Age and sex effects on synaptic density in healthy humans as assessed with SV2A PET

541 Objectives: Recently, we developed 11C-UCB-J, a PET ligand for synaptic vesicle glycoprotein 2A (SV2A) (1-3), which is expressed ubiquitously in presynaptic terminals. SV2A has comparable distribution to synaptophysin, a common post-mortem synaptic density marker. Thus, 11C‑UCB‑J may be a quantitative in vivo synaptic density marker, and we have found specific changes in epilepsy and Alzheimer’s disease vs. age-matched controls. To date, there has been no in vivo assessment of aging effects on synaptic density. Studies in postmortem human brains with no dementia have found some loss of presynaptic terminals in healthy aged subjects (> 60) compared to younger controls (4). Here, we used 11C-UCB-J PET imaging in healthy controls to assess the effect of age and sex on synaptic density. Methods: A total of 47 healthy controls (17 females, 30 males) were scanned with the HRRT (mean dose: 14 mCi). The age range was 24-83 y with 11 subjects >60 and no difference in mean age between females (46 y) and males (47 y). Arterial samples were used to measure the metabolite-corrected input function and plasma free fraction (fP). VT images were produced with the one-tissue model. Using the AAL template and individual MR images, 12 gray matter (GM) regions of interest (ROIs) were applied: frontal, occipital, parietal, and temporal cortices, anterior and posterior cingulate, hippocampus, precuneus, cerebellum, caudate, putamen, and thalamus. BPND was calculated with the centrum semiovale (CS) as reference region. A custom CS template was developed to minimize spill-in from GM. Age and sex effects in VT, K1, and BPND were evaluated with regional correlations and t tests without multiple comparison correction. A mixed linear model was also applied using an unstructured covariance matrix followed by post-hoc tests. Results: Average cortical VT was 18.2 mL/cm3 with 14% intersubject SD. Average cortical K1 was 0.34 mL/min/cm3 (14% SD). Plasma free fraction was 0.29±0.03. In CS, mean VT was 4.3 mL/cm3 (15% SD) and mean K1 was 0.11 (14% SD). There was a statistically significant reduction in CS VT (3.3% reduction per decade (RPD), p 0.10) and sex (15.6 (F), 14.7 (M), p > 0.15) effects were not significant. Thus, we used BPND as the primary outcome measure. In cortical regions, cerebellum, and putamen, there were no age-related changes in BPND (mean RPD ~ 1%). Hippocampus showed a non-significant 2% RPD. However, in caudate and thalamus, there were significant reductions in BPND with age (caudate: 8% RPD, p