A hybrid Bayesian-frequentist predictive design for monitoring multi-stage clinical trials

Abstract

In this article, we propose a hybrid-Bayesian frequentist approach using a Bayesian sequential prediction of the index of satisfaction. For interim analysis that addresses prediction hypothesis, such as futility monitoring with delayed outcomes, the prediction of satisfaction properly accounts for the amount of data remaining to be observed in a clinical trial and has the flexibility to incorporate additional information via auxiliary variables. The prediction of satisfaction design guarantees the type I error rate and does not require intensive computation or comprehensive simulation. The design is retrospectively applied to a lung cancer clinical trial.

KEYWORDS:

• Bayesian prediction
• interim monitoring
• prediction of satisfaction
• stochastic curtailment
• stopping rule

SUBJECT CLASSIFICATIONS:

• 62K05
• 62L10
• 62C10
• 62F15

Acknowledgments

We thank Zougab Nabil, professor at the University of Tizi-Ouzou, for his help. We are also extremely grateful to the Associate Editor and the anonymous referees for providing insights regarding the behavior of the proposed design and for their constructive comments.