EU-PEARL’s Elias Laurin Meyer was awarded the Arthur-Linder-Prize for developing simulations that will help design platform trials
Researcher Elias Laurin Meyer, a biostatistician at the Medical University of Vienna who is involved in the IMI-funded EU-PEARL project, received the award for his paper “Decision rules for identifying combination therapies in open-entry, randomized controlled platform trials”. The goal of platform trials generally is to test as many investigational treatments as possible over the shortest duration. Meyer and his colleagues investigated platform trial design in the context of combination therapies, in which one of the two therapies a common, backbone monotherapy.
Instead of investigating each combination of therapies in an individual trial, a platform trial allows information on the common backbone monotherapy and standard of care to be shared across the different groups in the trial. This means that the efficacy of therapies can be evaluated faster, and fewer patients are allocated to standard of care. However, there are still many unanswered questions about how to design platform trials - such as the impact of decision rules (the algorithm that leads to advancing or stopping a compound) and exact method of data sharing across different groups, among others. In platform trials, defining operating characteristics is not straightforward because the operating characteristics of standard clinical trials do not necessarily apply.
For a platform trial comparing combination therapies to different monotherapies and standard-of-care, the researchers defined a set of error rates and operating characteristics and then used these to compare a set of design parameters and assumptions using simulations. When setting up the simulations, they tried to make the trajectories as realistic as possible, e.g. in case one compound is found to be superior to standard-of-care, it could replace standard-of-care in future cohorts.
“Our results indicate that the method of data sharing, exact specification of decision rules and quality of the biomarker used to make interim decisions all strongly contribute to the operating characteristics of the platform trial. Together with the potential flexibility and complexity of a platform trial, which also impact the achieved operating characteristics, this implies that utmost care needs to be given to evaluation of different assumptions and design parameters at the design stage,” the paper states.
“From a statistical point of view,” Meyer told the EU-PEARL project, “platform trials constitute a new paradigm of clinical trial design, which in turn requires the development and/or re-calibration of statistical tools, since methods, software and quality criteria of more traditional trial designs might not be applicable.”
“My contributions (in EU-PEARL) will be mainly in aiding development of new statistical methods and software for the simulation of complex platform trials. Ideally, the tools and software provided will aid clinical trial design and help shed light on the impact of certain design choices and assumptions on the operating characteristics of the trial. This, for example, includes the required number of patients in the trial or the required number of patients on placebo/control treatment arms.”
“One goal is to find ways to reduce these numbers, leading to fewer patients treated with inefficacious/old treatments and faster decisions on the new experimental treatments, while maintaining statistical integrity of the design.”
The paper was drafted by Elias Laurin Meyer together with his co-authors Peter Mesenbrink, Cornelia Dunger-Baldauf, Ekkehard Glimm, Yuhan Li und his PhD supervisor Franz König. The prestigious Arthur-Linder-Prize is awarded every two years by the Austrian-Swiss Region of the International Biometric Society (IBS-ROeS) to support young scientists in honor of Prof. Arthur Linder for a research paper in the field of biometrics.
Article on the EU-PEARL website
The WHO’s COVID-19 clinical trial is ‘adaptive’. What does that mean?