186. Compartmental analysis of [11C]temozolomide in gliomas and normal brain tissue
R. Hinz1, A. Saleem2, C.S. Brock3, J.C. Matthews1, T. Jones2 and P.M. Price2
1Wolfson Molecular Imaging Centre, University of Manchester; 2Academic Department of Radiation Oncology, Christie Hospital NHS Trust, Manchester; 3Imperial College Healthcare NHS Trust, London, UK
Background and aims: This investigation presents a comparison of compartmental modelling with spectral analysis for the kinetic analysis of a radiolabelled anti-cancer drug [11C]temozolomide aiming at resolving inconsistencies in previously reported results.
Methods: 15 glioma patients underwent paired dynamic PET scans for 90 mins with [11C]temozolomide on the brain tomograph ECAT 953B. 3D imaging data and arterial plasma input functions were acquired and regions of interest defined for the whole tumour, for a tumour area with high signal (tumour rim) and on contralateral normal brain tissue.
Purpose: Two previously published reports analysing these [11C]temozolomide data with spectral analysis found that distribution volumes (VD) and plasma-to-tissue rate constants (K1) could differ as much as an order of magnitude dependent on the treatment of the fractional tissue blood volume.1,2 In an attempt to resolve these differences, we have characterised the kinetics of [11C]temozolomide in gliomas and normal brain tissue with reversible compartmental models.
Results: Consistent with the previous reports,1,2 the [11C]temozolomide kinetics in normal brain tissue is adequately described by a one-tissue compartment model with a free blood volume term. In gliomas, an additional exponential was identified with spectral analysis. In line with this, the fit of the tumour tissue-time activity curves with a two-tissue compartment model and a free blood volume term was significantly better than the fit with the one-tissue compartment model.
Population mean…
The figure shows mean values with one standard deviation bars for K1, VD and the mean residence time of the tracer in tissue (MRT) from our cohort of 15 subjects obtained with compartmental analysis. The population mean values for K1 are in close agreement with those obtained with spectral analysis in ref. 2, when the contribution from the blood volume was excluded, and are about an order of magnitude smaller than those in ref. 1, where the blood volume term was erroneously included in the calculation of K1. Contrary to the findings of VDnormal = 6.97 and VDtumour = 54.4 in ref. 1, our VD estimates are only slightly smaller than those reported in ref. 2. Consequently, the MRT estimates for normal tissue are slightly smaller than in ref. 2, but in contrast to ref. 1, clearly distinct to those obtained for tumour and tumour rim, respectively.
Conclusions: Compartmental modelling provides stable estimates of [11C]temozolomide kinetic parameters compared to spectral analysis due to its ability to constrain the parameter space independent of the choice of a functional base, thus reducing variability. Consistent with earlier reports,1,2 the increase in K1 best reflects the activity of the lesion suggesting changes of the [11C]temozolomide blood-brain-barrier penetration in malignant disease.
Population mean±one standard deviation.
221. Error analysis of dopamine D2 receptor occupancy study with agonist ligand [11C]MNPA
M. Shidahara1, H. Ito1, T. Otsuka1, Y. Ikoma2, C. Seki1, F. Kodaka1, R. Arakawa1, H. Takano1, H. Takahashi1, Y. Kimura1, I. Kanno1 and T. Suhara1
1Molecular Imaging Center, National Institute of Radiological Sciences, Chiba; 2Department of Investigative Radiology, National Cardio-Vascular Center Research Institute, Suita, Japan
Objectives: Occupancy of dopamine D2 receptors by antipsychotic drugs can be estimated from reduction in the observed binding potential (BPND). Because BPND varies widely in occupancy studies, accuracy of the measurement of wide range BPND should be confirmed. The purpose of this study is to investigate errors in quantitative analysis for estimating dopamine D2 receptor occupancy by antipsychotics with agonist ligand [11C]MNPA which has high affinity and selectivity to dopamine D2 receptors.1
Methods: Simulated TACs of [11C]MNPA with several noise levels were generated to investigate the bias and variation of parameter estimates caused by the statistical noise for non-linear least square (NLS) fitting and a simplified reference tissue model (SRTM) methods. A dynamic tracer concentrations was consisted of three target conditions (K1 = 0.44, k2 = 0.067, k3 = 0.02, 0.1 or 0.2, k4 = 0.18), and a reference (K1 = 0.44, k2 = 0.067) with a dynamic frame (20secs × 9, 1 mins × 5, 2 mins × 4, 4 mins × 11, 5 mins × 6, totally 90 mins) and a measured input function from human study.1 Three true BPND values were assumed; 1.111 (baseline), 0.556 (occupancy = 50%) and 0.111 (90%). Then Gaussian noise was added at the noise level 1%, 3%, 5%, 7% and 10%, and five hundred noisy data sets were generated for each.2 The reliability of BPND estimated by NLS and a SRTM and the calculated occupancy, dependency of scan durations (32, 44, 60, 75 and 90 mins) for SRTM were investigated.
Results: For NLS and SRTM methods, the bias of estimated BPND values became larger as the noise level increased and these bias of BPND were larger than those of occupancy (Figures 1A and 1B). In the case of small BPND, the bias became larger. For SRTM method, reliable and unbiased occupancy estimates of [11C]MNPA could be obtained by 60 mins with the relative standard deviation remaining less than 10%. However, shorten scan duration depredate the quantification of very small binding potential (Figure 1C).
(A) Noise level dependency: Bias of BP and occupancy estimated by NLS, (B) estimated by SRTM, (C) Scan duration dependency: Bias of BP and occupancy at a 3% noise level estimated by SRTM.
