Using earlier measures in a longitudinal sequence as a potential surrogate for a later one

The number of potential surrogate markers for clinical-trial endpoints is increasing rapidly, not in the least owing to the availability of biomarkers. At the same time, considerable development has taken place regarding statistical evaluation paradigms for such markers. As a consequence, such endpoints are given more extensive consideration for practice than previously had been the case. A particular but important instance is where the true endpoint is the ultimate assessment in a sequence of repeated measures. It is then appealing to consider earlier measures, either in isolation or several combined, as a potential surrogate endpoint. The length and cost reducing potential has to be weighed carefully against loss in precision and the risks of an inappropriate decision regarding a new compound's fate. Quantitative criteria to do so are developed, embedded in a meta-analytic framework. The methodology's behavior is assessed through simulations and applied to data from a pair of clinical trials, one in ophthalmology and one in schizophrenia.