Bayesian adaptive decision-theoretic designs for multi-arm multi-stage clinical trials

A comparison of alternative ranking methods in two-stage clinical trials with multiple interventions: An application to the anxiolysis for laceration repair in children trial

BACKGROUND/AIMS

Multi-arm, multi-stage trials frequently include a standard care to which all interventions are compared. This may increase costs and hinders comparisons among the experimental arms. Furthermore, the standard care may not be evident, particularly when there is a large variation in standard practice. Thus, we aimed to develop an adaptive clinical trial that drops ineffective interventions following an interim analysis before selecting the best intervention at the final stage without requiring a standard care.

METHODS

We used Bayesian methods to develop a multi-arm, two-stage adaptive trial and evaluated two different methods for ranking interventions, the probability that each intervention was optimal (Pbest) and using the surface under the cumulative ranking curve (SUCRA), at both the interim and final analysis. The proposed trial design determines the maximum sample size for each intervention using the Average Length Criteria. The interim analysis takes place at approximately half the pre-specified maximum sample size and aims to drop interventions for futility if either Pbest or the SUCRA is below a pre-specified threshold. The final analysis compares all remaining interventions at the maximum sample size to conclude superiority based on either Pbest or the SUCRA. The two ranking methods were compared across 12 scenarios that vary the number of interventions and the assumed differences between the interventions. The thresholds for futility and superiority were chosen to control type 1 error, and then the predictive power and expected sample size were evaluated across scenarios. A trial comparing three interventions that aim to reduce anxiety for children undergoing a laceration repair in the emergency department was then designed, known as the Anxiolysis for Laceration Repair in Children Trial (ALICE) trial.

RESULTS

As the number of interventions increases, the SUCRA results in a higher predictive power compared with Pbest. Using Pbest results in a lower expected sample size when there is an effective intervention. Using the Average Length Criterion, the ALICE trial has a maximum sample size for each arm of 100 patients. This sample size results in a 86% and 85% predictive power using Pbest and the SUCRA, respectively. Thus, we chose Pbest as the ranking method for the ALICE trial.

CONCLUSION

Bayesian ranking methods can be used in multi-arm, multi-stage trials with no clear control intervention. When more interventions are included, the SUCRA results in a higher power than Pbest. Future work should consider whether other ranking methods may also be relevant for clinical trial design.

A Bayesian-adaptive decision-theoretic approach can reduce the sample sizes for multiarm exercise oncology trials

OBJECTIVES

Adaptive designs may reduce trial sample sizes and costs. This study illustrates a Bayesian-adaptive decision-theoretic design applied to a multiarm exercise oncology trial.

STUDY DESIGN AND SETTING

In the Physical exercise during Adjuvant Chemotherapy Effectiveness Study (PACES) trial, 230 breast cancer patients receiving chemotherapy were randomized to supervised resistance and aerobic exercise (OnTrack), home-based physical activity (OncoMove) or usual care (UC). Data were reanalyzed as an adaptive trial using both Bayesian decision-theoretic and a frequentist group-sequential approach incorporating interim analyses after every 36 patients. Endpoint was chemotherapy treatment modifications (any vs. none). Bayesian analyses were performed for different continuation thresholds and settings with and without arm dropping and both in a 'pick-the-winner' and a 'pick-all-treatments-superior-to-control' setting.

RESULTS

Treatment modifications occurred in 34% of patients in UC and OncoMove vs. 12% in OnTrack (P = 0.002). Using a Bayesian-adaptive decision-theoretic design, OnTrack was identified as most effective after 72 patients in the 'pick-the-winner' setting and after 72-180 patients in the 'pick-all-treatments-superior-to-control' setting. In a frequentist setting, the trial would have been stopped after 180 patients, and the proportion of patients with treatment modifications was significantly lower for OnTrack than UC.

CONCLUSION

A Bayesian-adaptive decision-theoretic approach substantially reduced the sample size required for this three-arm exercise trial, especially in the 'pick-the-winner' setting.

Being Bayesian in the 2020s: opportunities and challenges in the practice of modern applied Bayesian statistics

Building on a strong foundation of philosophy, theory, methods and computation over the past three decades, Bayesian approaches are now an integral part of the toolkit for most statisticians and data scientists. Whether they are dedicated Bayesians or opportunistic users, applied professionals can now reap many of the benefits afforded by the Bayesian paradigm. In this paper, we touch on six modern opportunities and challenges in applied Bayesian statistics: intelligent data collection, new data sources, federated analysis, inference for implicit models, model transfer and purposeful software products. This article is part of the theme issue 'Bayesian inference: challenges, perspectives, and prospects'.

Adaptive treatment allocation and selection in multi-arm clinical trials: a Bayesian perspective

Background

Adaptive designs offer added flexibility in the execution of clinical trials, including the possibilities of allocating more patients to the treatments that turned out more successful, and early stopping due to either declared success or futility. Commonly applied adaptive designs, such as group sequential methods, are based on the frequentist paradigm and on ideas from statistical significance testing. Interim checks during the trial will have the effect of inflating the Type 1 error rate, or, if this rate is controlled and kept fixed, lowering the power.

Results

The purpose of the paper is to demonstrate the usefulness of the Bayesian approach in the design and in the actual running of randomized clinical trials during phase II and III. This approach is based on comparing the performance of the different treatment arms in terms of the respective joint posterior probabilities evaluated sequentially from the accruing outcome data, and then taking a control action if such posterior probabilities fall below a pre-specified critical threshold value. Two types of actions are considered: treatment allocation, putting on hold at least temporarily further accrual of patients to a treatment arm, and treatment selection, removing an arm from the trial permanently. The main development in the paper is in terms of binary outcomes, but extensions for handling time-to-event data, including data from vaccine trials, are also discussed. The performance of the proposed methodology is tested in extensive simulation experiments, with numerical results and graphical illustrations documented in a Supplement to the main text. As a companion to this paper, an implementation of the methods is provided in the form of a freely available R package ’barts’.

Conclusion

The proposed methods for trial design provide an attractive alternative to their frequentist counterparts.