Synaptic density in healthy human aging is not influenced by age or sex: a 11C-UCB-J PET study

Abstract Rationale : 11C-UCB-J binds to synaptic vesicle glycoprotein 2A, a protein ubiquitously expressed in presynaptic nerve terminals, and can therefore serve as in vivo proxy of synaptic density. There are discrepancies in post-mortem data on stability of synaptic density with healthy aging. In this study, healthy aging and sex as potential modifiers of 11C-UCB-J binding were investigated in healthy volunteers over 7 adult decades, assuming that the number of SV2A vesicles per synapse is not influenced by age or sex. Methods : 80 healthy volunteers underwent 11C-UCB-J PET and structural T1 and T2 MR imaging. Grey matter changes with aging were firstly evaluated by voxel-based morphometry (VBM). Parametric 11C-UCB-J standardized uptake value ratio (SUVR) images were calculated using the centrum semiovale as reference tissue. To correct for atrophy-related partial volume effects, a region-based voxel-wise type partial volume correction (PVC) was applied in FreeSurfer. The correlations of 11C-UCB-J binding with age and with sex were investigated by a voxel-based and volume-of-interest (VOI)-based approach, and with and without PVC to assess the contribution of underlying morphology changes upon aging. Results : Full results were available for 78 participants (19-85y; 33 M/45 F). VBM grey matter concentration changes with aging were most predominant in the perisylvian and frontal regions. After PVC, no significantly decreased 11C-UCB-J SUVR with aging was found in the voxelbased analysis, whereas the VOI-based analysis showed a slight decrease in the caudate nucleus (-1.7% decrease per decade, p= 0.0025) only. There was no association between sex and 11C-UCB-J SUVR, nor an interaction between aging and sex for this parameter. Conclusion : In vivo, PET using 11C-UCB-J does not support a cortical decrease of synaptic density with aging, whereas subcortically a small effect with aging in the caudate nucleus was observed. In addition, no association between synaptic density and sex was detected, which allows pooling of datasets of both genders.