PET evaluation of cholinergic system differences in progressive supranuclear palsy and age-matched controls using [18F]VAT.

BACKGROUND

The availability of binding sites for the vesicular acetylcholine transporter (VAChT) provides a measure of presynaptic cholinergic neuron terminals, which undergo degeneration in various conditions, including progressive supranuclear palsy (PSP). We used a VAChT-specific PET radiotracer, [18F]VAT, to determine whether the density of cholinergic terminals in PSP differs from that of cognitively intact healthy controls (HC). We further performed the analysis using the standardized uptake value ratio (SUVR) and partial volume correction (PVC) methods. Nine clinically diagnosed PSP participants (72 ± 8 years, 5 M/4F) and twenty sex- and age-matched HC (71 ± 8 years,12 M/8 F) had a 120-min dynamic [18F]VAT-PET scan. Distribution volume ratios (DVR) from Logan graphical analysis and SUVR were generated with eroded cerebral white matter as the reference region using the region of interest (ROI)-based approach. Spearman correlations between SUVR and DVR with clinical measurements, namely the Montreal Cognitive Assessment (MoCA), disease duration, and Unified Parkinson's Disease Rating Scale Part-III (UPDRS-III), were investigated. Statistical analysis was performed using the Mann-Whitney U test, with significance defined at p < 0.05. The data were corrected for multiple comparisons for target regions of caudate, putamen, accumbens, and entorhinal cortex, and without multiple comparisons for other ROIs in exploratory analysis.

RESULTS

Significant reduction of >9 % in DVR and SUVR values was observed in PSP in the entorhinal cortex (FDR-corrected p ≤ 0.023), while significant reduction of >10 % in the DVR values only was observed in the caudate (FDR-corrected p = 0.044). Exploratory analyses suggested that PSP patients had lower DVR and SUVR values in thalamus, hippocampus, caudal anterior cingulate, isthmus cingulate, lateral occipital cortex, lingual, pars orbitalis, pericalcarine, posterior cingulate, superior temporal, supramarginal, and transverse temporal regions compared to HCs. Exploratory analyses also revealed that DVR and SUVR of the brainstem correlated with disease duration (|ρ| > 0.877, p ≤ 0.003), whereas DVR and SUVR of the ventral diencephalon and brainstem correlated with MoCA scores (|ρ| > 0.48, p ≤ 0.008). No region correlated with UPDRS-III. SUVR (100-120 min) correlated well with DVR. PVC did not improve correlations with clinical measures.

CONCLUSIONS

Lower [18F]VAT relates to lower cognitive function; further studies with an increased sample size could strengthen the justification for cholinergic treatment approaches to target cognitive decline in PSP.