Correction to: Molecular Psychiatry advance online publication, 16 May 2017; doi: 10.1038/mp.2017.107 The authors’ affiliations were presented incorrectly. The correct affiliations appear below: JJ Weinstein1,2, E van de Giessen3, RJ Rosengard4, X Xu4, N Ojeil4, G Brucato2,4, RB Gil1, LS Kegeles2,4,M Laruelle2, M Slifstein1, A Abi-Dargham1

UNLABELLED:The synaptic vesicle glycoprotein 2A (SV2A) is found in secretory vesicles in neurons and endocrine cells. PET with a selective SV2A radiotracer will allow characterization of drugs that modulate SV2A (e.g., antiepileptic drugs) and potentially could be a biomarker of synaptic density (e.g., in neurodegenerative disorders). Here we describe the synthesis and characterization of the SV2A PET radiotracer (11)C-UCB-J ((R)-1-((3-((11)C-methyl-(11)C)pyridin-4-yl)methyl)-4-(3,4,5-trifluorophenyl)pyrrolidin-2-one) in nonhuman primates, including whole-body biodistribution. METHODS:(11)C-UCB-J was prepared by C-(11)C-methylation of the 3-pyridyl trifluoroborate precursor with (11)C-methyl iodide via the Suzuki-Miyaura cross-coupling method. Rhesus macaques underwent multiple scans including coinjection with unlabeled UCB-J (17, 50, and 150 μg/kg) or preblocking with the antiepileptic drug levetiracetam at 10 and 30 mg/kg. Scans were acquired for 2 h with arterial sampling and metabolite analysis to measure the input function. Regional volume of distribution (VT) was estimated using the 1-tissue-compartment model. Target occupancy was assessed using the occupancy plot; the dissociation constant (Kd) was determined by fitting self-blocking occupancies to a 1-site model, and the maximum number of receptor binding sites (Bmax) values were derived from baseline VT and from the estimated Kd and the nondisplaceable distribution volume (VND). RESULTS:(11)C-UCB-J was synthesized with greater than 98% purity. (11)C-UCB-J exhibited high free fraction (0.46 ± 0.02) and metabolized at a moderate rate (39% ± 5% and 24% ± 3% parent remaining at 30 and 90 min) in plasma. In the monkey brain, (11)C-UCB-J displayed high uptake and fast kinetics. VT was high (∼25-55 mL/cm(3)) in all gray matter regions, consistent with the ubiquitous expression of SV2A. Preblocking with 10 and 30 mg/kg of levetiracetam resulted in approximately 60% and 90% occupancy, respectively. Analysis of the self-blocking scans yielded a Kd estimate of 3.4 nM and Bmax of 125-350 nM, in good agreement with the in vitro inhibition constant (Ki) of 6.3 nM and regional Bmax in humans. Whole-body biodistribution revealed that the liver and the brain are the dose-limiting organs for males and females, respectively. CONCLUSION:(11)C-UCB-J exhibited excellent characteristics as an SV2A PET radiotracer in nonhuman primates. The radiotracer is currently undergoing first-in-human evaluation.

Antagonism of 5-hydroxytrypamine-6 (5HT6) receptors is associated with procognitive effects in preclinical species, suggesting a therapeutic potential for this mechanism in Alzheimer disease (AD) and other cognitive diseases. In a phase 2 dose study, SB742457, a novel 5HT6 antagonist, showed increasing procognitive effects in patients with AD as the dose increased, with a procognitive signal in AD patients at a dose of 35 mg/d superior to the other doses tested (5 and 15 mg/d). Methods: In this article, we describe the quantification and pharmacologic selectivity of a new 5HT6 PET ligand (11C-GSK215083) in healthy volunteers and its use to measure occupancies achieved at various doses of SB742457. Results: Kinetic analysis of 11C-GSK215083 uptake in the human brain demonstrated the multilinear model, MA2, to represent the method of choice when a blood input was available and the full tissue reference method when no input was available. Pharmacologic dissection of the in vivo 11C-GSK215083–specific binding showed the ligand bound mostly the 5HT6 in the striatum (blocked by SB742457 but not by the selective 5-hydroxytryptamine-2A (5HT2A) antagonist ketanserin) and the 5HT2A in the frontal cortex (blocked by both ketanserin and SB742457). Repeated administration of SB742457 (3, 15, and 35 mg/d) saturated the 5HT6 receptors at all doses. In the cortex, 5HT2A receptor occupancy was 24% ± 6% (3 mg/d), 35% ± 4% (15 mg/d), and 58% ± 19% (35 mg/d; mean ± SD), suggesting a progressive engagement of 5HT2A as the dose increased. Conclusion: Collectively, these data support the use of 11C-GSK215083 as a 5HT6 clinical imaging tool and suggest that blocking both the 5HT6 and the 5HT2A receptors may be required for the optimal therapeutic action of SB742457 in AD.

