Opinion on Adaptive Designs in Clinical Trials

[摘要] Starting from a couple of research papers in this type published inEurope’s peer-reviewed statistical journals, adaptive designs have mademuch progress in the development and implementation in the past 20years, which are anticipated to increase the information value ofclinical trial data in order to enable better decisions during the courseand speed up the development process in the context of ٽercecompetition and limited trial budgets. So far for now, main types ofadaptive deigns are: 1) adaptive randomization which allows changingrandomization probabilities using information from past treatmentassignment (such as the biased coin design), or covariate-adaptive, orresponse-adaptive or covariate-adjusted-adaptive; 2) adaptive doseresponse designs; 3) sample size re-estimation; 4) Treatment selectiondesigns; 5) group sequential designs. All areas in this topic areundergone active development because analytic derivations are notwell investigated for many methods. Meanwhile, new methods or oldmethods but applying to new problems is mushrooming in theliterature. In my opinion, topics having noticeably importance incoming years are dose-response designs and treatment selectiondesigns.