In vivo variation in same-day estimates of metabotropic glutamate receptor subtype 5 binding using [ 11 C]ABP688 and [ 18 F]FPEB

Abstract

Positron emission tomography tracers [¹¹C]ABP688 and [¹⁸F]FPEB target the metabotropic glutamate receptor subtype 5 providing quantification of the brain glutamatergic system in vivo. Previous [¹¹C]ABP688 positron emission tomography human test–retest studies indicate that, when performed on the same day, significant binding increases are observed; however, little deviation is reported when scans are >7 days apart. Due to the small cohorts examined previously (eight and five males, respectively), we aimed to replicate the same-day test–retest studies in a larger cohort including both males and females. Results confirmed large within-subject binding differences (ranging from −23% to 108%), suggesting that measurements are greatly affected by study design. We further investigated whether this phenomenon was specific to [¹¹C]ABP688. Using [¹⁸F]FPEB and methodology that accounts for residual radioactivity from the test scan, four subjects were scanned twice on the same day. In these subjects, binding estimates increased between 5% and 39% between scans. Consistent with [¹¹C]ABP688, mean absolute test–retest variability was previously reported as <12% when scans were >21 days apart. This replication study and pilot extension to [¹⁸F]FPEB suggest that observed within-day binding variation may be due to characteristics of mGluR5; for example, diurnal variation in mGluR5 may affect measurement of this receptor.

Keywords

Positron emission tomography metabotropic glutamate receptor subtype 5 test–retest

Introduction

Glutamate is the most prominent neurotransmitter in the brain, and impairment of glutamatergic transmission is implicated in many psychiatric illnesses. For example, the metabotropic glutamate receptor subtype 5 (mGluR5) has been implicated in schizophrenia, mood disorders, anxiety disorders and Parkinson’s Disease.¹ Positron emission tomography (PET) with the tracer (E)-3-((6-methylpyridin-2-yl)ethynyl)cyclohex-2-en-1-one-O-[¹¹C]methyloxime ([¹¹C]ABP688) allows measurement of mGluR5 expression and distribution in vivo and may enhance our understanding of these disorders. [¹¹C]ABP688 is a negative allosteric modulator (NAM) at mGluR5, binding with high affinity (K_D = 1.7 ± 0.2 nM)² and selectivity to this receptor. However, there is uncertainty about its test–retest reliability in humans.^3,4

Test–retest studies allow assessment of within-subject tracer performance. In rodents and non-human primates, [¹¹C]ABP688 test–retest studies indicate that this tracer provides reliable outcome measures, with low intrasubject variability in binding parameters. Two same-day studies in anesthetized rats showed that [¹¹C]ABP688 BP_ND (ratio at equilibrium of specifically bound tracer to that of nondisplaceable tracer in tissue) did not differ significantly in any region from test to retest.^5,6 In a study performed by our group in anesthetized Papio anubis on the same day, consistent [¹¹C]ABP688 volumes of distribution (V_T) were found between test and retest, with average percent differences of 4.3%–8.2% across all brain regions.⁷ A separate study in anesthetized Papio anubis confirmed no significant difference in BP_ND values between test and retest.⁸ However, a study in anesthetized rhesus monkeys performed on the same day found an average increase of 9% and 13% in [¹¹C]ABP688 V_T and BP_ND, respectively, from test to retest.⁹

Two human test–retest studies have been reported using this tracer in males; no studies have been reported in females. Our group showed intraindividual [¹¹C]ABP688 binding differences (between test and retest, completed on the same day) in eight male subjects.⁴ In most subjects, outcome measures (V_T and BP_ND, both from bolus delivery) significantly increased from test to retest across most regions, by 31.4% ± 30.4%.⁴ A different study using [¹¹C]ABP688 bolus compared to bolus–infusion in five healthy adult males with a mean of 30 days (range: 7–97 days) between scans showed little variability in V_T values after modeling methods for the bolus–infusion scan were optimized.³

The disparity in study results may reflect the challenges of imaging the glutamatergic system. As most neurons are glutamatergic, aspects of study design such as timing of scans or use of anesthesia (for preclinical studies), which may affect glutamate neurotransmission or glutamate receptor levels, could affect [¹¹C]ABP688 binding. However, as our previous study was the first report of this phenomenon, the goals of the current study were to determine whether the findings in males were replicable, and to determine test–retest differences in [¹¹C]ABP688 binding in females for the first time. A significant advantage of this replication is that the study is performed using a different PET scanner and different PET facility (for radiochemistry and plasma analysis) than the original study, allowing assessment of whether scanner or facility affected outcome.

Furthermore, we aimed to determine whether the previous results were specific to [¹¹C]ABP688 by using another tracer targeting at mGluR5 NAM site, 3-[¹⁸F]fluoro-5-[(pyridin-3-yl)ethynyl]benzonitrile ([¹⁸F]FPEB, K_D = 0.11 ± 0.04 nM—0.15 ± 0.02 nM).¹⁰ Our recent [¹⁸F]FPEB test–retest study involved a bolus–infusion protocol with seven healthy male subjects, scanned 3–11 weeks apart. Mean absolute test–retest variability of V_T and BP_ND were <15%.¹¹ This is in agreement with two other [¹⁸F]FPEB studies. One involved a test–retest component completed by five healthy men, scanned between 2 and 16 days apart (except for one subject).¹² In this study, average test–retest variability of V_T was <10% and average BP_ND variability was <10%.¹² In another study of four males and five females, scans were at the same time of day but separated by six months. Absolute test–retest variability of V_T in that study ranged from 8% to 13%.¹³ In a preclinical study involving test–retest evaluation in six anaesthetized male rats, variability in BP_ND was found to be <10% on average for most modeling techniques and scan durations tested.¹⁴ Same-day test–retest studies with ¹⁸F tracers are challenging, because residual radioactivity from the test scan will be present in the retest scan. Therefore, in the present study, we used previously developed modeling techniques that account for these residual effects¹⁵ and determined whether same-day binding changes in a pilot study of [¹⁸F]FPEB were consistent with those observed for [¹¹C]ABP688.

As [¹¹C]ABP688 and [¹⁸F]FPEB are the first to provide quantification of the human glutamatergic system, it is imperative to understand the effect of study design and tracer on study results. Here, we provide the first comparison of the same-day test–retest results for these tracers in female and male subjects.

