Adaptive platform trials provide a framework to simultaneously study multiple treatments in adisease. They are multi-arm trials in which interventions can enter as they become available and leave for futility or efficacy based on interim analyses or a final analysis. Theattractiveness of platform trials compared to separate parallel-group trials is not only due tooperational aspects as a shared trial infrastructure and more efficient patient recruitment, but also results from the possibility to share control groups, to efficiently identify non-efficacious treatments, and to allow direct comparisons between experimental treatment arms.

However, the flexibility of the framework comes with challenges for statistical inference and the interpretation of trial results. The main challenges concern the adaptivity of platform trials (as decisions on adding or dropping arms cannot be fully pre-specified and may have an impact on current trial arms), multiplicity issues (due to multiple interventions, subgroups and interim analyses) and the use of shared controls (concurrent or non-concurrent controls, blinding for interventions that have different routes of administration, change in the standard of care). We will provide an overview of statistical approaches that have recently been proposed to address these issues.

Underlying papers:

Roig, Marta Bofill, et al. "Treatment-control comparisons in platform trials including non-concurrent controls." arXiv preprint arXiv:2407.13546 (2024). (To appear in: Statistics in Biopharmaceutical Research).
Bofill Roig, Marta, et al. "Optimal allocation strategies in platform trials with continuous endpoints."Statistical Methods in Medical Research 33.5 (2024): 858-874.
Krotka, Pavla, et al. "Statistical modeling to adjust for time trends in adaptive platform trials utilizing non-concurrent controls." arXiv preprint arXiv:2403.14348 (2024).
Zehetmayer, Sonja, Martin Posch, and Franz Koenig. "Online control of the False Discovery Rate in group-sequential platform trials." Statistical Methods in Medical Research 31.12 (2022): 2470-2485.

Personal website of Martin Posch