Interim Treatment Selection Using Exact Binomial Distribution in Clinical Trials

• Activity Number: 505
• Type: Contributed
• Date/Time: Wednesday, August 12, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Bob Zhong* and Gordon Lan and Surya Mohanty and Jose Pinheiro and Sudhakar Rao and Kyle Wathen
• Companies: Johnson & Johnson and Johnson & Johnson and Johnson & Johnson and Johnson & Johnson and Janssen R&D and Johnson & Johnson
• Keywords: pick-a-winner ; dose selection ; binomial distribution

Abstract

We consider a study starting with two treatment groups and with two planned interim analyses for safety, futility, and dose selection. The inferior treatment group(s) may be dropped at both interim analyses due to safety or futility. A winning treatment may be selected at the first interim analysis due to superior efficacy. If two treatments pass both safety and futility threshold at the second interim analysis, one treatment has to be selected then. The selected treatment will continue to the end of the study. This design is based on the concepts of adaptation, winner selection and multiple comparisons. In a study with such a design, the probability of correct treatment selection has to be considered and the multiplicity due to treatment selection has to be adjusted at the final analysis. The treatment selection probability, type 1 error, and power is provided in this paper using the exact distribution.