Materials and methods

Subjects

This study was approved by the Institutional Review Boards of Columbia, Stony Brook and Yale Universities, the Yale Radiation Safety Committee and by the Yale-New Haven Hospital Radioactive Drug Research Committee. All subjects signed informed consent. This study was performed according to the Ethical Principles and Guidelines for the Protection of Human Subjects of Research (Belmont Report). After completing the informed consent process, inclusion criteria were assessed (for both cohorts) by the following: demographic and medical history, Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders (SCID), review of systems, physical examination, routine blood tests and urine toxicology. The inclusion criteria consisted of (1) adults 18 to 55 years and (2) capacity to provide written informed consent. Exclusion criteria included: (1) lifetime history of alcohol or substance abuse or dependence, (2) presence and/or history of a clinically significant major neurological or psychiatric disorder, (3) an Axis I or Axis II disorder diagnosis, as assessed by the SCID, (4) presence and/or history of first-degree relative with history of major depression, schizophrenia, schizoaffective disorder, or suicide attempt, (5) current or recent pregnancy, breastfeeding or plans to conceive during the course of the study, (6) laboratory tests with clinically significant abnormalities or positive urine toxicology screen, (7) anticoagulant or anti-platelet treatment, other than aspirin, within 10 days of PET scanning, (8) history of head trauma with prolonged loss of consciousness (>5 min), any neurological condition including stroke or seizure (excluding a single childhood febrile seizure), or a history of migraine headaches as determined by a physician.

A total of 14 healthy adults were enrolled in the [¹¹C]ABP688 study. One subject completed the protocol, but arterial input function data were not available during the retest scan. One subject had highly asymmetric brain anatomy including asymmetric ventricles and [¹¹C]ABP688 binding patterns. One subject did not complete the protocol due to the inability to complete the baseline PET scan. All three subjects were excluded from the analysis.

Five healthy adults were enrolled in the [¹⁸F]FPEB study. One subject completed the first scan, but the arterial line failed prior to the second scan. Of the remaining (two males, two females), the average age was 26 ± 6.5 years.

Positron emission tomography

All PET scanning was performed at the Yale University PET Center on the High-Resolution Research Tomograph (HRRT; Siemens, Knoxville, TN). A 6-min transmission scan was acquired before injection. Head motion was recorded during the scan using a commercial optical tracking system, the Polaris Vicra device (Northern Digital Inc., Waterloo, ON, Canada). Images were reconstructed and corrected for attenuation, scatter, randoms and motion using the MOLAR algorithm.¹¹ Subjects were not required to fast on the day of the PET scans or between scans.

[¹¹C]ABP688 synthesis and PET scanning

High specific activity [¹¹C]ABP688 was produced from the reaction of [¹¹C]methyl iodide with desmethyl-ABP688 using the loop method as performed previously.⁹ The average radiochemical and chemical purities are 97% and 93%, respectively (n = 20). The average E/Z conformational isomers of [¹¹C]ABP688 in the final PET drug product is 70:1. The E/Z ratio is determined using analytical radio-HPLC area percent, by comparing the areas under the radioactive peaks of the [¹¹C]ABP688 conformers present in the final PET drug product solution for each produced batch. This E/Z value represents the average ratio (range: 42:1–98:1 for n = 20) produced using the mild radiolabeling method developed by Nabulsi.⁹ [¹¹C]ABP688 was administered as a bolus over 1-min, and emission data were collected for 60-min in list mode and binned into 6 × 0.5 min, 3 × 1 min, 2 × 2 min and 10 × 5 min frames.⁷

Following the test [¹¹C]ABP688 scan, subjects were given a break (3.4 ± 0.7 h for same-day subjects) and then completed the retest [¹¹C]ABP688 scan. One exception was Subject 1 who, due to equipment failure, participated in the retest scan two days after the test scan. Vital signs (heart rate, blood pressure and oxygen saturation, SPO₂) were obtained during scans via the arterial line at intervals between 0.5 and 5 min. Due to equipment problems, one subject received heart rate measurements at irregular intervals.

[¹⁸F]FPEB synthesis and PET scanning

High specific activity [¹⁸F]FPEB (177 ± 48 MBq/nmol at EOS) was prepared as previously described.¹¹ The average radiochemical and chemical purities were both 99% (n = 4). [¹⁸F]FPEB was injected as a 1-min bolus, and emission data were recorded for 120 min in list mode and binned into 6 × 0.5 min, 3 × 1 min, 2 × 2 min and 22 × 5 min frames. The test and retest scans were performed on the same day, 2.7 ± 0.5 h apart. Vital signs (heart rate, blood pressure and oxygen saturation, SPO₂) were read four times during the scan (at approximately 15, 30, 60 and 90 min). Since [¹⁸F]FPEB was injected twice on the same day, and fluorine-18 has a long half-life of 110 min, the injected dose was intentionally limited during the test scan to reduce contamination of the retest scan data by leftover activity from the first injection. Prior to the retest injection, a 15-min emission (3 × 5 min frames) scan was performed to measure residual activity from the test injection and correct the retest data from the effects of that residual activity (background correction, see below).

Magnetic resonance imaging

In order to perform delineation of anatomical regions on the PET data, images were coregistered with T1-weighted MRIs acquired on a 3T scanner at Columbia University (n = 5, Signa Advantage system; GE Healthcare, Waukesha, WI, USA), Stony Brook University (n = 6, MAGNETOM Trio Tim; Siemens Healthcare, Malvern, PA, USA) or Yale University (n = 4, Trio MR scanner; Siemens, Erlangen, Germany).

Input function measurement

Prior to PET imaging, catheters were inserted in the radial artery and forearm veins for arterial blood sampling and radioisotope injection, respectively. For [¹¹C]ABP688, blood activity was measured continuously for the first 7 min after radiotracer administration, and manually at 9, 12, 15, 20, 25, 30, 40, 50 and 60 min. For [¹⁸F]FPEB, blood activity was measured continuously for the first 7 min after radiotracer administration, and manually at 8, 12, 15, 20, 30, 45, 60, 75, 90, 105 and 120 min. Radioactivity was analyzed as described previously.¹¹ After applying a fixed correction for external dispersion, automated samples were smoothed by convolution with a Gaussian function ([¹¹C]ABP688 only) and the automated and manual values were merged. An HPLC assay of arterial blood samples (at 0, 4, 12, 30 and 60 for [¹¹C]ABP688; 0, 3, 8, 15, 30, 60, 90 and 120 for [¹⁸F]FPEB) was used to establish unmetabolized parent compound levels.^9,11,16

For [¹¹C]ABP688, unmetabolized parent fraction levels were fitted with a Hill function, which is described by three parameters (A, B and C), in which percent parent compound = A(t^B/[t^B+ C]) + 1, where t is time.¹⁷ The input function was calculated as the product of the interpolated parent fraction and the merged plasma counts. These combined data were then fitted as the combination of a straight line and the sum of three exponentials, describing the function before and after the peak.

For [¹⁸F]FPEB, the unmetabolized parent fraction levels were fit with a sum of exponential function (with three to five parameters, the number of parameters being automatically adjusted to minimize $χ^{2}$ ) constrained to be ≤1. The late merged plasma counts were fitted by a sum of exponential functions (with the fit starting time and the number of parameters automatically adjusted to minimize $χ^{2}$ ) and replaced by the fitted values to reduce noise in late plasma counts. The input function was calculated as the product of the interpolated parent fraction and the smoothed merged plasma counts.

