaglia, C.
2021-76 101
Tauscher, J.
2021-10 30
Taylor, S.F.
2021-117 146
Telu, S.
2020-18 218
2020-38 239
2021-37 57
Tempest, P.
2020-46 248
Teodoro, R.
2021-104 132
2021-84 109
Terry-Lorenzo, R.
2020-31 231
Therriault, J.
2020-11 210
2020-43 245
2020-58 261
2021-107 135
2021-112 141
2021-119 148
2021-120 149
2021-135 164
2021-147 176
2021-148 177
2021-150 180
2021-152 182
2021-155 185
2021-156 186
2021-160 189
2021-165 195
2021-53 75
2021-75 99
Thibault, E.
2021-59 83
Thiel, A.
2021-108 136
2021-163 193
Thiessen, J.D.
2020-23 223
Thomas, A.
2021-91 117
Thomsen, M.B.
2021-126 155
2021-158 188
Timmers, T.
2021-101 129
2021-125 154
2021-127 156
Tinaz, S.
2021-132 161
2021-42 62
Tippler, M.
2020-45 247
2020-49 251
2021-99 127
Tissot, C.
2020-11 210
2020-43 245
2020-58 261
2021-107 135
2021-112 141
2021-119 148
2021-120 149
2021-135 164
2021-147 176
2021-148 177
2021-150 180
2021-152 182
2021-155 185
2021-156 186
2021-160 189
2021-162 192
2021-165 195
2021-75 99
Tjerkaski, J.
2020-56 258
Tolf, A.
2020-52 254
Tollefson, S.
2021-116 145
2021-61 85
Tomassen, J.
2021-102 130
Tomljanovic, Z.
2020-40 242
2021-154 184
2021-77 102
Tong, J.
2021-09 29
2021-16 37
2021-36 56
Tonietto, M.
2021-91 117
Torrado-Carvajal, A.
2020-12 211
2020-54 256
2021-151 181
Torres, K.
2020-04 202
2021-139 169
Tournier, N.
2021-100 127
2021-66 90
2021-78 103
Toussaint, M.
2021-104 132
Toussaint, P.-J.
2021-108 136
2021-48 69
Toyonaga, T.
2020-15 215
2021-133 162
2021-139 169
2021-143 172
2021-35 55
2021-73 97
Treaba, C.A.
2021-06 26
Tremblay, R.
2020-45 247
2020-49 251
2021-99 127
Tremblay, S.
2021-43 63
Tresse, C.
2020-26 226
Truillet, C.
2021-66 90
Tseng, C.-E.J.
2021-06 26
2021-115 144
2021-17 38
2021-69 93
Tsujikawa, T.
2021-20 41
Tsukada, H.
2021-67 92
Tudorascu, D.
2020-37 238
2020-51 253
Tullis, T.
2020-25 225
Tuncel, H.
2021-125 154
2021-127 156
2021-27 47
Tuominen, L.
2021-41 61
2021-43 63
2021-55 78
Tuosto, M.
2020-20 220
Turkheimer, F.E.
2020-02 200
2020-20 220
2020-35 236
2020-41 243
2021-115 144
2021-31 51
2021-38 58
2021-62 86
2021-63 87
Tyndale, R.
2021-95 122
U
Uribe, C.
2021-153 183
2021-166 196
V
Valentine, H.
2021-52 74
Vallabhajosula, S.
2020-30 230
Vallee, I.
2021-76 101
Valli, M.
2021-153 183
2021-166 196
van Berckel, B.N.M.
2021-101 129
2021-102 130
2021-125 154
2021-127 156
2021-27 47
2021-86 111
van der Flier, W.M.
2021-101 129
2021-125 154
2021-127 156
van der Landen, S.
2021-102 130
van Dyck, C.H.
2021-131 160
Van Laere, K.
2020-14 213
2020-44 246
van Leeuwenstijn, M.
2021-101 129
Van Weehaeghe, D.
2020-14 213
Vanduffel, W.
2020-44 246
2021-145 174
Vanmechelen, E.
2021-112 141
Varlow, C.
2021-07 27
Varnäs, K.
2021-140 169
Varrone, A.
2020-07 205
2021-81 105
Vartananian, O.
2021-76 101
Vasdev, N.
2021-07 27
2021-09 29
2021-16 37
2021-36 56
2021-70 94
2021-76 101
2021-95 122
Verfaillie, S.C.J.