Conclusions: Dopamine D2 receptor occupancy by antipsychotics can be estimated precisely by SRTM method with an optimal scan duration with [11C]MNPA.
384. Correlation between intraindividual serotonin transporter and serotonin 1A receptor in normal male subjects: a positron emission tomography study
H. Takano, H. Ito, H. Takahashi, R. Arakawa, M. Miyoshi, F. Kodaka, T. Otsuka and T. Suhara
Clinical Neuroimaging Team, Molecular Neuroimaging Group, National Institute of Radiological Sciences, Chiba, Japan
Background and aims: The central serotonergic (5-HT) system is closely involved in regulating various mental functions associated with mood, anxiety, impulsive behavior etc. To date, several reports have addressed the comparison between pre- and postsynaptic functions of 5-HT;1,2 however, these studies are limited in their applicability because of their small sample size. The aim of this study was to investigate the intraindividual relationship between 5-HTT and 5-HT1A receptors using PET in a relatively large group of normal male subjects.
Methods: Dynamic PET scans were performed on 23 young, healthy males with both [C-11]DASB and [C-11]WAY-100635 to measure the binding of 5-HTT and 5-HT1A receptors, respectively. A 2-parameter multilinear reference tissue model and a reference tissue model were used to calculate the binding potential (BPND) of [C-11]DASB and [C-11]WAY-100635, respectively, on a voxel-by-voxel basis, with the cerebellum as the reference region. All PET images were anatomically standardized to the template magnetic resonance image and smoothed with an 8-mm Gaussian filter using SPM2. Furthermore, the volumes-of-interest (VOIs), namely, the raphe nucleus, thalamus, striatum, parahippocampal gyrus, insula, temporal cortex, base of the frontal cortex, and the convexity of the frontal cortex were manually traced on the standardized and smoothed BPND images. Pearson's coefficient correlation was used to compare the bindings of both the tracers in the same VOIs; corrections for multiple comparisons among VOIs were not performed.
Results: The BPND values of each VOI are presented in Table (mean±s.d.).
DASB
WAY
Raphe Nuclei
2.46±0.53
1.48±0.46
Thalamus
1.67±0.28
0.68±0.28
Striatum
1.46±0.17
1.38±0.49
Parahippocampus
0.89±0.16
5.13±0.98
Insula
0.89±0.09
5.26±0.97
Anterior Cingulate
0.48±0.09
4.1±0.63
Temporal
0.25±0.05
4.24±0.79
Frontal base
0.26±0.03
3.33±0.64
Frontal convexity
0.24±0.05
3.29±0.58
Pearson's correlation coefficient indicated significant inverse correlations between BPND of [C-11]DASB and that of [C-11]WAY-100635 in the insula (r = −0.46, P = 0.028), base side of the frontal cortex (r = −0.47, P = 0.023), and the convex side of the frontal cortex (r = −0.46, P = 0.029). No significant correlation was observed in other regions.
Conclusions: The distribution of 5-HTT and 5-HT1A receptors in the brain was consistent with the findings of post-mortem studies. In addition, we found inverse correlations between BPND values of 5-HTT and 5-HT1A in the insula and frontal cortex. These relationships might indicate a complementary nature between 5-HTT and 5-HT1A functions in such regions. Further investigations are required to elucidate the relationship between the activity of the pre- and postsynaptic serotonergic system in relation to mental functions.
402. Quantitative assessment of density and affinity for dopamine D2 receptors with multiple injections of [11C]raclopride
Y. Ikoma1,2, H. Watabe1, T. Hayashi1, K. Minato2 and H. Iida1
1Department of Investigative Radiology, National Cardiovascular Center Research Institute, Suita; 2Graduate School of Information Science, Nara Institute of Science and Technology, Ikoma, Japan
Objectives: Positron emission tomography with [11C]raclopride is widely used to investigate the dopamine D2 receptor system that is related with several neurological and psychiatric disorders in the brain. The dopamine D2 receptor density (Bmax) and affinity (Kd) can be determined using a graphical analysis based on Scatchard plot with PET scans.1,2 However, in this estimation, multiple PET scans are required under variable specific activity of administered [11C]raclopride, so that they are often obtained on separate day. To shorten the study period of Bmax and Kd measurement, we have developed a multiple injection simplified reference tissue model (MI-SRTM) that provides the change of binding potential (BP) value from a single session of PET scanning in conjunction with multiple injections of [11C]raclopride.3 In this study, we established the method for estimating Bmax and Kd with multiple-injection approach, and validated the proposed method by performing numerous simulations and studies on monkeys with PET and [11C]raclopride.
Methods: In the simulation studies, to determine the injection protocol of our multiple-injection approach, time-activity curves (TACs) of the striatum and cerebellum were generated using plasma input function and two-tissue compartment 5-parameter model including K1, k2, kon, Bmax, and koff for various mass of administered raclopride, and a relation between injected mass and BP estimated by MI-SRTM was investigated. Next, TACs were simulated with various Bmax and kon with proposed multiple-injection protocol. In these TACs, Bmax and Kd were estimated by a graphical analysis of Scatchard plot, and estimated values were compared with true values. In the studies on three monkeys, PET scans were performed using PCA-2000A and three injections of [11C]raclopride, and Bmax and Kd were measured by the graphical analysis.
