Patterns of grey matter volume predict who worsens in secondary progressive MS

Whole-brain and single-region MRI measures are only modestly associated with disability in multiple sclerosis. Network-based measures — which treat the brain as a set of regions that change together — have looked more promising, and this study tested whether they can actually forecast decline.

Published in the Journal of Neurology, Neurosurgery & Psychiatry and senior-authored by Dr Arman Eshaghi, the work used cross-sectional structural MRI plus baseline and longitudinal EDSS, Nine-Hole Peg Test and Symbol Digit Modalities Test data from a clinical trial in 988 people with secondary progressive MS. T1-weighted scans were processed into grey matter probability maps and spatial independent component analysis identified patterns of covarying regional volume; the same analysis was repeated in 400 healthy controls for comparison.

Fifteen patterns emerged, and several tracked clinical outcomes more closely than whole-brain grey matter, deep grey matter or lesion volume did. A predominantly basal ganglia component had the strongest baseline correlation with the Symbol Digit Modalities Test and predicted cognitive worsening (HR 1.29, 95% CI 1.09–1.52, p < 0.005); two further components predicted worsening on the Nine-Hole Peg Test. For trial design, that matters — a prognostic marker measurable at baseline lets you enrich a study with the participants most likely to show change.

Read the paper: Predicting disability progression and cognitive worsening in multiple sclerosis using patterns of grey matter volumes.