A Bayesian Optimal Design Advancing the Precision Medicine Approach
There is recognized heterogeneity within any given tumor‐type from patient to patient (inter‐patient heterogeneity), and within an individual (intra‐patient heterogeneity).  Traditional designs are challenged by heterogeneity and that has led to the advent of targeted therapies that enable precision medicine. Precision medicine allows physicians to identify individual patient characteristics or tumor types that might be most susceptible to a given therapy and represents a huge advance in cancer care.
Currently, a lot of effort is being put in to look for more pairs of targeted drugs and corresponding biomarkers.  There are only few effective subgroup treatment pairs (STPs) identified as there is a lack of statistical methods for subgroup discovery and analysis. With the lessons learned from studies conducted over the years, the field has shifted to methods that allow new subgroup discovery during and after the clinical trial.
I am excited to report, Cytel introduces a new module in East Bayes, Subgroup Enrichment and Analysis – SCUBA, which performs trial simulation to examine the operating characteristics of the subgroup cluster-based Bayesian adaptive (SCUBA) design (Guo et al., 2017).
Using random partitions and semiparametric Bayesian models, SCUBA provides coherent and probabilistic assessment of potential patient subgroups and their associated targeted therapies.  This design is applicable to Phase II randomized and controlled trials with one or more treatment arm and a common control arm. Baseline continuous biomarkers are measured for each patient, based on which subgroups will be estimated. Each STP can be used for future confirmatory studies for regulatory approval.
SCUBA helps to resolve the issue of identifying proper subgroups defined as patients whose biomarker values fall into specific ranges. It can be used as a population enrichment design that allows enrollment in predicted optimal treatment arm or a data analysis method that estimates subgroups of patients at the end of the trial.
This novel approach has the capability to handle a trial with multiple treatment arms and provide desirable subgroups for each arm. It advances the precision medicine approach with the power of statistical modeling and inference. SCUBA can also be easily applied to simpler trials where only one treatment arm and one control are investigated, and it can handle a single biomarker as well.
Click below to learn more about this new module in East Bayes.
 Catenacci, D. V. (2015). Next‐generation clinical trials: Novel strategies to address the challenge of tumor molecular heterogeneity. Molecular Oncology, 9(5):967-996
 Mullard, A. (2015). NCI-MATCH trial pushes cancer umbrella trial paradigm. Nature Reviews Drug Discovery, 14(8):513.
Pantelis is Vice President, Customer Success for Cytel, Inc. based in Geneva. He joined the company in January 2013. Before that, he was a Principal Biostatistician at Merck Serono as well as a Professor of Statistics at Carnegie Mellon University for 12 years. His research interests lie in the area of adaptive designs, mainly from a Bayesian perspective, as well as hierarchical model testing and checking although his secret passion is Text Mining. He has served as Managing Editor of the journal “Bayesian Analysis” as well as editorial boards of several other journals and online statistical data and software archives.