Imaging biomarkers of cortical neurodegeneration underlying cognitive impairment in Parkinson’s disease

• DOI: 10.1007/s00259-025-07070-z
• Publication Date: 2025-01-31T07:49:26Z
• Authors: Jesús Silva-Rodrí...; Miguel Ángel Labr...; Sandra Castro-Lab...; Laura Muñoz-Delgado; Pablo Franco-Rosado; Ana María Castell...; Daniel Macías-García; Silvia Jesús; Astrid D. Adarmes...; Fátima Carrillo; Juan Francisco Ma...; David García-Solís; Florinda Roldán-Lora; Pablo Mir; Michel J. Grothe

Imaging biomarkers bear great promise for improving the diagnosis and prognosis of cognitive impairment in Parkinson's disease (PD). We compared ability three commonly used neuroimaging modalities to detect cortical changes PD patients with mild (PD-MCI) dementia (PDD). 53 cognitively normal (PD-CN), 32 PD-MCI, 35 PDD underwent concurrent structural MRI (sMRI), diffusion-weighted (dMRI), [18F]FDG PET. extracted grey matter volumes mean diffusivity (MD, dMRI), standardized uptake value ratios ([18F]FDG PET) 52 regions included a neuroanatomical atlas. assessed group differences using ANCOVA models further applied cross-validated machine learning approach identify modality-specific brain that are most indicative status their diagnostic accuracy separation receiver operating characteristic analyses. In sMRI, atrophy temporal posterior-parietal areas allowed separating from PD-CN (AUC = 0.77 ± 0.07), but was poor PD-MCI (0.57 0.10). dMRI showed pronounced medial lobe, which provided excellent performance 0.87 0.06), more modest still significant 0.71 0.09). Finally, PET revealed hypometabolism posterior-occipital regions, highest accuracies both 0.89 0.05) 0.78 0.05). statistical comparisons, (p < 0.001) 0.031) outperformed sMRI detecting PD-MCI. Among tested modalities, accurate associated PD, especially at early stages. Diffusion measurements may represent promising MRI-based alternative.