Surrogate endpoints in clinical trials: definition and operational criteria

摘要: I discuss the idea of using surrogate endpoints in context clinical trials to compare two or more treatments interventions respect some 'true' endpoint, typically a disease occurrence. In order that treatment comparison based on response variable have meaningful implication for corresponding true endpoint comparison, rather restrictive criterion is proposed use adjective 'surrogate'. Specifically, propose yield valid test null hypothesis no association between and response. This essentially requires 'capture' any relationship notion can be operationalized by requiring rate at follow-up time independent treatment, given preceding history variable. then this operational examples accompanying papers setting aimed primary secondary prevention cancer.