Results: In the simulation study with various injection mass, BP estimates became smaller as the injected mass increased. Based on this result, we planned a protocol for three injections with 50-mins intervals of 1.5, 10, and 30 nmol/kg raclopride so that the BP estimates would be high, middle, and small. Using this injection protocol, Bmax and Kd could be estimated by the graphical analysis. Both of Bmax and Kd were overestimated. However, in TACs with various Kd and fixed Bmax, there was a good correlation between true and estimated Kd, and estimated Bmax was constant. Meanwhile, in TACs with various Bmax and fixed Kd, a good correlation was observed between true and estimated Bmax, and there was little variation in estimated Kd. The estimation error became larger in the case where the Kd was larger or Bmax was smaller. In the monkey studies, estimated Bmax and Kd were about 32 and 16 pmol/mL, respectively, and both values were close to those previously obtained by the conventional method.2
Conclusions: The dopamine D2 receptor density and affinity could be estimated by 150-mins PET scan with three injections of [11C]raclopride with 50-mins intervals.
472. Electroconvulsive therapy induces alterations in dopamine receptor and transporter binding in the striatum of healthy and MPTP-lesioned non-human primates
A.M. Landau1,2, M.M. Chakravarty1, S. Jivan2, A.P. Zis2, A. Gjedde1 and D.J. Doudet1,2
1PET Center and Center of Functionally Integrative Neuroscience, Aarhus University Hospital, Aarhus, Denmark; 2Medicine/Neurology, University of British Columbia, Vancouver, BC, Canada
Objectives: Parkinson's disease (PD) is a progressive neurodegenerative disorder. Depression is one of the most disabling co-morbidities associated with PD and both PD and depression are poorly treated in advanced patients. Electroconvulsive therapy (ECT), a form of electric brain stimulation, is one of the most effective therapies used to treat depression and has additional beneficial effects on the motor symptoms of PD. Little is known about the mechanisms of action of ECT in PD patients. Rodent studies do not allow long-term longitudinal studies of regional changes in monoamine neurotransmission and ethical concerns reduce the feasibility of performing extensive longitudinal studies in human PD subjects. The aim of this work is to explore the neural mechanisms underlying the therapeutic effects of ECT, focusing on dopaminergic neurotransmission in the lesioned and unlesioned striatum of non-human primates.
Methods: The 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) neurotoxin was used to induce hemi-parkinsonism (unilateral lesion) in 4 rhesus monkeys and full parkinsonism (bilateral lesion) in 3. Six monkeys were left unlesioned. ECT was performed under anaesthesia using a clinical ECT device. Electrodes were placed bilaterally and 6 ECT treatments were administered over a 3 week period. Animals received a PET scan before the initiation of ECT treatment and then 24 to 48 h, 1 week and 6 weeks after the end of the course of ECT. Monkeys were scanned using SCH23390, a dopamine D1 receptor ligand; raclopride, a dopamine D2 receptor antagonist; and methylphenidate, which binds to the dopamine transporter (DAT). PET data were coregistered to an average MR from 12 different monkeys and binding potentials were determined using the Logan model for the 30 to 60 mins time period of the scan. Left and right sides of each brain region were averaged in unlesioned and bilaterally lesioned monkeys; whereas sides were kept separate for the analysis in unilaterally lesioned monkeys.
Results: In unlesioned monkeys, ECT induced a significant increase in dopamine D1 and a decrease in dopamine D2 receptor binding in the striatum. DAT binding increased at 24 to 48 h and 1 week following the final ECT treatment but returned to baseline levels by 6 weeks. In full PD and the lesioned side of hemi-PD monkeys, no effects were observed in DAT binding and, like in unlesioned monkeys, increases in D1 binding were again observed. Decreases were still observed in raclopride binding of D2 receptors in full PD monkeys.
Conclusion: These data are in agreement with studies in unlesioned and parkinsonian rodents showing increases in dopamine D1 receptor binding in response to ECT. Effects of ECT on D2 receptors and DAT have not been previously investigated. Better understanding of the mechanism of action of ECT may lead to refinements in its clinical therapeutic applications in depression and PD.
649. Cognitive function and intelligence related to serotonergic neurotransmission: a brain PET study in young healthy volunteers with 11C-DASB
K. Madsen1,2, D. Erritzoe1,2, G. Zornhagen2,3, E.L. Mortensen2,3, G.M. Knudsen1,2 and S.G. Hasselbalch1,2
1Neurobiology Research Unit, Rigshospitalet; 2Center for Integrated Molecular Brain Imaging; 3University of Copenhagen, Copenhagen, Denmark
Objectives: Studies of central serotonergic activity in healthy volunteers have indicated that long-term memory and executive functions are sensitive to manipulations. Diseases involving the serotonin system often also involves cognitive dysfunction. However, only few PET serotonin studies have involved cognitive research in healthy volunteers. In this study was a positive correlation hypothesized between regional BPND of the serotonin transporter (SERT) and intelligence, primarily in left hemisphere regions.
Methods: A bolus injection of 11C-DASB was followed by a 90 mins dynamic PET scan in a GE Advance scanner in 32 healthy volunteers (25 males, mean age 25.8 years, range 20 to 37). Volumes of interest (VOIs) were delineated automatically on co-registered 3T MRIs. Time-activity curves were extracted from the VOIs' gray matter voxels, and BPND was modeled using Ichise's MRTM2 method.
Long term memory was measured with Rey Auditory Verbal Learning Test and Rey's Complex Figure. Executive function was measured with Trail Making B Test and Stroop. General intelligence was measured with The Danish Adult Reading Test (DART), and The Word and Number Test (WNT) (N = 24). Educational years were scored. Correlations were performed in 4 bilateral fronto-striatal VOIs with age and day-light hours as covariates.
