Statistical Methods for Adding New Arms to an Ongoing Response-Adaptive Trial

• Activity Number: 437
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Matteo Cellamare* and Lorenzo Trippa and Steffen Ventz
• Companies: and Dana-Farber Cancer Institute and Dana-Farber Cancer Institute
• Keywords: Response-adaptive design ; Clinical trials ; Bayesian design ; Randomization ; Adaptive designs

Abstract

Most designs for phase II clinical trial assume that all experimental treatments are available at the beginning of the trial. If a new experimental drug becomes available to be tested, a new trial has to be conducted. Only few recent studies include in the protocol the possibility of new additional arms after the study has already started. This can reduce the cost of drug development and speed up the process of screening drugs. Nevertheless, a methodological formalization of this type of designs has not been discussed in the literature. We propose a design to manage the dynamic input of new drugs throughout the course of the trial and combine the approach with Bayesian response-adaptive randomizations. The goal is to add new experimental drugs, available once the study has already enrolled a group of patients, and to test their effectiveness through the recruitment of a sufficient number of patients.