A Novel Design for Decision Rules Based on Statistical Testing Strategies in a Definitive Go/No-Go Clinical Study

• Activity Number: 647
• Type: Contributed
• Date/Time: Thursday, August 8, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Ming Zhou and Larry Z. Shen*+
• Companies: Amylin Pharma., LLC (a Bristol-Myers Squibb Co.) and Amylin Pharma., LLC (a Bristol-Myers Squibb Co.)
• Keywords: Statistical Testing ; Decision Making ; Wilson Score ; Binary Endpoint ; Sample Size

Abstract

In drug development a denitive go/no-go clinical study is sometimes conducted before major investment is made to advance the development into the next stage. Such a study sometimes focuses on a key binary endpoint such as a particular adverse event or failure of a delivery device. However, such a study is often constrained by the limited budget and scope, which limits the Sponsor's ability to draw clear conclusions. This paper proposes a novel design that uses Wilson condence intervals to devise statistical testing strategies that allow the decision makers to arrive at certain denite conclusions within the constraint of the scope of the study.