Initial Evaluation of 18F-GE-179, a Putative PET Tracer for Activated N-Methyl D-Aspartate Receptors

N-methyl D-aspartate (NMDA) ion channels play a key role in a wide range of physiologic (e.g., memory and learning tasks) and pathologic processes (e.g., excitotoxicity). To date, suitable PET markers of NMDA ion channel activity have not been available. F-18-GE-179 is a novel radioligand that selectively binds to the open/active state of the NMDA receptor ion channel, displacing the binding of H-3-teno-cyclidine from the intrachannel binding site with an affinity of 2.4 nM. No significant binding was observed with 10 nM GE-179 at 60 other neuroreceptors, channels, or transporters. We describe the kinetic behavior of the radioligand in vivo in humans. Methods: Nine healthy participants (6 men, 3 women; median age, 37 y) each underwent a 90-min PET scan after an intravenous injection of F-18-GE-179. Continuous arterial blood sampling over the first 15 min was followed by discrete blood sampling over the duration of the scan. Brain radioactivity (KBq/mL) was measured in summation images created from the attenuation- and motion-corrected dynamic images. Metabolite-corrected parent plasma input functions were generated. We assessed the abilities of 1-, 2-, and 3-compartment models to kinetically describe cerebral time-activity curves using 6 bilateral regions of interest. Parametric volume-of-distribution (V-T) images were generated by voxelwise rank-shaping regularization of exponential spectral analysis (RS-ESA). Results: A 2-brain-compartment, 4-rate-constant model best described the radioligand's kinetics in normal gray matter of subjects at rest. At 30 min after injection, 37% of plasma radioactivity represented unmetabolized F-18-GE-179. The highest mean levels of gray matter radioactivity were seen in the putamina and peaked at 7.5 min. A significant positive correlation was observed between K-1 and V-T (Spearman rho = 0.398; P = 0.003). Between-subject coefficients of variation of V-T ranged between 12% and 16%. Voxelwise RS-ESA yielded similar VTS and coefficients of variation. Conclusion: F-18-GE-179 exhibits high and rapid brain extraction, with a relatively homogeneous distribution in gray matter and acceptable between-subject variability. Despite its rapid peripheral metabolism, quantification of F-18-GE-179 V-T is feasible both within regions of interest and at the voxel level. The specificity of F-18-GE-179 binding, however, requires further characterization with in vivo studies using activation and disease models.