Assessment of the impact of outcome definitions on power in multiple sclerosis clinical trials.

Background: Clinical trials for disease progression in multiple sclerosis (MS) often use a confirmed increase on the expanded disability status scale (EDSS) as the primary outcome. Objective: To compare potential outcome measures of disease progression for MS clinical studies. Methods: Data were simulated with clinical visits including measurement of the EDSS and relapses completed every 6 months. At each visit, the EDSS could increase due to a relapse, disease progression, or other reasons. Six definitions of disease progression were used based on the number of measurements required for the progression (1, 2, or 3) and whether concurrent relapse could have occurred. The power for each definition was calculated as the proportion of simulated datasets with a statistically significant treatment effect. Results: All the approaches maintained the type I error rate when there was no overall treatment effect. The approach with progression at a single visit and no relapses led to the greatest power when there was only a treatment effect on progression and there were no increases due to other causes, and this scenario may be directly applicable to a progressive MS trial. For other scenarios, the best approach included a confirmation visit. Conclusions: Choosing the best outcome definition for a clinical trial depends on the assumed mechanism of action of the treatment and the patient population.