Online Program Home
My Program

Abstract Details

Activity Number: 123 - Binary and Ordinal Outcome Regression
Type: Contributed
Date/Time: Monday, July 30, 2018 : 8:30 AM to 10:20 AM
Sponsor: Biometrics Section
Abstract #330897 Presentation
Title: A Bayesian Logistic Model with Covariate to Identify Optimal Dose for Heterogeneous Population in Phase I Oncology Trial
Author(s): Xin Wei* and Michael Branson
Companies: Celgene Corporation and celgene corporation
Keywords: Oncology Clinical Trial; dose finding; MTD; Bayesian Logistic Regression; adaptive design; CRM

Phase I oncology FIH trial aims at identifying the maximum tolerated dose (MTD) for patients that can be used in later trial. Previous work developed a Bayesian logistic model based continuous reassessment method (BLRM) that improves the precision of MTD identification. Our work demonstrates the further improvement of optimal dose finding for a heterogeneous population by modeling the patient risk factor as covariate. In addition, we use simulation to show that the non-parsimonious model does not compromise the ability of model to pick the MTD and other operating characteristics in the context of 3-parameter logistic regression for dose/toxicity relationship.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program