Randomization and External Controls - When to Use What in a Drug Development Landscape


Using synthetic control arms may offer a potentially faster alternative to other trial designs, e.g. RCTs. We will review the benefits of randomization, comment on the recent surge of single-arm trials in oncology, provide an overview over methodology to deal with non-randomized comparisons such as propensity scores, will discuss dynamic borrowing from joint control arms, and will initiate a discussion on where the place of non- or adaptively randomized comparisons can be in the drug development landscape.

Roche RxDx summit (invited talk)
Munich, Germany