ABSTRACT
A surrogate endpoint is intended to replace a clinical endpoint for the evaluation of new treatments when it can be measured more cheaply, more conveniently, more frequently, or earlier than that clinical endpoint. A surrogate endpoint is expected to predict clinical benefit, harm, or lack of these. A quantitative assessment of the strength of evidence for surrogacy requires the demonstration of the prognostic value of the surrogate for the clinical endpoint, and evidence that treatment effects on the surrogate reliably predict treatment effects on the clinical endpoint. We review the statistical approaches that have been proposed to assess the extent to which these conditions are fulfilled. In particular, we focus on the so-called meta-analytic approach, which uses data from several randomized clinical trials. The concepts and methods are illustrated with two meta-analyses: one in advanced colorectal cancer and one in advanced gastric cancer.
