Seminars and Events - Pharm Stat/Eli Lilly Seminar

Statistical Applications in Pharmaceutical Research and Development
Department of Statistics
Purdue University

Monday, October 1, 2001
4:30 PM in LAEB B222

Dr. Doug Faries
Eli Lilly & Company

will speak on

Statistical Issues in Phase II Trials: Reaching a Product Decision

Abstract

The goal of Phase II trials is to provide sufficient data for making a decision whether to invest in a Phase III clinical trial program. Conducting Phase III involves a tremendous expense and consumes the bulk of limited research dollars and resources. Entering Phase III on limited data results in a high risk for failure along with significant losses in dollars and resources that could have been applied to developing other compounds. However, competitive pressures result in the need to start Phase III trials as soon as possible and delays in Phase III trials mean delays in getting improved medicines to patients.

This seminar will review the process, issues, and challenges facing a pharmaceutical company during Phase II trials and the statisticians' role in addressing these issues. The process of reaching a decision to enter Phase III begins with development of critical success factors -- that is, a summary of all key characteristics the compound must possess in order to be a viable candidate for Phase III trials. As in all phases of drug development, appropriate interpretation of data and communication of results are vital statistical contributions. In addition, it is important to have statistical involvement in the planning stages of developing critical success factors, decision phase clinical plans, and individual study protocols. This includes appropriately powering the studies such that all critical success factors are addressed, performing timely interim analyses, and the potential use of surrogate variables. In summary, statisticians have an opportunity to play a critical role in making correct decisions on Phase II trials and making that decision at the most appropriate time.