2021-101 129
2021-125 154
2021-27 47
Vergie, I.
2021-76 101
Verhaeghe, J.
2020-22 222
2021-01 21
Vermeulen, I.
2020-44 246
Veronese, M.
2020-02 200
2020-20 220
2020-35 236
2020-41 243
2021-115 144
2021-31 51
2021-38 58
2021-60 84
2021-62 86
2021-63 87
2021-91 117
Victorsson, P.
2020-57 260
Videbech, P.
2021-47 67
Visser, D.
2021-101 129
2021-125 154
2021-127 156
Visser, P.J.
2021-102 130
Vitali, P.
2021-150 180
2021-156 186
Vitaro, F.
2020-45 247
2020-49 251
2021-99 127
Volpi, T.
2020 Abstract 199
2021-105 133
2021-114 143
2021-65 89
Vraka, C.
2021-123 152
Vyas, P.
2020-13 212
W
Wadhwa, P.
2021-83 108
Wadsak, W.
2020-05 203
2021-123 152
2021-72 96
Wagner, F.
2020-44 246
2020-59 262
Wagstyl, K.
2021-108 136
Wahlestedt, J.N.
2021-124 154
Wainstein, G.
2021-55 78
Wakabayashi, Y.
2021-81 105
Wall, M.
2021-113 142
2021-67 92
Wang, C.
2021-06 26
Wang, J.
2021-136 165
Wang, L.
2020-55 257
Wang, Y.-T.
2020-43 245
2020-58 261
2021-107 135
2021-119 148
2021-135 164
2021-148 177
2021-150 180
2021-152 182
2021-160 189
2021-75 99
Wang, Z.
2021-96 123
Ward, C.G.
2020-42 244
Warsh, J.
2021-36 56
2021-76 101
2021-95 122
Watling, S.E.
2021-36 56
2021-76 101
2021-95 122
Wayne, C.
2021-77 102
Weerasekera, A.
2021-115 144
Wegener, G.
2021-126 155
2021-158 188
2021-47 67
2021-51 72
Wegener, T.
2021-03 23
2021-13 33
2021-88 114
Wehrli, F.W.
2020-47 249
Weiwer, M.
2020-59 262
Weltings, E.
2021-102 130
2021-127 156
Wenzel, B.
2021-104 132
2021-84 109
Wey, H.-Y.
2020-55 257
2021-130 159
2021-149 178
2021-41 61
Widström, C.
2020-36 237
Wikström, J.
2021-24 45
Wilson, O.
2021-39 59
Wilton, D.K.
2020-59 262
Wimberley, C.
2021-86 111
Wimmer, N.R.Z.
2021-06 26
2021-69 93
Windhorst, A.D.
2021-101 129
2021-102 130
2021-125 154
2021-27 47
Winkeler, A.
2021-100 127
Winkler, D.
2021-123 152
Winterdahl, M.
2021-126 155
2021-47 67
2021-49 70
2021-51 72
Wolf, J.
2021-139 169
Wolters, E.E.
2021-102 130
2021-125 154
2021-127 156
2021-27 47
Wolz, R.
2021-86 111
Wong, C.Y.W.
2021-82 106
Wong, D.F.
2020-13 212
2021-52 74
2021-56 79
Wong, R.
2020-31 231
Woodcock, E.A.
2020-50 252
2021-57 81
Wu, Y.
2021-146 176
Wyffels, L.
2021-01 21
X
Xie, Z.
2020-12 211
Xin, Y.
2021-74 98
Xiong, M.
2021-90 116
Xu, Y.
2021-22 43
Y
Yamada, T.
2021-18 39
2021-21 42
2021-92 119
Yan, X.
2020-18 218
2020-40 242
2021-111 140
2021-154 184
2021-83 108
Yang, J.
2020-03 201
2021-136 165
Yaqub, M.
2021-127 156
2021-86 111
2021-91 117
Yokell, D.L.
2021-109 138
Yoo, C.-H.
2021-130 159
Young, A.J.
2020-42 244
2021-19 40
Yous, S.
2021-11 31
Yousaf, T.
2021-62 86
Yu, Z.
2020-51 253
Z
Zürcher, N.R.
2021-115 144
2021-17 38
Zaharchuk, G.
2021-44 64
Zakiniaeiz, Y.
2021-12 32
2021-73 97
Zaman, S.
2020-37 238
Zammit, M.
