(Vienna, 11 August 2026) Computer simulations can help to improve the design of clinical trials during the planning phase. This is the conclusion reached by an international team of experts led by the Medical University of Vienna in a research article published in "Nature Reviews Drug Discovery". The authors demonstrate how simulations can be used to systematically test, compare and refine different trial designs in advance. At the same time, they set out recommendations on how such simulations should be conducted in a robust and transparent manner.
Moderne klinische Studien bedienen sich immer häufiger adaptiver Studiendesigns, Biomarker-gesteuerter Rekrutierung, Masterprotokolle und kontinuierlicher Datenmonitorings. Diese innovativen Ansätze können die Forschungsfortschritte beschleunigen und die Ergebnisse verbessern, führen jedoch auch zu einer zusätzlichen Komplexität bei der Planung und Durchführung der Studie. Ein internationales Expert:innenteam aus Statistik, klinischer Forschung, Arzneimittelentwicklung und Regulierung zeigt in der aktuellen Publikation, dass und wie Computersimulationen helfen können, verschiedene Studiendesigns unter realistischen Annahmen bereits vor Beginn einer Studie zu vergleichen und ihre Auswirkungen auf den Studienverlauf und -erfolg abzuschätzen. So lassen sich mögliche Schwächen früh erkennen und Studienpläne gezielt verbessern.
Modern clinical trials increasingly use adaptive trial designs, biomarker-guided recruitment, master protocols and continuous data monitoring. These innovative approaches can accelerate research and improve outcomes, but they also introduce additional complexity to both planning and conduct of the trial. In this latest publication, an international team of experts from the fields of statistics, clinical research, drug development and regulatory sciences demonstrates that computer simulations can help compare different trial designs under realistic assumptions even before a trial begins, and assess their impact on the course and success of the trial. This allows potential weaknesses to be identified at an early stage and trial protocols to be improved in a targeted manner.
In such simulations, the entire course of a clinical trial – from recruitment through randomisation and data generation to interim analyses and final decision-making – is modelled using virtual patients. In doing so, different assumptions – for example, regarding the efficacy of a treatment, the recruitment of trial participants or various trial procedures – can be played through. The aim is to assess which trial design will ultimately provide the most reliable answer to the research question before a single participant has been enrolled. "The more complex clinical trials become, the more important it is to understand their operating characteristics as comprehensively as possible before they start. Simulations make it possible to systematically examine different scenarios and make decisions about the trial design on a transparent basis," says Franz König from the Center for Medical Data Science at MedUni Vienna, the last author of the publication.
In addition to the benefits of simulations, the article also describes the prerequisites that must be met for their scientifically sound use. The authors recommend that simulations be conducted in a transparent and reproducible manner, that realistic scenarios be taken into account, and that results be presented in such a way that they are comprehensible to clinical researchers, regulatory authorities, funding bodies and patients alike. The aim is to establish a shared understanding of how a trial design is likely to perform under different conditions and what can be learned at the end. "Our publication aims to contribute to the development of study designs that are scientifically sound, ethically sound, operationally feasible and, ultimately, capable of providing patients with safe and effective therapies as efficiently as possible," says first author Elias Laurin Meyer (Berry Consultants and, at the time of the research, also the Center for Medical Data Science, MedUni Vienna), summarising the paper’s relevance.
Publikcation: Nature Reviews Drug Discovery
Design, simulate, refine: simulation-guided clinical trials for accelerated drug development
Elias Laurin Meyer, Tim Friede, Andrew P. Grieve, Christopher Jennison, Michael Krams, Elizabeth Lorenzi, Husseini K. Manji, Tobias Mielke, Jessica R. Overbey, Kert Viele, Marc K. Walton & Franz König
https://www.nature.com/articles/s41573-026-01481-9