A Bayesian three-parameter logistic model for early- and late-onset DLTs in oncology Phase I studies. Issue 2 (3rd March 2016)

• Record Type: Journal Article
• Authors: Zheng, Wei; Zhao, Yang; Lu, Yuefeng; Miao, Harry; Liu, Hengchang
• Is Part Of: Journal of biopharmaceutical statistics. Volume 26:Issue 2(2016)
• Page Start: 339
• Page End: 351
• Publication Date: 2016-03-03
• Subjects: Bayesian -- CRM -- late toxicity -- Phase I; Pharmacy -- Statistical methods -- Periodicals; Drugs -- Testing -- Statistical methods -- Periodicals; Biometry -- Periodicals; Biopharmaceutics -- Periodicals; Pharmacokinetics -- Periodicals; 615.19
• DOI: 10.1080/10543406.2014.1003433
• Languages: English
• ISSNs: 1054-3406

Abstract

ABSTRACT: We introduce a three-parameter logistic model to analyze the dose limiting toxicity (DLT) as a time-to-event endpoint in oncology Phase I trials. In the proposed model, patients are allowed to stay on trial without the constraint of a maximum follow-up time. Our model accommodates late-onset DLT as well as early-onset DLT, by both of which the dose recommendation is informed. A Bayesian approach is used to incorporate prior knowledge of the test treatment into dose recommendation. Simulation examples show that our proposed model has good operating characteristics in assessing the maximum tolerated dose (MTD).