Results: A positive correlation was found between WNT and BPND in left caudate (R = 0.586, P = 0.0026, see Figure), right caudate (R = 0.578, P = 0.0398), left ventrolateral prefrontal cortex (R = 0.495, P = 0.0139), a tendency was found in left dorsolateral prefrontal cortex (R = 0.366, P = 0.0784), but none in putamen. A positive correlation was also found between education years and left caudate (R = 0.438, P = 0.012) and a tendency was found in right caudate (R = 0.526, P = 0.096). No correlations between regional BPND and DART, specific memory and executive function tests were found. All p-values are uncorrected.
SERT binding in left caudate in relation to WNT.
Conclusion: Our results showed a positive correlation between SERT binding primarily in the left hemisphere regions and intelligence measured with WNT. This was supported by same the pattern for education years found in caudate. Our results indicate that a putative link between the SERT level and general intelligence must be mediated through other mechanisms than mere memory and executive functions. However, an efficient 5-HT reuptake system seems to be of importance for general intelligence.
746. Effect of levodopa and DA agonist treatment on DA release patterns and DTBZ binding in the 6-OHDA Parkinson's rat model
V. Sossi1, K. Dinelle1, D. Doudet2 and R. De La Fuente Fernandez2
1Physics and Astronomy, University of British Columbia; 2Medicine, University of British Columbia, Vancouver, BC, Canada
Background: There is still controversy on the effect of levodopa and dopamine (DA) agonist treatment on Parkinson's disease (PD) progression as evaluated with imaging studies and on their impact on changes in synaptic DA levels. The latter point is of great clinical interest due to the established link between oscillations in synaptic DA levels and propensity to develop treatment-related motor complications. Here we investigated the effect of chronic levodopa and pramipexole treatment on the binding of the 11C-dihydrotetrabenazine (DTBZ) tracer to the vesicular transporter of type 2 (VMAT2), taken as a marker of dopaminergic integrity, and on changes in synaptic levodopa-derived synaptic DA levels (delta (DA) as measured by a double 11C-Racloprice (RAC) scan using the unilateral 6-OHDA rat model of PD.
Methods: 28 unilaterally 6-OHDA lesioned rats underwent a DTBZ, and two RAC scans (baseline and 45 mins after I.P. administration of levodopa/benserazide (50 mg/kg, 15 mg/kg prepared). After being assigned to three degree-of-denervation-matched treatment groups (saline (S, N = 7), pramipexole (PRA, N = 10) and levodopa (LD, N = 11)) and undergoing treatment for approximately 4 weeks, they underwent an identical scanning protocol and imaging measures were compared. Dopaminergic denervation as estimated from the pre-treatment DTBZ scan ranged between 17% and 96%. ANOVA analysis was used to compare pre- and post-treatment results expressed in terms of BPND for the control and lesioned side respectively.
Results: On the control side, no treatment effect was found for DTBZ binding together with a significant overall reduction (P = 0.003). On average the PRA group showed a slightly higher reduction (∼7% versus ∼3% for the other two groups). On the lesioned side a significant treatment effect was observed (P = 0.043) with the PRA group only showing a slight decrease. After adjusting for baseline DTBZ and RAC binding values, a treatment effect (P = 0.018) was observed for delta (DA) on the lesioned side with no change for the PRA and LD treated groups and a trend towards an increase in the S group. On the control side a significant treatment effect was observed (P = 0.032) with a trend toward an increase of delta (DA) in the PRA treated group.
Conclusion: The most important finding of this study is that once denervation is present, neither pharmacological treatment changes delta (DA), thus neither treatment should influence the probability of developing motor complications. In addition, it shows that DTBZ might not be an objective marker to accurately monitor disease progression: the fact that a reduction in DTBZ binding on the control side was observed for all treatment groups, including saline, indicates that such reduction likely does not arise as a consequence of a possible treatment-induced terminal damage but rather from competition with DA arising either from increased synthesis (with the corollary that the control side might be susceptible to some of the regulatory changes occurring on the lesion side) or pharmacologically increased availability thus rendering this marker sensitive to either compensatory or pharmacological effects consistently with previous findings. Care must be thus taken when using PET markers to objectively determine either treatment effects or disease progression.
750. Possible effects of family history of alcoholism on the test-retest variability of striatal D2 availability
K. Yoder1,2; M. Normandin1,3, C. Cox4, C. Herring1, K. Perry1, D. Garzon4, D. Kareken4, E. Morris1,5 and D. Albrecht1
1Radiology; 2Stark Neurosciences Research Institute, Indiana University School of Medicine, Indianapolis; 3Weldon School of Biomedical Engineering, Purdue University, West Lafayette; 4Neurology, Indiana University School of Medicine; 5Biomedical Engineering, Indiana University-Purdue University Indianapolis, Indianapolis, Indiana, USA
A commonly used method for studying relative changes in endogenous dopamine concentration is the two-scan paradigm with [11C]raclopride (RAC). In a 2-scan design, D2 receptor binding potential (BP) during a challenge state is compared to BP estimated from a resting or baseline state. Any detected changes in BP are attributed to changes in striatal dopamine levels. One of the inherent assumptions of this method is that basal dopamine tone is relatively stable in individuals. Indeed, the majority of available literature supports the notion that intra-individual RAC BP is consistent, with most reports citing between-scan differences in baseline BP in healthy subjects of approximately 10%.1–5 However, the test-retest variability (TRV) has not been determined for individuals who have a family history of alcoholism (FHA).
Objective: The TRV of baseline RAC binding was determined as part of an ongoing study investigating the reproducibility of dopaminergic responses to alcohol.