2020-25 225
2020-37 238
Zanderigo, F.
2020-07 205
2020-24 224
2020-30 230
2020-32 232
2021-02 22
Zandi, A.
2021-146 176
Zanotti-Fregonara, P.
2020-18 218
2020-20 220
2021-111 140
2021-83 108
Zetterberg, H.
2021-112 141
2021-120 149
2021-150 180
Zhang, A.
2021-111 140
Zhang, L.
2021-22 43
2021-81 105
Zhang, Y.
2021-151 181
Zhao, M.Y.
2021-44 64
Zheng, M.-Q.
2021-22 43
Zhou, W.
2020-55 257
Zhou, Y.
2021-42 62
Zhu, C.
2020-03 201
Zhu, Y.
2021-106 134
Zientek, F.
2020-33 233
Zilles, K.
2021-108 136
2021-122 151
Zimmer, E.R.
2020-10 208
2021-112 141
2021-147 176
Zimmermann, M.
2021-161 190
Zoghbi, S.S.
2020-18 218
2020-38 239
2021-111 140
2021-81 105
2021-83 108
Zou, J.
2020-40 242
2021-154 184
Zukotynski, K.
2021-163 193
Keyword Index
[11C]-(R)-PK11195
2021-85 110
[11C]carfentanil
2021-130 159
[11C]-carfentanil
2021-79 103
[11C]carfentanil positron emission tomography
2021-117 146
[11C]Cimbi-36
2020-34 234
2021-14 170
[11C]colchicine
2021-70 94
[11C]CPPC
2021-146 176
[11C]DTBZ
2021-153 183
[11C]HD-800
2021-70 94
[11C]Martinostat
2021-06 26
[11C]MK-6884
2021-109 138
[11C]NOP-1A
2021-116 145
[11C]PBR2
2021-110 139
[11C]PBR28
2020-07 205
2021-61 85
[11C]-PBR28
2020-40 242
[11C]PE2I
2020-36 237
[11C]Raclopride
2021-34 54
[11C]Ro15-4513
2021-114 143
[11C]-UCB-J
2020-04 202
2021-133 162
2021-67 92
[11C]UCM1014
2021-11 31
[11C]UCM765
2021-11 31
[11C]verubulin
2021-70 94
[124I]IBETA
2021-87 112
[125I]IAZA
2021-167 197
[125I]-IPPI
2021-80 104
[15O]
2020-39 240
[15O]-oxygen
2020-47 249
[15O]water
2021-82 106
[15O]-water
2021-89 115
[18F]Bavarostat
2020-14 213
2020-44 246
[18F]-BCCP-EF
2021-67 92
[18F]EKZ-001
2020-14 213
2020-44 246
[18F]FDG brain PET
2021-138 168
[18F]FDG PET
2021-105 133
[18F]-FDG PET
2020-01 199
[18F]FDG SUVR
2021-65 89
[18F]FDOPA PET
2020-41 243
[18F]FDS PET imaging
2021-100 127
[18F]FEPPA
2021-141 170
[18F]flortaucipir
2021-125 154
[18F]Flotaza
2021-87 112
[18F]fluoroagomelatine
2021-11 31
[18F]FPEB
2021-140 169
[18F]-MK-6240
2020-40 242
[18F]NOS
2021-19 40
[18F]SynVesT-1
2021-90 116
[³H]ABP688
2021-161 190
[³H]UCB-J
2021-158 188
5
5-HT2A receptor
2021-14 170
5HT2A receptors
2021-47 67
5HT2C
2020-34 234
5XFAD
2021-58 82
A
A2A receptor
2021-104 132
acetylcholine receptors
2021-84 109
AD
2020-40 242
addiction
2020-50 252
2021-49 70
adenosine A1 receptor
2021-30 50
adenosine receptor
2021-104 132
Adenosine receptor
2021-28 49
adrenergic receptor
2020-21 221
aging
2021-119 148
2021-50 71
Aging
2021-131 160
alcohol-use disorder
2021-116 145
alpha2 adrenoceptors
2021-46 66
alpha7
2021-84 109
alpha-synuclein
2021-106 134
2021-15 184
2021-93 120
Alzheimer’s disease
2020-22 222
2020-46 248
2020-53 255
2021-07 27
2021-101 129
2021-102 130
2021-112 141
2021-118 146
2021-122 151
2021-125 154
2021-127 156
2021-135 164
2021-144 173
2021-147 176
