MRI lesions as a surrogate for relapses in multiple sclerosis: a meta-analysis of randomised trials

摘要: Summary Background A meta-analysis of randomised trials in relapsing-remitting multiple sclerosis published 2009 showed a quantitative relation between the treatment effects detected on MRI lesions and clinical relapses. We aimed to validate that using data from large independent set sclerosis. Methods searched Medline for assessed disease-modifying drugs Sept 1, 2008, Oct 31, 2012. extracted relapses each trial, correlation log transformed relative measures these was with weighted linear regression analysis. The R 2 value estimated quantify strength correlation, we used an interaction test difference slope previously equation. also ran several sensitivity analyses. Findings identified 31 eligible trials, which provided 18 901 patients equation derived studies concurrent (slope=0·52; =0·71), much same as (p =0·45). Analysis tested phase 3 over short follow-up periods (6–9 months) can predict longer (12–24 months), reported were within 95% prediction intervals eight nine trials. Interpretation Our findings indicate effect be accurately predicted by therapy lesions, implying use markers primary endpoints future treatments considered, specific situations, such testing generics or biosimilars well known mechanism action paediatric already approved adults. Funding None.