Methods: Data are reported from 7 subjects who underwent 2 baseline RAC scans on separate days. Scan start times were between 1–3:30 p.m. Subjects were male, right-handed smokers who had >1 blood relative with a history of alcoholism. Three subjects were social drinkers (SD), and 4 were nontreatment-seeking alcoholics (NTS). Subjects received a nicotine patch to prevent craving. Timing and dose of last alcohol exposure were controlled by administering a controlled IV infusion of alcohol 5 h prior to scanning (target breath alcohol concentration (BrAC): 60 mg%; BrAC returned to zero prior to baseline RAC). RAC BP was estimated for right and left caudate, putamen, and ventral striatum using conventional reference region methods.6 TRV was calculated as |BP1−BP2|/[(BP1+BP2)/2].
Results: Across all subjects and all regions, the average TRV was very high (19.4±0.04%). There was no consistency in directionality of change in BP between scan days, either across all subjects or between SD and NTS. Exploratory analyses were conducted to identify putative sources of the high TRV. SD subjects had a significantly higher TRV in the left putamen relative to NTS (P<0.02). Light smokers (≤15 cigarettes/day) tended to have greater TRV in the left ventral striatum than heavy smokers (>15/day; P = 0.08). The number of alcoholic relatives was significantly correlated with TRV in the left caudate and putamen (n = 7; P<0.05). There were no interaction effects between smoking and FHA in any region. No variables were associated with TRV in the right striatum; small sample size may have precluded detection of any effects.
Conclusions: Striatal DA tone may be highly labile in smokers with a FHA. It is unclear what is contributing to the high TRV in this sample; a larger cohort is needed for more definitive conclusions. These preliminary data suggest that, without knowledge of the TRV in a population, ‘negative results’ from DA challenge data should be interpreted with caution.
834. Displacement of 18F-fallypride binding with haloperidol in a within-subject design using a beta microprobe
G. Warnock1, D. Goblet1, C. Lemaire1, F. Giacomelli1, M. Bahri1, X. Langlois2, A. Luxen1 and A. Plenevaux1
1Centre de Recherches du Cyclotron, Universite De Liege, Liège; 2Johnson & Johnson Pharmaceutical Research & Development, Division of Janssen Pharmaceutica N.V., Beerse, Belgium
The dopamine D2 receptor is of interest for its role in the action of antipsychotic drugs, and therefore its implied role in the pathology of psychotic illness. A linear relationship exists between clinical potency and affinity for D2 receptors for many antipsychotic drugs.
18F-Fallypride is a high affinity dopamine D2/D3 radioligand used for quantifying striatal but also extrastriatal dopamine D2/D3 receptors with positron emission tomography (PET) and related techniques. We have successfully accomplished the automation of 18F-fallypride radiosynthesis, purification and formulation. The full synthesis has been accomplished in a total time of 80 mins with a radiochemical yield of 40%. 18F-fallypride can be obtained with excellent radiochemical purity (>96%) and enantiomeric excess (94%), with specific activities of 1 to 3 Ci/μmol.
In the present studies we have used a beta microprobe system to measure 18F-fallypride binding in the striatum of Sprague-Dawley rats, using the cerebellum as a reference region. For displacement of 18F-fallypride, the antipsychotic haloperidol was chosen for its high affinity for D2 receptors.
Beta microprobe systems have recently become commercially available as an affordable alternative to PET. These systems utilise implantable probes consisting of a small (0.25 to 1 mm) scintillation crystal bonded to a fibre optic cable, which can be stereotactically located in a specific region of the rodent brain. Scintillation in the crystal is carried by fibre optic to a photomultiplier tube from which a time-activity curve is generated.
18F-fallypride was injected intravenously through a catheter in the femoral vein and activity in the striatum and cerebellum measured for 90 mins. After 30 mins a bolus of saline was injected as a control before the injection of haloperidol (0.025 to 0.25 mg/kg) 30 mins later. Specific striatal binding was calculated as [total binding striatum]-[binding cerebellum]. The effect of saline and haloperidol on specific binding was calculated as the percentage change in specific binding from the 30 or 60 mins maximum.
Haloperidol displaced 18F-fallypride binding in the striatum in a clear dose- and time-dependent manner. Measurement of binding for a longer duration could be useful to confirm the maximum change in binding after haloperidol treatment. These results confirm the usefulness of the beta microprobe system for the measurement of D2 binding using 18F-fallypride.
851. [F-18]MEFWAY and [F-18]MPPF binding to the 5-HT1A receptor in the nonhuman primate
B. Christian1, N. Vandehey1, J. Moirano1, D. Murali1, D. Wootin1, O. Dejesus1, T. Barnhart1, A. Converse2, R. Nickles1, T. Oakes2, J. Mukherjee3, R. Davidson2 and M. Schneider4
1Medical Physics; 2Waisman Brain Imaging Lab, University of Wisconsin-Madison, Madison, Wisconsin-Madison; 3University of California Irvine, Irvine, California; 4Kinesiology, University of Wisconsin-Madison, Madison, Wisconsin, USA
Introduction: Preliminary studies suggest [F-18]MEFWAY will offer significant improvements as a PET biomarker for the 5-HT1A system.1 The objective of this work was to compare the binding of 5-HT1A antagonists [F-18]MEFWAY and [F-18]MPPF in the rhesus monkey.
Methods: Dynamic microPET scans of MEFWAY and MPPF were acquired on 2 anesthetized rhesus monkeys. Arterial plasma samples were drawn throughout the 2 h scan to assay parent compound. Time activity curves were obtained for the regions of the anterior cingulate (AC), hippocampus complex (HP), raphe nuclei (RN) and prefrontal cortex (PFC) the obtain a measure of binding potential (BPND) with the cerebellum (CBM) as a reference region using the Logan DVR and MRTM2 methods.
