Characterization of [11C]RO5013853, a novel PET tracer for the glycine transporter type 1 (GlyT1) in humans

NeuroImage 75 (2013) 282–290

Contents lists available at SciVerse ScienceDirect

NeuroImage journal homepage: www.elsevier.com/locate/ynimg

Characterization of [ 11C]RO5013853, a novel PET tracer for the glycine transporter type 1 (GlyT1) in humans Dean F. Wong a, b, c, d, e,⁎, 1, Susanne Ostrowitzki f, 1, Yun Zhou d, Vanessa Raymont d, Carsten Hofmann f, Edilio Borroni f, Anil Kumar d, Nikhat Parkar f, James R. Brašić d, John Hilton d, Robert F. Dannals d, Meret Martin-Facklam f a The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, 601 North Caroline Street, Room 3245, Johns Hopkins Outpatient Center, Baltimore, MD 21287 -0807, USA b Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore, MD 21287–0807, USA c Department of Environmental Health Sciences, Johns Hopkins University School of Medicine, Baltimore, MD 21287–0807, USA d Department of Psychiatry, Johns Hopkins University School of Medicine, Baltimore, MD 21287 -0807, USA e Honorary Professor of Neuroscience and Pharmacology, University of Copenhagen, Denmark f F. Hoffmann-La Roche Ltd., Pharmaceutical Division, Grenzacherstrasse 124, CH-4070 Basel, Switzerland

a r t i c l e

i n f o

Article history: Accepted 14 November 2011 Available online 1 December 2011 Keywords: Positron emission tomography (PET) Receptors N-methyl-D-aspartate (NMDA) Glycine transporter type 1 (GlyT1) Glycine reuptake inhibitor (GRI) Radioligand assays Schizophrenia

a b s t r a c t We characterize a novel radioligand for the glycine transporter type 1 (GlyT1), [11 C]RO5013853, in humans. Ten healthy male volunteers, 23–60 years of age, were enrolled in this PET study; seven subjects participated in the evaluation of test-retest reliability and three subjects in whole body dosimetry. Subjects were administered intravenous bolus injections of approximately 1100 MBq (30 mCi) [11 C]RO5013853 with a high specific activity of about 481 GBq (13 Ci)/μmol. Standard compartmental model analysis with arterial plasma input function, and an alternative noninvasive analysis method which was evaluated and validated by occupancy studies in both baboons and humans, were performed. Mean parameter estimates of the volumes of distribution (VT) obtained by a 2-tissue 5-parameter model were higher in the cerebellum, pons, and thalamus (1.99 to 2.59 mL/mL), and lower in the putamen, caudate, and cortical areas (0.86 to 1.13 mL/mL), with estimates showing less than 10% difference between test and retest scans. Tracer retention was effectively blocked by the specific glycine reuptake inhibitor (GRI), bitopertin (RG1678). [ 11 C]RO5013853 was safe and well tolerated. Human dosimetry studies showed that the effective dose was approximately 0.0033 mSv/MBq, with the liver receiving the highest absorbed dose. In conclusion, quantitative dynamic PET of the human brain after intravenous injection of [11C]RO5013853 attains reliable measurements of GlyT1 binding in accordance with the expected transporter distribution in the human brain. [11C]RO5013853 is a radioligand suitable for further clinical PET studies. Full characterization of a novel radiotracer for GlyT1 in humans is provided. The tracer has subsequently been used to assess receptor occupancy in healthy volunteers and to estimate occupancy at doses associated with best efficacy in a clinical trial with schizophrenic patients with predominantly negative symptoms. © 2011 Elsevier Inc. All rights reserved.

Introduction Glycine transporter type 1 (GlyT1) is a target of interest in the development of novel drugs for schizophrenia. GlyT1 inhibition is one of the key strategies to address N-methyl-D-aspartate receptor (NMDAR) hypofunction which is postulated to play an important role in the pathophysiology of schizophrenia (Harrison and Weinberger, 2005; Hashimoto, 2011; Krystal et al., 1994; Millan, 2005; Sanger, 2004; ⁎ Corresponding author at: 601 N. Caroline St., JHOC Room 3245, Johns Hopkins Medical Institutions, Baltimore, MD 21287–0807, USA. Fax: + 1 410 955 0696. E-mail address: [email protected] (D.F. Wong). 1 These two authors contributed equally. 1053-8119/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2011.11.052

Stone, 2009). Promising clinical findings were recently reported in schizophrenic patients with predominantly negative symptoms who were treated with the glycine reuptake inhibitor (GRI) bitopertin (RG1678) (Umbricht et al., 2010). Positron emission tomography (PET) data indicate that low to medium target occupancy was associated with the best efficacy (Umbricht et al., 2011). A radioligand specific for GlyT1 with appropriate molecular properties for a PET imaging probe is a key tool for the clinical development of a GRI. Such a radioligand allows assessment of brain penetration of candidate drugs and enables the quantification and characterization of transporter occupancy. RO5013853 ([5-methanesulfonyl-2-((S)-2,2,2-trifluoro-1methyl-ethoxy)-phenyl]-[5-(tetrahydro-pyran-4-yl)-1,3-dihydroisoindol-2-yl]-methanone) was identified as a high affinity ligand for

D.F. Wong et al. / NeuroImage 75 (2013) 282–290

GlyT1 suitable for PET radiochemistry and was successfully radiolabeled yielding [ 11C]RO5013853 (Pinard et al., 2011). Furthermore, it was selected as the better of two PET ligands evaluated in a study in non-human primates presented in this issue (Borroni et al., 2013– this issue). Here we report the full characterization of [ 11C]RO5013853 as a novel PET radioligand for GlyT1 in humans. Objectives were as follows: (1) to examine and analyze the specific uptake and kinetics of [ 11 C]RO5013853 in the human brain; (2) to assess the test–retest reliability of quantitative PET measurements between two sessions; (3) to determine human radiation dosimetry following injection of a single microdose of [ 11 C]RO5013853; (4) to report rodent radiation dosimetry for suitability of the microdose of the radiotracer for human use; and (5) to report the acute safety and tolerability of [ 11C] RO5013853 administered intravenously at microdose levels in healthy humans.

283

Table 1 lists the radioactivity dose (MBq [mCi]), mass (μg), mass per kg (ng/kg), and specific activity (GBq [Ci]/μmol) for individual subjects and scans. For the first PET session, subjects (n= 7) received an average dose of 1111.11 ± 52.17 MBq (30.03 ± 1.41 mCi) (mean± SD), with mass of 1.62 ± 0.91 μg, mass per kg of 19.57± 11.70 ng/kg, and specific activity of 496 ± 388 GBq (13 ± 10 Ci)/μmol. The administered dose for the second PET session averaged 1082.99 ± 51.8 MBq (29.27 ± 1.40 mCi), with mass of 1.31± 0.45 μg, mass per kg of 15.19 ± 4.98 ng/kg, and specific activity of 459± 178 GBq (12 ± 5 Ci)/μmol. The administered doses, masses, and specific activities of [11C] RO5013853 were not significantly different across the two sessions. The injected radioactivity was consistent with ALARA (as low absorbed radiation as possible) principles for the scientific question asked and based on rodent and verified by human radiation dosimetry. The maximal allowed mass dose per injection was 20 μg RO5013853 which fulfilled the definition of a microdose (Food and Drug Administration, 2006).

