Cholinergic innervation topography in GBA-associatedde novoParkinson’s disease patients

Abstract

AbstractThe most common genetic risk factors for Parkinson’s disease areGBA1mutations, encoding the lysosomal enzyme glucocerebrosidase. Patients withGBA1mutations (GBA-PD) exhibit earlier age of onset and faster disease progression with more severe cognitive impairments, postural instability, and gait problems. The GBA-PD features suggests more severe cholinergic system pathologies. PET imaging with the vesicular acetylcholine transporter ligand [18F]-F-fluoroethoxybenzovesamicol ([18F]FEOBV PET) provides the opportunity to investigate cholinergic systems changes and their relationship to clinical features of GBA-PD.One hundred and twenty three newly diagnosed, treatment-naive Parkinson’s disease subjects – with confirmed presynaptic dopaminergic deficits on PET imaging – were included, all part of the Dutch Parkinson Cohort (DUPARC) study. FullGBA1sequencing of saliva samples was performed to evaluateGBA1variants. Patients underwent extensive neuropsychological assessment assessing all cognitive domains, motor evaluation with the Unified Parkinson’s disease Rating Scale, brain MRI, dopaminergic PET to measure striatal-to-occipital ratios of the putamen and [18F]FEOBV PET. We investigated differences in regional cholinergic innervation between GBA-PD carriers and non-GBA1mutation carriers (non-GBA-PD), using voxel-wise and volume-of-interest (VOI)-based approaches. The degree of overlap between t-maps from two-sample t-test models was quantified using the Dice similarity coefficient.Seventeen (13.8%) subjects had aGBA1mutation. No significant differences were found in the clinical features and dopaminergic ratios between GBA-PD and non-GBA-PD at diagnosis. Lower [18F]FEOBV binding was found in both the GBA-PD and non-GBA-PD group compared to controls. Dice (P< 0.05, cluster size 100) showed a good overlap (0.6233) between the GBA-PD and non-GBA-PD maps. GBA-PD patients showed more widespread reduction in [18F]FEOBV binding than non-GBA-PD when compared to controls in occipital, parietal, temporal, and frontal cortices (P< 0.05, FDR-corrected). In VOI analyses (Bonferroni corrected), the left cuneus, entorhinal cortex, fusiform gyrus and supramarginal gyrus were more affected in GBA-PD.De novoGBA-PD show a distinct topography of regional cholinergic terminal ligand binding, including both higher and lower binding regions. Although the Parkinson’s disease groups were not distinguishable clinically, in comparison to healthy controls, GBA-PD showed more extensive cholinergic denervation compared to non-GBA-PD. Our results suggest thatde novoGBA-PD and non-GBA-PD show differential patterns of cholinergic system changes before clinical differences are observable.