Optimal Adaptive Promising Zone Designs

• Activity Number: 30 - Statistical Considerations in Adaptive Designs
• Type: Contributed
• Date/Time: Sunday, August 7, 2022 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Lingyun Liu* and Cyrus Mehta and Apurva Bhingare and Pralay Senchaudhuri
• Companies: Vertex Pharmaceuticals and Cytel and Bristol Myers Squibb and Cytel Corporation
• Keywords: Adaptive Design; Promising Zone ; Interim Analysis; Type 1 Error ; Sample Size Re-estimation

Abstract

The statistical methodology for adaptive designs has developed and matured over the past two decades. Although it is possible to modify the sample size of an on-going trial adaptively without inflating its type-1 error, the question remains whether there is an optimal way for doing so. We develop optimal decision rules for sample size re-estimation in two-stage adaptive group sequential trials. It is difficult for the sponsors to make the up-front commitment to adequately power a study to detect the smallest clinically meaningful treatment effect. It is easier to justify sample size increase if the interim data enter a so-called promising zone that ensures with high probability that the trial will succeed. We have considered promising zone designs that optimize unconditional power and promising zone designs that optimize conditional power. Where there is reluctance to base the sample size re-estimation rule on the smallest clinically meaningful treatment effect, we propose a Bayesian option where a prior distribution is assigned to the unknown treatment effect, which is integrated out of the objective function with respect to its posterior distribution at the interim analysis.