Abstract:
|
Biomarker-directed targeted designs have been developed recently for pharmaceutical development aimed at patient subpopulation with a specific etiology. To integrate multistage testing into targeted designs enhances its flexibility by enabling sequential monitoring and stochastic curtailment. We propose a multistage adaptive design for targeted trials with either normally distributed or binary endpoint. The study has demonstrated that the design improves study efficiency, information accumulation and conditional power, compared with its untargeted counterpart. Furthermore, the study has indicated that biomarker sensitivity and specificity influences level of heterogeneity of targeted study population, and therefore impacts trial efficiency, power, information accruement and stochastic curtailment. When performance of a biomarker is imperfect, conditional/predictive power at an earlier stage may be over-estimated, resulting invalid early stopping decision. Thus biomarker performance needs to considered in statistical planning. In summary, the design provides targeted trials with flexibility in multistage testing and early stopping while retaining the rigor of the study design.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.