PET with selective adenosine 2A receptor (A2A) radiotracers can be used to study a variety of neurodegenerative and neuropsychiatric disorders in vivo and to support drug-discovery studies targeting A2A. The aim of this study was to describe the first in vivo evaluation of 18F-MNI-444, a novel PET radiotracer for imaging A2A, in healthy human subjects. Methods: Ten healthy human volunteers were enrolled in this study; 6 completed the brain PET studies and 4 participated in the whole-body PET studies. Arterial blood was collected for invasive kinetic modeling of the brain PET data. Noninvasive methods of data quantification were also explored. Test–retest reproducibility was evaluated in 5 subjects. Radiotracer distribution and dosimetry was determined using serial whole-body PET images acquired over 6 h post-radiotracer injection. Urine samples were collected to calculate urinary excretion. Results: After intravenous bolus injection, 18F-MNI-444 rapidly entered the brain and displayed a distribution consistent with known A2A densities in the brain. Binding potentials ranging from 2.6 to 4.9 were measured in A2A-rich regions, with an average test–retest variability of less than 10%. The estimated whole-body radiation effective dose was approximately 0.023 mSv/MBq. Conclusion: 18F-MNI-444 is a useful PET radiotracer for imaging A2A in the human brain. The superior in vivo brain kinetic properties of 18F-MNI-444, compared with previously developed A2A radiotracers, provide the opportunity to foster global use of in vivo A2A PET imaging in neuroscience research.

1792 Objectives SV2A is a 90-kDa protein widely distributed in the cerebral cortex and the target of action for the antiepileptic drug levetiracetam (LEV). Although abundant evidence pointed to the importance of SV2A in epilepsy and other brain diseases, the cellular mechanisms of these diseases are largely unknown due to the lack of suitable in vivo probes. 18F-UCB-H is a potent SV2A ligand and a previous study in rodents indicated its potential as a PET radiotracer. Here we report a simplified synthesis of 18F-UCB-H and its evaluation in rhesus monkeys. Methods 18F-UCB-H was prepared by isotopic exchange using 19F-UCB-H as the starting material. Imaging experiments were carried out on the Focus 220 PET scanner with generation of arterial input function and metabolite analysis by HPLC. Brain regional time-activity curves (TACs) were analyzed by one-tissue (1T) or 2-tissue (2T) compartmental models, as well as the multilinear analysis (MA1) method to derive binding parameters. Results 18F-UCB-H was synthesized in 45 ± 13% radiochemical yield and specific activity of 0.44 ± 0.24 Ci/umol at the end of synthesis (n = 4). In rhesus monkeys, the metabolism of 18F-UCB-H was moderate, with ~ 40% of parent remaining at 30 min after tracer injection. In the brain, 18F-UCB-H displayed fast kinetics (regional activity peak times frontal cortex (FNT) > striatum (STR)> cerebellum (CER) ~ thalamus (THA) > pons (PON) ~ brainstem (BST). Both 1T and MA1 produced reliable estimates of regional distribution volume (VT). MA1 VT values were 16.5, 16.0, 15.2, 14.6, 12.9, 12.4, 11.0 and 10.8 for CIN, OCC, FNT, STR, CER, THA, PON and BST, respectively. Pretreatment with LEV (10 mg/kg, iv, n=2) resulted in 79% occupancy, indicating binding specificity of the tracer. Conclusions 18F-UCB-H is a specific tracer for PET imaging of SV2A in non-human primates.