Free fraction (f_P) measurements were performed using an ultrafiltration technique.¹⁸ For [¹¹C]ABP688, all measured free fraction values were low (2.2% ± 0.6%) and considered unreliable, as has been shown previously.⁷ [¹⁸F]FPEB f_P values were higher (5.4% ± 0.7%) and reproducible (% difference = −0.3% ± 0.3%, n = 4).

Image analysis

[¹¹C]ABP688 image analysis was performed using MATLAB (The MathWorks, Natick, MA). Subsequent frames of each PET study were registered to the eighth frame using the FMRIB linear image registration tool (FLIRT), version 5.0 (FMRIB Image Analysis Group, Oxford, UK), to correct for residual subject motion that may not have been accounted for by the Polaris Vicra system. The mean PET image was then coregistered to the subject’s MRI using a semi-automated technique, as used previously.⁴ Probabilistic regions of interest from manually defined atlases were transferred to the subject’s MRI using nonlinear registration techniques.⁴ Following delineation, cortical regions of interest were gray matter masked by voxel-wise multiplication (regional probability multiplied by the probability of that the voxel is in grey matter, as assessed by Statistical Parametric Mapping, SPM5; Institute of Neurology, University College of London, London, England). Time activity curves were generated from the mean of the measured activity, weighted by regional label probabilities, over the time course of the PET acquisition.

[¹⁸F]FPEB image analysis was performed using IDL 8.0 (Exelis, Boulder, CO). To correct for residual subject motion that may not have been accounted for by the Polaris Vicra system, frames were realigned to the average image of the first 10 min after injection with FLIRT (version 5.3). The motion-corrected average image of the first 10 min after injection was then rigidly coregistered to subject’s MRI using FLIRT (version 5.3). The subject’s MRI was then coregistered to a template MRI using bioimagesuite (http://bioimagesuite.yale.edu) and nonlinear transforms.¹⁹ Regions of interest were defined using the automated anatomical labeling template²⁰ for SPM2.

Background correction for [¹⁸F]FPEB retest scans

Background correction was applied to both the arterial input data and brain time activity curves in order to correct for the residual activity from the test scans. Specifically, an initial condition term was added to the kinetic model for the time activity curve fits, and the plasma curve was modified since the initial activity was >0 and the initial parent fraction in the retest scan was low.

For the correction of time activity curves, a 15-min scan was acquired just before the retest injection. Each regional time activity curve was computed using the original 120-min scan plus the additional 15-min scan; then the last 15-min of each time activity curve was fitted with a single exponential function, extrapolated to the start time of the retest scan, and the decay factor between the two injection times was applied to obtain the effective initial activity just prior to the time of the retest scan $C_{T} (0^{-})$ in each region. These concentrations were used to set the initial concentration of the two tissue compartments of the 2TC model. Taking into account the non-zero initial concentration, the 2TC model equation for the retest scan was

C_{T} (t) = φ_{1} C_{P} (t) \otimes e^{- θ_{1} t} + φ_{2} C_{P} (t) \otimes e^{- θ_{2} t} + C_{1} (0^{-}) e^{- θ_{1} t} + C_{2} (0^{-}) e^{- θ_{2} t}

(1)

where

C_{P} (t)

is the arterial input function,

θ_{1}

and

θ_{2}

are the two eigenvalues of the linear system of partial differential equations of the 2TC model and

φ_{1}

and

φ_{2}

are the linear coefficients associated to these two eigenvalues (equations to compute

θ_{1}

θ_{2}

φ_{1}

and

φ_{2}

from the 2TC model rate constants K₁, k₂, k₃ and k₄ can be found in Gunn et al.²¹). Using

θ_{1}

θ_{2}

φ_{1}

and

φ_{2}

, the 2TC model is mathematically equivalent to a model having two compartments in parallel with input rates

φ_{1}

and

φ_{2}

, respectively, and washout rates

θ_{1}

and

θ_{2}

, respectively. The concentrations

C_{1} (0^{-})

and

C_{2} (0^{-})

in equation (1) are the two initial concentrations in these two parallel compartments. These two compartments were assumed to be at equilibrium with the plasma compartment, and thus

C_{1} (0^{-})

and

C_{2} (0^{-})

were set to

C_{T} (0^{-}) \times φ_{1} / θ_{1} / (φ_{1} / θ_{1} + φ_{2} / θ_{2})

and

C_{T} (0^{-}) \times φ_{2} / θ_{2} / (φ_{1} / θ_{1} + φ_{2} / θ_{2})

, respectively. Thus, as in the conventional 2TC model, there were four floating parameters to fit.

For the arterial input curve correction, the end of total plasma radioactivity concentration ( $C_{TPR}$ ) curve from the test injection was fit with a single exponential and extrapolated to the time of the retest injection, and the decay factor between the two injection times was applied to obtain the effective total plasma activity just prior to the time of the retest injection $C_{TPR} (0^{-})$ . Similarly, the parent fraction curve was extrapolated to obtain the unchanged fraction in plasma just prior to the time of the retest injection $f_{U} (0^{-})$ . The metabolite data from the retest injection were processed as usual, but without including the fraction data from the saline standard (SS) sample ( $f_{U}^{SS}$ , a value near 1.0) at this point of the analysis, to obtain the measured retest parent fraction curve $f_{U}^{M} (t)$ . These measured fractions are correct for all times after the first metabolite sample time ( $t_{FS}$ ), but the values must be corrected before $t_{FS}$ . Between the retest injection time ( $t_{RI}$ ) and $t_{FS}$ , assuming the residual total plasma concentration and unchanged fraction from the test injection do not change during this short interval, the corrected parent fraction curve ( $f_{U}^{C} (t)$ ) can be expressed as follow

f_{U}^{C} (t) = \frac{f_{U} (0^{-}) C_{TPR} (0^{-}) + f_{U}^{R} (t) (C_{TPR} (t) - C_{TPR} (0^{-}))}{C_{TPR} (t)} for t \in [t_{RI}, t_{FS}]

(2)

where

f_{U}^{R} (t)

is the parent fraction of the newly injected radioactivity in the retest scan. During that interval,

f_{U}^{R} (t)

is assumed to vary linearly, i.e.

f_{U}^{R} (t) = f_{U}^{SS} + m \times t

(3)

At $t_{FS}$ , $f_{U}^{C}$ is set to be equal to the $f_{U}^{M}$ , thus, solving for the early slope m yields

m = \frac{f_{U}^{SS} - \frac{f_{U}^{M} (t_{FS}) \times C_{TPR} (t_{FS}) - f_{U} (0^{-}) C_{TPR} (0^{-})}{C_{TPR} (t_{FS}) - C_{TPR} (0^{-})}}{t_{FS}}

(4)

Effects of applying this correction are addressed in the Supplementary Material (Section I).

