This talk investigated the statistical design features of a breast cancer risk reduction Bayesian adaptive trial. Most adaptive designs have interim analyses that allow for early stopping, typically utilizing only the primary endpoint. A drawback to this approach is that the study may not have enough data for adequate comparisons of a single, key secondary endpoint. In this talk, we investigate a trial design that stops early only if a criterion is met for primary and secondary endpoints. The approach focuses the final analysis on the primary endpoint but ensures adequate data for the secondary analysis. We presented operating characteristics including power, trial duration, and type I error rate, and discussed the value and risks of modeling Bayesian adaptive designs with primary and secondary endpoints, comparing against alternative designs. Our approach balances trial speed and the need for information on the single, key secondary endpoint.
Prof. Byron Gajewski’s statistical methodological research interests center around Bayesian data analysis and its application to cancer prevention, emergency medicine, neurology, nursing, health professions, and other related fields. He is co-director of the Biostatistics & Informatics Shared Resource at the University of Kansas Cancer Center. He is an active faculty member in the Biostatistics Ph.D. program teaching and mentoring.