Background: Atomoxetine (ATX), a drug for treatment of depression and ADHD, has a high affinity for the norepinephrine transporter (NET); however, our previous study showed it had a blocking effect similar to fluoxetine on binding of 11 a selective serotonin transporter (SERT) ligand. Whether the therapeutic effects of ATX are due to inhibition of either or both transporters is not known. Here we report our comparative PET imaging studies with [11C]lvIRB (a NET ligand) and [11C]AFM (a SERT ligand) to evaluate in vivo IC50 values of ATX in monkeys. Methods: Rhesus monkeys were scanned up to four times with each tracer with up to four doses of ATX ATX or saline (placebo) infusion began 2 h before each PET scan, lasting until the end of the 2-h scan. The final infusion rates were 0.01-0.12 mg/kg/h and 0.045-1.054 mg/kg/h for the NET and SERT studies, respectively. AD(plasma levels and metabolite-corrected arterial input functions were measured. Distribution volumes (VT) and IC50 values were estimated.Results: ATX displayed dose-dependent occupancy on both NET and SERT, with a higher occupancy on NET: IC50 of 31 10 and 99 21 ng/mL plasma for NET and SERT, respectively. At a clinically relevant dose (1.0-1.8 mg/kg, approx. 300-600 ng/mL plasma), ATX would occupy >90% of NET and >85% of SERT. This extrapolation assumes comparable free fraction of AIX in humans and non-human primates.Conclusion: Our data suggests that AIX at clinically relevant doses greatly occupies both NET and SERT. Thus, therapeutic modes of ATX action for treatment of depression and ADHD may be more complex than selective blockade of NET. (C) 2013 Published by Elsevier Inc.

Background and PurposeThis study aimed to investigate the relationship between the plasma concentration (PK) of the novel histamine H3 receptor antagonist, GSK239512, and the brain occupancy of H3 receptors (RO) in healthy human volunteers.Experimental ApproachPET scans were obtained after i.v. administration of the H3‐specific radioligand [11C]GSK189254. Each subject was scanned before and after single oral doses of GSK239512, at 4 and 24 h after dose. PET data were analysed by compartmental analysis, and regional RO estimates were obtained by graphical analysis of changes in the total volumes of distribution of the radioligand, followed by a correction for occupancy by the high affinity radioligand. The PK/RO relationship was analysed by a population‐modelling approach, using the average PK of GSK239512 during each scan.Key ResultsFollowing administration of GSK239512, there was a reduction in the brain uptake of [11C]GSK189254 in all regions, including cerebellum. RO at 4 h was higher than at 24 h, and the PK/RO model estimated a PK associated with 50% of RO of 0.0068 ng·mL−1. This corresponds to a free concentration of 4.50 × 10−12 M (pK = 11.3).Conclusions and ImplicationsThe affinity of GSK239512 for brain H3 receptors in humans in vivo is much higher than that expected from studies in vitro, and higher than that observed in PET studies in pigs. The study illustrates the utility of carrying out PET studies in humans early in drug development, providing accurate quantification of GSK239512 RO in vivo as a function of time and dose.

356 Objectives Adenosine 2a (A2a) receptors co-localize with dopamine D2 receptors in basal ganglia. A2a receptor antagonists potentiate D2 response to dopamine and dopamine agonists, leading to hypotheses that A2a antagonists could be useful for treatment of neurodegenerative movement disorders such as Parkinson’s and Huntington’s disease. Existing A2a PET tracers are C-11 labeled and not optimal for imaging in humans. This study is the initial evaluation of [18F]MNI-444, a novel F-18 PET imaging ligand targeting A2a receptors, in healthy volunteers (HV). Methods [18F]MNI-444 PET brain imaging (test/retest) was completed in 4 male HV (mean age 31 yrs). A dose of 10 mCi [18F]MNI-444 was injected followed by 210 min PET imaging that included arterial blood sampling. Also, one whole-body PET scan was performed in a female HV over 6 hours. Brain PET data were modeled with Logan graphical analysis to estimate distribution volume VT. Binding potential BPND was derived using cerebellum as reference region. BPND was also estimated using SRTM and non-invasive Logan. For the whole-body study, radiation absorbed doses and effective dose (ED) were estimated with OLINDA/EXM 1.0. Results [18F]MNI-444 displayed regional brain uptakes in accordance with expected A2a receptors distribution: highest in striatum and lowest in cerebellum, with BPND of ~4.0-5.0 in the putamen and globus pallidus. Reproducibility was 20-25% for VT and Conclusions Preliminary data suggest [18F]MNI-444 is a promising PET radiotracer for imaging A2a receptors in human that can be quantified non-invasively. Additional test-retest and whole body imaging are underway, and full tracer characterization will be presented.