Outcome measure calculation and analysis

Regional outcome measures were calculated using an unconstrained two-tissue compartment (2TC) model. Given the unreliable f_P values for [¹¹C]ABP688 and lack of a reference region,¹⁰ V_T was the main outcome measure. For each subject, V_T of nine mid- to high-binding regions—anterior cingulate, medial prefrontal cortex, orbitofrontal cortex, ventral striatum, parietal lobe, putamen, caudate, amygdala and hippocampus—and the low-binding cerebellum was calculated.

Since V_T includes both specific and non-specific binding, and non-specific binding varies by tracer, it is less informative to compare change in V_T across tracers. For this reason, and to remove any effects of measurement of radiotracer activity in blood or free fraction, we also estimated BP_ND using two techniques in addition to the conventional BP_ND estimate. See Supplementary Material (Section II) for details.

A linear mixed effect model with region as a fixed effect and subject as a random effect was used to compare test–retest differences. The dependence structure used in these models was compound symmetry. All binding measures were log-transformed for better model fitting. Analyses were performed using SAS 9.3 (SAS Institute Inc., Cary, NC).

Results

[¹¹C]ABP688 test–retest results

There were no significant differences between scans (test scan, retest scan, p value) in the injected dose (564 ± 209 MBq, 574 ± 189 MBq, 0.64), specific activity (270 ± 275 MBq/nmol, 253 ± 248 MBq/nmol, 0.58) or mass (1.2 ± 1.1 µg, 1.1 ± 0.9 µg, 0.36). None of the three parameters used to fit the subjects’ metabolite values were significantly different between test and retest scans (p > 0.15 in all cases). However, clearance values, calculated as the injected dose divided by the extrapolated area under the metabolite-corrected arterial input function²², were significantly different across scans (test: 99.2 ± 14.5 L/h, retest: 131.3 ± 33.0 L/h, p = 0.02). Furthermore, K₁ values were higher on average during the retest scan (test: 0.26 ± 0.17, retest: 0.33 ± 0.15 across all regions and subjects, p < 0.01). Large differences in binding between test and retest scans were observed (Figure 1 (left)).

Figure 1.

V_T from test (x axis) and retest (y axis) scans using [¹¹C]ABP688 (left) and [¹⁸F]FPEB (right). The solid red line is the line of identity. All scans were performed on the same day except [¹¹C]ABP688 Subject 1.

Percent difference in [¹¹C]ABP688 V_T (V_TPD = 100 × [V_T^retest—V_T^test]/V_T^test), averaged across all mid- to high-binding regions within a subject, ranged from −23% ± 2% to 108% ± 5% (Figure 2 (left)). As evidenced by the standard deviations, within-subject variability of percent differences across regions was small. Because of this, for each subject, the magnitude of the average percent difference was within 2% of the absolute percent difference. (Using V_T/f_P as the outcome measure produced results that were similar, with average percent differences ranging from −28% ± 1% to 189% ± 9% across subjects.) Most subjects (7 of 11) had higher binding in the retest scan compared to the test scan, similar to our previous same-day test–retest study (seven of eight showing increases). Using a mixed linear effects model, the test–retest difference was found to be significant (p < 0.01). Of the subjects whose binding was lower in the retest scan (Subjects 1, 3, 4 and 10), average V_TPD was less than 8%, except for Subject 1 who participated in PET scans on two different days. On average, females showed greater changes in V_T between test and retest than males (p = 0.02) despite a lack of difference in their baseline V_T (p = 0.80).

Figure 2.

Average percent difference (between test and retest scans) in V_T (V_TPD = 100 × [V_T^retest—V_T^test]/V_T^test) for [¹¹C]ABP688 (left) and [¹⁸F]FPEB (right). V_TPD was measured and averaged over nine regions: anterior cingulate, medial prefrontal cortex, orbitofrontal cortex, ventral striatum, parietal lobe, putamen, caudate, amygdala and hippocampus. Standard deviation of the percent differences across these regions is represented by error bars. Subjects are presented in chronological order for each cohort. Blue bars represent male subjects and red bars represent female subjects. All subjects participated in test and retest scans on the same-day except for Subject 1 of the [¹¹C]ABP688 cohort, whose scans were two days apart.

[¹⁸F]FPEB test–retest results

The injected dose, intentionally limited during the first scan, was 63 ± 6 MBq for the test scan and 155 ± 20 MBq for the retest scan. Since the two injections were performed using a single radiosynthesis, the specific activity at TOI was higher during the test scan (136 ± 35 MBq/nmol) than during the retest scan (44 ± 11 MBq/nmol), and finally the injected mass was lower for the test scan (110 ± 30 ng versus 820 ± 110 ng). This could tend to reduce specific binding in the retest scan.

Clearance values were again significantly different across scans (test: 96.8 ± 10.3 L/h, retest: 128.0 ± 9.7 L/h, p = 0.01). K₁ values were slightly higher during the retest scan (test: 0.37 ± 0.07, retest: 0.39 ± 0.11 across all regions and subjects, p = 0.26) though the differences did not reach significance.

Similar to [¹¹C]ABP688, significant differences in binding between test and retest scans were observed (Figure 1 (right), p < 0.01, linear mixed effects model). Percent difference in [¹⁸F]FPEB V_T in mid- to high-binding regions ranged from 5% ± 5% to 39% ± 9% (Figure 2 (right)). (Using V_T/f_P as the outcome measure produced results that were almost identical, with average percent differences also ranging from 5% ± 5% to 39% ± 9% across subjects.) All four subjects had higher binding in the retest scan compared to the test scan.

Comparison across tracers

V_TPD cannot directly be compared across tracers because of the different amount of non-specific binding. To account for this, an estimate of binding potential (called BP_ND′ since the non-displaceable binding was estimated) was calculated for each tracer (see Supplementary Material, Section II). Figure 3 shows average percent difference in BP_ND′ (100 × [BP_ND′^retest—BP_ND′^test]/BP_ND′^test), averaged across all regions within a subject). For comparison, our previous [¹¹C]ABP688 test–retest cohort is also included.⁴ From the linear mixed effects model, BP_ND′ values were significantly higher in the retest scan for both [¹¹C]ABP688 and [¹⁸F]FPEB (p < 0.01 for both). As with V_T, for each subject, the magnitude of the average percent difference in BP_ND′ was within 2% of the absolute percent difference.

Figure 3.

