A calibrated power prior approach to borrow information from historical data with applications to biosimilar clinical trials

Dr. Haitao Pan

Haitao Pan, PhD

This article proposes a Bayesian adaptive design for trials to evaluate biosimilar products. To take advantage of the abundant historical data on the efficacy of the reference product that is typically available at the time that a biosimilar product is developed, the paper proposes the calibrated power prior, which allows the design to borrow information adaptively from the historical data according to the congruence between the historical data and the new data collected from the current trial. A new measure, the Bayesian biosimilarity index, is also proposed for measuring the similarity between the biosimilar product and the reference product. During the trial, we evaluate the Bayesian biosimilarity index in a group-sequential manner on the basis of the accumulating interim data and stop the trial early when there is enough information to conclude or reject the similarity. Extensive simulation studies show that the design proposed has higher power than do traditional designs. The proposed method was used to design a biosimilar trial for treating rheumatoid arthritis.

Full Citation

Pan H, Yuan Y, Xia J. A calibrated power prior approach to borrow information from historical data with application to biosimilar clinical trials. Journal of the Royal Statistical Society: Series C (Applied Statistics), 66(5),:979-996, 2017.