Futility Monitoring with Conditional Power: A Rationale for the Choice of Theta Assumption Based on Type II Error Probability

• Activity Number: 128
• Type: Contributed
• Date/Time: Monday, August 4, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Thomas Dobbins*+
• Companies: Merck
• Keywords: group sequential test ; tests of futility ; type II error probability ; conditional power ; adaptive trial

Abstract

For a clinical trial incorporating a group sequential test that allows early stopping for efficacy or futility (GSTEF), the two main frequentist methods of evaluating futility are type II error spending and conditional power. Dobbins (2013) partitioned the total type II error probability of a K-group GSTEF into the sum of its component type II error probabilities of futility and efficacy. This partitioning is utilized to show that the conditional power method can be completely specified as a GSTEF using type II error spending and non-binding futility boundaries. In particular, we obtain the exact relationship between conditional power and the type II error probability of tests of futility of a K-group GSTEF. Consequently, the choice of theta assumption for conditional power can be linked directly to a unique type II error spending sequence for the tests of futility. We consider an example that illustrates the correspondence, and express the choice of theta in terms of published type II error spending functions. Thus, we provide a rationale for the choice of theta in the conditional power method based on the size of the corresponding type II error of the tests of futility.