Commentary on “Improving the performance of Bayesian logistic regression model with overdose control in oncology dose‐finding studies”

Abstract

The Bayesian logistic regression model (BLRM) design is a variation of the continuous reassessment method (CRM). Due to the use of an excessively tight escalation with overdose control (EWOC) rule, BLRM has high tendency to underdose patients and low accuracy to identify the maximum tolerated dose (MTD). We congratulate Zhang, Chiang, and Wang-hereafter referred to as ZCW-for proposing elegant and practical approaches to addressing this important problem, and we thank the co-editor, Dr. Lisa McShane, for inviting our discussion.

Rejoinder: Improving the performance of Bayesian logistic regression model with overdose control in oncology dose‐finding studies

“…We are grateful to the discussants for their thoughtful comments: Drs. Ying Yuan and Yixuan Zhao 1 (hereafter YZ); Drs. Paul Frankel, Elizabeth Garrett‐Mayer, and Mark Krailo 2 (hereafter FGK); Drs.…”