Table 1.
Types of biomarker-based designs, classifications, and examples.
| Design Types | Number of disease types within a single protocol | Number of molecular profiles | Number of targeted therapies | Design Features | Real-World Example Trials |
|---|---|---|---|---|---|
| Enrichment or Targeted | 1 | 1 (e.g. marker positive only) | ≥ 1 | -Strong biologic rationale that marker negative patients are unlikely to benefit -Reliable assay -Statistical efficiency (i.e. reduced sample size requirements) -Recommended for rare prevalence markers and rare diseases |
-N9831 -TOGA |
| Marker-Stratified or Marker-by-Treatment Interaction | 1 | >1 (e.g. marker positive and marker negative) | ≥1 | -Insufficient evidence of a biomarker’s ability to predict treatment effect to justify exclusion of a subpopulation from randomization | -INTEREST -MARVEL |
| Modified Marker Strategy | ≥1 | >1 | ≥1 | -Typically used in settings with one or more approved therapies, and the interest is in identifying marker subgroups that may have the most benefit -Overlap between the marker based and the non-marker based arms can result in large sample size -Similar to a marker strategy design, except that it includes multiple molecular profiles matched with multiple targeted agents -Can include multiple tumor types -Tests for overall strategy, and not for individual marker-treatment pair |
-SHIVA -M-PACT |
| Umbrella | 1 | >1 | >1 | -Existence of national network of clinical sites doing molecularly targeted clinical trials using a common genomic screening platform -Flexible design for the adding/dropping of sub trials based on new emerging data -Use of central clinical laboratory for molecular profiling for a large cohort of patients -Can be logistically complex to set up and implement -Careful statistical consideration needed when adding new or removing existing sub trials |
-FOCUS4 -LUNG-MAP -ALCHEMIST |
| Bayesian Biomarker- Adaptive | 1 | >1 | >1, one per molecular subtype | -Strong scientific rationale, and preliminary evidence for the molecular marker-drug pairing -Reliable assay, with rapid turn-around times -Short term, reliable endpoint to make the adaptation meaningful -Sufficient infrastructure set up and real time data availability |
-BATTLE -I-SPY2 |
| (with adaptive randomization) | |||||
| Basket | >1 | >1 | >1, one per molecular subtype | -Strong scientific rationale for the molecular marker-drug pairing -Reliable assay -Availability of a sufficient number of drugs targeting multiple pathways -Single protocol for multiple disease cohorts -Assess for signals of efficacy for each individual marker-drug pairing, and sometimes within each disease cohort -Statistical design principles not well established |
-NCI-MATCH |