Abstract
46
Introduction: The GluN2B subunit of the NMDA receptor complex is a therapeutic target for a range of neuropsychiatric disorders including dementia and schizophrenia. Ongoing efforts to develop GluN2B-specific radiotracers for clinical use have produced several promising compounds. The objective of this work was to compare the in vivo imaging properties in rhesus macaque of three candidate radiotracers: (R)-18F-OF-Me-NB1 [1], (R)-11C-NR2B-Me [2], and (S)-18F-OF-NB1 [3]. Methods: Baseline and blocking scans for each radiotracer were performed in 2 animals on a Focus 220 scanner for 120 min. Blocking scans were performed using the selective GluN2B antagonist Co-101,244 (0.25 mg/kg) or the selective σ1 receptor antagonist FTC-146 at 2 doses (0.027 or 0.125 mg/kg). Metabolite-corrected input functions and plasma free fractions (fp) were determined from arterial blood samples. Regional time-activity curves (TACs) were extracted across the brain. Volume of distribution (VT) was determined for each radiotracer using one- and two-tissue compartment models (1TCM and 2TCM) and multilinear analysis-1 (MA1). Receptor occupancy for each blocking drug and nondisplaceable volume of distribution (VND) were determined from occupancy plots. Guo plots of regional VT values were used to estimate relative affinity and binding potential (BPND) of each pair of radiotracers [4]. Estimated BPND was determined using VND estimates from GluN2B blocking scans (BPND = VT/VND - 1).
Results: fp was lower for (R)-18F-OF-Me-NB1 and (R)-11C-NR2B-Me (1.5% and 0.6%, respectively) than for (S)-18F-OF-NB1 (15%). Each radiotracer showed high uptake throughout cortical and subcortical grey matter and cerebellum (Fig A). Tissue clearance was faster for (R)-18F-OF-Me-NB1 compared to the other two compounds (Fig B). For each radiotracer, TACs were well-described by the 1TCM. TACs for (R)-18F-OF-Me-NB1and (S)-18F-OF-NB1, but not (R)-11C-NR2B-Me, were also well-fitted by MA1. VT values from the 1TCM were lowest for (R)-18F-OF-Me-NB1 (8.2 - 12.8 mL/cm3) and highest for (R)-11C-NR2B-Me (16.9 - 38.8 mL/cm3). In each case, VT was highest in cingulate and frontal cortex, low to moderate in cerebellum, and lowest in occipital lobe and white matter. Co-101,244 reduced VT values in all regions with target occupancy of >75% for all tracers. VND was lowest for (R)-18F-OF-Me-NB1 (6.4 mL/cm3) and highest for (R)-11C-NR2B-Me (11.2 mL/cm3). FTC-146 pre-treatment also reduced uptake and VT values of all three tracers with 30-56% reduction in specific binding (see Table), suggesting a possible interaction of the radiotracers and/or GluN2B target with σ1 receptors. Guo plots indicated higher GluN2B affinity of (R)-11C-NR2B-Me and (S)-18F-OF-NB1 than (R)-18F-OF-Me-NB1 (Fig C). Consistent with this affinity rank order, average cortical BPND value was higher for (R)-11C-NR2B-Me (2.64) and (S)-18F-OF-NB1 (2.53), and lower for (R)-18F-OF-Me-NB1 (0.65).
Conclusions: Three candidate GluN2B radiotracers showed promising imaging properties in non-human primates, with similar brain distribution and selectivity profiles. (S)-18F-OF-NB1 had highest fp of 15%, while (R)-11C-NR2B-Me and (S)-18F-OF-NB1 showed similar kinetics and cortical BPND values of 2-3, suggesting that both are good candidates for evaluation in humans. However, displaceable binding in the cerebellum, which has been shown in rodent studies to have negligible density of GluN2B during adulthood, is a significant concern as it suggests these radiotracers could have an additional, as-yet-unidentified target across the brain. Additional pharmacologic and genetic studies are exploring this apparent off-target binding. The nature of the interaction with σ1 receptors also requires further study.
References: 1. Haider et al. J Nucl Med 2019; 60:259-266. 2. Dick et al. J Nucl Med 2019; 60 (supp 1):115. 3. Ahmed and Zheng et al. J Nucl Med 2020; 61 (supp 1): 263. 4. Guo et al. JCBFM 2014; 34:1162-1128. Funding: NIMH U01MH107803
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.