657 – Clinical Trial Design V
Phase I Dose Finding in Clinical Trials---An Interface for Designs
Yunfei Wang, PhD
George Washington University and Children’s National Medical Center
Many dose finding methods have been published which can be categorized into two groups: nonparametric methods and parametric methods. We present brief review of these different designs to provide a general picture for those investigators or statisticians who will use these methods. The nonparametric traditional 3+3 design is most often used in Phase I clinical trial because it is simple and easy to use though advanced parametric methods outperform the traditional design. We developed an interface that would enable the investigators easily to use the advanced methods including continual reassessment methods (CRM), varieties of modified CRM methods such as escalation with overdose control (EWOC), two parameter logistic CRM together with option for 3+3 simulation, and other method such as modified toxicity probability interval (mPTI), to run simulation for a Phase I study design or to calculate the next dose for an ongoing Phase I study.