Each week, we will be publishing layman’s abstracts of new articles from our prestigious portfolio of journals in statistics. The aim is to highlight the latest research to a broader audience in an accessible format.

The article featured today is from Pharmaceutical Statistics, with the full article now available to read in Early View here.

Takahashi, A, Suzuki, T. Bayesian optimization design for dose‐finding based on toxicity and efficacy outcomes in phase I/II clinical trials. Pharmaceutical Statistics. 2020; 1– 18. https://doi.org/10.1002/pst.2085

In oncology phase I trials, the main goal is to identify a maximum tolerated dose under an assumption of monotonicity in dose–response relationships, whereas some investigational agents such as biologic agents require an optimal dose that satisfies an expected efficacy rate under a tolerable toxicity rate instead of a maximum tolerated dose. Unlike cytotoxic agents, the monotonicity is no longer applied to such agents that potentially draw unimodal or flat dose-efficacy curves. Along with that drug development for a variety of modes of action has increased, optimal-dose estimation designs have been receiving increased attention. This article proposes a Bayesian optimization design for identifying optimal doses in phase I/II trials. The proposed design models dose–response relationships nonparametrically, and its dose selection process considers uncertainties of posterior distributions. The operating characteristics of the proposed design were compared against those of three other designs through simulation studies. These include an expansion of Bayesian optimal interval design, the parametric model-based EffTox design, and the isotonic design. In simulations, the proposed design performed well and provided results that were more stable than those from the other designs, in terms of correct determinations of optimal doses.