A Free Gift: An Adaptive Strategy in a Single Arm Trial Using an Exact Test Through the Binomial Distribution
View Presentation View Presentation
*Jin Wang, Abbott Vascular 


For medical product development within the same generation, usually a small incremental change is made. Therefore, single-arm trial designs are commonly implemented to test the performance of the new product against an Objective Performance Criterion (OPC), which is a fixed value developed by historical data or literature review. When the primary endpoint is an event rate, to be conservative, regulatory agencies usually require the use of an exact test through binomial distribution when the sample size is modest. Utilizing the idea by Proschan and Hunsberger, this article proves that it can be a free gift to add an adaptive component to a regular fixed sample size design so that when the result in the first stage is marginal, the adaptive feature can be activated without any penalty. A hypothetical example is used to illustrate the application of this method.