UNLABELLED:Motor symptoms in Parkinson disease (PD) are caused by a loss of dopamine input from the substantia nigra to the striatum. Blockade of adenosine 2A (A(2A)) receptors facilitates dopamine D(2) receptor function. In phase 2 clinical trials, A(2A) antagonists (istradefylline, preladenant, and tozadenant) improved motor function in PD. We developed a new A(2A) PET radiotracer, (18)F-MNI-444, and used it to investigate the relationship between plasma levels and A(2A) occupancy by preladenant and tozadenant in nonhuman primates (NHP). METHODS:A series of 20 PET experiments was conducted in 5 adult rhesus macaques. PET data were analyzed with both plasma-input (Logan graphical analysis) and reference-region-based (simplified reference tissue model and noninvasive Logan graphical analysis) methods. Whole-body PET images were acquired for radiation dosimetry estimates. Human pharmacokinetic parameters for tozadenant and preladenant were used to predict A(2A) occupancy in humans, based on median effective concentration (EC(50)) values estimated from the NHP PET measurements. RESULTS:(18)F-MNI-444 regional uptake was consistent with A(2A) receptor distribution in the brain. Selectivity was demonstrated by dose-dependent blocking by tozadenant and preladenant. The specific-to-nonspecific ratio was superior to that of other A(2A) PET radiotracers. Pharmacokinetic modeling predicted that tozadenant and preladenant may have different profiles of A(2A) receptor occupancy in humans. CONCLUSION:(18)F-MNI-444 appears to be a better PET radiotracer for A(2A) imaging than currently available radiotracers. Assuming that EC(50) in humans is similar to that in NHP, it appears that tozadenant will provide a more sustained A(2A) receptor occupancy than preladenant in humans at clinically tested doses.

Schizophrenia might be considered a neurodevelopmental disease. However, the fundamental process(es) associated with this disease remain(s) uncertain. Many lines of evidence suggest that schizophrenia is associated with excessive stimulation of dopamine D2 receptors in the associative striatum, with a lack of stimulation of dopamine D1 receptors in prefrontal cortex, and with modifications in prefrontal neuronal connectivity involving glutamate transmission at N-methyl aspartate (NMDA) receptors. This article, whilst briefly discussing the current knowledge of the disease, mainly concentrates on the NMDA hypofunction hypothesis. However, there are also potential consequences for a Dopamine imbalance on NMDA function. Thus, it is proposed that schizophrenia has a complex aetiology associated with strongly interconnected aberrations of dopamine and glutamate transmission.

GSK1144814 is a potent, insurmountable antagonist at human NK1 and NK3 receptors. Understanding the relationship between plasma pharmacokinetics and receptor occupancy in the human brain, was crucial for dose selection in future clinical studies. GSK1144814 occupancy data were acquired in parallel with the first-time-in-human safety and tolerability study. [11C]GR-205171 a selective NK1 receptor PET ligand was used to estimate NK1 occupancy at several time-points following single dose administration of GSK1144814. The time-plasma concentration–occupancy relationship post-single dose administration was assessed, and used to predict the plasma concentration–occupancy relationship following repeat dose administration. Repeat dose predictions were tested in a subsequent cohort of subjects examined following approximately 7 and 14 days dosing with GSK1144814. GSK1144814 was shown to demonstrate a dose-dependent occupancy of the NK1 receptor with an estimated in vivo EC50 ~0.9 ng/mL in the human brain. A direct relationship was seen between the GSK1144814 plasma concentration and its occupancy of the brain NK1 receptor, indicating that in future clinical trials the occupancy of brain receptors can be accurately inferred from the measured plasma concentration. Our data provided support for the further progression of this compound and have optimised the likely therapeutic dose range.

3,4-Methylenedioxymethamphetamine (MDMA), the main psychoactive component of the recreational drug ecstasy, is a potent serotonin (5-HT) releaser. In animals, MDMA induces 5-HT depletion and toxicity in 5-HT neurons. The aim of this study was to investigate both presynaptic (5-HT transporter, SERT) and postsynaptic (5-HT2A receptor) markers of 5-HT transmission in recently abstinent chronic MDMA users compared with matched healthy controls. We hypothesized that MDMA use is associated with lower SERT density and concomitant upregulation of 5-HT2A receptors. Positron emission tomography studies using the SERT ligand [11C]DASB and the 5-HT2A receptor ligand [11C]MDL 100907 were evaluated in 13 current and recently detoxified MDMA users and 13 matched healthy controls. MDMA users reported a mean duration of ecstasy use of 8 years, regular exposure, and at least 2 weeks of abstinence before the scans. SERT and 5-HT2A receptor availability (binding potential, BPND) were analyzed with a two-tissue compartment model with arterial input function. Current recreational MDMA use was significantly associated with lower SERT BPND and higher 5-HT2A receptor BPND in cortical, but not subcortical regions. Decreased SERT BPND was regionally associated with upregulated 5-HT2A receptor BPND. In light of the animal literature, the most parsimonious interpretation is that repeated exposure to MDMA in humans, even in moderate amounts, leads to damage in 5-HT neuron terminals innervating the cortex. Alterations in mood, cognition, and impulse control associated with these changes might contribute to sustain MDMA use. The reversibility of these changes upon abstinence remains to be firmly established.

