The FOCUS4 design for biomarker stratified trials

Richard Kaplan

Abstract

Randomised clinical trials (RCTs) remain the gold standard of evidence for the benefit of new therapeutics but standard designs fit awkwardly with key developments in biomarker-stratified drug development. Firstly, the unprecedented number of new agents being developed in oncology (usually with specific targets for which there may be predictive biomarkers) mandates a need for new trial designs that are more efficient in screening out new agents with modest likelihood of benefit, concentrating resources on the most promising ones. The multi-arm multi-stage (MAMS) design developed some years ago addresses this need. Secondly, biomarker-stratified trials, when tackled one biomarker/drug pairing at a time, are inherently highly inefficient. The FOCUS4 trial design was developed to overcome this problem, using a platform that incorporates multiple parallel biomarker-stratified RCTs in individual cohorts, and capable of adapting its design in response to developing evidence.