2021-148 177
2021-150 180
2021-154 184
2021-155 185
2021-156 186
2021-159 188
2021-160 189
2021-162 192
2021-163 193
2021-167 197
2021-168 198
2021-18 39
2021-21 42
2021-27 47
2021-45 66
2021-48 69
2021-53 75
2021-58 82
2021-59 83
2021-74 98
2021-80 104
2021-86 111
2021-87 112
2021-92 119
2021-97 124
2021-98 125
amnestic MCI
2021-48 69
Amphetamine
2021-34 54
amplitude of low frequency fluctutations
2021-67 92
Amyloid
2020-11 210
2020-37 238
2021-101 129
2021-155 185
2021-156 186
2021-159 188
2021-162 192
2021-164 194
2021-166 196
2021-167 197
2021-168 198
2021-20 41
2021-32 52
2021-45 66
2021-53 75
2021-58 82
2021-86 111
2021-87 112
2021-98 125
Amyotrophic Lateral Sclerosis
2021-17 38
Anesthetics
2021-51 72
Animal model
2021-01 21
anosmia
2021-154 184
Antidepressant treatment response prediction
2020-19 219
APOE
2020-43 245
2021-148 177
arachidonic acid
2020-30 230
Arterial Spin Labeling
2020-05 203
2020-39 240
Astrocytes
2020-10 208
2021-45 66
asymmetry
2021-85 110
aterial input function
2020-52 254
atlas
2021-41 61
Atlas
2021-04 24
ATP-binding cassette
2021-66 90
Attenuation Correction
2020-55 257
autism
2021-139 169
autism spectrum disorder
2021-69 93
Autism spectrum disorder
2020-13 212
Autism spectrum disorder (ASD)
2021-56 79
Autism Spectrum Disorders
2020-09 207
autoradiography
2021-108 136
2021-140 169
2021-158 188
2021-54 76
2021-93 120
Autoradiography
2021-04 24
2021-118 146
2021-161 190
AV-133
2020-31 231
B
basal ganglia encephalitis
2021-77 102
bayes factor
2020-27 227
Bayesian
2020-29 229
2021-94 121
Bayesian modeling
2021-42 62
Behavioral Challenge
2020-06 204
2021-157 187
Behavioral psychology
2021-56 79
between-subject variance
2021-137 167
binding
2021-85 110
binding potential
2021-96 123
Binding Potential
2021-73 97
binding potential (BPND)
2021-141 170
biodistribution
2021-123 152
Biograph
2020-32 232
biomarker
2021-01 21
2021-147 176
biomarkers
2021-98 125
Biomarkers
2021-155 185
bipolar disorder
2020-02 200
2020-30 230
blood biomarker
2021-112 141
Blood Brain Barrier
2021-100 127
2021-62 86
blood-free PET quantification
2021-02 22
BMI
2021-131 160
Bolus-infusion
2021-14 170
BP quantification
2020-48 250
Braak stages
2021-135 164
Brain
2020-54 256
2021-30 50
brain activation
2020-04 202
brain development
2020-15 215
Brain imaging
2021-22 43
brain mapping
2021-103 131
Brain PET
2020-16 216
2021-129 158
2021-73 97
C
caffeine
2021-28 49
cAMP
2021-81 105
cannabis
2021-79 103
Carbon-11
2020-10 208
2021-78 103
carfentanil
2021-41 61
CBD
2021-97 124
CBF water PET
2020-52 254
Centiloids
2020-37 238
Cerebral Amyloid Angiopathy
2021-164 194
cerebral blood flow
2021-101 129
2021-125 154
2021-20 41
2021-24 45
2021-86 111
2021-89 115
cerebral cortex
2021-135 164
Cerebral cortex parcellation
2020-08 206
cerebral metabolic rate of glucose