Materials and methods PET study Subjects and design This single-center, open label, non-randomized study was conducted in accordance with the Declaration of Helsinki Principles. The study received ethical approval from the Johns Hopkins School of Medicine Institutional Review Board (Baltimore, MD) and Chesapeake Research Review, Inc. (Columbia, MD) and was conducted under an exploratory U.S. IND following appropriate animal toxicology studies of the unlabeled ligand and radiation dosimetry. All participants provided written, informed consent after receiving oral and written descriptions of study procedures and aims. The study was conducted in 10 healthy male volunteers, 23 to 60 years of age, with a body mass index ranging from 21.6 to 29.7 kg/m 2. Eligibility criteria included: no significant medical or surgical history; no history of head trauma with prolonged loss of consciousness; no neurologic conditions; no history of alcohol and/or drug abuse or addiction within the last 2 years; heart rate between 40 and 100 beats per minute; no family history of congenital long QT syndrome or sudden death; and a clinically unremarkable brain magnetic resonance imaging (MRI) scan at screening. Subjects were to be medication-free for 14 days or 5 times the elimination half-life (whichever was longer) prior to the first PET scan and throughout the study. Subjects entered the clinic the day before the PET scan and remained in-house until the day after the PET scan. Seven subjects were to be scanned twice, in order to study test-retest reliability of outcome parameters, and 3 subjects underwent whole body PET scans for evaluation of human radiation dosimetry. Safety assessments Safety assessments included: continuous medical monitoring during and following the radioligand injection for approximately 24 hours, adverse event reporting, blood chemistry profile, including liver and renal function tests, complete blood count, electrocardiogram, urinalysis, and a physical examination.

Blood sampling and metabolite analysis For derivation of the plasma input function, arterial blood was initially sampled frequently with increasingly prolonged intervals throughout the scan for 90 minute post injection. Total radioactivity was measured in more than 30 samples using a gamma counter that was cross-calibrated against the PET activity measurements. Selected blood samples collected at 0, 5, 15, 30, 45, 60, and 90 minutes were analyzed by HPLC for the presence of [ 11 C]RO5013853 and its radiolabeled metabolites using a general method developed previously for PET radiotracers (Hilton et al., 2000). PET scanning Following placement of a radial arterial catheter (for the input function) and a venous catheter (radioligand injection), seven

Table 1 Injected radioactivity dose, mass, mass per kg, and specific activity per subject per scan [n = 7 subjects]. Data subject

ID

Scan number

Injected dose (MBq) [mCi]

Mass (μg)

Mass per kg (ng/kg)

Specific activity (GBq/μmol) [Ci/μmol]

1

101

1

1095.94 [29.62] 1084.47 [29.31] 1079.29 [29.17] 1175.86 [31.78] 1135.53 [30.69] 1091.5 [29.50] 1013.80 [27.40] 1072.26 [28.98] 1143.67 [30.91] 1025.64 [27.72] 1161.80 [31.40] 1107.04 [29.92] 1147.37 [31.01] 1025.27 [27.71]

2.83

30.43

1.65

17.74

2.35

30.92

1.73

22.76

2.47

33.38

0.72

9.73

1.16

13.18

1.47

16.70

0.43

4.67

1.08

11.74

0.93

9.69

1.74

18.12

1.18

14.75

0.76

9.50

191 [5] 324 [9] 227 [6] 335 [9] 227 [6] 746 [20] 430 [12] 359 [10] 1301 [35] 468 [13] 614 [17] 313 [8] 480 [13] 667 [18]

2 2

102

1 2

3

106

1 2

4

107

1 2

Radioligand [ 11 C]RO5013853 is a highly selective ligand for GlyT1 and was synthesized as previously described (Pinard et al., 2011). Briefly, an appropriate desmethyl precursor was radiolabeled with [ 11C]methyl iodide and purified by reverse-phase high-performance liquid chromatography (HPLC). A typical synthesis produced 4132.9 ± 1465.2 MBq (111.7 ± 39.6 mCi) (mean ± SD, hereafter) of [ 11C] RO5013853. The end of synthesis specific radioactivity was 494 ± 296 GBq (13 ± 8 Ci)/μmol. Analytical HPLC showed a radiochemical purity of 100%. [ 11C]RO5013853 was provided as a sterile, pyrogenfree injectable solution (14:1 normal saline:ethanol).

5

109

1 2

6

108

1 2

7

110

1 2

284

D.F. Wong et al. / NeuroImage 75 (2013) 282–290

subjects received two single injections of [ 11C]RO5013853 each, 1–3 weeks apart. In five of these subjects, both test and retest scans were acquired on the GE Advance PET scanner. To examine whether a high-resolution research tomograph (HRRT®, CPS Innovations, Inc., Knoxville, TN, USA) scanner could improve the quantitative PET measurements of [ 11C]RO5013853 uptake in brain tissue, an additional two subjects' first scan was acquired on the HRRT and the second on the GE Advance PET scanner. The head of the subject was fitted with a thermoplastic mask to minimize head motion. A transmission scan with a Ge-68 (GE Advance) or Cs-137 (HRRT) point source was initially performed for attenuation correction. The 90-minute dynamic emission scan of the GE Advance scanner was acquired (30 frames: 4 × 0.25, 4 × 0.5, 3 × 1, 2 × 2, 5 × 4, and 12 × 5 min) in three-dimensional (3D) mode and started with radioligand injection over 1 minute. Dynamic images were reconstructed using filtered back-projection with a ramp filter (35 slices with image matrix 128 × 128; pixel size: 2 × 2 mm 2; interslice spacing 4.25 mm), resulting in a spatial resolution of 5.5 × 5.5 × 4.25 mm full-width at half-maximum at the center of the field of view. For the dynamic HRRT scans, the images were reconstructed in 3D mode with high-resolution span-3 modality with the same frame schedule as the one used for the GE Advance scanner. The statistical 3D reconstruction of each frame sinogram used six iterations of the OP­OSEM algorithm (16 subsets), followed by 2 mm Gaussian post-smoothing. Crystal dead time, decay, and calibration factor corrections were applied, producing a quantitative dynamic image (207 slices with image matrix 256 × 256; pixel size: 1.22 × 1.22 mm 2, inter-slice spacing 1.22 mm), resulting in a spatial resolution of 2.5 × 2.5 × 2.5 mm full-width at half-maximum at the center of the field of view (Rahmim et al., 2004). HRRT has been used and validated in the Johns Hopkins PET center over the past 5 years (Horti et al., 2006; Wilcox et al., 2008; Wong et al., 2010). MRI data Each participant underwent a brain MRI scan (GE 1.5 T Sigma; spoiled gradient-recalled [SPGR]) sequence and a double echo (proton density and T2-weighted) sequence as part of the eligibility assessment for this study. Additionally, the MRI (matrix size: 256 × 256, pixel size 0.938 × 0.938 mm 2, 124 slices with 1.5 mm slice thickness) was used for co-registration with the PET data to enhance anatomical definitions of regions of interest (ROIs). Image analysis The means of 90-minute dynamic PET images were used for MRI to PET co-registration using Statistical Parametric Mapping 2 software (SPM2; Friston, 2002). The ROIs including cerebellum, pons, thalamus, putamen, caudate, and cingulate, occipital, parietal, temporal, orbital frontal, prefrontal, and superior frontal cortices were manually drawn on the co-registered MRI images (Zhou et al., 2007). The ROI time-activity curves (TACs) were obtained by applying the ROIs to the dynamic PET images. A standardized uptake value (SUV) was calculated for the normalization of ROI TACs: SUVðTACÞ ¼ TACðMBq=mLÞ=ðinjected tracer doseðMBqÞ=body weightðgÞÞ ¼ TACðμCi=mLÞ=ðinjected tracerdoseðμCiÞ=body weightðgÞÞ:

Tracer kinetic modeling The ROI tracer kinetics were analyzed by compartmental models with the measured metabolite-corrected plasma input function. A 1tissue 3-parameter (1T3P; K1, k2′, VP) as well as 2-tissue 5­parameter (2T5P; K1, k2, k3, k4, VP) model in series configuration were used to fit the measured ROI TACs (Huang et al., 1986; Innis et al., 2007; Koeppe et al., 1991; Schmidt and Turkheimer, 2002). To reduce the variation of tracer total distribution volume (VT) resulting from the estimates of k4, a nonlinear model fitting algorithm with k4 was coupled over

all ROIs for each dynamic PET study (Cunningham et al., 2004; Zhou et al., 2007, 2010). The final selection of the appropriate tracer kinetic model was based on the Akaike information criterion (AIC; Akaike, 1976; Carson et al., 1993, 1997; Gunn et al., 2001; Turkheimer et al., 2003). A lower AIC predicts a better fit; therefore, the model associated with the lower AIC is considered the best model. VT in tissue was calculated after model fitting as VT = K1/k2′ and VT = (K1/k2)(1 + k3/k4) for 1T3P and 2T5P, respectively. The nondisplaceable binding potential (BPND), an index of tracer specific binding in ligand-receptor PET studies was estimated as: BPND = VT/VT(reference tissue) − 1 (Innis et al., 2007; Koeppe et al., 1991), where the superior frontal cortex was used as a reference tissue (as this region exhibits minor specific binding only). This method was evaluated in full human occupancy studies (unpublished data). BPND was also estimated by a simplified reference tissue model (SRTM; Lammertsma and Hume, 1996; Zhou et al., 2003). Parametric image of VT To evaluate voxel-wise tracer distribution in brain, a multigraphical analysis method (Zhou et al., 2010) using a relative equilibrium-based graphic analysis (RE plot; Zhou et al., 2009) with the Gjedde-Patlak plot (GP plot; Gjedde, 1981; Patlak et al., 1983; Patlak and Blasberg, 1985) (RE–GP plots) was used to generate parametric VT images. The novel graphic analysis method is a modelindependent quantitative approach. Reproducibility metrics The test-retest reliability (percent difference) of an outcome variable P was calculated as: 100  jPtest –Pretest j Ptest where P can be K1, VT, and BP ROI estimates, and Ptest and Pretest are estimated from the first and second dynamic PET scans, respectively. Specificity of [ 11C]RO5013853 In a second study (unpublished data), healthy volunteer subjects were dosed with a high-affinity, selective GRI, bitopertin (RG1678) (Alberati et al., 2010; Pinard et al., 2010), until steady state plasma concentrations were reached. Subjects underwent dynamic PET scanning with a standard protocol pre-dose and on the last day of treatment at steady state. Included for illustrative purposes here are PET data from one subject (baseline and post-baseline after once daily administration with 175 mg bitopertin over 12 days) to show proof-ofprinciple for bitopertin to block [ 11 C]RO5013853 binding. Radiation dosimetry (rodents and humans) Rodent radiation dosimetry was carried out by intravenous injection of the radiotracer into the tail vein in CD-1 mice (triplicates). After sacrifice and organ dissection, radioactivity was measured in a gamma spectrometer at varying sacrifice times at 5, 15, 30, 60, 90, 120, and 180 minutes post injection. Results were used to estimate the human effective dose in this study. For human dosimetry, three subjects received an average injected activity of 1076.7 MBq (range: 1065.6–1084.1) [29.1 mCi (range: 28.8–29.3)] followed by wholebody PET-computed tomography scans on a GE Discovery VCT. Radioactivity concentrations were measured by PET at three to four time points up to 60 min after radioligand injection. Time integrals of activity (Stabin and Siegel, 2003) were entered into the OLINDA/

D.F. Wong et al. / NeuroImage 75 (2013) 282–290

285

EXM® software (Vanderbilt University, Nashville, TN, USA; Stabin et al., 2005) using the adult male model for both species. Autoradiographic studies in healthy human brain Tritiated RO5013853 was used for this study. Sections (10 μm) from a healthy human brain (48 year old female, University of Auckland) were cut in a cryostat and thaw-mounted on adhesion glass slides (Paul Marienfeld GmbH, Lauda-Königshofen, Germany). Brain sections (3 sections per region of interest) were incubated at room temperature for 10 min in Ringer buffer (NaCl 120 mM, KCl 5 mM, CaCl2 2 mM, MgCl2 1 mM, Tris–HCl 50 mM, pH 7.4), followed by 60 min in Ringer buffer containing 1 nM radioligand. Non-specific binding (NSB) was measured using an additional two sections incubated with Ringer buffer containing 1 nM radiotracer and 1 μM bitopertin (60 min at 37 °C). Sections were rinsed 2 × 5 min and 1 × 15 min in Ringer buffer and dipped three times in distilled water at 4 °C. Slide-mounted brain sections were dried and exposed to a Fuji Imaging Plate (BAS-TR 2025, Fujifilm, Dielsdorf, Switzerland) with a [ 3H]microscale (RPA-510, GE Healthcare, Glattbrugg, Switzerland) for five days. The imaging plate was scanned in a Fujifilm high-resolution plate scanner (BAS-5000, Bucher Biotec AG, Basel, Switzerland). The total amount of radioligand bound to the brain areas of interest (TB) was measured using the MCID™ image analysis program (version 7; InterFocus Imaging GmbH, Mering, Germany) and expressed as fmol of bound radioligand per mg of protein. The amount of radioligand specifically bound to the GlyT1 (SB) was calculated according to the formula SB = TB − NSB. NSB was very low and ranged from 0.9% to 4% of the TB measured. Results Safety [ 11C]RO5013853 was well tolerated and no clinically relevant changes in vital signs, electrocardiogram, or laboratory parameters were observed. Four out of ten subjects reported a total of six adverse events of mild intensity. Two of the adverse events were judged remotely related (diarrhea, abdominal pain), and four unrelated to [ 11C]RO5013853. The highest single mass dose administered of 2.83 μg RO5013853 was well below the maximal allowed dose of 20 μg and consistent with the definition of microdose. Whole-body radiation dosimetry (rodent and human) The mouse radiation dosimetry demonstrated that the liver had the highest uptake of any single organ, around 65%, and received a dose of approximately 0.05 mSv/MBq, as did the small intestine. The radiation effective dose was calculated to be 0.0047 mSv/MBq (0.0174 rem/mCi). The human dosimetry data fit well with one or two exponential functions. Most organs appeared to receive around 0.002–0.005 mSv/ MBq (0.074 to 0.019 rem/mCi). The liver appeared to receive the highest dose of 0.016 mSv/MBq (0.059 rem/mCi), and the gallbladder 0.01 mSv/MBq (0.037 rem/mCi). The effective dose was 0.0033 mSv/ MBq (0.012 rem/mCi), therefore approximately 1.4 fold lower than the estimated effective dose in mice. Time activity curves (TACs) Average SUVs over time from test-retest scans (two scans on each of five participants, all performed on the GE scanner) showed that the tracer rapidly enters the brain. The peak was reached at around 1 min, with an initial fast decline in SUV followed by a slower elimination phase in the areas of high GlyT1 binding

Fig. 1. Mean time activity curves (TACs) in standardized uptake values (SUVs) for representative regions of test-retest scans performed on GE Advance scanner (n = 10) in five healthy volunteers demonstrating initial rapid decline in SUV followed by slower elimination phase. Other cortical regions examined (cingulate, occipital, parietal, temporal, orbital frontal, and prefrontal cortices) had TACs (data not shown) similar to that of the superior frontal cortex.