Average percent difference (between test and retest scans) in BP_ND′ (BP_ND′PD = 100 × [BP_ND′^retest—BP_ND′^test]/BP_ND′^test) for [¹¹C]ABP688 and [¹⁸F]FPEB. CU1-8 are [¹¹C]ABP688 subjects from a previous cohort, scanned at Columbia University.⁴ YU1-11 are the [¹¹C]ABP688 cohort examined in this paper. FPEB1-4 were scanned with [¹⁸F]FPEB. Both YU ABP688 and FPEB cohorts were scanned using the HRRT at the Yale PET Center. BP_ND′PD was measured and averaged over nine regions: anterior cingulate, medial prefrontal cortex, orbitofrontal cortex, ventral striatum, parietal lobe, putamen, caudate, amygdala and hippocampus. Standard deviation of the percent differences across these regions is represented by error bars. Blue bars represent male subjects and red bars represent female subjects.

Correlation to physiological parameters

Vital sign (heart rate, diastolic and systolic blood pressure and SPO₂) measurements were averaged over each scan for both cohorts. The correlation between average percent difference in V_T and average of the vital signs in the test scan, retest scan and percent difference between test and retest was evaluated. To increase statistical power, eight subjects from our previously published test–retest cohort were included (CU1-8, Figure 3). The only significant correlation of all of the vital signs examined was found between percent difference in V_T and percent change in heart rate (r = 0.54, p = 0.02 in the combined cohort). This relationship remained with the addition of the [¹⁸F]FPEB cohort (r = 0.53, p = 0.01; the difference in average heart rate between test and retest scans was 3.9% ± 10.5% across cohorts). (Note that there were no significant differences in average heart rate between scan 1 and scan 2 for any cohort, all p values >0.22.) As mentioned above, there were significant differences in K₁ and clearance between test and retest scans. The percent difference in heart rate between scans was related to both the percent difference in K₁ (r = 0.66, p < 0.01) and clearance (r = 0.49, p = 0.02) across all cohorts.

Discussion

The goals of this study included: (1) repeating a same-day [¹¹C]ABP688 test–retest study in a new cohort and different PET scanner in order to verify same-day within-subject binding changes in males; (2) determining same-day [¹¹C]ABP688 binding changes in females; and (3) examining whether same-day binding changes were observed in a pilot cohort of [¹⁸F]FPEB subjects.

Similarities in test–retest patterns

[¹¹C]ABP688 and [¹⁸F]FPEB are both NAMs that are disubstituted acetylenes containing a 2-pyridinyl substituent.²³ The binding of these ligands is blocked completely by the administration of NAMs such as 2-methyl-6-(2-phenylethynyl)pyridine (MPEP) or 3-[(2-methyl-1,3-thiazol-4-yl)ethynyl]pyridine (MTEP)²⁴ suggesting that they bind to the mGluR5 at the same, or close to the same, NAM site in the seven transmembrane domain of the receptor. Therefore, it seems very likely that these two radioligands quantify the same pool of mGluR5, and it is not surprising that both ligands show a very similar pattern of increase during the retest scan for the same-day test–retest paradigm, although the underlying mechanism remains unclear.

Interestingly, glutamate, which binds instead to the N-terminal orthosteric binding site, does not directly impact the binding of [¹¹C]ABP688 or [¹⁸F]FPEB in a membrane preparation (Novartis, personal communication, Lin et al., in preparation). Although MPEP, a close analog of ABP and FPEP, does cross cell membranes sufficiently at 10 µM to bind to the intracellular mGluR5,²⁵ in vitro studies suggest that the affinity of FPEB and ABP688 for intracellular receptor is vastly reduced, most consistent with the plasma membrane impeding their intracellular access (Lin et al., in preparation). Thus, we expect that binding of radiotracer concentrations of ABP and FPEB will be greatly reduced by internalization of mGluR5.

In this study, across tracers, binding differences between test and retest studies were correlated with percent change in both K₁ and clearance. Although clearance rates increased between test and retest scans, this would not be expected to affect K₁ or V_T retest scan values if the input function is measured accurately. Therefore, observed binding differences are most likely due to either a systemic error in test or retest input function measurement, or single or multiple physiological changes affecting all variables—K₁, clearance and V_T. Though further studies with larger cohorts will be required to definitely address this issue, the former hypothesis is less likely given that the relationship between the percent difference in V_T and K₁/clearance exists in [¹¹C]ABP688 studies performed at two different facilities, and, further, that the same relationship was observed in the [¹⁸F]FPEB studies.

One potential mechanism regarding the latter hypothesis is alterations in glutamatergic activity that cause up or down regulation of glutamatergic receptors at the cell surface.^1,26 There is a wealth of evidence suggesting glutamatergic regulation of the circadian rhythm and sleep-wake cycle.^27–29 de Prado et al.²⁷ specifically showed decreases in brain glutamate levels during the light cycle and increases during the dark cycle. Thus, it is possible that during the morning scan, there were increases in glutamate levels which in turn downregulate mGluR5 (leading to quantification of lower receptor availability), whereas the decreases in glutamate during the day would lead to mGluR5 upregulation (leading to the quantification of higher receptor availability). Relatedly, a recent clinical trial of an mGluR5 NAM suggested that metabolism of this medication in blood differed in the morning as compared to the evening in healthy females.³⁰ The authors hypothesized circadian regulation of endogenous ligands leads to the differences in the absorption of this medication. Furthermore, observing changes in PET tracer binding in response to varying glutamate levels is consistent with our previous work showing decreases in mGluR5 availability upon pharmacologically induced increases in endogenous glutamate in vivo in human subjects using PET.³¹ Also, a recent study with [¹¹C]ABP688 in rodents showed circadian variation in mGluR5 binding.³²

Injections of glutamate in rodents are also known to affect heart rate.^33,34 It is therefore possible that underlying glutamate transmission changes between test and retest scans affect heart rate and clearance, in addition to PET binding. Although changes in heart rate should not directly affect delivery of tracer to the brain (due to cerebral autoregulation), diurnal variation in cerebral blood flow has been demonstrated in humans³⁵ and rats.^36,37 Recent work has established that neurotransmitter-, and in particular, glutamate-mediated signaling, plays a key role in regulation of cerebral blood flow.³⁸ Therefore, we hypothesize that the underlying brain glutamate transmission changes also influenced tracer delivery (K₁), resulting in correlated V_T and K₁ differences.

