Industry Perspective on Biomarker Development and Qualification

    loading  Checking for direct PDF access through Ovid


Biomarkers are molecular, histologic, radiographic, or physiologic measurements assessed in bodily fluid or tissue as indicators of normal biological processes, pathogenic processes, or responses to an exposure or intervention. Biomarkers are used at all stages of drug development and are classified as diagnostic, monitoring, pharmacodynamic/response, predictive, prognostic, safety, or susceptibility biomarkers, depending on their use. Surrogate endpoints are biomarkers that substitute for hard endpoints and are often used in phase II and phase III trials to provide signs of efficacy and to increase the efficiency of clinical development in terms of cost and time. To provide harmony and consistency for different terms used to describe biomarkers, the US Food and Drug Administration (FDA) and National Institutes of Health (NIH) Biomarker Working Group has published a glossary called BEST (Biomarkers, EndpointS, and other Tools, that serves as a “living” resource for the entire biomedical community.1
Industry and society at large face great problems with drug attrition and increasing costs, driven largely by failures in the development pipeline. All told, less than 10% of drugs that enter phase I trials end up earning approval by the FDA.2 Because efficacy has consistently been the greatest reason for attrition over the last decade,2 companies now dedicate a significant effort to identifying translational biomarkers that provide metrics for drug exposure at the site of action, pharmacological activity, and optimal patient selection and stratification. Precision oncology has employed biomarkers to match patients who have particular genetic or molecular signatures with targeted treatments that have an increased likelihood of success. Although drug attrition due to safety is less common, these issues still remain consistently high in phase II and phase III studies. Reliable translational biomarkers do not exist for many safety endpoints such as vascular injury, while other routinely used biomarkers that assess preclinical safety fail to detect toxicity in humans either because of the genetic heterogeneity or because of the limitations of the animal models.3 A new generation of biomarkers are needed that not only allow monitoring for safety risks in nonclinical and clinical studies but also demonstrate reversibility of toxicity. Such translational safety biomarkers would allow testing of promising drug candidates and their mechanisms in humans, especially when benefit/risk considerations are favorable, that may otherwise be abandoned because of nonclinical safety issues.
To accelerate biomarker science and enable their use as drug development tools, the FDA has created several ways of working with companies and other stakeholders to collectively assess biomarker performance. Commonly, individual companies work directly with the FDA during an Investigational New Drug (IND) application or a New Drug/Biologics License Application (NDA/BLA) to reach agreements on a case‐by‐case basis about whether particular biomarkers can be used in a drug development program. In an effort to establish a more generalizable, publicly‐accessible and collaborative path for biomarker assessment, the FDA introduced the Biomarker Qualification Program (BQP) in 2007.4 The goal of BQP is to qualify biomarkers for stated contexts of use that can allow for specific interpretations and applications of biomarkers in multiple drug programs without re‐review. The process is also intended to encourage widespread adoption and further study of the biomarker, as well as the development of the appropriate evidentiary standards for biomarker use. Since the inception of BQP, the FDA has qualified six biomarker applications composed of 14 biomarkers that include molecular and imaging biomarkers for nonclinical/clinical safety as well as patient selection and response.
The biomarker discovery and development pipeline requires a large investment of time and resources (Figure1).
    loading  Loading Related Articles