Abstract:
|
Despite the development of innovative clinical trial designs, the standard 3+3 design continues to be the most common phase I dose finding trial design proposed by researchers to determine the optimal dose (MTD). Review criticisms of this design are increasing as awareness of the alternative designs grows. Unfortunately, as statisticians, it is sometimes difficult to convince the researcher to use the newer methods available. The goal of this study is to develop a persuasive argument to present to researchers to convince them to use a design other than the standard 3+3 design. The focus will be on trials with limited dose ranges (3-4 doses). Limited dose trials are common when an experimental drug is being tested in a new patient population or is being combined with other approved drugs. In this study, several phase I designs (standard 3+3, CRM, EWOC and BOIN) will be compared under various scenarios that preserve safety and maintain rapid accrual. Based on the simulation results, a "best" design, with minimal number of subjects being exposed to sub-therapeutic doses of the agent, will be recommended.
|