(Fig. 1). Cerebellum, pons, and thalamus showed the highest and cortical areas the lowest SUVs. Distribution of GlyT1 in human brain (autoradiography) The regional distribution of RO5013853 binding sites was further investigated at higher resolution by autoradiography using tritiated RO5013853. High binding density was observed in the pons, the superior and inferior colliculi, and various thalamic nuclei (data not shown). In the cerebellum, a higher density of [3H]RO5013853 binding sites was detected in the granular layer and a lower density in the molecular layer and the white matter (Fig. 2A). Moderate binding site density was seen in the caudate nucleus, the dentate gyrus, the putamen, and the CA 1–3 regions of the hippocampus (Fig. 2B). Low binding was observed in all cortical regions investigated. Co-incubation of the radioligand with the specific GRI, bitopertin (RG1678), completely abolished the binding, confirming the specificity of RO5013853 for GlyT1 (data not shown). Radiotracer metabolism Fig. 3 shows the average metabolite ratios for all dynamic scans (two scans on each of 7 participants). At 30 and 90 min after [ 11C] RO5013853 injection, ~ 30% and ~ 50% of the radiotracer were metabolized, respectively. Both scans showed very similar plasma metabolite profiles, with b4% difference in the mean metabolite ratio throughout the entire duration of the scan. All metabolites were more polar than the parent as they eluted rapidly in the HPLC. Kinetic model for quantification of binding Fig. 4 shows a representative observed TAC from a single study for the superior frontal cortex and thalamus with fitted curves as predicted by the 1T3P and the 2T5P models. Note that the predicted curves fit the data points better for the 2T5P model than of the 1T3P model, particularly in the early phase of tracer kinetics (paired ttest for AICs, p b 0.001 for all ROIs; Table 2). Therefore, we used the 2T5P model for further data analysis. Parameter estimates Overall, the estimates of K1 and VT were consistent across both GE Advance and HRRT scanners. Estimates of K1 were fairly homogenous across all regions of interest (ROIs), with values between 0.02 and

286

D.F. Wong et al. / NeuroImage 75 (2013) 282–290

Fig. 2. Pseudocolor images of [3H]RO5013853 autoradiographs of human brain sections. Red indicates high binding site density whereas blue indicates low binding sites density (see calibration scale). Two brain regions are shown in this figure, the cerebellum (A) and the hippocampus (B). In the cerebellum, the highest density of binding sites is observed in the granular layer (gl), with lower densities observed in the white matter and the molecular layer. In the hippocampus, moderate densities of binding sites are observed in the dentate gyrus (DG) and in the CA1, CA2 and CA3 regions.

0.08 mL/min/mL (0.04 ± 0.01 [mean ± SD hereafter], n = 12 × 14) for data from both scanner types. VT ranged from 1.86 to 2.89 mL/mL in cerebellum, pons, and thalamus (pons being highest: 2.59 ± 0.23 mL/mL, n = 10; GE data), from 0.94 to 1.26 mL/mL in striatum (putamen and caudate; 1.12 ± 0.09 mL/mL, n = 2 × 10), and from 0.78 to 1.09 mL/mL in the cortical areas (0.90 ± 0.06 mL/mL, n = 7 × 10; Fig. 5). In the two subjects also scanned on the HRRT, VT ranged from 1.61 to 2.15 mL/mL in cerebellum, pons, and thalamus and was approximately 0.86 mL/mL in the cortical areas. BPND derived by the 2T5P model in cerebellum, pons, and thalamus was 1.26 ± 0.19, 1.93 ± 0.30, and 1.43 ± 0.22, respectively (n = 10; GE data). For comparison, BPND values in the two subjects also scanned on the HRRT were 0.85–0.99 in cerebellum, 1.44–1.47 in pons, and 1.29–1.45 in thalamus. The 2T5P GE values were overall comparable to the BPND derived by SRTM (1.02 ± 0.14 [cerebellum], 1.50 ± 0.18 [pons], and 1.47 ± 0.32 [thalamus], [n = 10]). In all other areas, BPND was below 0.30 for both methods used with high coefficient of variation and were thus not reliable.

(GE data) ranged from 14.6 ± 12.8% in the thalamus to 26.0 ± 20.7% in the caudate (n = 5). VT estimates from GE data showed a percentage difference over all ROIs of 4.3 ± 2.9% (n = 5 × 12), with ranges from 2.4 ± 1.8% in the cerebellum to 6.1 ± 2.7% in the prefrontal cortex (n = 5). The percentage difference for the BPND estimates of cerebellum, pons, and thalamus were 12.6 ± 12.3%, 7.9 ± 6.9%, 14.4 ± 11.3%, and 7.3 ± 7.3%, 6.0 ± 2.6%, 26.1 ± 13.3% for the 2T5P and SRTM models, respectively (n = 5).

Parametric images Consistent with the estimates of VT using kinetic modeling, voxelwise parametric mapping showed that the cerebellum, pons, and thalamus were the regions with the highest VT, while the cortical regions had the lowest VT (Fig. 6).

Test-retest reliability

Transporter specificity of the tracer

The parameters K1 and VT were used for test–retest analysis. The percentage difference between the two scans for the K1 estimates

Following 12 days of once daily dosing with 175 mg bitopertin, a drug candidate that binds selectively and with high affinity to GlyT1, tracer binding is almost completely blocked (GlyT1 occupancy by bitopertin: 92% in the thalamus, Fig. 7; and unpublished data).

Fig. 3. Metabolite ratio of [11C]RO5013853 (%) in plasma of healthy volunteers for all dynamic scans (n = 7 each, mean ± SD) as well as the average curve for all scans showing that about 30% and 50% of the radiotracer were metabolized after 30 and 90 min, respectively.

Fig. 4. Representative time activity curve (TAC) from a single dynamic PET scan for the superior frontal cortex and the thalamus and the fitted curves as predicted by the 1tissue 3-parameter (1T3P) and 2-tissue 5-parameter (2T5P) compartment model.

