Adaptive sequential sampling plans are often used in the monitoring of clinical trials in order to address the ethical and efficiency issues inherent in human testing of a new treatment or preventive agent for disease. Group sequential stopping rules are perhaps the most commonly used approaches, but in recent years, a number of authors have proposed more flexible adaptive methods such as unblinded sample size re-estimation, adaptive enrichment, and response adaptive randomization.
This module describes some of the special issues that can arise from a statistical and/or regulatory standpoint in any sequential RCT. Particular emphasis is placed on issues that can arise with time to event endpoints. Topics include:
- Analysis of RCT data in the presence of adaptive enrichment, response adaptive randomization or adaptive selection of doses;
- The use of sequential designs when treatment effects might be time-varying (e.g. non proportional hazards), and
- The avoidance of operational bias with unblinded adaptation.