ABSTRACT
The use of adaptive design methods has received a great deal of attention in past decades due to its appealing feature of flexibility and potential efficiency in clinical research and development based on accrued data. Practically, adaptive design may provide the investigators a second chance to modify or re-evaluate the trial while the study is still ongoing. However, it is then a concern that the actual patient population after the adaptations could deviate from the originally target patient population. In addition, the overall type I error rate may not be controlled after adaptations and therefore the results are difficult to interpret. A recently released draft FDA guidance on adaptive design clinical trials emphasizes that the adaptive design must be a prospectively planned opportunity and should be based on information collected within the study, with or without formal statistical hypothesis testing. In this article, the relative advantages, limitations, and feasibility of commonly considered adaptive designs in clinical trials are examined. Statistical concerns and regulatory perspectives when implementing adaptive designs are also discussed.