D.F. Wong et al. / NeuroImage 75 (2013) 282–290 Table 2 Mean (SD) AIC values by each ROI for the GE advance scanner (n = 10). Regions

Models 1T3P

Cerebellum Pons Thalamus Putamen Caudate CingulateCx OcciptalCx ParietalCx TemporalCx OrbitalFrontalCx PreFrontalCx SuperiorFrontalCx

− 223.88 − 214.64 − 179.37 − 187.24 − 181.50 − 196.50 − 203.96 − 213.38 − 222.21 − 217.07 − 214.99 − 209.77

2T5P (8.92) (8.03) (13.79) (9.99) (14.49) (10.69) (6.74) (8.31) (10.74) (9.05) (12.18) (7.97)

− 268.11 (25.51) − 239.29 (17.62) − 196.62 (21.08) − 198.89 (15.03) − 187.49 (13.24) − 215.77 (18.82) − 235.24 (16.25) − 253.18 (23.99) − 252.92 (23.06) − 244.18 (18.09) − 239.80 (24.41) − 235.23 (23.63)

Discussion Since the successful imaging of dopamine and D2/D3, serotonin 5HT2A, and mu opiate receptors (Frost et al., 1985; Wagner et al., 1983; Wong et al., 1984) and subsequent imaging of dopamine (Wong et al., 1993) and serotonin transporters (Szabo et al., 1995), the imaging of glutamate/glycine receptors has been an area of intense interest in basic and translational neuroscience. The imaging of this system including the GlyT1 is of great value for understanding normal and pathological receptor distribution and expression levels in neuropsychiatric illnesses. In addition, imaging is paramount in the development of putative novel drugs involving these targets, as target occupancy measurements are now considered an integral part of clinical drug development in neuroscience. In this paper, we document for the first time feasibility and reliability of the novel radiotracer [ 11C]RO5013853, a highly selective inhibitor of the GlyT1 in humans. A recent clinical trial showed an improvement in negative symptoms in patients with schizophrenia who were treated with the GRI, bitopertin (RG1678) for 8 weeks (Umbricht et al., 2010). Using this novel tracer, it was shown that low to medium target occupancy seems sufficient to achieve the strongest clinical effects, while high occupancy levels were associated with a lack of efficacy (Umbricht et al., 2011). This finding underscores the significant contribution PET studies can make to the understanding of a drug's mechanism of action. Microdose levels of [ 11C]RO5013853 were safe and well tolerated in this first in human study. [ 11 C]RO5013853 is primarily retained in the cerebellum, the pons, and the thalamus, with little uptake in

Fig. 5. Tracer distribution volume (VT, mean ± SD) for scans acquired on the GE advance scanner for each ROI, showing highest values for cerebellum, pons, and thalamus.

287

cortical regions. This is consistent with human autoradiography data presented here, and with the known distribution of the GlyT1 in non-human primate and pig brain PET studies (Borroni et al., 2013– this issue; Hamill et al., 2011; Passchier et al., 2010; Toyohara et al., 2011) and also with results from previous immunohistochemical, autoradiography, and tissue distribution studies in rodents and nonhuman primates (Borroni et al., 2013–this issue; Cubelos et al., 2005; Toyohara et al., 2011; Zafra et al., 1995; Zeng et al., 2008). Examination with the gold standard arterial input function demonstrated characteristics associated with many successful PET radiotracers, including reasonable brain uptake (VT was 2.24 ± 0.29 mL/ mL in target areas [means over cerebellum, pons, and thalamus, n = 3 × 10] versus 0.95 ± 0.11 mL/mL in others, n = 9 × 10) and a relatively high test/retest reliability (b10% difference between scans for VT, a parameter used to measure tissue tracer uptake and to derive tracer specific binding in tissue BPND). We determined that quantification with the 2T5P model was suitable for all kinetic analyses, which is consistent with prior baboon studies presented in this issue (Borroni et al., 2013–this issue). We subsequently found in a related receptor occupancy study that full range of dose occupancy up to near 100% could be achieved (unpublished data) and that a reference region approach is quite acceptable, with the superior frontal cortex having been identified as the reference region. Test-retest reliability of BPND derived from the SRTM was comparable to the standard compartmental model. This suggests that in future studies using this tracer, data can appropriately be quantified using the SRTM method, and thus no arterial sampling would be required. Characterization of the kinetics of another GlyT1 PET tracer, [ 11C] GSK931145, in dynamic PET with standard compartmental analysis was published recently (Gunn et al., 2011, Synapse). In this study, the regional distribution of tracer binding in humans (increased uptake in cerebellum, brain stem, and thalamus) was similar to that shown here. Furthermore, the authors reported that the [ 11C] GSK931145 cerebral K1 value was 0.025 ± 0.0097 mL/min/mL in human studies which was lower than the [ 11C]RO5013853 K1 value (0.037 ± 0.0076 mL/min/mL, n = 12 ROIs × 10 scans) obtained in our study. The direct comparison of VT and BPND estimates between [ 11 C]GSK931145 (Gunn et al., 2011, Synapse) and [ 11 C]RO5013853 from this study are listed in Supplemental Table 1. As compared to [ 11C]GSK931145, the [ 11C]RO5013853 tracer has higher VT with remarkably higher test-retest reliability. The BPND estimates of [ 11C] RO5013853 in cerebellum, pons, and thalamus are comparable to those in [ 11C]GSK931145 studies (see Supplemental Table 1). Thus initial brain uptake, binding, and test–retest reliability are better with [ 11 C]RO5013853. In a brief, preliminary report from a second GlyT1 PET tracer, [ 18F] CFpyPB, the highest binding was reported in the brain stem, followed by cerebellum, thalamus and midbrain, and white matter, and the lowest binding in the striatum and cortical gray matter, with VT ranging from 7 down to 2 mL/mL. Test–retest variability was noted as below 12% (n = 3, Sanabria-Bohorquez et al., 2009; Williams et al., 2008). In comparison, for [ 11C]RO5013853, while we observed a magnitude of binding that was overall lower (VT ranged from approximately 3 down to 2 mL/mL in cerebellum, pons, and thalamus, and less than 1 mL/mL in cortical areas) and slightly different ranking in regional biodistribution sites (i.e., thalamus higher than cerebellum), we obtained a substantially stronger VT test–retest reliability (mean percentage difference over all regions below 5%). Specificity of [ 11C]RO5013853 for GlyT1 was shown, as tracer binding is blocked by prior administration of a selective GRI (bitopertin, Fig. 7; and unpublished data). Additional favorable characteristics of this novel radiotracer include (i) the results from human radiation dosimetry which allow at least three injections of 1110 MBq (30 mCi) each to total 3330 MBq (90 mCi) under local institutional and FDA rules and

288

D.F. Wong et al. / NeuroImage 75 (2013) 282–290

Fig. 6. Transverse parametric images of tracer distribution volume (VT; [mL/mL], means, n = 10) generated by applying the RE-GP plots to the [11C]RO5013853 dynamic PET images acquired from GE Advance scanner, with the pons and the thalamus showing the highest VT, and the cortical regions, the lowest VT. The parametric images of VT were spatially normalized to the standard space (voxel size: 2 × 2 × 2 mm3) using SPM2, and the mean of 90-min dynamic PET images were used to determine the parameters of spatial normalization.