The passage of drugs in and out of the brain is controlled by the blood-brain barrier (BBB), typically, using either passive diffusion across a concentration gradient or active transport via a protein carrier. In-vitro and preclinical measurements of BBB penetration do not always accurately predict the in-vivo situation in humans. Thus, the ability to assay the concentration of novel drug candidates in the human brain in vivo provides valuable information for de-risking of candidate molecules early in drug development. Here, positron emission tomography (PET) measurements are combined with in-vitro equilibrium dialysis assays to enable assessment of transport and estimation of the free brain concentration in vivo. The PET and equilibrium dialysis data were obtained for 36 compounds in the pig. Predicted P-glycoprotein (P-gp) status of the compounds was consistent with the PET/equilibrium dialysis results. In particular, Loperamide, a well-known P-gp substrate, exhibited a significant concentration gradient consistent with active efflux and after inhibition of the P-gp process the gradient was removed. The ability to measure the free brain concentration and assess transport of novel compounds in the human brain with combined PET and equilibrium dialysis assays can be a useful tool in central nervous system (CNS) drug development.

The D(2)/D(3) agonist radioligand [(11)C]-(+)-PHNO is currently the most suitable D(3) imaging agent available, despite its limited selectivity for the D(3) over the D(2). Given the collocation of D(2) and D(3) receptors, and generally higher densities of D(2), the separation of D(2) and D(3) information from [(11)C]-(+)-PHNO PET data are somewhat complex. This complexity is compounded by recent data suggesting that [(11)C]-(+)-PHNO PET scans might be routinely performed in non-tracer conditions (with respect to D(3) receptors), and that the cerebellum (used as a reference region) might manifest some displaceable binding signal. Here we present the modelling and analysis of data from two human studies which employed an adequate dose range of selective D(3) antagonists (GSK598809 and GSK618334) to interrogate the [(11)C]-(+)-PHNO PET signal. Models describing the changes observed in the PET volume of distribution (V(T)) and binding potential (BP(ND)) were used to identify and quantify a [(11)C]-(+)-PHNO mass dose effect at the D(3), and displaceable signal in the cerebellum, as well as providing refined estimates of regional D(3) fractions of [(11)C]-(+)-PHNO BP(ND). The dose of (+)-PHNO required to occupy half of the available D(3) receptors (ED(50)(PHNO,D3)) was estimated as 40ng/kg, and the cerebellum BP(ND) was estimated as 0.40. These findings confirm that [(11)C]-(+)-PHNO human PET studies are in fact routinely performed under non-tracer conditions. This suggests that (+)-PHNO injection masses should be minimised and tightly controlled in order to mitigate the mass dose effect. The specific binding detected in the cerebellum was modest but could have a significant effect, for example on estimates of D(3) potency in drug occupancy studies. A range of methods for the analysis of future [(11)C]-(+)-PHNO data, incorporating models for the effects quantified here, were developed and evaluated. The comparisons and conclusions drawn from these can inform the design and analysis of future PET studies with [(11)C]-(+)-PHNO.

D1 receptors are the main mediators of dopamine transmission in the cortex and subserve cognitive functions that are affected in patients with schizophrenia. Prior imaging studies have suggested abnormalities in the expression of these receptors in schizophrenia, but no conclusive picture has emerged yet. One source of discrepancy may have been prior antipsychotic exposure. We used positron emission tomography (PET) and a D1 radiotracer, [11C]NNC112, in drug naïve (DN, n = 12) and drug free (DF, n = 13) patients with schizophrenia and 40 healthy control subjects (HC, n = 40 total, n = 24 per comparison group) matched for age, gender, ethnicity, parental socioeconomic status and cigarette smoking. We measured the binding potential BPP, corrected for partial volume effects. The outcome measure was obtained in cortical and striatal subregions outlined on coregistered individual MRIs. Partial volume effect corrected BPP measures were significantly higher in DN vs controls in cortical regions. No such increases were found in the DF versus controls comparison. Furthermore, in the DF group, DF interval correlated positively with cortical BPP. We conclude that upregulation of D1 receptors in schizophrenia is related to the illness itself and may be corrected and normalized by chronic antipsychotic treatment.