Article
Lots of tools, lots of designs – a proposal on how to create a basket trial via a modular construction kit
Search Medline for
Authors
-
Moritz Pohl
- Institute of Medical Biometry and Informatics, University of Heidelberg, Heidelberg, Germany
-
Johannes Krisam
- Institute of Medical Biometry and Informatics, University of Heidelberg, Heidelberg, Germany
-
Meinhard Kieser
- Institute of Medical Biometry and Informatics, University of Heidelberg, Heidelberg, Germany
| Published: | February 26, 2021 |
|---|
Outline
Text
Throughout the last decade, basket trials have become a virulent research field for medical statistics. Several different designs have been proposed utilising numerous different statistical techniques with a plethora of variable parameters, which consequently resulted in a broad field of tools in various combinations. The current literature contains designs which apply frequentist techniques (e.g. statistical tests, Simon two-stage designs) [1], [2], Bayesian methods (e.g. hierarchical models, similarity measures) [3], [4] and mixtures of both [5].
The purpose of this talk is to give an overview of fundamental mathematical and statistical tools which are applied in basket trial designs. The intentions behind each tool will be presented as well as their advantages and shortcomings. Moreover, we will present a categorization of the tools and how they can be simplified. This approach results in a modular construction kit which allows practitioners (i.e. statisticians planning trials based on the underlying medical challenges) to create new basket trials in an effective and plausible way. The kit contains the different proposed tools for interim assessment (futility and/or efficacy), the techniques to share information between baskets, and tools for the final evaluation of the trial. This opens also the option to combine, e.g., interim tools from one proposed design with sharing techniques of another design jointly with a final analysis strategy from a third design.
The construction kit will bring clarity to the evolving field of basket trial designs, it will simplify similar tools to one tool category, and will consequently offer researchers the opportunity to design their own basket trial using different components from the kit. This construction kit will hopefully motivate medical researchers to design more tailored basket trials for the assessment of efficacy of targeted treatments and might contribute to the required evolution of clinical trials in the new era of personalised medicine.
The authors declare that they have no competing interests.
The authors declare that an ethics committee vote is not required.
References
- 1.
- Cunanan K M, Iasonos A, Shen R, Begg C B, Gönen M. An efficient basket trial design. Statistics in Medicine. 2017; 36(10): 1568-1579.
- 2.
- Zhou H, Liu F, Wu C, Rubin E, Giranda V, Chen C. Optimal two-stage designs for exploratory basket trials. Contemporary Clinical Trials. 2019; 85: 105807. DOI: 10.1016/j.cct.2019.06.021
- 3.
- Berry S M, Broglio K R, Groshen S, Berry D A. Bayesian hierarchical modeling of patient subpopulations: Efficient designs of Phase II oncology clinical trials. Clinical Trials. 2013; 10(5): 720–734.
- 4.
- Fujikawa, K, Teramukai, S, Yokota, I, Daimon, T. A Bayesian basket trial design that borrows information across strata based on the similarity between the posterior distributions of the response probability. Biometrical Journal. 2020; 62: 330–338.
- 5.
- Liu R, Liu Z, Ghadessi M, Vonk R. Increasing the efficiency of oncology basket trials using a Bayesian approach. Contemporary Clinical Trials. 2017; 63: 67-72. DOI: 10.1016/j.cct.2017.06.009