which indicate a lower human effective dose 0.00032 rem/MBq (0.012 rem/mCi) compared to the radiotracers [ 11C]GSK931145 0.00041 rem/MBq (0.015 rem/mCi; Murthy et al., 2008; Bullich et al., Online First™, 21.August 2010) and [ 18F]CFpyPB 0.0025 rem/ MBq (0.093 rem/mCi; Sanabria-Bohorquez et al., 2009), (ii) modest radiotracer metabolism, and (iii) a reproducible synthesis which is a key element to any PET study where timely availability and reliability of the tracer are critical. There are some limitations to the radiotracer, however, as the overall brain uptake is sufficient but modest. For [ 11C]RO5013853, this limitation can be overcome by injection of as much as 1110 MBq (30 mCi) per PET scan, which was considered safe given results from the radiation dosimetry studies. We also tested whether using the HRRT rather than the GE Advance scanner would benefit the imaging results from these studies. The HRRT scanner has improved resolution compared to the GE Advance (2.5 mm compared to ~5 mm full width at half maximum resolution) with comparable sensitivity; however, the HRRT is composed of LSO scintillator-crystals which emit radioactivity. In phantom studies, careful consideration of the LSO background activity in the field of view indicates that for any ROI in which concentration becomes around 370–555 Bq/mL (10–15 nCi/mL), background activity may potentially interfere with the measurements (it is possible to compensate for the resulting bias, but noise levels introduced

by background contamination will remain substantial). Thus, if the ROI has a low radioactive concentration (less than around 7400 Bq/ mL [200 nCi/mL]) early on in a 90 min [ 11C] PET study, the activity of the ROI could fall below the background threshold at the end of the study, and the observed TAC would not be accurate because of contamination with the background activity. Because of the modest brain uptake of [ 11C]RO5013853 (although good regional brain distribution, and occupancy [Fig. 7; unpublished data), we elected to continue studies with the GE Advance to avoid possible artifacts in our future occupancy studies. This does not exclude the HRRT for future use, however. Further studies may be advisable to determine if the HRRT could be useful to discriminate GlyT1 binding in small brain regions when an even higher radioactivity is injected. For the current application, i.e., GlyT1 occupancy, larger brain regions are of primary interest, and hence scanning at high resolution (HRRT) was not beneficial. Conclusions We found that [ 11C]RO5013853 is a suitable PET imaging agent for human studies of the GlyT1 transporter. [ 11C]RO5013853 readily penetrates the blood–brain barrier and binds to brain regions known to express GlyT1. Brain signal is modest; however, test-retest reliability for estimated key parameters is sufficiently high to afford reasonable precision in tracer binding determinations and is superior to other existing human GlyT1 PET tracers. Finally, evidence is presented that shows that radioligand binding can be effectively blocked by a highly specific GlyT1 inhibitor. [ 11C]RO5013853 can thus be used in further occupancy studies. Supplementary materials related to this article can be found online at doi:10.1016/j.neuroimage.2011.11.052. Acknowledgments

Fig. 7. Tracer distribution volume VT (mL/mL) axial images of a representative participant at baseline (middle) and following once daily dosing over 12 days with 175 mg of bitopertin at steady state (right) at the level of the thalamus demonstrating almost complete blockage of tracer binding. T1-weighted MRI (left) used for co-registering PET images to better delineate ROIs.

Special thanks to Arman Rahmin, Ph.D, the HRRT physicist in charge, and Maria Thomas, Ph.D, for scientific editorial assistance, as well as to Ron Goldwater, MD, Principal Investigator at PAREXEL, and Michael Stabin, Ph.D. (Vanderbilt University) for analysis of dosimetry data. This study was funded by F. Hoffmann-La Roche,

D.F. Wong et al. / NeuroImage 75 (2013) 282–290

contract to JHU. JHU faculty receive salary support through a number of sponsored research sources including Wong NIH career award K24 DA000412, and none receive direct funding from Roche except via sponsored JHU contracts. Support for third-party editorial assistance for this manuscript, furnished by Veronica Porkess, Ph.D. (Complete HealthVizion) and by archimed medical communication ag, was provided by F. Hoffmann-La Roche Ltd. References Akaike, H., 1976. An information criteria (AIC). Math. Sci. 14, 5–9. Alberati, D., Moreau, J.-L., Mory, R., Pinard, E., Wettstein, J.G., 2010. Pharmacological evaluation of a novel assay for detecting glycine transporter 1 inhibitors and their antipsychotic potential. Pharmacol. Biochem. Behav. 97, 185–191. Borroni, E., Zhou, Y., Ostrowitzki, S., Alberati, D., Kumar, A., Hainzl, D., Hartung, T., Hilton, J., Dannals, R.F., Wong, D.F., 2013. Pre-clinical characterization of [11C]R05013853 as a novel radiotracer for imaging of the glycine transporter type 1 by positron emission tomography. NeuroImage 75, 291–300 (this issue). Bullich, S., Slifstein, M., Passchier, J., Murthy, N.V., Kegeles, L.S., Kim, J.H., Xu, X., Gunn, R.N., Herance, R., Gispert, J.D., Gutierrez, A., Farre, M., Laruelle, M., Catafau, A.M., 2010. Biodistribution and radiation dosimetry of the glycine transporter-1 ligand 11C-GSK931145 determined from primate and human whole-body PET. Mol. Imaging Biol. (Aug 21. [Epub ahead of print]). Carson, R.E., Channing, M.A., Blasberg, R.G., Dunn, B.B., Cohen, R.M., Rice, K.C., Herscovitch, P., 1993. Comparison of bolus and infusion methods for receptor quantitation: application to [18F]cyclofoxy and positron emission tomography. J. Cereb. Blood Flow Metab. 13, 24–42. Carson, R.E., Breier, A., de Bartolomeis, A., Saunders, R.C., Su, T.P., Schmall, B., Der, M.G., Pickar, D., Eckelman, W.C., 1997. Quantification of amphetamine-induced changes in 11C raclopride binding with continuous infusion. J. Cereb. Blood Flow Metab. 17, 437–447. Cubelos, B., Gimenez, C., Zafra, F., 2005. Localization of the GLYT1 glycine transporter at glutamatergic synapses in the rat brain. Cereb. Cortex 15, 448–459. Cunningham, V.J., Matthews, J.C., Gunn, R.N., Rabiner, E.A., Gee, A.D., 2004. Identification and interpretation of microparameters in neuroreceptor compartmental models. Neuroimage 22, T13–T14. Food and Drug Administration, 2006. Guidance for Industry, Investigators, and Reviewers: Exploratory IND Studies. http://www.fda.gov/downloads/Drugs/ GuidanceComplianceRegulatoryInformation/Guidances/UCM078933.pdf. (Last accessed: 21st June, 2011). Friston, K., 2002. Statistical parametric mapping 2. http://www.fil.ion.ucl.ac.uk/spm/ software/spm2/. (Last accessed: 21st June, 2011). Frost, J.J., Wagner Jr., H.N., Dannals, R.F., Ravert, H.T., Links, J.M., Wilson, A.A., Burns, H.D., Wong, D.F., McPherson, R.W., Rosenbaum, A.E., 1985. Imaging opiate receptors in the human brain by positron tomography. J. Comput. Assist. Tomogr. 9, 231–236. Gjedde, A., 1981. High- and low-affinity transport of D-Glucose from blood to brain. J. Neurochem. 36, 1463–1471. Gunn, R.N., Gunn, S.R., Cunningham, V.J., 2001. Positron emission tomography compartmental models. J. Cereb. Blood Flow Metab. 21, 635–652. Gunn, R.N., Murthy, V., Catafau, A.M., Searle, G., Bullich, S., Slifstein, M., Ouellet, D., Zamuner, S., Herance, R., Salinas, C., Pardo-Lozano, R., Rabiner, E.A., Farre, M., Laruelle, M., 2011. Translational characterization of [(11) C]GSK931145, a PET ligand for the glycine transporter type 1 (GlyT-1). Synapse 65, 1319–1332 (Electronic publication ahead of print 2011 Aug 12). Hamill, T.G., Eng, W., Jennings, A., Lewis, R., Thomas, S., Wood, S., Street, L., Wisnoski, D., Wolkenberg, S., Lindsley, C., Sanabria-Bohorquez, S.M., Patel, S., Riffel, K., Ryan, C., Cook, J., Sur, C., Burns, H.D., Hargreaves, R., 2011. The synthesis and preclinical evaluation in rhesus monkey of [(1)F]MK-6577 and [(1)(1)C]CMPyPB glycine transporter 1 positron emission tomography radiotracers. Synapse 65, 261–270. Harrison, P.J., Weinberger, D.R., 2005. Schizophrenia genes, gene expression, and neuropathology: on the matter of their convergence. Mol. Psychiatry 10, 40–68. Hashimoto, K., 2011. Glycine transporter-1: a new potential therapeutic target for schizophrenia. Curr. Pharm. Des. 17, 112–120. Hilton, J., Yokoi, F., Dannals, R.F., Ravert, H.T., Szabo, Z., Wong, D.F., 2000. Columnswitching HPLC for the analysis of plasma in PET imaging studies. Nucl. Med. Biol. 27, 627–630. Horti, A.G., Fan, H., Kuwabara, H., Hilton, J., Ravert, H.T., Holt, D.P., Alexander, M., Kumar, A., Rahmim, A., Scheffel, U., Wong, D.F., Dannals, R.F., 2006. 11CJHU75528: a radiotracer for PET imaging of CB1 cannabinoid receptors. J. Nucl. Med. 47, 1689–1696. Huang, S.C., Barrio, J.R., Phelps, M.E., 1986. Neuroreceptor assay with positron emission tomography: equilibrium versus dynamic approaches. J. Cereb. Blood Flow Metab. 6, 515–521. Innis, R.B., Cunningham, V.J., Delforge, J., Fujita, M., Gjedde, A., Gunn, R.N., Holden, J., Houle, S., Huang, S.C., Ichise, M., Iida, H., Ito, H., Kimura, Y., Koeppe, R.A., Knudsen, G.M., Knuuti, J., Lammertsma, A.A., Laruelle, M., Logan, J., Maguire, R.P., Mintun, M.A., Morris, E.D., Parsey, R., Price, J.C., Slifstein, M., Sossi, V., Suhara, T., Votaw, J.R., Wong, D.F., Carson, R.E., 2007. Consensus nomenclature for in vivo imaging of reversibly binding radioligands. J. Cereb. Blood Flow Metab. 27, 1533–1539. Koeppe, R.A., Holthoff, V.A., Frey, K.A., Kilbourn, M.R., Kuhl, D.E., 1991. Compartmental analysis of [11C]flumazenil kinetics for the estimation of ligand transport rate and receptor distribution using positron emission tomography. J. Cereb. Blood Flow Metab. 11, 735–744.

