Dose-Finding Strategies in Combination Therapy

• Activity Number: 416
• Type: Topic Contributed
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Priyam Mitra* and Helen Zhou and Yun Shen
• Companies: Rutgers University and Bristol-Myers Squibb and Bristol-Myers Squibb
• Keywords: Drug Combination ; Cancer Phase I-II Trials ; Bayesian Hierarchical Models ; Adaptive Design

Abstract

The early phase clinical studies in oncology drug development are focused on the toxicity and sometimes efficacy of a new treatment or a combination of treatments. Many statistical methodologies have been developed to determine the MTD (maximum tolerated dose)in single drug treatment. However, this problem could be more complicated in combination therapy where MTD may not be unique, as using multiple treatments give rise to a dose toxicity surface where multiple MTDs may exist. In this paper, we explore the combination of two treatments using Bayesian hierarchical model. We study the properties of varieties of copula models to make sure the copula model covers all general combination cases including additive and synergistic effects. In a typical drug combination trial, the toxicity profile of each individual treatment may already been well studied before the combination study. We utilize these prior information to develop our combination model framework. The theoretic framework is extended to combination of more than two treatments and a combination of ordinal toxicity measures. A simulation study as well as a case study based on a combination oncology trial results is presented.