TY - JOUR T1 - Kinetic modeling of CB1 PET tracer [<sup>11</sup>C]OMAR in rhesus monkeys and humans JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 216 LP - 216 VL - 51 IS - supplement 2 AU - Marc Normandin AU - David Weinzimmer AU - Jim Ropchan AU - David Labaree AU - Kuo-Shyan Lin AU - Neale Mason AU - Richard Carson AU - Deepak D'Souza AU - Alexander Neumeister AU - Yiyun Huang Y1 - 2010/05/01 UR - http://jnm.snmjournals.org/content/51/supplement_2/216.abstract N2 - 216 Objectives [11C]OMAR was recently developed for PET imaging of cannabinoid type 1 (CB1) receptors. Here we investigate quantification methods for [11C]OMAR in monkeys and humans. Methods Three rhesus monkeys underwent a total of 18 scans on the Focus220. Data were acquired for 120 min after 4.2±0.6 mCi [11C]OMAR. In 5 scans, unlabeled OMAR was co-injected at doses up to 1.7 mg/kg. Ten human subjects underwent a total of 18 scans. Data were collected for 120 min on the HRRT after bolus of 18±1 mCi. In all studies, arterial input functions were obtained. Regional time activity curves (TACs) were extracted from the dynamic PET data. Distribution volume (VT) was estimated by 1- and 2-tissue models with (1Tv, 2Tv) or without (1T, 2T) vascular component. Logan graphical (LGA) and multilinear analysis (MA1) were applied with a range of t* values. Binding constants ED50 and EC50 (with regard to mass dose or late plasma level, respectively) were estimated by fitting a one-site model to VT values in monkeys. Results TACs from monkeys and humans were fitted poorly by 1T and 1Tv. 2T fits were good in both species. In monkeys, 2Tv usually estimated negligible blood volume (Vb) and VT nearly identical to 2T. In humans, 2Tv improved fits slightly in some regions with Vb=3±2% and VT 1±4% lower than 2T. In both species, MA1 with t*&gt;20 min yielded VT values highly correlated with 2T (in humans: y=0.98x+0.05, R2=0.98). LGA was reliable in humans (y=0.96x+0.06, R2=0.95) but gave many low outliers in monkeys. VT was lower in humans than monkeys (~2 vs ~10 in high binding regions), as were K1 values (~0.05 vs ~0.2 1/min). In monkeys, ED50 was ~200μg/kg and EC50 ~40 nM with small variations between methods. Conclusions 2T was the best compartmental model for analysis of [11C]OMAR data in monkeys as well as in humans, where a vascular component sometimes improved fits. Graphical methods gave VT estimates similar to 2T in humans, but MA1 performed much better than LGA in monkeys ER -