Drug Evaluation Committee Simulation Study for Planning Adaptive Design
February 2025
Discussions regarding adaptive design are currently underway under ICH-E20. While the flexibility provided by adaptive design is expected to offer benefits such as increased opportunities for trial participants to receive better treatment, more efficient drug development, and the optimal utilization of available resources, if statistical methods are not used appropriately, undesirable phenomena may occur from the perspective of statistical validity, such as an increase in the Type I error rate,bias in point estimates, and a decrease in the coverage probability of confidence intervals relative to the nominal confidence coefficient—all of which are undesirable from the perspective of statistical validity.
When actually designing an adaptive trial, it is important to evaluate the design’s performance characteristics during the planning stage, which is typically done through simulations under hypothetical scenarios.This deliverable summarizes the methods and procedures for simulations used to evaluate the performance characteristics of trial plans when designing adaptive designs, as well as the metrics to be evaluated through simulation and the methods for reporting the results.
The intended audience for this document includes biostatisticians involved in the planning and conduct of clinical trials, clinical staff planning clinical trials, and regulatory affairs staff responsible for regulatory affairs.
We hope this document will contribute to a better understanding of adaptive designs and the proper conduct of clinical trials.
Japan Pharmaceutical Manufacturers Association (JPMA) Drug Evaluation Committee
Data Science Subcommittee FY2023 Continuing Task Force 1