Discussion of “Adaptive and Model-Based Dose-Ranging Trials: Quantitative Evaluation and Recommendations”

Abstract

Dose-finding is one of the most important steps during the drug development process in all disease areas. Failure to determine the appropriate dose leads to lower success rates of clinical trials in the late phases of development. Use of an adaptive design for dose-finding would help sponsors investigate a broader range of doses efficiently, thereby increasing the success rate of clinical trials and facilitating the clinical drug development process. Although adaptive dose-ranging (ADR) approaches have been widely studied, a clear summarization and discussion of the methods is lacking. This paper summarizes several current methods and is a good reference for biostatisticians/investigators who design ADR trials. An increasing number of clinical trial consultation meetings are being conducted to discuss ADR trials within the Pharmaceuticals and Medical Devices Agency (PMDA). To date, these meetings primarily discuss the necessity and the advantages of such trials. However, the methodology of ADR will be discussed in detail in the near future, making this paper a useful resource. We begin with a brief mention of our experiences and perspectives in reviewing ADR trials based on the PMDA clinical trial consultation meetings (Section 2). In Section 3, there is a description of the simulation results and their findings with respect to the ADR approaches of the paper. In Section 4, we conclude our discussion with some recommendations.