Bayesian adaptive designs in a breast cancer trial with a delayed binary outcome

Background Clinical trials using Bayesian adaptive designs can be more efficient than those using traditional fixed designs, but there is a vast range of possible design approaches described in literature. We performed virtual re-executions of a breast cancer clinical trial with a delayed binary outcome to demonstrate what would have happened if the trial had used various Bayesian adaptive designs instead, and to explore which specific designs yielded the most benefits. Methods We retrospectively “re-executed” a randomised controlled trial that compared two chemotherapy regimens for women with metastatic breast cancer (ANZ 8614) using Bayesian adaptive designs. We used computer simulations to estimate the power and sample sizes of a large number of different candidate designs and shortlisted designs with the either highest power or the lowest average sample size. Using the real-world data, we explored what would have happened had ANZ 8614 been conducted using these shortlisted designs. Results Adaptive designs that prioritised small sample size reduced the average sample size by up to 50% when there was no clinical effect and by up to 21% at the target clinical effect. Adaptive designs that prioritised high power yielded up to an absolute increase of 6.5% in power without a corresponding increase in type I error. The performance of the adaptive designs when applied to the real-world ANZ 8614 data was consistent with the simulations. Conclusion Bayesian adaptive designs improved power or lowered the average sample size substantially when applied to this data set. When designing new oncology trials, researchers should consider whether a Bayesian adaptive design may be beneficial.