5-HT1A tracers in rhesus monkey.
Results: In the arterial plasma, both radiotracers cleared from the plasma at a similar rate, with a slow component of 0.0088 and 0.0093 min−1 for MEFWAY and MPPF. The CBM to plasma ratios plateaued at 3:1 for MEFWAY and 1.5:1 for MPPF. Throughout the regions of the brain, MEFWAY revealed significantly higher binding.
Using the ROI data, BPND was (MEFWAY and MPPF): AC—12 and 3.6, HP—11 and 3.1, RN—3.8 and 1.3, PFC—7.9 and 2.5 with less than 5% difference between the Logan and MRTM2 methods. Stable measurements of MEFWAY BPND were obtained using less than 90 mins of dynamic data.
Conclusions: MEFWAY demonstrates a dramatic improvement as a PET biomarker for the 5-HT1A site, yielding almost 3-fold increases in BPND over MPPF. Interestingly, MPPF demonstrated approximately double the BPND in rhesus monkeys over literature values in humans.2 Further investigation is required to see if this difference is species or methods specific.
Research support: AA017706, AG030524, AA012277.
855. Sensitivity of 11C-raclopride small animal PET to striatal dopamine release induced by alcohol or cocaine: effects of anesthesia/protocol
1IU School of Medicine, Indianapolis; 2Purdue University, West Lafayette; 3Purdue School of Eng & Tech, IUPUI, Indianapolis, Indiana, USA
Objectives: Microdialysis experiments have shown that drugs of abuse, such as alcohol and cocaine, produce an increase in extracellular dopamine (DA) in the striatum of rats.1,2 This study examined if similar changes in striatal DA can be detected in Wistar or P rats using small animal PET. We also sought to establish optimal doses and times for imaging the dopaminergic response to drugs of abuse; we varied the time of IV cocaine administration, the type and mode of anesthesia delivery, and the dose of IP alcohol. We gauged DA release by calculating change in raclopride [RAC] binding potential (BP).
Drug-induced DA release measured as BP decrease.
Methods: PET images were acquired in drug-naïve male Wistar and P rats. All scans were performed on IndyPETIII (∼1 mm FWHM in-plane resolution). Prior to each scan, all animals were anesthetized and secured on a stereotaxic holder. Time activity curves were extracted from the striatum and cerebellum and BPs were calculated as a measure of D2 receptor availability.3,4
Only Wistars were given cocaine and both Wistars and P rats were given alcohol. In cocaine studies, Wistars received up to three 11-C RAC scans (one rest and two cocaine) and were anesthetized with isoflurane or a ketamine/xylazine cocktail. Ket/Xyl boosters were given ∼20 mins after 11C-RAC injection. These animals received 2.0 mg/kg cocaine intravenously 10 mins (‘early cocaine’) or 25 mins (‘late cocaine’) after tracer injection. In alcohol studies, Wistars and P rats received up to three 11C-RAC scans (rest, alcohol, saline) and were anesthetized with isoflurane only. These animals received 3.0 g/kg ethanol (20%v/v), 1.5 g/kg ethanol (15%v/v), or 2.25 g/kg ethanol (15%v/v) IP 5 mins before tracer injection. Venous blood samples were collected from animals 10 mins after tracer injection for determination of blood alcohol content (BAC).
Results: There was a significant decrease in BP in Wistars given 3 g/kg ethanol IP, as well as in Wistars under ketamine/xylazine given 2 mg/kg cocaine IV. There was no significant decrease in BP due to cocaine if Wistars were anesthetized with isoflurane. There was no detectable change in BP in P rats who received IP alcohol at either 1.5 or 2.25 g/kg ethanol. In P rats, decreased BP was positively correlated with BAC (r = 0.78, P<0.04). In fact, large decreases in BP were observed mainly in rats with BACs above 200 mg%. No significant change in BP was found between rest and saline scans.
Conclusions: To maximize detectability of alcohol-induced DA release in the striatum, it may be necessary to achieve BAC levels of at least 200 mg%. Cocaine-induced DA release may be quenched by isoflurane, which would be consistent with work of Votaw et al.5 No effect of cocaine injection time was observed, but timing of ketamine boosters may have interacted with ‘late’ cocaine.6
872. Investigating amphetamine-stimulated dopamine release in methamphetamine users using the PET D2/3 agonist ligand [11C]PHNO
I. Boileau1, S. Houle1, D. Wilkins2, P. Selby1, P. Rusjan1, A. Wilson1 and S. Kish1
1CAMH, Toronto, ON, Canada; 2University of Utah, Salt Lake City, Utah, USA
Background: PET/[11C]raclopride neuroimaging data suggests a hypo-dopaminergic state (decreased D2/3-receptor density and dopamine release) in the striatum of withdrawn addicted subjects. The goal of the present study was to use PET and the recently developed D3 receptor-preferring agonist, [11C]PHNO to assess dopamine receptor density and dopaminergic response to an open challenge with amphetamine in recently withdrawn methamphetamine users (MA).
Methods: 14 MA and 14 matched-controls (HC) underwent [11C]PHNO scans during a non-placebo-controlled baseline and after a low oral-dose of amphetamine (0.4 mg/kg).
