Simplified Quantification of <sup>11</sup>C-UCB-J PET Evaluated in a Large Human Cohort
Mika Naganawa, Jean-Dominique Gallezot, Sjoerd J. Finnema, David Matuskey, Adam P. Mecca, Nabeel Nabulsi, David Labaree, Jim Ropchan, Robert T. Malison, Deepak Cyril D’Souza, Irina Esterlis, Kamil Detyniecki, Christopher H. van Dyck, Yiyun Huang, Richard E. Carson
Abstract
<sup>11</sup>C-UCB-J ((<i>R</i>)-1-((3-(<sup>11</sup>C-methyl-<sup>11</sup>C)pyridin-4-yl)methyl)-4-(3,4,5-trifluorophenyl)pyrrolidin-2-one) is a PET tracer for synaptic vesicle glycoprotein 2A, which may be a marker of synaptic density. To simplify the scan protocol, SUV ratios (SUVRs) were compared with model-based nondisplaceable binding potential (<i>BP</i><sub>ND</sub>) to select the optimal time window in healthy and neuropsychiatric subjects. <b>Methods:</b> In total, 141 scans were acquired for 90 min. Arterial blood sampling and metabolite analysis were conducted. SUVR-1 (centrum semiovale reference region) was computed for six 30-min windows and compared with 1-tissue-compartment model <i>BP</i><sub>ND</sub>. Simulations were performed to assess the time dependency of SUVR-1. <b>Results:</b> Greater correlation and less bias were observed for SUVR-1 at later time windows for all subjects. Simulations showed that the agreement between SUVR-1 and <i>BP</i><sub>ND</sub> is time-dependent. <b>Conclusion:</b> The 60- to 90-min period provided the best match between SUVR-1 and <i>BP</i><sub>ND</sub> (−1% ± 7%); thus, a short scan is sufficient for accurate quantification of <sup>11</sup>C-UCB-J–specific binding.