It is appropriate when the aim of the study is to give treatment recommendations for each biomarker-defined subgroup separately at the same time.

1. In order to control the overall type I error rate of the designs at the overall level of significance α (Type I error) it is required to allocate this overall α between the test for the biomarker-positive subgroup and the test for the biomarker-negative subgroup using the Bonferroni correction method for multiple testing; e.g., if we choose the value of 0.025 for the global significance level α, then we could choose the values of α_1=0.010 and α_2=0.015 for testing the biomarker-negative and biomarker-positive subgroups respectively.
2. These trial designs are powered in such a way so as to detect the treatment effect in each biomarker-defined subgroup separately. A higher portion of the type I error rate can be given for the test within the biomarker-positive subgroup in order to maximize the power of the trial to identify the treatment effect in this subpopulation. However, even if there is a slight increase in the type I error probability spent on the test of one of the biomarker-defined subgroups, the power would probably not change much.
3. As in the sequential subgroup-specific designs, the probability of rejecting either the null hypothesis of no treatment effect in the biomarker-positive subgroup or in the biomarker-negative effect under the global null hypothesis is less than or equal to the overall type I error rate α.
4. The probability of rejecting the null hypothesis of no treatment effect in the biomarker-negative subpopulation when the treatment benefit is only restricted to biomarker-positive patients is less than or equal to α. The significance levels a can be considered as one-sided or two-sided significance levels.

• Same formula proposed for marker stratified designs could be considered to achieve sufficient power in each biomarker-defined subgroup simultaneously.
• In order to control the overall type I error rate of the designs at the overall level of significance α it is required to allocate this overall α between the test for the biomarker-positive subgroup and the test for the biomarker-negative.
• For biomarker-positive subgroup the reduced significance level α₁=α-α₂ can be used whereas the reduced significance level α₂=α-α₁ can be used for biomarker-negative subgroup.

Advantages

• Allow for the estimation of treatment effect in biomarker-positive and biomarker-negative subgroups.
• Do not require larger sample size than the overall/biomarker-positive designs when the prevalence of the biomarker-positive patients is small.

Limitations

• Have less power when there is homogeneity of treatment across the different biomarker defined subgroups as compared to the overall/biomarker-positive designs.
• Allocate the overall level α between the two biomarker-defined subgroup tests which means that it will be more difficult to achieve statistical significance in the biomarker-positive subgroup.

• Freidlin, B.; Sun, Z.; Gray, R.; Korn, E.L. Phase III clinical trials that integrate treatment and biomarker evaluation. J. Clin. Oncol. 2013, 31, 3158–3161. [Google Scholar] [CrossRef] [PubMed]
• Freidlin, B.; Korn, E.L. Biomarker enrichment strategies: Matching trial design to biomarker credentials. Nat. Rev. Clin. Oncol. 2014, 11, 81–90. [Google Scholar] [CrossRef] [PubMed]
• Freidlin, B.; Korn, E.L.; Gray, R. Marker sequential test (MaST) design. Clin. Trials 2014, 11, 19–27. [Google Scholar] [CrossRef] [PubMed]