More Than a Question of Correlation: Characterization of the Evidentiary Basis for Biomarker Surrogates Used in European Marketing Authorizations

Abstract

Traditionally, clinical outcomes measuring how a patient feels, functions, or survives are preferred endpoints in clinical trials; however, some may take a long time to manifest in slowly developing diseases. Biomarkers, if properly validated, can serve as surrogate endpoints, acting as substitutes for clinical outcomes. This study investigated the extent and type of evidence demonstrating the correlation between biomarkers and clinical outcomes available in the literature before these biomarkers were first used as surrogate primary endpoints in pivotal studies supporting marketing authorization of cardiovascular and metabolic medicines in the European Union. Fourteen biomarkers newly used as surrogate endpoints between 2008 and 2023 were identified from European public assessment reports, and systematic literature searches were conducted for each of the identified biomarkers in PubMed and Embase to review existing evidence for the correlation between biomarker surrogates and clinical outcomes. We found that such correlation had often not been established in the literature prior to their use as a primary endpoint, as validation studies were identified for only four of the biomarkers. Of the 14 identified novel biomarker surrogates, all but one (93%) were used in trials for medicines with an orphan designation. Regulators seem to accept the use of surrogates in trials for rare diseases, despite limited validation, and consider the totality of evidence submitted to support authorization of the medicine. Presentation of complete and explicit justification for the choice and acceptability of biomarker surrogates in public documentation, such as European public assessment reports, should be encouraged.