Bayesian Dose Escalation in Oncology Phase I Trials: Overview, Regulatory Implications, and S/Ws

• Activity Number: 636
• Type: Topic Contributed
• Date/Time: Thursday, August 7, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Shibao Feng*+ and Tom Parke
• Companies: Onyx Pharmaceuticals and Tessella
• Keywords: N-CRM ; Two Parameter Logistic Regression ; Phase I ; Oncoloy ; CRM ; MTD

Abstract

Bayesian two parameter logistic regression model based dose finding method has gained popularity since the publication of Neuschwander et al (2008). The method tends to incur fewer toxic events, and yield more accurate estimate for maximum tolerable dose (MTD) as compared to the standard Phase I 3+3 dose escalation design. It also overcomes the limitations of original continual reassessment method (CRM) for finding the MTD. In this talk, we will review the Bayesian method and CRM for assessing MTD in Oncology phase I clinical trials. CRM examples from literatures will be examined. Regulatory implications in applying the Bayesian method will also be discussed. We will further discuss statistical software (FACTS and R BCRM package) currently available for the Bayesian method implementation.