2020-32 232
Cerebral metabolism
2020-47 249
Cerebrovascular disease
2021-44 64
cerebrovascular reactivity
2021-44 64
CerePET
2020-32 232
Chemogenetics
2021-23 44
Childhood Trauma
2021-136 165
choroid plexus
2021-63 87
Choroid Plexus
2021-62 86
chronic pain
2021-151 181
chronic traumatic encephalopathy
2021-76 101
Clearing agent
2021-32 52
click-reaction
2021-32 52
clinical assessment
2021-107 135
Clinical trials
2021-56 79
CMRGlu
2021-105 133
CMRO2
2020-47 249
cocaine
2021-158 188
cognition
2021-125 154
2021-55 78
cognitive decline
2021-147 176
2021-166 196
cognitive reserve
2021-165 195
Compartmental
2020-24 224
compartmental modeling
2020-29 229
compartmental modelling
2020-21 221
computer-aided diagnosis
2020-36 237
connectivity
2021-144 173
Connectivity
2021-53 75
constant infusion
2020-22 222
cortical thickness
2021-168 198
Covid
2021-162 192
COVID-19
2021-115 144
cox-2
2020-59 262
Crispr/Cas9
2021-128 157
cross-sectional
2020-46 248
CRP
2020-35 236
csf
2021-160 189
CSF
2021-120 149
CSF1R
2021-21 42
cyclooxygenase 1
2020-38 239
cyclooxygenase 2
2020-38 239
cyclooxygenase-1 and -2
2021-09 29
cyclooxygenase-2
2021-111 140
D
D3 receptors
2020-42 244
Data Harmonization
2021-73 97
data-driven Methods:2021-121 149
DaTscans
2021-42 62
Deep Learning
2020-41 243
Dementia
2020-39 240
2021-150 180
demyelination
2021-142 171
denoising
2021-121 149
depression
2020-35 236
Depression
2021-35 55
2021-43 63
2021-62 86
Deschloroclozapine (DCZ)
2020-18 218
designer receptor exclusively activated by designer drugs (DREADD)
2020-18 218
diprenorphine
2021-41 61
Direct-4D reconstruction
2021-137 167
disease progression
2021-122 151
Disease progression models
2021-42 62
Displacement
2021-14 35
docosahexaenoic acid
2021-61 85
Dopamine
2020-06 204
2020-42 244
2020-45 247
2021-05 25
2021-12 32
2021-145 174
2021-157 187
2021-25 46
2021-34 54
2021-49 70
2021-99 127
Dopamine D2-receptor
2020-57 260
Dopamine release
2021-121 149
2021-60 84
Dopamine synthesis
2021-60 84
2021-71 95
Dopamine transporter
2021-132 161
Dopaminergic degeneration
2021-15 36
Dopaminergic system
2021-03 23
2021-13 33
dorsal attention network
2021-152 182
dosimetry
2021-123 152
Down syndrome
2020-37 238
DREADDs
2021-23 44
drug-drug interaction
2021-66 90
DTBZ
2021-29 49
DTI
2020-57 260
2021-17 38
dynamic (DVR/BPND)
2021-127 156
Dynamic functional connectivity
2020-01 199
dynamic PET
2021-114 143
Dynamic PET
2020-35 236
dynamics
2021-55 78
E
Early Detection
2021-74 98
early life stress
2021-99 127
early stopping
2020-27 227
ebselen
2020-02 200
EC50
2021-129 158
EEG
2021-33 53
electroconvulsive stimulation
2021-47 67
electroencephalography
2020-03 201
2021-46 66
Endocannabinoid System
2021-95 122
enzyme
2020-26 226
2021-10 30
epigenetics
2021-69 93
epilepsy
2021-40 60
epilepsy.