Comparison to previous mGluR5 test–retest studies

The rodent and non-human primate literature consistently show that [¹¹C]ABP688 and [¹⁸F]FPEB test–retest variability is low. However, previous [¹¹C]ABP688 results in humans have been disparate, and only in male subjects. Although results of the present study—showing high variability in binding between test and retest scans (Figure 1)—agree with our previous work,⁴ they are not comparable to the other published test–retest study,³ most likely because the latter study design involved imaging sessions that occurred many days apart (at least 7, with an average of 30). For example, a second scan in the same session may be more taxing, and therefore affect glutamate to a greater degree, than two scans separated by multiple days, or diurnal variations, as explained above, may play a role (scan times were not reported in that work). Partially validating the hypothesis that same-day studies are affected by glutamate variation, we are the first to show in a pilot study that [¹⁸F]FPEB appears to behave similarly, i.e. in same-day test–retest studies, binding is significantly higher in the second scan, whereas two scans days to weeks apart showed good test–retest variability.^11–13 (In our previous [¹⁸F]FPEB study, scans were separated by 1.1 ± 2.7 h in terms of time of day of injection.¹¹ Furthermore, Leurquin-Sterk et al., specifically ensured that their test and retest [¹⁸F]FPEB studies, though separated by six months, differed by less than 30 min, on average, in terms of time of day.¹³)

In our previous test–retest study, we observed increased [¹¹C]ABP688 binding in seven of the eight male subjects from test to retest. In the present study, increases were observed in 7 of 11 subjects, while three subjects showed modest decreases in binding (<10%) and one subject (Subject 1) showed an average decrease of 22.1% (this subject was scanned two days apart). [¹⁸F]FPEB was increased in all four subjects during the second scan. Given that (1) significant test–retest changes are not observed when scans are separated by multiple days, (2) the same-day test–retest binding changes are mostly in a consistent direction (increase at retest) and (3) both tracers behave similarly, it is most likely that there are physiological changes in same-day studies (e.g., glutamate modulation) rather than poor reliability of these tracers. This may not have been observed in preclinical studies due to the effects of anesthesia. For example, at subanesthetic doses, ketamine (an NMDA receptor antagonist) leads to a large surge in glutamate but at anesthetic doses there are no changes in glutamate levels.³⁹

Implications for same-day studies

To optimize energy resources, numerous biological processes exhibit variation over light–dark (or feeding–fasting) cycles.⁴⁰ Diurnal variation in cytokine activity,⁴¹ amyloid beta plaques⁴², serotonin,⁴³ dopamine, glutamate and GABA⁴⁴ have been reported. This variation can affect same-day PET studies, and initial studies have reported such effects. For example, a recent same-day study using [¹¹C]PBR28, a translocator protein (TSPO) ligand, reported a significant increase in V_T between morning and afternoon scans, potentially due to circadian variation in cytokines which are known to affect TSPO levels.⁴⁵ Diurnal (and seasonal) variation was also reported in same-day PET studies of the serotonin 1A receptor using [¹¹C]WAY-100635 and serotonin transporter using [¹¹C]MADAM.⁴⁶ As these tracers were shown to be insensitive to changes in endogenous serotonin, the hypothesized mechanism was diurnal changes in receptor and transporter expression. Similarly, a preliminary study reported age-dependent diurnal changes in binding to the dopamine D_2/3 receptors using [¹¹C]FLB 457 but not [¹¹C]raclopride.⁴⁷ Authors of this study posited that this effect could be due to tracer sensitivity to changing dopamine levels. The above studies show that same-day variation can arise from multiple sources, including diurnal variation in the receptor itself, endogenous ligands that occupy the target or up- or down-regulate the receptor and/or signals upstream that control levels of the receptor and/or ligands. However, mGluR5 PET studies, so far, are unique in the magnitude of same-day differences observed and the replication of results in a second cohort. In the case of [¹¹C]PBR28, for example, the difference in absolute variation in binding of same-day subjects versus those scanned two to five days later was not significant (21% ± 21% and 16% ± 12%, respectively).⁴⁵ Furthermore, diurnal variation was not observed in other same-day studies of the serotonin 1A receptor,⁴⁸ transporter,^49–51 dopamine transporter,^22,52 or D_2/3 receptors.⁵³ This does not indicate that variation does not exist; however, it is possible that the changes are subtler due to the systems or tracers used. As such, mGluR5 may represent an ideal system in which to probe such differences and develop strategies to quantify or minimize them as needed.

Relationship to challenge studies

Our recent study showed an effect of a ketamine-induced glutamate surge on [¹¹C]ABP688 binding in male and female subjects.³¹ These subjects participated in two scans on the same day; however, they received an infusion of ketamine during the second (retest) scan. The results indicated a significant decrease in [¹¹C]ABP688 binding during this scan. Given the findings in the present study, it is likely that the effect of the ketamine challenge was substantially underestimated. This may be particularly significant for the female sample, which showed an average V_TPD increase of up to 107.7%. Thus, future studies performed with drug challenges will need to consider the data presented herein when interpreting their results.

Conclusion

Study results show increases in [¹¹C]ABP688 and [¹⁸F]FPEB binding for most subjects between same-day test–retest studies in both sexes. The results are consistent with previous same-day test–retest studies for [¹¹C]ABP688, and binding patterns were similar across tracers. Specifically, both tracers show low variability when test and retest scans are separated by days to weeks, but significant differences in binding (mostly increases on retest) when scans are performed on the same day. As the [¹⁸F]FPEB pilot cohort was added to show consistency with [¹¹C]ABP688 findings, full validation of these findings will require a larger cohort. Future work is also needed to understand the source of these changes and to ensure changes are not due to systematic errors, although the likelihood of this is small given the effect was observed across facilities and tracers. As both of these tracers target mGluR5, one potential source driving these binding changes may be changes in glutamate levels, which do not appear to affect tracer binding directly but affect mGluR5 surface localization and thus ligand accessibility. If this hypothesis is validated, this work illustrates the importance for control of experimental design procedures, potentially including time of day, for glutamatergic tracers. Furthermore, it may open up potential future areas of research, such as monitoring glutamatergic changes in health and disease.

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Support provided by K01MH092681 (Esterlis), VA National Center for PTSD (Esterlis), K01MH091354 (DeLorenzo), K05DA022413 (Javitch) and R01MH054137 (Javitch).

Acknowledgements

The authors wish to thank Dr Ana Franceschi and Ms Grace Lawal for their diligent work transcribing, organizing and analyzing study data, and Dr Kyle Lapidus for his input on manuscript content. We acknowledge the biostatistical consultation and support from the Biostatistical Consulting Core at the School of Medicine, Stony Brook University.

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr Mann receives royalties for commercial use of the C-SSRS from the Research Foundation for Mental Hygiene. F Gasparini is an employee of Novartis Pharma AG. The other authors have no conflicts of interest to declare.

Authors’ contribution

Christine DeLorenzo, Jean-Dominique Gallezot and Irina Esterlis performed the outlined studies, performed the analysis and wrote/edited the majority of the manuscript. John Gardus and Beata Planeta performed PET image analysis and provided editorial comments. Jie Yang and R. Todd Ogden performed and wrote up the statistical analysis. Nabeel Nabulsi, David C. Labaree and Yiyun H. Huang performed the PET chemistry and wrote up the tracer synthesis and plasma analysis sections. Fabrizio Gasparini and Xin Lin provided background analysis for theoretical foundations of the manuscript and editorial comments. J. John Mann, Jonathan A. Javitch, Ramin V. Parsey and Richard E. Carson provided oversight of PET studies, reviewed intermediate data, provided guidance for study design and analysis and extensive editorial comments.

