White matter microstructure in Parkinson’s disease: investigating associations with cognitive functioning and plasma p-tau217

Parkinson’s disease (PD) is traditionally framed as a neurodegenerative disorder primarily affecting grey matter, since the defining pathology is loss of dopaminergic neurons in the substantia nigra. However, increasing evidence suggests that white matter (WM) microstructural alterations may occur early in the disease course and contribute to disease progression. Several studies have identified an association between WM microstructural changes and cognitive decline, suggesting that they could possibly contribute to cognitive impairment. Furthermore, elevated plasma levels of plasma phosphorylated tau-217 (p-tau217) have been associated with both accelerated cognitive decline and reduced WM integrity. Based on these findings, the present study aims to investigate the role of WM microstructural alterations in PD, evaluate its association with cognitive functions and plasma p-tau217 levels, and compare these results with those observed in a cohort of dementia-free older adults. A total of 68 patients with PD and 45 older adults were included in the study. All participants underwent a neuropsychological assessment, evaluating six cognitive domains: executive functions, attention and working memory, visuospatial abilities, memory, language and social cognition. Based on the cognitive performance, each participant was classified as: PD with normal cognition (PD-NC), with mild cognitive impairment (PD-MCI) and with dementia (PDD). Older adults were classified as cognitively unimpaired (CU). WM microstructure was investigated using Neurite Orientation Dispersion and Density Imaging (NODDI), from which two metrics were extracted: Intracellular Volume Fraction (ICVF), reflecting neurite density, and Orientation Dispersion Index (ODI), reflecting neurite orientation dispersion. Plasma p-tau217 was also examined to investigate its relationship with cognitive impairment and WM microstructural alterations. Three main findings emerged. First, between-group differences in NODDI-derived measures showed focal higher ODI within left cerebellar WM in cognitively impaired PD (PD-CI) compared to CU participants. Second, lower ICVF in long longitudinal association pathways, as well as occipital and corona radiations, was strongly associated with cognitive performance within the PD spectrum across all six cognitive domains. Conversely, in the PD spectrum, ODI showed more spatially restricted and lateralized associations. Third, no significant associations were observed between NODDI metrics and plasma p-tau217. These findings support the involvement of WM microstructural alterations in cognitive dysfunction across the PD spectrum. The distinct results obtained for ICVF and ODI highlight that the two metrics capture distinct aspects of WM microstructure, with ICVF primarily reflecting the neurite density and ODI reflecting the angular dispersion of axonal orientations, which appear to be differently related to cognitive functioning in PD.