2021-33 53
ER176
2021-77 102
executive function
2021-152 182
externalizing traits
2020-49 251
2021-99 127
extracerebral signal
2021-59 83
F
F18-CPFPX
2021-30 50
Fatty Acid Hydrolase
2021-95 122
FAZIN3
2020-53 255
FDG
2020-19 219
2021-159 188
2021-24 45
2021-43 63
FDG-PET
2021-33 53
2021-48 69
fear
2021-25 46
first-in-human
2020-14 213
2021-146 176
florbetapir
2020-22 222
fluorine-18
2021-07 27
2021-104 132
fMRI
2021-67 92
Focused Ultrasound
2021-100 127
FPEB
2020-02 200
2021-35 55
Free water
2021-163 193
frontal cortex
2021-96 123
frontotemporal dementia
2021-119 148
FTC146
2021-37 57
Fully quantitative
2020-19 219
function
2021-55 78
functional connectivity
2021-72 96
Functional connectivity
2021-128 157
functional MRI
2020-05 203
2021-20 41
functional PET
2020-05 203
2021-149 178
2021-71 95
G
GABA
2021-04 24
gender
2021-30 50
Gene editing
2021-128 157
gene expression
2021-54 76
Genes
2021-128 157
genetics
2021-102 130
Geometric Transfer Matrix
2020-29 229
Glia
2020-12 211
glial activation
2021-120 149
glial cells
2021-151 181
glioma
2021-82 106
GLT-1
2020-10 208
glucose metabolism
2020-30 230
glucose metabolism
2021-72 96
Glucose metabolism
2021-24 45
GluN2B
2021-22 43
GluN2B subunit
2021-123 152
glutamate
2020-02 200
2020-45 247
2021-52 74
2021-60 84
glycogen synthase kinase-3
2021-09 29
Graph theory
2020-01 199
gray matter volume
2021-96 123
grey matter volume
2021-60 84
H
HDAC
2021-69 93
HDAC6
2020-14 213
2020-44 246
Healthy
2021-115 144
healthy ageing
2021-90 116
healthy brain
2020-30 230
heterogeneity
2021-82 106
Hierarchical clusering
2020-08 206
Hippocampus
2021-161 190
histone deacetylases
2021-06 26
HIV
2020-54 256
HSCT
2020-52 254
HT1A receptors
2020-42 244
human
2021-10 30
2021-39 59
Human embryonic stem cells
2021-124 154
Human hippocampus
2021-118 146
Huntington’s Disease
2021-01 21
I
I-124
2021-58 82
IDIF
2021-105 133
2021-110 139
image analysis
2021-80 104
Image Analysis
2021-64 88
Image Derived Input Function
2020-29 229
image quality
2021-138 168
image-derived input function
2021-89 115
image-derived input function (IDIF)
2021-141 170
imaging biomarker
2021-112 141
Immune activation
2020-09 207
2020-28 228
Immune system
2021-155 185
independent component analysis
2021-48 69
inflammation
2020-38 239
2021-151 181
Inflammation
2020-12 211
injected radioactivity
2020-27 227
Internalization
2021-34 54
Inter-Scanner
2021-73 97
IPPI Tau
2021-118 146
irreversible
2021-10 30
irreversible kinetics
2021-02 22
isoflurane
2021-51 72
K
kainic acid
2021-33 53
kinetic modeling
2021-02 22
2021-08 28
2021-140 169
Kinetic Modeling
2020-06 204
2021-145 174
2021-157 187
kinetic modelling
2020-07 205
2020-52 254
2021-90 116
Kinetic modelling
2020-16 216
kinetic models
2020-18 218
Knee osteoarthritis
2020-12 211
L
Lassen Plot
2021-129 158
lateral ventricles
2021-63 87
Lateralization
2021-75 99
LDAEP
2020-03 201
learning and memory
2021-50 71
lesions
2021-06 26
Lewy Bodies
2021-64 88
limbic system
2020-13 212
linear-parametric neurotransmitter PET
2021-05 25
Lipopolysaccharide
2021-16 37
longitudinal
2021-133 162
longitudinal [18F]flortaucipir PET
2021-127 156
LSD1
2021-10 30
lysine-specific histone demethylase
2021-10 30
M
M4 mAChR
2021-109 138
macrophage colony stimulating factor 1 receptor
2021-146 176
MAGL
2020-26 226
Major Depressive Disorder
2020-19 219
2021-136 165
MAO-A
2021-98 125
mapping
2021-152 182
MDD
2021-81 105
MDL100907
2021-47 67
medial temporal lobe
2021-75 99
Melatonin receptors
2021-11 31
membrane transporter
2021-66 90
mental health
2021-99 127
Mesial Temporal Lobe Epilepsy
2021-161 190
metabolic connectivity mapping