Supplementary material

Supplementary material for this paper can be found at

References

Krystal

Mathew

D’Souza

, et al. Potential psychiatric applications of metabotropic glutamate receptor agonists and antagonists. CNS Drugs 2010; 24: 669–693.

Ametamey

Kessler

Honer

, et al. Radiosynthesis and preclinical evaluation of 11C-ABP688 as a probe for imaging the metabotropic glutamate receptor subtype 5. J Nucl Med 2006; 47: 698–705.

Burger

Deschwanden

Ametamey

, et al. Evaluation of a bolus/infusion protocol for 11C-ABP688, a PET tracer for mGluR5. Nucl Med Biol 2010; 37: 845–851.

DeLorenzo

Kumar

Mann

, et al. In vivo variation in metabotropic glutamate receptor subtype 5 binding using positron emission tomography and [11C]ABP688. J Cereb Blood Flow Metab 2011; 31: 2169–2180.

Elmenhorst

Aliaga

Bauer

, et al. Test-retest stability of cerebral mGluR(5) quantification using [(1)(1)C]ABP688 and positron emission tomography in rats. Synapse 2012; 66: 552–560.

Wyckhuys

Verhaeghe

Wyffels

, et al. N-acetylcysteine- and MK-801-induced changes in glutamate levels do not affect in vivo binding of metabotropic glutamate 5 receptor radioligand 11C-ABP688 in rat brain. J Nucl Med 2013; 54: 1954–1961.

DeLorenzo

Milak

Brennan

, et al. In vivo positron emission tomography imaging with [(11)C]ABP688: binding variability and specificity for the metabotropic glutamate receptor subtype 5 in baboons. Eur J Nucl Med Mol Imaging 2011; 38: 1083–1094.

Miyake

Skinbjerg

Easwaramoorthy

, et al. Imaging changes in glutamate transmission in vivo with the metabotropic glutamate receptor 5 tracer [11C] ABP688 and N-acetylcysteine challenge. Biol Psychiatry 2011; 69: 822–824.

Sandiego

Nabulsi

Lin

, et al. Studies of the metabotropic glutamate receptor 5 radioligand [(1)(1)C]ABP688 with N-acetylcysteine challenge in rhesus monkeys. Synapse 2013; 67: 489–501.

10.

Patel

Hamill

Connolly

, et al. Species differences in mGluR5 binding sites in mammalian central nervous system determined using in vitro binding with [18F]F-PEB. Nucl Med Biol 2007; 34: 1009–1017.

11.

Park

Sullivan

Planeta

, et al. Test-retest reproducibility of the metabotropic glutamate receptor 5 ligand [F]FPEB with bolus plus constant infusion in humans. Eur J Nucl Med Mol Imaging 2015; 42(10): 1530–1541.

12.

Wong

Waterhouse

Kuwabara

, 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 Med 2013; 54: 388–396.

13.

Leurquin-Sterk

Postnov

de Laat

, et al. Kinetic modelling and long-term test-retest reproducibility of the mGluR5 PET tracer F-FPEB in human brain. Synapse 2016; 70(4): 153–162.

14.

de Laat

Leurquin-Sterk

Celen

, et al. Preclinical evaluation and quantification of 18F-FPEB as a radioligand for PET imaging of the metabotropic glutamate receptor 5. J Nucl Med 2015; 56: 1954–1959.

15.

Carson

Kiesewetter

Jagoda

, et al. Muscarinic cholinergic receptor measurements with [18F]FP-TZTP: control and competition studies. J Cereb Blood Flow Metab 1998; 18: 1130–1142.

16.

Sullivan

Lim

Labaree

, et al. 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 Metab 2012; 33(4): 532–541.

17.

Ogden

Mann

, et al. Optimal metabolite curve fitting for kinetic modeling of 11C-WAY-100635. J Nucl Med 2007; 48: 926–31.

18.

Gallezot

Nabulsi

Neumeister

, et al. Kinetic modeling of the serotonin 5-HT(1B) receptor radioligand [(11)C]P943 in humans. J Cereb Blood Flow Metab 2010; 30: 196–210.

19.

Papademetris

Jackowski

Schultz

, et al. Integrated intensity and point-feature nonrigid registration. Med Image Comput Comput Assist Interv 2001; 3216: 763–770.

20.

Tzourio-Mazoyer

Landeau

Papathanassiou

, et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage 2002; 15: 273–289.

21.

Gunn

Cunningham

. Positron emission tomography compartmental models. J Cereb Blood Flow Metab 2001; 21: 635–652.

22.

Hirvonen

Johansson

Teras

, et al. Measurement of striatal and extrastriatal dopamine transporter binding with high-resolution PET and [11C]PE2I: quantitative modeling and test-retest reproducibility. J Cereb Blood Flow Metab 2008; 28: 1059–1069.

23.

Fuchigami

Nakayama

Yoshida

. Development of PET and SPECT probes for glutamate receptors 2015. ScientificWorldJournal 2015, pp. 716514.

24.

Wang

Tueckmantel

Zhu

, et al. Synthesis and preliminary biological evaluation of 3-[(18)F]fluoro-5-(2-pyridinylethynyl)benzonitrile as a PET radiotracer for imaging metabotropic glutamate receptor subtype 5. Synapse 2007; 61: 951–961.

25.

Vincent

Cornea

Jong

, et al. Intracellular mGluR5 plays a critical role in neuropathic pain. Nat Commun 2016; 7: 10604.

26.

Niciu

Henter

Sanacora

, et al.

Glial abnormalities in substance use disorders and depression: does shared glutamatergic dysfunction contribute to comorbidity?

World J Biol Psychiatry 2014; 15: 2–16.

27.

de Prado

Castañeda

Galindo

, et al. Melatonin disrupts circadian rhythms of glutamate and GABA in the neostriatum of the aware rat: a microdialysis study. J Pineal Res 2000; 29: 209–216.

28.

Fuller

Gooley

Saper

. Neurobiology of the sleep-wake cycle: sleep architecture, circadian regulation, and regulatory feedback. J Biol Rhythms. 21: 482–493.

29.

Meng

Yuan

Zheng

, et al. Effects of endogenous melatonin on glutamate and GABA rhythms in the striatum of unilateral 6-hydroxydopamine-lesioned rats. Neuroscience 2015; 286: 308–315.

30.

Farha

Bruggeman

Baljé-Volkers

. Metabotropic glutamate receptor 5 negative modulation in phase i clinical trial: potential impact of circadian rhythm on the neuropsychiatric adverse reactions—do hallucinations matter? ISRN Psychiatry 2014. 2014: 652750.

31.

DeLorenzo

DellaGioia

Bloch

, et al. In vivo ketamine-induced changes in [11C]ABP688 binding to metabotropic glutamate receptor subtype 5. Biol Psychiatry 2014; 77: 266–275.

32.

