Dataset Information

A Bayesian Sequential Design for Clinical Trials with Time-to-Event Outcomes.

ABSTRACT: There is increasing interest in Bayesian group sequential design because of its potential to improve efficiency in clinical trials, to shorten drug development time, and to enhance statistical inference precision without undermining the clinical trial's integrity or validity. We propose a Bayesian sequential design for clinical trials with time-to-event outcomes and use alpha spending functions to control the overall type I error rate. Bayes factor is adapted for decision-making at interim analyses. Algorithms are presented to make decision rules and to calculate power of the proposed tests. Sensitivity analysis is implemented to evaluate the impact of different choices of prior parameters on choosing critical values. The power of tests, the expected event size of the proposed design, and the quality of estimators are studied through simulations, and compared with the frequentist group sequential design. Simulations show that at fixed total number of events, the proposed design can achieve greater power and require smaller expected event size when appropriate priors are chosen, compared with the frequentist group sequential design. The feasibility of the proposed design is further illustrated on a real data set.

SUBMITTER: Zhu L

PROVIDER: S-EPMC7100880 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A Bayesian Sequential Design for Clinical Trials with Time-to-Event Outcomes.

Zhu Lin L Yu Qingzhao Q Mercante Donald E DE

Statistics in biopharmaceutical research 20190722 4

There is increasing interest in Bayesian group sequential design because of its potential to improve efficiency in clinical trials, to shorten drug development time, and to enhance statistical inference precision without undermining the clinical trial's integrity or validity. We propose a Bayesian sequential design for clinical trials with time-to-event outcomes and use alpha spending functions to control the overall type I error rate. Bayes factor is adapted for decision-making at interim analy ...[more]

PMID: 32226580

Similar Datasets

Project description:Currently, a growing placebo response rate has been observed in clinical trials for antidepressant drugs, a phenomenon that has made it increasingly difficult to demonstrate efficacy. The sequential parallel comparison design (SPCD) is a clinical trial design that was proposed to address this issue. The SPCD theoretically has the potential to reduce the sample-size requirement for a clinical trial and to simultaneously enrich the study population to be less responsive to the placebo.Because the basic SPCD already reduces the placebo response by removing placebo responders between the first and second phases of a trial, the purpose of this study was to examine whether we can further improve the efficiency of the basic SPCD and whether we can do so when the projected underlying drug and placebo response rates differ considerably from the actual ones.Three adaptive designs that used interim analyses to readjust the length of study duration for individual patients were tested to reduce the sample-size requirement or increase the statistical power of the SPCD. Various simulations of clinical trials using the SPCD with interim analyses were conducted to test these designs through calculations of empirical power.From the simulations, we found that the adaptive designs can recover unnecessary resources spent in the traditional SPCD trial format with overestimated initial sample sizes and provide moderate gains in power. Under the first design, results showed up to a 25% reduction in person-days, with most power losses below 5%. In the second design, results showed up to a 8% reduction in person-days with negligible loss of power. In the third design using sample-size re-estimation, up to 25% power was recovered from underestimated sample-size scenarios.Given the numerous possible test parameters that could have been chosen for the simulations, the study's results are limited to situations described by the parameters that were used and may not generalize to all possible scenarios. Furthermore, dropout of patients is not considered in this study.It is possible to make an already complex design such as the SPCD adaptive, and thus more efficient, potentially overcoming the problem of placebo response at lower cost. Ultimately, such a design may expedite the approval of future effective treatments.

Dataset Information

A Bayesian Sequential Design for Clinical Trials with Time-to-Event Outcomes.

Publications

A Bayesian Sequential Design for Clinical Trials with Time-to-Event Outcomes.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets