Abstract 764: Multiple testing in a phase II trial design for the detection of tumor subtypes

Abstract

Listen

Translate article

Abstract Tumor subtypes may be especially responsive to a particular treatment. Phase II designs that incorporate tumor subtypes typically require prospective tumor subtyping and stratified patient enrollment. These features are cumbersome, discourage patient and investigator participation, and select for patients with relatively indolent disease who can wait for subtyping results. Previously we described a two stage phase II design that permits tumor subtyping at the end of stage 1 and that does not require stratified enrollment (Clin Cancer Res 2011;17:5538). In this design results from stage 1 are used to select an appropriate stage 2 objective. The stage 2 objective may be either to estimate the response rate of either the tumor or a subtype or a formal test of the hypothesis that the response rate for a subtype is greater than the overall response rate. Previously, although our design accommodated multiple subtypes in stage 1, it allowed further testing of only a single subtype in stage 2. Here we expand upon the previous design by describing a method to evaluate multiple tumor subtypes in stage 2 with appropriate adjustments for multiple testing. As there are practical limits on sample size, these adjustments tend to reduce power. We identify strategies to limit loss of power by taking advantage of tumor biology. For example, when tumor subtypes are mutually exclusive or nested, some comparisons can be made without further “alpha spending”. We explore the promise of our approach for the efficient identification of clinically significant tumor subtypes more responsive to a specific targeted therapy. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 764. doi:1538-7445.AM2012-764