Each week, we publish lay 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 here.
, , , et al. Evaluating hybrid controls methodology in early-phase oncology trials: A simulation study based on the MORPHEUS-UC trial. Pharmaceutical Statistics. 2023; 1–15. doi:10.1002/pst.2336
Drug development involves four phases. The main goal of the first two phases is to explore the safety of the novel drug and as proof of concept for its efficacy. In such trials, the number of subjects both being assigned to the treatment or control involved in these phases is usually small, for example, around 20 in total. The statistical estimates derived from such a dataset are thus unstable. To address this issue, one can combine the data that contains subjects who received controls from historical trials. However, the characteristics of the population from historical trials may differ from the current trial, which should be considered in estimating the effects of the treatment. When borrowing from two historical trials, the heterogeneity that arises from the two historical trials should also be taken into account. This work provides various statistical methods that specifically address the two layers of heterogeneity. It next evaluates the methods via a comprehensive simulation study, which is based on the real data from a Roche MORPHEUS-UC (urothelial carcinoma) trial. The results are informative to the choice of methods used in the early phase oncology trials in a similar setting. In addition, the simulation serves as a template of evaluating those methods in different settings.
More Details