Multiplicity Adjustment in Two Dose-Response Study Examples

• Activity Number: 620 - Dose-Finding for Monotherapy and Combination Therapy in Oncology and Other Complex Studies
• Type: Contributed
• Date/Time: Thursday, August 3, 2017 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Li He* and Joseph Heyse
• Companies: Merck Research Laboratories and Merck & Co., Inc.
• Keywords: Dose-response Studies ; Multiple Scalings ; FWER ; Correlation

Abstract

We consider adjusting multiplicity by controlling the familywise error rate (FWER) in two dose-response study examples where test statistics arise from count data. In dose-response studies, it is often necessary to consider multiple dose scalings in order to explore different dose response relationships. Significance of dose-response trend corresponding to each of the dose scalings is then evaluated simultaneously. In light of correlation among the different scalings, we adjust multiplicity based on the exact joint distribution of trend scores. We explore the connection of this method to the multiple comparisons and modeling techniques in dose-response studies. In the second example, we explore multiplicity adjustment that incorporates the correlation among the test statistics in the scenario where single doses are compared to the common control.