Validation of surrogate markers in multiple randomized clinical trials with repeated measures

Abstract

While the practice of looking at multiple endpoints is by no means recent in clinical research, the validity of using one endpoint as a surrogate for another one has been raised and studied only over the last decade or so. Past of the recent literature on the validation of biomarkers as surrogate endpoints proposes to undertake the validation exercise in a multi-trial context which led to a definition of validity in terms of the quality of both trial level and individual level association between the surrogate and the true endpoint (Buyse et al, 2000). These authors concentrated on continuous responses. When both the surrogate and true endpoints are measured repeatedly over time, one is confronted with the modelling of bivariate longitudinal data. In this work, we show how such a joint model can be implemented in the context of surrogate marker validation. In addition, a further challenge consists of summarizing the concept of “surrogacy” in simple yet meaningful measures. We propose the use of the so-called variance reduction factor. The methodology is illustrated on data from a meta-analysis of five clinical trials comparing antipsychotic agents for the treatment of chronic schizophrenia.