Stepwise CRM based on a setup rule for Phase I clinical trials
by
Jian Tao
Key Laboratory for Applied Statistics of MOE (KLAS), Northeast Normal University, Changchun, Jilin, P.R. China, 130024
Coauthors: Ning-Zhong Shi

The clinical trial of experimental drug is normally done in three phases. A phase I clinical trial is primarily concerned with assessing the drug's safety, and one of the main objectives is to estimate the maximum tolerated dose of the dose. The continual reassessment method (CRM) is a Bayesian procedure for identifying the maximum tolerated dose. The CRM has been shown to have significant advantages over the traditional Up-and-Down method. However, its acceptance may be in doubt, the reason is that the enrolled subjects may experiencing excess toxicity. To overcome this disadvantage, an improved stepwise CRM based on a setup rule is proposed. Comparing with two other methods, the stepwise CRM reduces the total number of enrolled subjects and avoids severe toxicity exposure at higher dose levels. As a result, it is expected that the new method is acceptable to clinical trials.

Date received: February 10, 2007