2021-72 96
metabolism
2021-149 178
Metabolism
2021-136 165
Metabotropic glutamate receptor 5 (mGluR5)
2021-143 172
mGlu5
2020-49 251
mglur5
2021-35 55
mGluR5
2021-07 27
2021-161 190
mGuR5
2021-39 59
mHTT
2021-01 21
mice
2021-90 116
microdialysis
2021-52 74
microglia
2020-59 262
2021-146 176
2021-154 184
2021-18 39
2021-21 42
2021-77 102
Microparameters
2021-105 133
MicroPET imaging
2021-33 53
Microstructure
2021-163 193
Microtubules
2021-70 94
mild cognitive impairment
2021-45 66
mild traumatic brain injury
2021-76 101
Mild Traumatic Brain Injury
2021-95 122
minipigs
2021-51 72
Minipigs
2021-124 154
2021-126 155
mitochondrial dysfunction
2021-50 71
mk2640
2021-160 189
MK-6240
2021-59 83
modeling
2020-26 226
modelling
2021-38 58
Modelling
2020-24 224
2020-56 258
Molecular connectivity
2021-114 143
Molecular Imaging
2021-83 108
money
2020-27 227
Monkey
2021-23 44
Monoamine Oxidase
2021-64 88
Monoamine Oxidase A
2020-53 255
Monoamine oxidase B
2021-16 37
morphine
2020-50 252
Morphine
2021-57 81
mouse imaging
2021-09 29
MRI
2020-17 217
2020-39 240
2021-136 165
mRNA
2020-08 206
2021-04 24
2021-68 92
MRS
2021-115 144
mu opioid receptor
2021-79 103
Mu opioid receptor
2021-117 146
Multicenter collaboration
2020-28 228
Multilevel
2020-24 224
multi-modal
2021-122 151
multimodal analysis
2021-20 41
Multimodal imaging
2021-68 92
multimodal neuroimaging
2021-113 142
2021-67 92
Multiple regression modelling
2021-65 89
Multiple sclerosis
2021-06 26
multiple system atrophy
2021-93 120
multi-site PET
2021-08 28
Multi-tracer
2020-16 216
multi-tracer PET
2021-88 114
mu-opioid receptor
2021-130 159
myelin
2021-103 131
N
net influx rate
2021-02 22
Network
2021-53 75
neurobiology
2021-103 131
neurodegeneration
2021-104 132
2021-122 151
2021-147 176
2021-63 87
Neurodevelopmental disabilities
2021-56 79
neuroepigenetics
2021-06 26
neuroimaging
2021-109 138
2021-52 74
Neuroimaging
2021-74 98
2021-83 108
neuroimmune signaling
2020-50 252
Neuroimmune System
2021-57 81
Neuroinflamation
2021-17 38
neuroinflammation
2020-35 236
2020-59 262
2021-111 140
2021-144 173
2021-168 198
2021-92 119
Neuroinflammation
2020-54 256
2021-115 144
2021-148 177
2021-16 37
2021-162 192
2021-19 40
2021-62 86
neuronal plasticity-related proteins
2021-26 46
neuropathic
2021-151 181
neuropharmacology
2021-66 90
neuroplasticity
2021-72 96
neuropsychiatric disorders
2021-123 152
neuropsychology
2020-11 210
neuropsychologytau
2021-107 135
neuroreceptor template
2021-38 58
neurotransmitter release
2021-25 46
neurotransmitters
2021-108 136
Neutrophils
2021-155 185
next-generation
2021-05 25
NHP
2020-55 257
2021-109 138
Nicotine
2020-17 217
2021-50 71
Nicotine patch
2021-12 151
Nicotinic receptors
2020-17 217
Nicotinic Receptors
2020-33 233
Nifene
2021-118 146
NMDA receptor
2021-123 152
2021-22 43
N-methyl-D-aspartate
2021-52 74
nondisplaceable binding potential
2020-13 212
non-human primate
2020-31 231
nonverbal memory
2021-75 99
Norepinephrine
2021-131 160
norepinephrine transporter
2021-26 46
novel pattern analysis
2021-13 164
Novel pattern analysis
2021-03 23
NR2B/NMDA receptor
2021-37 57
NR2B-SMe
2021-37 57
Nucleus accumbens
2021-117 146
O
Obesity
2020-33 233
2021-24 45
Occupancy
2020-31 231
2021-129 158
olfaction
2021-154 184
oncology
2021-09 29
opioid
2021-49 70
2021-79 103
opioid receptor
2021-41 61
Opioid use disorder
2020-50 252
2021-57 81
optimization
2021-138 168
optogenetics
2021-134 163
oxidative stress
2021-106 134
2021-50 71
oxygenation
2021-44 64
P
Pain
2020-12 211
PAM
2021-109 138
PANDAS
2021-77 102
parametric images
2020-36 237
parametric imaging
2021-05 25
parametric mapping