Results: MA abuse was associated with a trend for increased baseline D2/3-receptor density corresponding to 15%(P = 0.1) and 16%(P = 0.06) in the ventral striatum and globus pallidus respectively and 9% in the dorsal striatum (P = 0.2). Relative to HC, MA had greater amphetamine-induced increases in heart-rate, and subjective effects (P<0.01) together with decreases in [11C]PHNO binding potential slightly above HC value in the ventral striatum (−21% MA versus −13% HC, P = 0.1) and globus-pallidus (−30% MA versus −14% HC, P = 0.04) but not in the dorsal striatum.
Conclusion: Contrasting with the current PET/[11C]raclopride imaging literature, these preliminary [11C]PHNO data suggest increased D2/3 receptors density and greater apparent DA release in MA users relative to HC. This could be explained by pharmacological differences between the in-vivo binding of [11C]PHNO and [11C]raclopride (consistent with our recent finding in Parkinson's disease, Brain, 2008). Specifically, its in-vivo distribution as well as its higher vulnerability to endogenous DA relative to [11C]raclopride could be related to some degree to its selectivity for D3 compared to D2 receptors.
912. Visualization of dopamine dynamics from PET images
E. Morris1,2,3, C. Constantinescu4, J. Sullivan2,3, M. Normandin2,3, K. Yoder2, S. Risacher2 and L. Christopher5
1Biomedical Engineering, Indiana University-Purdue University at Indianapolis; 2Radiology, Indiana University School of Medicine, Indianapolis; 3Biomedical Engineering, Purdue University, West Lafayette, IN; 4Psychiatry & Human Behavior, University of California Irvine, Irvine, California; 5Electrical Engineering, Indiana University-Purdue University at Indianapolis, Indianapolis, Indiana, USA
Introduction: We recently introduced multiple strategies (‘ntPET’) for extracting temporal patterns of brain dopamine (DA) fluctuations from dynamic PET using 11C-raclopride.1–5 Each of these modeling and parameter estimation methods yields a collection of time-concentration curves for endogenous DA. When the collection of DA curves is dense (i.e., one curve for every voxel), we can produce images of DA concentration at each time-frame; the 4D data set can be thought of as a volumetric DA ‘movie’. Viewing a single slice of the movie or viewing parametric maps that we derive from it can reveal clusters of brain regions whose DA activity is synched to a task performance or other DA-ergic stimulus.
Methods: A male subject underwent 3 bolus PET studies. One study was a ‘rest’ condition, the remaining two were ‘task’ conditions. The latter two sessions involved an identical right-handed finger-tapping task lasting 10 mins. Once, the task was initiated near the midpoint (25 mins) of the scan session. Another time, the task was initiated near the end (40 mins). Data were processed according to non-parametric ntPET4 to produce DA movies. We generated mages of the DA peak time at each voxel for each of the two tasks.5 Finally, we created an image of the time-difference between peak times for ‘early’ and ‘late’ tasks, respectively. This image was segmented with a Guassian mixture model6 to identify clusters of temporally like-responding voxels.
Results: The segmented image of time-difference between peaks of DA activity for the two tasks is shown in the Figure for 12 contiguous (axial) striatal slices The units in the image are minutes and the largest cluster of striatal voxels (red) involved in the right-handed task are on the left. The time-difference for the dominant cluster is 14.6±0.08 (SEM) min. Analysis of a control study yielded no clusters of comparable size or contiguity.
Conclusion: Given that ‘early’ and ‘late’ tasks were identical except for timing, we would expect the DA-ergic response of voxels involved in execution of the task to be delayed by 15 mins in the late task relative to the early one. This appears to be true of the largest cluster of (primarily contralateral) voxels whose peak activity shifts in time by 15 mins from early to late task performance. We believe our images of DA timing are the first of their kind and may enhance our ability to identify clusters and sequencing responsible for responses to motor and other DA-ergic stimuli such as drugs of abuse.
Acknowledgements: R21 AA015077.
917. Imaging serotonin receptors using 18F-MEFWAY in aging
T. Dinh, S. Faheem, N. Rana, N. Saigal, S. Pandey, M.L. Pan, E. Head and J. Mukherjee
University of California Irvine, Irvine, California, USA
Objectives: Serotonin 5HT1A receptors have been implicated in aging. Effects of aging on the 5HT1A receptor have been studied using post-mortem tissue as well as by PET imaging (e.g., Costes et al1). Results have varied from no effect to reductions and observed gender effects. Using the selective 5HT1A radiotracer 18F-trans-Mefway5 we have evaluated postmortem in the aging canine model. Ages of the dogs varied from 1 to 15.4 years (equivalent to ∼15 to 90 human years).
Methods: Radiosynthesis of 18F-Mefway was carried out with a specific activity of >2 Ci/mmol. Brain tissue was obtained from the cerebellum (CB) and temporal cortex (COR) of beagle dogs of various ages. Sections (10 μm thick) of the tissue were obtained including the cortex (COR) and cerebellum (CB). Sections were preincubated at room temperature for 15 mins in 50 mmol/L Tris-HCl buffer (pH 7.4). Incubation with 1 to 2 μCi/cc of 18F-Mefway at 37°C took place for 1 h. Nonspecific binding was with 10 μm serotonin. Post incubation, sections were washed twice for 3 mins in cold buffer, submerged quickly in cold distilled water and laid out to dry and exposed to phosphor films. Optiquant imaging software was used to quantify binding to the various cortical layers.
18F-Mefway binding in canine. (A) 18F-Mefway binding to cortical layers in canine model (14.8 yr old); (B) Scan of brain slice in image (A) showing cortical layers; (C) Binding profile in cortical layers showing maximal binding layers I–II followed by V–VI.
