Predicting Individual Responses to Therapy in Multiple Sclerosis

Investigators used MSBase (a data repository from a global cohort spanning 34 countries) to develop models to predict individual treatment responses for each of seven commonly used disease-modifying therapies. Models were based on demographic, clinical, and paraclinical features. For each treatment, 27 clinical predictors were evaluated in a subset of some 8500 patients, followed by validation in a separate cohort of 1200, and then an independent external validation in a cohort of 3000. Response was analyzed regarding disability progression, disability regression, relapse frequency, conversion to secondary progressive disease, change in disease burden, and treatment discontinuation.

Median follow-up was 8 years, with Expanded Disability Stat…