289

Krystal, J.H., Karper, L.P., Seibyl, J.P., Freeman, G.K., Delaney, R., Bremner, J.D., Heninger, G.R., Bowers Jr., M.B., Charney, D.S., 1994. Subanesthetic effects of the noncompetitive NMDA antagonist, ketamine, in humans. Psychotomimetic, perceptual, cognitive, and neuroendocrine responses. Arch. Gen. Psychiatry 51, 199–214. Lammertsma, A.A., Hume, S.P., 1996. Simplified reference tissue model for PET receptor studies. Neuroimage 4, 153–158. Millan, M.J., 2005. N-methyl-D-aspartate receptors as a target for improved antipsychotic agents: novel insights and clinical perspectives. Psychopharmacology 179, 30–53. Murthy, N.V., Passchier, J., Gunn, R.N., Searle, G.E., Bullich, S., Suarez, M., Herance, R., Farre, M., Herdon, H., Porter, R., Sutherland, S., Fagg, R., Neve, M., Slifstein, M., Laruelle, M., Catafaui, A.M., 2008. [11C]GSK931145: a new pet ligand for glycine transporter 1. Neuroimage 41, T21. Passchier, J., Gentile, G., Porter, R., Herdon, H., Salinas, C., Jakobsen, S., Audrain, H., Laruelle, M., Gunn, R.N., 2010. Identification and evaluation of [11C]GSK931145 as a novel ligand for imaging the type 1 glycine transporter with positron emission tomography. Synapse 64, 542–549. Patlak, C.S., Blasberg, R.G., Fenstermacher, J.D., 1983. Graphical evaluation of blood-tobrain transfer constants from multiple-time uptake data. J. Cereb. Blood Flow Metab. 3, 1–7. Patlak, C.S., Blasberg, R.G., 1985. Graphical evaluation of blood-to-brain transfer constants from multiple-time uptake data. Generalizations. J. Cereb. Blood Flow Metab. 5, 584–590. Pinard, E., Alanine, A., Alberati, D., Bender, M., Borroni, E., Bourdeaux, P., Brom, V., Burner, S., Fischer, H., Hainzl, D., Halm, R., Hauser, N., Jolidon, S., Lengyel, J., Marty, H.P., Meyer, T., Moreau, J.L., Mory, R., Narquizian, R., Nettekoven, M., Norcross, R.D., Puellmann, B., Schmid, P., Schmitt, S., Stalder, H., Wermuth, R., Wettstein, J.G., Zimmerli, D., 2010. Selective GlyT1 inhibitors: discovery of [4-(3-fluoro-5-trifluoromethylpyridin-2-yl)piperazin-1-yl][5-methanesulfonyl-2-(( S)-2,2,2-trifluoro-1methylethoxy)phenyl]methanone (RG1678), a promising novel medicine to treat schizophrenia. J. Med. Chem. 53, 4603–4614. Pinard, E., Burner, S., Cueni, P., Hartung, T., Norcross, R.D., Schmid, P., Waldmeier, P., Zielinski, G., Ravert, H.T., Holt, D.P., Dannals, R.F., 2011. Radiosynthesis of [5[11C]methanesulfonyl-2-((S)-2,2,2-trifluoro-1-methyl-ethoxy)-phenyl]-[5-(tetrahydro-pyran-4-yl)-1,3-dihydro-isoindol-2-yl]-methanone ([11C]RO5013853), a novel PET tracer for the glycine transporter type I (GlyT1). J. Labeled Comp. Rad. 54, 702–707. Rahmim, A., Lenox, M., Reader, A.J., Michel, C., Burbar, Z., Ruth, T.J., Sossi, V., 2004. Statistical list-mode image reconstruction for the high resolution research tomograph. Phys. Med. Biol. 49, 4239–4258. Sanabria-Bohorquez, S., Van Laere, K., Hamill, T., Koole, M., Bormans, G., De Lepeleire, I., Reynders, T., Pickett, C., Van Hecken, A., Burns, D., 2009. Evaluation of the novel glycine transporter 1 (GlyT1) tracer [18F]CFpyPB: Dosimetry and brain quantification in human. J. Nucl. Med. 50, S1211. Sanger, D.J., 2004. The search for novel antipsychotics: pharmacological and molecular targets. Expert Opin. Ther. Targets 8, 631–641. Schmidt, K.C., Turkheimer, F.E., 2002. Kinetic modeling in positron emission tomography. Q. J. Nucl. Med. 46, 70–85. Stabin, M.G., Siegel, J.A., 2003. Physical models and dose factors for use in internal dose assessment. Health Phys. 85, 294–310. Stabin, M.G., Sparks, R.B., Crowe, E., 2005. OLINDA/EXM: the second-generation personal computer software for internal dose assessment in nuclear medicine. J. Nucl. Med. 46, 1023–1027. Stone, J.M., 2009. Imaging the glutamate system in humans: relevance to drug discovery for schizophrenia. Curr. Pharm. Des. 15, 2594–2602. Szabo, Z., Kao, P.F., Scheffel, U., Suehiro, M., Mathews, W.B., Ravert, H.T., Musachio, J.L., Marenco, S., Kim, S.E., Ricaurte, G.A., 1995. Positron emission tomography imaging of serotonin transporters in the human brain using [11C](+)McN5652. Synapse 20, 37–43. Toyohara, J., Ishiwata, K., Sakata, M., Wu, J., Nishiyama, S., Tsukada, H., Hashimoto, K., 2011. In vivo evaluation of carbon-11-labelled non-sarcosine-based glycine transporter 1 inhibitors in mice and conscious monkeys. Nucl. Med. Biol. 38, 517–527. Turkheimer, F.E., Hinz, R., Cunningham, V.J., 2003. On the undecidability among kinetic models: from model selection to model averaging. J. Cereb. Blood Flow Metab. 23, 490–498. Umbricht, D., Yoo, K., Youssef, E., Dorflinger, E., Martin-Facklam, M., Bausch, A., Arrowsmith, R., Alberati, D., Marder, S., Santarelli, L., 2010. Glycine transporter type 1 (GLYT1) inhibitor RG1678: positive results of the proof-of-concept study for the treatment of negative symptoms in schizophrenia. Neuropsychopharmacology 35, S320–S321. Umbricht, D., Martin-Facklam, M., Pizzagalli, E., Youssef, K., Yoo, K., Dorflinger, E., Bausch, A., Arrowsmith, R., Alberati, D., Santarelli, L., 2011. Glycine transporter type 1 (GlyT1) inhibitor RG1678: results of the proof-of-concept study for the treatment of negative symptoms in schizophrenia. Schizophr. Bull. 37, S324. Wagner Jr., H.N., Burns, H.D., Dannals, R.F., Wong, D.F., Langstrom, B., Duelfer, T., Frost, J.J., Ravert, H.T., Links, J.M., Rosenbloom, S.B., Lukas, S.E., Kramer, A.V., Kuhar, M.J., 1983. Imaging dopamine receptors in the human brain by positron tomography. Science 221, 1264–1266. Wilcox, K.M., Zhou, Y., Wong, D.F., Alexander, M., Rahmim, A., Hilton, J., Weed, M.R., 2008. Blood levels and DA transporter occupancy of orally administered methylphenidate in juvenile rhesus monkeys measured by high resolution PET. Synapse 62, 950–952. Williams, D.L., Jennings, A., Lewis, R., Street, L., Paten, S., Wisnoski, D.D., Wolkenberg, S.E., Lindsey, C.W., Zhao, Z., Krause, S.M., Ryan, C., Michener, M.S., Zeng, Z., Miller, P.A., Riffel, K., O'Malley, S.S., Sanabria, S., Eng, W., Gibson, R.E., Sur, C., Cook, J.J., Har-

