Objective: To identify structural MRI-based progression subtypes in multiple system atrophy (MSA) and determine whether these imaging-defined subtypes are associated with distinct clinical features, molecular imaging patterns, and milestone risks.

Background: MSA is a rapidly progressive synucleinopathy with marked clinical heterogeneity. The conventional MSA-P/MSA-C classification does not fully capture underlying biological diversity or prognostic variability. More informative imaging-based stratification may improve disease characterization and risk assessment.

Method: We included 161 patients fulfilling the 2022 Movement Disorder Society diagnostic criteria for MSA and 81 healthy controls. Baseline regional brain volumes derived from longitudinal FreeSurfer processing were screened to identify candidate atrophy biomarkers. An event-based disease progression model implemented in pySuStaIn was applied to infer latent atrophy subtypes and stages. Baseline clinical scales, FDG-PET, DAT-PET, and longitudinal milestone outcomes were compared across subtypes. Follow-up MRI was used to assess longitudinal subtype and stage stability.

Results: Sixteen MRI biomarkers were selected for modelling. Three distinct atrophy progression subtypes were identified: a cerebellum-led subtype, a basal ganglia-led subtype, and a brainstem-led mixed subtype with early multisystem involvement. Among 161 patients, 83 were assigned to subtype 1, 32 to subtype 2, and 32 to subtype 3; 14 were classified as stage 0. Event-based model stage correlated with disease duration at baseline. Longitudinal MRI showed high stability, with 90.6% of follow-up scans retaining the same subtype or progressing from stage 0 to a specific subtype, without stage reversal. Clinically, subtype 2 showed the greatest parkinsonian and disability burden, subtype 1 showed more prominent cerebellar features, and subtype 3 showed earlier severe urinary symptoms and orthostatic dizziness. FDG-PET demonstrated selective striatal metabolic differences, and DAT-PET showed lower posterior-to-anterior putamen ratio in subtype 2 than in subtypes 1 and 3.

Conclusion: Structural MRI identified three biologically and clinically distinct progression subtypes in MSA. Imaging-based progression subtyping may provide a more informative framework than conventional motor phenotype for risk stratification, prognostic assessment, and future trial enrichment in MSA.

Spatiotemporal progression patterns

KM curves for key milestone events

« Back to 2026 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/novel-structural-mri-based-progression-subtypes-and-prognostic-study-in-multiple-system-atrophy/