Clinical relevance of MRI-informed motor progression phenotypes in early Parkinson’s disease

Parkinson’s disease (PD) exhibits significant variability in disease progression, making individual trajectory prediction challenging. Previously, we defined motor progression phenotypes based on longitudinal OFF-medication changes in the Movement Disorder Society–Unified Parkinson’s Disease Rating Scale Part III (MDS-UPDRS-III) and demonstrated their predictability at baseline using MRI-informed machine learning. OFF-medication assessments are rare in routine care. Therefore, this study evaluated whether these phenotypes correspond to standard clinical outcomes. Eighty-eight early PD patients from the Parkinson’s Progression Markers Initiative were classified as “faster” or “slower” progressors based on 48-month OFF-medication ΔMDS-UPDRS-III scores. A support vector machine incorporating baseline structural MRI and clinical features classified these phenotypes with 89% accuracy. Phenotypes were compared across independent 48-month outcomes, including MDS-UPDRS Part II, Schwab & England Activities of Daily Living (S&E ADL), levodopa equivalent daily dose (LEDD), and Hoehn & Yahr (HY) staging. Faster progressors exhibited significantly greater functional declines in MDS-UPDRS Part II and S&E ADL scores and higher LEDD requirements at 48 months. Clinically meaningful deterioration consistent with published thresholds occurred more frequently among faster progressors for MDS-UPDRS Part II and HY staging. These findings validate the clinical relevance of MRI-informed progression phenotyping for prognostic stratification in early PD, highlighting its potential to identify biologically distinct trajectories.