Biomarker Qualification at the European Medicines Agency: A Look Under the Hood

Understanding the process for qualification of biomarkers by regulatory agencies can help drive additional work on their discovery, development, and adoption. By providing data on successful and unsuccessful qualification efforts, Bakker et al. provide insights into regulatory activities that can help increase the likelihood of success of future biomarker programs. In particular, the authors document the relatively low proportion of submissions that attained qualification and the types of issues that arose during the qualification process.

The use of drug development tools (DDTs), such as biomarkers, facilitates drug development by adding efficiency or additional certainty to the evidence‐generation process for novel therapeutics. Qualification of such tools by regulatory bodies facilitates their use by drug developers by ensuring support for the tool within its context of use. Bakker et al. ¹ provide insights into a complex process where the internal deliberations of regulatory bodies charged with evaluating potential new therapies for patients and tools that can facilitate drug development programs are not available to the public. By accessing, evaluating, and describing the data contained in an internal European Medicines Agency (EMA) database, the authors quantified the volume of biomarker qualification procedures and trends over time, the disease areas of focus for biomarker developers, the contexts of use for biomarkers submitted for qualification, and, importantly, the issues identified in the qualification procedures. These data, not publicly available and unable to be accessed elsewhere, provide valuable perspective when preparing for a new biomarker qualification effort.

Bakker et al. found that of 86 biomarker qualification procedures that started between January 2008 and December 2020, only 13 resulted in qualified biomarkers. Insight into what caused some projects to not be qualified is provided through their finding of commonly raised issues related to biomarker properties (92%), general study design (79%), and assay validity (77%). These findings will help biomarker developers focus on specific aspects of their development plan that can address issues that may slow down the qualification process.

Bakker et al. determined that most biomarker qualifications were targeted for a context of use related to patient selection, stratification, and enrichment and that fewer qualification efforts were aimed at biomarkers intended to assess efficacy in drug trials. While the authors hypothesize that this focus may reflect a targeting of trial efficiency improvements by biomarker qualifiers, there may be other potential interpretations. The authors propose another potential explanation—namely, the higher burden of evidence required for the qualification of biomarkers intended to assess efficacy, i.e., surrogate end point biomarkers, compared with biomarkers intended for other uses. In the absence of more specific information regarding the motivation of entities submitting biomarker qualification proposals, one cannot draw a firm conclusion as to the primary driver; however, there are clues as to the possible explanation within the data presented.

Bakker et al. noted that prior to 2014 most of the biomarker qualification procedures were linked to a single company and a specific drug development program, while most procedures in later years were linked to consortia and intended for general use. The authors note an important limitation in their analysis—namely that the data presented come from an analysis of only one of the two potential avenues that biomarker developers can pursue to have their biomarker assessed for acceptability by regulators. In addition to the qualification procedures described by Bakker et al., biomarkers can also be assessed on a case‐by‐case basis in conjunction with an application to market a novel therapeutic. It is likely, though only supported anecdotally, that more biomarkers have been deemed acceptable by regulators through this pathway, given the numerous successful biomarker‐guided cancer treatments. ² Further analysis of these “hidden” biomarker acceptances could provide additional insight into the evidentiary standards for biomarker development among regulatory agencies as it is reasonable to assume that many of these are biomarkers have efficacy‐related contexts of use. The data available for US Food and Drug Administration (FDA) biomarker qualifications are constrained similarly to the EMA data: while biomarkers intended for use with a specific drug development program or a class of drugs can be submitted for qualification, most biomarker qualification submissions received by the FDA are intended for general use as shown in the publicly accessible qualification data set. ³ While biomarkers assessed as companion diagnostics or otherwise tied to a specific marketed therapeutic are described in other data sources, these may not be accessible to the public.

One important consideration when evaluating the learnings from Bakker et al. is a recognition that oversight of biomarker qualification and use by regulators for drug development is guided by the statutes authorizing the regulatory activities in a region. The insights gleaned into biomarker qualification procedures of the EMA can be largely applied to qualification at the FDA since the scientific basis for assessing these submissions is shared between the EMA and the FDA, but the procedures in each jurisdiction are necessarily different.

Bakker et al. has provided valuable insights into the nature of biomarker evaluation and qualification by the EMA. Similar data is available for biomarkers submitted to the FDA for qualification and made publicly accessible thanks to the transparency provisions of the statute that modernized drug development tool qualification in the United States. In the United States, the biomarker qualification process was formalized in the 21st Century Cures Legislation, ⁴ signed into law in December 2016. This legislation established a process for qualification based on three regulatory submissions to the FDA: a Letter of Intent that initiates the process, a Qualification Plan that describes how the data supporting the biomarker will be gathered, and a Full Qualification Package that contains the data required to make a regulatory determination on the qualification of the biomarker. In the 5 years since the enactment of this modernized process, the FDA has received 118 Letters of Intent for drug development tools (which include biomarkers, clinical outcome assessments, and animal models for use under the animal rule). Of the 118, 63 were accepted; 51 Qualification Plans and 7 Full Qualification Packages were submitted, resulting in six qualifications (of which two were biomarkers). ⁵ The FDA’s process also allows for the provision of Letters of Support for promising biomarkers not yet accepted into the qualification program. In the same time frame the FDA has issued 14 Letters of Support. ⁶

To maximize the ability of drug development programs to take advantage of potentially useful biomarkers requires partnership among stakeholders and open communication, to the extent possible. Partnership is necessary as biomarker developers need resources and access to data sets and biologic samples to validate their biomarkers. Communication is necessary so that biomarker developers have access to all the information they need to validate their biomarkers and a clear understanding of evidentiary standards for use of biomarkers in drug development programs. Bakker et al. contributes to open communication by providing insight into the qualification process in the EMA.

Bridging the knowledge gap around the evidentiary and procedural requirements for acceptance of drug development tools by regulatory authorities will become even more critical as the industry moves from more traditional tools, like soluble or imaging biomarkers, to new modalities that leverage emerging science and technology. The EMA has established the Innovation Task Force (ITF), with the accompanying ITF meeting pathway to provide a venue to discuss new technologies such as complex trial designs, digital technologies, and new approach methodologies to replace the use of animals in the testing of medicines. ⁷ The FDA offers two venues for engagement on scientific advancements in drug development: Critical Path Innovation Meetings, and the Innovative Science and Technology Approaches for New Drugs Pilot Program. Critical Path Innovation Meetings can be used for informal scientific discussions of innovative approaches to drug development and trial design, ⁸ while the Innovative Science and Technology Approaches for New Drugs Pilot Program is designed to expand DDT types by encouraging development of DDTs that are out of scope for existing DDT qualification (i.e., biomarker and clinical outcome assessment) programs through qualification or other potential outcomes. ⁹ These initiatives in the European Union and the United States will encourage innovation by allowing incorporation of new scientific advances into drug development tools with the ultimate goal of providing important new therapies for patients.

FUNDING

No funding was received for this work.

CONFLICT OF INTEREST

The authors declared no competing interests for this work.

DISCLAIMER

This commentary reflects the views of the authors and should not be construed to represent the FDA’s views or policies.