Calibration of the Continual Reassessment Method
by
Shing M. Lee
Mailman School of Public Health, Columbia University, USA
Coauthors: Ying Kuen Cheung (Columbia University)

The Continual Reassessment Method (CRM) is an adaptive model-based design used to estimate the maximum tolerated dose (MTD) in dose finding clinical trials. The method has been shown to be consistent under certain conditions [1]. In situations where the CRM is not consistent, indifference intervals have been proposed as a way to evaluate the dose response models used with the CRM and the appropriateness of the “doses” selected for the experiment given the target probability of toxicity [2]. This talk will describe ways to select the scaled doses for experimentation given the target probability of toxicity, the dose-response model and the desired indifference interval around the true probability of toxicity.

Reference:

[1] Shen LZ, O’Quigley J. (1996) Consistency of the Continual Reassessment Method under model misspecification. Biometrika, 83, 395-405.

[2] Cheung YK, Chappell R. (2002) A simple technique to evaluate model sensitivity in the Continual Reassessment Method. Biometrics, 58, 671-674.

PDF

Date received: March 9, 2007