Andrew John Lees

Parkinson’s disease is a common and debilitating neurodegenerative disorder, with over half of patients progressing to postural instability, dementia, or death within 10 years of diagnosis. However, the onset and rate of progression to poor outcomes is highly variable, underpinned by heterogeneity in underlying pathological processes. Quantitative and sensitive measures predicting poor outcomes will be critical for targeted treatment, but most studies to-date have been limited to a single modality or assessed patients with established cognitive impairment. Here, we used multimodal neuroimaging and plasma measures in 98 patients with Parkinson’s disease and 28 age-matched controls followed-up over 3 years. We examined: grey matter (cortical thickness and subcortical volume), white matter (fibre-cross section, a measure of macrostructure; and fibre density, a measure of microstructure) at whole-brain and tract level; structural and functional connectivity; and plasma levels of neurofilament light chain (NFL) and phosphorylated tau (p-tau) 181. We evaluated relationships with subsequent poor outcomes, defined as development of mild cognitive impairment, dementia, frailty or death at any time during follow-up, in people with Parkinson’s disease. We show that extensive white matter macrostructural changes are already evident at baseline assessment in people with Parkinson’s disease who progress to poor outcomes (n = 31): with up to 19% reduction in fibre cross-section in multiple tracts, and a subnetwork of reduced structural connectivity strength, particularly involving connections between right fronto-parietal and left frontal, right fronto-parietal and left parietal, and right temporo-occipital and left parietal modules. In contrast, grey matter volumes and functional connectivity were preserved in people with Parkinson’s disease with poor outcomes. NFL, but not p-tau181 levels were increased in people with Parkinson’s disease with poor outcomes, and correlated with white matter loss. These findings suggest that imaging sensitive to white matter macrostructure and plasma NFL may be useful early markers of poor outcomes in Parkinson’s disease. As new targeted treatments for neurodegenerative disease are emerging, these measures show important potential to aid patient selection for treatment and improve stratification for clinical trials.