2020-20 220
Parametric-MOLAR
2021-137 167
Parkinson’s disease
2020-36 237
2020-53 255
2021-13 164
2021-153 183
2021-166 196
2021-19 40
2021-29 49
2021-42 62
2021-46 66
2021-64 88
2021-88 114
2021-93 120
Parkinson’s disease
2021-03 23
2021-106 134
2021-124 154
2021-132 161
Partial volume correction
2021-08 28
Partial Volume Correction
2020-29 229
PBIF
2021-110 139
PD rodent model
2021-15 181
PDE4B
2021-81 105
perfusion
2021-44 64
2021-82 106
PET
2020-21 221
2020-51 253
2021-26 46
PET [18F]flortaucipir
2021-76 101
PET imaging
2020-02 200
2021-01 21
2021-07 27
2021-09 29
2021-124 154
2021-131 160
2021-15 151
2021-164 194
PET reporter systems
2021-83 108
PET tracer evaluation
2021-93 120
PET/CT
2021-19 10
2021-58 82
PET/CT studies
2021-87 112
PET/fMRI
2021-128 157
PET/MR
2020-05 203
2021-24 45
PET/MRI
2020-06 204
2020-47 249
2021-145 174
2021-157 187
2021-20 41
2021-34 54
2021-44 64
2021-89 115
PET-BIDS
2021-91 117
PET-fMRI
2020-33 233
PET-imaging
2021-156 186
PET-MR
2021-17 38
PET-MRI
2021-149 178
2021-69 93
PF-06445974
2021-81 105
phantom
2020-51 253
Pharmacokinetic
2020-22 222
2020-24 224
2020-56 258
2021-100 127
2021-103 131
2021-66 90
2021-94 121
Photoacoustic
2021-52 74
PiB
2020-51 253
2021-159 188
Pick’s Disease
2021-97 124
piglets
2021-84 109
polygenic score
2021-96 123
population variability
2021-38 58
portable PET device
2020-32 232
Positron emission tomography (PET)
2021-56 79
positron-emission tomography
2021-111 140
Posttraumatic stress disorder
2021-143 172
2021-95 122
Post-traumatic Stress Disorder
2021-36 56
Precision medicine
2021-31 51
Preclinical
2020-23 223
2021-126 155
preclinical alzheimer’s
2021-152 182
pre-clinical imaging
2020-15 215
2020-59 262
pre-targeting
2021-32 52
principal component analysis
2021-121 149
Progressive Supranuclear Palsy
2020-46 248
propagation
2021-135 164
propofol
2021-51 72
Pro-radiotracer
2020-10 208
Proteome
2021-68 92
Proteomics
2021-143 172
PSAMs
2021-23 44
pseudo-CT
2020-55 257
psychiatry
2021-99 127
Psychosis
2020-28 228
2021-31 51
p-tau
2021-160 189
p-Tau biomarkers
2021-112 141
PVC
2020-51 253
Q
Quality Control
2020-41 243
quantification
2021-127 156
2021-138 168
2021-86 111
quantification Methods:
2020-31 231
QuPath
2021-64 88
2021-80 104
R
R
2020-56 258
R1
2021-101 129
raclopride
2021-25 46
radiochemistry
2021-78 103
Radiotracer
2021-22 43
reactive oxygen species
2021-106 134
receptor desensitization
2021-130 159
receptor trafficking
2021-130 159
receptors
2021-108 136
reconstruction
2020-51 253
2021-108 136
2021-138 168
reference region
2021-140 169
refractory
2021-40 60
Rejection sensitivity
2021-117 146
relationship
2020-30 230
reliability
2020-21 221
REM Sleep Behaviour Disorder
2021-153 183
Reproducible
2020-56 258
Resting state
2021-65 59
Resting state brain networks
2020-01 199
Reward Circuitry
2021-57 81
reward processing
2021-71 95
rheumatoid arthritis
2020-38 29
RIPK1
2021-18 39
2021-92 119
Rodent models
2021-143 172
ROStrace
2021-106 134
S
scanning window
2021-29 49
schizophrenia
2021-113 142
Schizophrenia
2020-28 228
2020-57 260
sequential testning
2020-27 227
serotonin
2021-47 67
Serotonin
2020-34 234
serotonin 1A
2020-03 201
serotonin receptor
2021-96 123
serotonin transporter
2020-03 201
Sex difference
2020-43 245
sigma receptor
2021-37 57
signal classification
2021-05 25
signal standardization
2021-114 143
Simulations
2021-129 158
simultaneous estimation
2020-07 205
simultaneous estimation (SIME)
2021-141 170
Simultaneous Estimation (SimE) of the arterial input function
Figure 1. 
APOE modulates the relationships between neuroinflammation biomarkers and tau burden
