AN URN MODEL FOR ODDS-RATIO-BASEDRESPONSE-ADAPTIVE PHASE III CLINICAL TRIALSFOR TWO OR MORE TREATMENTS

摘要

Adaptive data-dependent allocation designs are used in phase III clinical trials havingtwo or more competing treatments with sequential entrance of patients, in order toallocate a larger number of patients to the better treatment. The odds ratio is a popularconcept for biomedical practitioners; hence, odds-ratio-based adaptive designs couldbe very useful in practice. Rosenberger et al. (2001) introduced an odds-ratio-basedtwo-treatment response-adaptive design; however, they did not study the properties indetails. In this article, we describe these designs by means of urn models and providelimiting results for them. Some properties of the design are also studied numerically.We compare the performance of the proposed design with some possible competitorswith respect to a few criteria. A real dataset is used to illustrate the applicability of theproposed design. Thus, we provide a base for using odds-ratio-based response-adaptivedesigns in practice. We extend our design for covariates and also for more than twotreatments. In particular, we study the three-treatment design in this article.