|
Abstract:
|
Multiple sclerosis (MS) is the most common form of inflammatory demyelinating central nervous system disease and it is a chronic, progressive disease. Disability improvement has been used as an endpoint to evaluate the effects of remyelination compounds in MS clinical studies, and is usually assessed via time-to-event analysis (eg, Kaplan Meier method). However, disability improvement can be a transient condition for a progressive disease, such as MS. Current approach focusing on time to improvement is not able to consider the duration of disability improvement. In this presentation, we will first describe prevalence of response which can be utilized to evaluate both occurrence and duration of transient response simultaneously. Estimation and inference procedure for prevalence of response will be provided. Interpretation of the prevalence of response will be discussed. Potential extensions will be considered including covariate adjustment and stratification. Finally, we will apply this approach to a clinical study in MS to evaluate the duration of disability improvement in comparison with time-to-event analysis.
|
