Bayesian Two-Stage Single-to-Double Arm Design in Phase II Clinical Trials

• Activity Number: 522
• Type: Topic Contributed
• Date/Time: Wednesday, August 7, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Guosheng Yin*+
• Companies: University of Hong Kong
• Keywords: Bayesian posterior error rates ; Expected sample size ; Phase II clinical trial ; Single-to-double arm comparison ; Two-stage design

Abstract

Conventional phase II clinical trial design uses either a single-arm or two-arm comparison to examine the drug's therapeutic effects. Both single-arm and multi-arm evaluations have their own merits; for example, single-arm phase II trials may require a smaller sample size, while multi-arm trials may lead to more objective comparison. In general, patients tend to be more willing to participate in a trial if they know that they will receive the new (experimental) drug. To bridge the single- and two-arm schemes in a phase II clinical trial, we propose a Bayesian two-stage design, in which stage 1 takes a single-arm comparison of the experimental drug with the standard response rate and stage 2 imposes a multi-arm comparison of the experimental drug with the control arm. The design is calibrated using the error rates from both frequentist and Bayesian perspectives. We conduct extensive simulations on the calculations of these error rates to demonstrate and examine the operational characteristics of our proposed method.