Structural and diffusion imaging in olfactory-related brain regions in Parkinson's disease: predictors of clinical...

Olfactory dysfunction is a prevalent non-motor symptom in Parkinson's disease (PD). Structural and diffusion MRI studies suggest that olfactory-related brain regions undergo significant neurodegenerative changes in PD. The current study aims to explore the longitudinal structural and diffusion imaging in olfactory-related regions in PD over four years. The relationships between baseline imaging and fluid biomarkers, and subsequent cognitive and clinical changes were also explored. We analyzed 97 newly diagnosed early-stage PD patients from the Parkinson's Progression Markers Initiative (PPMI) who underwent T1-weighted MRI, diffusion tensor imaging (DTI), and clinical assessments at baseline, two years, and four years. Structural and diffusion measures of olfactory-related regions were extracted using FreeSurfer and ExploreDTI. Baseline fluid biomarkers were also evaluated. Baseline mean diffusivity (MD) displayed significant associations with further cognitive changes across multiple regions, including amygdala, orbitofrontal cortex (OFC), insula, and thalamus (β=-0.214 to - 0.422). Baseline structural measures, including amygdala volume and entorhinal and OFC thickness, were associated with subsequent changes in cognitive and Unified Parkinson's Disease Rating Scale (UPDRS) scores (β = 0.264 to 0.402). Moreover, baseline serum neurofilament-light chain levels predicted multiple cognitive score changes (β= - 0.603 to - 0.331). Longitudinal analyses revealed a significant MD increase in the thalamus along with gradual reductions in volume and cortical thickness in the amygdala, insula, OFC, and entorhinal cortex (FDR-adjusted p < 0.05). Linear mixed-effect models further confirmed thalamic diffusion metrics as a predictor of cognitive deterioration. Olfactory-related regions exhibit progressive neurodegeneration in early PD, contributing to worsening cognition and disease severity. Baseline imaging and fluid biomarkers may serve as prognostic tools in PD.