Results: Binding of 18F-Mefway in the various cortical layers were observed and were in agreement with reported 5HT1A distribution.2 Serotonin displaced >90% of binding. Binding in CB was very low; each animal was used as reference in ratio measurements. In the youngest dog (1 Yr) ratio COR/CB = 19. Highest binding was in layers I, II (100%) and least in III, IV (31%), similar to that found in humans. Over the age range evaluated, no significant reduction in 18F-Mefway was observed.
Conclusions: The canine model offers the ability to study aging effects. It has also been studied as a potential animal model for human AD.3 Further studies are underway in additional old animals. Evaluation of D2/D3 receptors using 18F-fallypride in the canine model is also underway in order to correlate our findings of aging in humans.4
Research Supported by NIH R21AG030524.
969. Dual-assessment of functional and pathological changes in Alzheimer's disease using single 11C-PIB PET
S.J. Kim, Y.K. Kim, J.S. Lee, E.J. Yoon, S.E. Kim and D.S. Lee
Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, South Korea
Objectives: In most studies, the association between functional (regional cerebral glucose metabolism) and pathological information (brain amyloid plaque load) have been separately investigated using two different traceres, 11C-PIB and 18F-FDG PET in assessment of of Alzheimer's disease (AD). In this study, we investigated the possibility of functional R1 map (relative flow delivery, K1 region of interest/K1 reference region) from PIB PET as alternative functional map in stead of FDG PET.
Methods: Thirty eight subjects (8 AD and 12 MCI patients, 18 normal volunteers) underwent dynamic 11C-PIB PET scans with 90 mins scan duration and 18F-FDG scans with 30 mins after injection. Parametric images of R1 for 11C-PIB PET were generated using multilinear reference tissue models (MRTM2).1 To compare the brain regions with perfusion and metabolic abnormalities on 11C-PIB and FDG PET quantitatively, R1 map and FDG images underwent z-transformation with respect to each normal average and SD. The association of the number of abnormal voxels and mean z-value between R1 map obtained from PIB PET and glucose metabolism obtained from FDG PET was evaluated in the entire brain excluding cerebellum.
Results: Z-values of each group were well reflected the severity of AD. At a z threshold of −2.97 for both R1 PIB and FDG, the number of abnormal voxels across the entire brain showed a correlation, and the mean z-values of abnormal voxels between them were highly correlated (r = 0.86, P<0.0001).
Conclusions: The relative perfusion obtained from PIB PET was well correlated with those of metabolic abnormality in the entire brain. This study may suggest that a single PIB PET study with R1 map could provide the sufficient information in the distribution of pathologic and metabolic abnormalities in disease progression.
Correlation mean z-value between FDG and PIB PET.
1034. Abnormal [11C]ABP688 binding in patients with temporal lobe epilepsy
E. Kobayashi1, J. Hall1, M.-C. Guiot2, F. Andermann1, F. Dubeau1, M. Diksic3, G. Massarweh3, J.-P. Soucy3 and P. Rosa-Neto4
1Department of Neurology and Neurosurgery; 2Department of Pathology; 3Brain Imaging Center, Montreal Neurological Institute, McGill University; 4Translational Neuroimaging Laboratory—MCSA, Douglas Hospital, McGill University, Montreal, QC, Canada
Objective: Immunohistochemistry studies suggested overexpression of metabotropic glutamate receptor type 5 (mGLUR5) in the epileptogenic hippocampus of patients with mesial temporal lobe epilepsy (MTLE). We performed in vivo mGUR5 binding studies using [11C]ABP688, a recently developed Positron Emission Tomography (PET) radioligand which binds specifically to mGUR5 as a potential clinical tool for identification of the seizure focus in a group of patients with unilateral MTLE.
Design/methods: We studied 9 patients (3 M, 6 F, mean age = 39yo, range, 22 to 67) with unilateral MTLE (5 left, 4 right). Six patients had MRI evidence of hippocampal atrophy. PET images were obtained in a Siemens HR+ scanner following a venous injection of 10 mCi of [11C]ABP688 with high specific activity. Parametric images representing binding potentials (BPND) were generated using the cerebellum as reference tissue. BPND parametric images were co-registered using the MRI and subsequently non-linearly resampled to a standard stereotaxic space. We used voxel-by-voxel statistical parametric analysis: BPND differences between patients (right and left MTLE) and controls (7 healthy volunteers, 4 F, mean age = 37, range 19 to 71) were computed. R1 maps were evaluated.
[11C]ABP688 BPND was reduced in the epileptogenic hippocampus for left and right MTLE patients (maximum t = 4.74 for left and t = 8.37 for right) as compared to controls. No differences in the contralateral hippocampus were seen. Finally, there was no significant differences on R1 maps.
Results:Figure 1 shows a typical [11C]ABP688 binding study in MTLE patient.
Conclusions/relevance: Our finding of reduced mGLUR5 binding in the epileptogenic hippocampus contrast with previous demonstration of hippocampal mGLUR5 upregulation. The discrepancy between our in vivo receptor binding study and the post mortem immunohistochemistry might be explained by abnormalities on mGLUR5 structure and function. It remains to be elucidated whether this is a cause or consequence of the seizures, but we were able to demonstrate that mGLUR5 binding studies can have clinical application in the identification of the abnormal hippocampus in MTLE patients.
Study supported by: Savoy Foundation for Epilepsy and the American Epilepsy Society Early Career Physician Scientist Award.
References
1.
MeikleSR. Cancer Chemother Pharmacol1998;42:183–93.
IchiseMasanoriLiowJeih-SanLuJian-Qiang. Linearized reference tissue parametric imaging methods: Application to [11C]DASB positron emission tomography studies of the serotonin transporter in human brain. JCBFM23:p1096–1112.