Elmenhorst

Mertens

Kroll

, et al. Circadian variation of metabotropic glutamate receptor 5 availability in the rat brain. J Sleep Res. Epub ahead of print 30 June 2016. DOI: 10.1111/jsr.12432 .

33.

Ruit

Neafsey

. Cardiovascular and respiratory responses to electrical and chemical stimulation of the hippocampus in anesthetized and awake rats. Brain Res 1988; 457: 310–321.

34.

Kannan

Hayashida

Yamashita

. Increase in sympathetic outflow by paraventricular nucleus stimulation in awake rats. Am J Physiol 1989; 256: R1325–R1330.

35.

Conroy

Spielman

Scott

. Daily rhythm of cerebral blood flow velocity. J Circadian Rhythms 2005; 3: 3.

36.

Endo

Jinnai

Endo

, et al. Diurnal variation of cerebral blood flow in rat hippocampus. Stroke 1990; 21: 1464–1469.

37.

Wauschkuhn

Witte

Gorbey

, et al. Circadian periodicity of cerebral blood flow revealed by laser-Doppler flowmetry in awake rats: relation to blood pressure and activity. Am J Physiol Heart Circ Physiol 2005; 289: H1662–H1668.

38.

Attwell

Buchan

Charpak

, et al. Glial and neuronal control of brain blood flow. Nature 2010; 468: 232–243.

39.

Chowdhury

Behar

Cho

, et al. ¹H-[¹³C]-nuclear magnetic resonance spectroscopy measures of ketamine’s effect on amino acid neurotransmitter metabolism. Biol Psychiatry 2012; 71: 1022–1025.

40.

Gooley

. Circadian regulation of lipid metabolism. Proc Nutr Soc. Epub ahead of print 26 May 2016. DOI: 10.1017/S0029665116000288 .

41.

Nakao

. Temporal regulation of cytokines by the circadian clock. J Immunol Res 2014. 2014: 614529.

42.

Huang

Potter

Sigurdson

, et al. Effects of age and amyloid deposition on Abeta dynamics in the human central nervous system. Arch Neurol 2012; 69: 51–58.

43.

Ferraro

Steger

. Diurnal variations in brain serotonin are driven by the photic cycle and are not circadian in nature. Brain Res 1990; 512: 121–124.

44.

Castaneda

de Prado

Prieto

, et al. Circadian rhythms of dopamine, glutamate and GABA in the striatum and nucleus accumbens of the awake rat: modulation by light. J Pineal Res 2004; 36: 177–185.

45.

Collste

Forsberg

Varrone

, et al. Test-retest reproducibility of [(11)C]PBR28 binding to TSPO in healthy control subjects. Eur J Nucl Med Mol Imaging 2016; 43: 173–183.

46.

Matheson

Schain

Almeida

, et al. Diurnal and seasonal variation of the brain serotonin system in healthy male subjects. Neuroimage 2015; 112: 225–231.

47.

Cervenka

Halldin

Farde

. Age-related diurnal effect on D2 receptor binding: a preliminary PET study. Int J Neuropsychopharmacol 2008; 11: 671–678.

48.

Parsey

Slifstein

Hwang

, et al. Validation and reproducibility of measurement of 5-HT1A receptor parameters with [carbonyl-11C]WAY-100635 in humans: comparison of arterial and reference tisssue input functions. J Cereb Blood Flow Metab 2000; 20: 1111–1133.

49.

Ogden

Ojha

Erlandsson

, et al. In vivo quantification of serotonin transporters using [(11)C]DASB and positron emission tomography in humans: modeling considerations. J Cereb Blood Flow Metab 2007; 27: 205–217.

50.

Frankle

Slifstein

Gunn

, et al. Estimation of serotonin transporter parameters with 11C-DASB in healthy humans: reproducibility and comparison of methods. J Nucl Med 2006; 47: 815–826.

51.

Kim

Ichise

Sangare

, et al. PET imaging of serotonin transporters with [11C]DASB: test-retest reproducibility using a multilinear reference tissue parametric imaging method. J Nucl Med 2006; 47: 208–214.

52.

DeLorenzo

Kumar

Zanderigo

, et al. Modeling considerations for in vivo quantification of the dopamine transporter using [(11)C]PE2I and positron emission tomography. J Cereb Blood Flow Metab 2009; 29: 1332–1345.

53.

Narendran

Mason

May

, et al. Positron emission tomography imaging of dopamine D(2)/(3) receptors in the human cortex with [(1)(1)C]FLB 457: reproducibility studies. Synapse 2011; 65: 35–40.

54.

Delforge

Syrota

Bottlaender

, et al. Modeling analysis of [11C]flumazenil kinetics studied by PET: application to a critical study of the equilibrium approaches. J Cereb Blood Flow Metab 1993; 13: 454–468.

55.

Koeppe

Raffel

Snyder

, et al. Dual-[11C]tracer single-acquisition positron emission tomography studies. J Cereb Blood Flow Metab 2001; 21: 1480–1492.

56.

Kågedal

Cselényi

Nyberg

, et al. A positron emission tomography study in healthy volunteers to estimate mGluR5 receptor occupancy of AZD2066—estimating occupancy in the absence of a reference region. Neuroimage 2013; 82: 160–169.

57.

Guo

Owen

Rabiner

, et al. A graphical method to compare the in vivo binding potential of PET radioligands in the absence of a reference region: application to [(1)(1)C]PBR28 and [(1)(8)F]PBR111 for TSPO imaging. J Cereb Blood Flow Metab 2014; 34: 1162–1168.

58.

Hesselgrave

Parsey

. Imaging the serotonin 1A receptor using [11C]WAY100635 in healthy controls and major depression. Philos Trans R Soc Lond B Biol Sci 2013; 368: 20120004.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.69 MB

0.00 MB

In vivo variation in same-day estimates of metabotropic glutamate receptor subtype 5 binding using [ 11 C]ABP688 and [ 18 F]FPEB

Abstract

Keywords

Introduction

Materials and methods

Subjects

Positron emission tomography

[11C]ABP688 synthesis and PET scanning

[18F]FPEB synthesis and PET scanning

Magnetic resonance imaging

Input function measurement

Image analysis

Background correction for [18F]FPEB retest scans

Outcome measure calculation and analysis

Results

[11C]ABP688 test–retest results

[18F]FPEB test–retest results

Comparison across tracers

Correlation to physiological parameters

Discussion

Similarities in test–retest patterns

Comparison to previous mGluR5 test–retest studies

Implications for same-day studies

Relationship to challenge studies

Conclusion

Footnotes

Funding

Acknowledgements

Declaration of conflicting interests

Authors’ contribution

Supplementary material

References

Supplementary Material

[¹¹C]ABP688 synthesis and PET scanning

[¹⁸F]FPEB synthesis and PET scanning

Background correction for [¹⁸F]FPEB retest scans

[¹¹C]ABP688 test–retest results

[¹⁸F]FPEB test–retest results