290

D.F. Wong et al. / NeuroImage 75 (2013) 282–290

greaves, R., Burns, D., Hamill, T.G., 2008. A novel glycine transporter-1 PET tracer: from discovery to clinic. Abstr. Neurosci. 885, 15. Wong, D.F., Wagner Jr., H.N., Dannals, R.F., Links, J.M., Frost, J.J., Ravert, H.T., Wilson, A.A., Rosenbaum, A.E., Gjedde, A., Douglass, K.H., 1984. Effects of age on dopamine and serotonin receptors measured by positron tomography in the living human brain. Science 226, 1393–1396. Wong, D.F., Yung, B., Dannals, R.F., Shaya, E.K., Ravert, H.T., Chen, C.A., Chan, B., Folio, T., Scheffel, U., Ricaurte, G.A., 1993. In vivo imaging of baboon and human dopamine transporters by positron emission tomography using [11C]WIN 35,428. Synapse 15, 130–142. Wong, D.F., Kuwabara, H., Horti, A.G., Raymont, V., Brasic, J., Guevara, M., Ye, W., Dannals, R.F., Ravert, H.T., Nandi, A., Rahmim, A., Ming, J.E., Grachev, I., Roy, C., Cascella, N., 2010. Quantification of cerebral cannabinoid receptors subtype 1 (CB1) in healthy subjects and schizophrenia by the novel PET radioligand [11C]OMAR. Neuroimage 52, 1505–1513. Zafra, F., Gomeza, J., Olivares, L., Aragon, C., Gimenez, C., 1995. Regional distribution and developmental variation of the glycine transporters GLYT1 and GLYT2 in the rat CNS. Eur. J. Neurosci. 7, 1342–1352. Zeng, Z., O'Brien, J.A., Lemaire, W., O'Malley, S.S., Miller, P.J., Zhao, Z., Wallace, M.A., Raab, C., Lindsley, C.W., Sur, C., Williams Jr., D.L., 2008. A novel radioligand for glycine transporter 1: characterization and use in autoradiographic and in vivo brain occupancy studies. Nucl. Med. Biol. 35, 315–325.

Zhou, Y., Endres, C.J., Brašic, J.R., Huang, S.C., Wong, D.F., 2003. Linear regression with spatial constraint to generate parametric images of ligand-receptor dynamic PET studies with a simplified reference tissue model. Neuroimage 18, 975–989. Zhou, Y., Resnick, S.M., Ye, W., Fan, H., Holt, D.P., Klunk, W.E., Mathis, C.A., Dannals, R., Wong, D.F., 2007. Using a reference tissue model with spatial constraint to quantify [11C]Pittsburgh compound B PET for early diagnosis of Alzheimer's disease. Neuroimage 36, 298–312. Zhou, Y., Ye, W., Brašic, J.R., Crabb, A.H., Hilton, J., Wong, D.F., 2009. A consistent and efficient graphical analysis method to improve the quantification of reversible tracer binding in radioligand receptor dynamic PET studies. Neuroimage 44, 661–670. Zhou, Y., Ye, W., Brašic, J.R., Wong, D.F., 2010. Multi-graphical analysis of dynamic PET. Neuroimage 49, 2947–2957.

DATA QUOTED AS UNPUBLISHED: Martin-Facklam, M., Pizzagalli, F., Zhou, Y., Ostrowitzki, S., Raymont, V., Brašić, J.R., Parkar, N., Umbricht, D., Dannals, R.F., Goldwater, R., Wong, D.F., submitted for publication. Glycine transporter Type 1 (GlyT1) occupancy by RG1678: A positron emission tomography study in healthy volunteers.