Adaptive Penalized D-Optimal Designs for Dose Finding Based on Continuous Efficacy and Toxicity

作者：Padmanabhan S Krishna; Hsuan Francis; Dragalin Vladimir^*

来源：Statistics in Biopharmaceutical Research, 2010, 2(2): 182-198.

DOI：10.1198/sbr.2009.0053

• Adaptive dose finding
• Efficacy and toxicity outcomes
• Optimal clinical trial design
• Risk-benefit profile modeling

When a new drug is under development, a conventional dose-finding study involves learning about the dose-response curve in order to bring forward right doses of the drug to late-stage development. We propose an adaptive procedure for dose-finding in clinical trials in the presence of both efficacy and toxicity endpoints. We use the principles of optimal experimental designs for bivariate continuous endpoints.
However, instead of using the traditional D-optimal design, which favors collective ethics but neglects the individual ethics, we consider the penalized D-optimal design that achieves an appropriate balance between the efficient treatment of patients in the trial (by penalizing allocation of patients to ineffective or toxic doses) and the precise estimation of the model parameters to be used in the identification of the target dose. This is compared with the traditional fixed allocation design in terms of allocation of subjects and precision of the identified dose-response curve and selection of the target dose.

• 出版日期2010-5