- These designs provide an alternative strategy to analysing biomarker-stratified designs.
- It is an indirect way of evaluating both biomarker and treatment by testing the treatment effect in the entire study population and in the biomarker-positive subgroup separately.
- Three approaches are included in the biomarker-positive and overall strategies; the parallel assessment, the sequential assessment and the fall-back designs.
- Despite the fact that the biomarker-positive subgroup and overall strategy designs allow the treatment effect to be tested in the biomarker-positive subpopulation and provides good statistical power when the treatment effect is homogeneous across subgroups, these designs are usually considered problematic and its use is not often recommended.
- A major concern is that when the benefit of the novel treatment is limited to the biomarker-positive patients, it is possible that the designs might lead to a wrong recommendation of treatment for the biomarker-negative patients. This might happen because when there is no treatment effect in the biomarker-negative subgroup, there might be an observed effect in the entire population due to the potentially large effect in the biomarker-positive patients. This concern is particularly pronounced in the sequential version of the designs, which first tests the biomarker-positive subgroup and then, if it is positive, it tests the overall population.