A Bayesian Approach to Three Levels Surrogacy Modeling for Multiple Sclerosis

• Activity Number: 602
• Type: Topic Contributed
• Date/Time: Thursday, August 2, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Luca Pozzi*+ and David I Ohlssen and Heinz Schmidli
• Companies: University of California at Berkeley and Novartis and Novartis Pharma AG
• Keywords: Meta-analysis ; Multiple Sclerosis ; Bayesian ; Surrogacy

Abstract

In Clinical Trials, a surrogate endpoint is one that can be used instead of the outcome of interest in the evaluation of an experimental treatment. The development of new therapies for the treatment of multiple sclerosis is one area where surrogacy used in various stages of clinical research. While the aim of treatments in MS is to prevent long term irreversible disability, evaluating drug effect on disability progression would require a large sample of patients and many years of follow-up. To reduce study size and duration, clinical relapses are usually used as primary endpoints in phase III trials. In the case of phase II studies, the primary outcomes are typically based on magnetic resonance imaging (MRI). Recent work provided a systematic review and meta-analysis examining the role of both MRI and relapse as trial level surrogate outcomes for disability. However, classical meta-regression assumes that the independent variable is measured without error. The aim of this work is to re-examine the same data using a Bayesian trial level surrogate outcome meta-analysis model. Further, we extend this model to three levels, simultaneously linking both MRI and relapses to disability.