Application of gamma process to two-agent combinations with delayed toxicity. Issue 1 (2nd January 2021)

• Record Type: Journal Article
• Authors: Yada, Shinjo; Hamada, Chikuma
• Is Part Of: Communications in statistics. Volume 50:Issue 1(2021)
• Page Start: 153
• Page End: 163
• Publication Date: 2021-01-02
• Subjects: Missing data -- Data augmentation algorithm -- Gamma process -- Delayed toxicity -- Phase I clinical trial; Mathematical statistics -- Periodicals; Mathematical statistics -- Data processing -- Periodicals; Digital computer simulation -- Periodicals; 519.5
• DOI: 10.1080/03610918.2018.1554107
• Languages: English
• ISSNs: 0361-0918

Abstract

Abstract: Phase I trials investigate the drugs used on humans for the first time. In cancer treatment, drug safety is examined based on the toxicity. Herein, we propose a dose-finding method for oncological phase I drug combinations with delayed toxicity. The toxicity data of the patients who do not complete the follow-up period and do not experience toxicity are considered missing data. We use data augmentation to impute missing data while modeling the time required to experience toxicity using a hazard function, modeled using a gamma process. Simulation results demonstrate that the proposed method shortens trial durations without affecting performance.