The start of 2025 has brought a wave of regulatory guidance from the FDA, including the finalized Bioanalytical Method Validation for Biomarkers – Guidance for Industry issued on January 21. Although the document is less than three pages long, it has sparked significant discussion and debate on within the bioanalytical community.
For reference, the guidance is available on the U.S. Department of Health & Human Services Guidance Portal (Bioanalytical Method Validation for Biomarkers | Guidance Portal). [Note: It was moved from its original publication site on the FDA's media website.] Notably, the guidance is issued as a finalized document with the usual caveats of "contains nonbinding recommendations" and represents the current thinking of the agency. While there were discussions about this guidance being drafted at the FDA, I am not aware of any consultation process outside the agency, which is sometimes experienced within the bioanalytical community for similar guidance. Comments on this guidance can still be submitted at any time via regulations.gov using Docket No. FDA-2017-D-6821.
The bioanalytical community has expressed a range of opinions about this new guidance. It is generally seen as retiring the FDA BMV 2018 Guidance while maintaining the FDA's stance on biomarker bioanalysis as stipulated in the 2018 Guidance. The requirement for high standards in biomarker bioanalysis for safety, efficacy, and product labeling should not come as a surprise. For about a decade, FDA representatives have pointed out sub-standard bioanalytical data for biomarkers in regulatory submissions that do not meet the established standards of bioanalysis in BMV doctrine. However, as has been reiterated ad nauseam by others, herein lies the root of the issue; BMV has evolved around xenobiotic drug bioanalysis while biomarkers fundamentally differ from drug analytes.
The European Bioanalytical Forum (EBF) quickly released a position statement on the new FDA Biomarker Bioanalysis Guidance, which resonated with many (EBF-comment-to-FDA-BMV-for-BM-January-2025.pdf). Two fundamental points were highlighted in their critique:
These points are difficult to dispute. In an earlier Fit The Curve blog article, I supported the case that we cannot discuss biomarkers without referencing COU. The only counterargument I can envisage is that all bioanalysis should be suitable for its intended purpose, and therefore the FDA guidance inherently assumes that COU applies. However, it creates confusion when one regulatory guidance specific to biomarkers directs to another that explicitly excludes biomarkers. For example, questions have arisen in online forums about whether incurred sample reanalysis is now expected by the FDA for biomarker data? Does this guidance apply to diagnostic biomarker assays and clinical laboratory-based work? Given the nuances of biomarker analyte biology, stability chemistry, reference ranges, and matrices, it is clear that "biomarkers are not drugs" and we should not treat them as if they were.
The application of biomarker assays in drug development extends far beyond the limited scope of bioanalytical assays designed for drug quantitation in biological samples. The criteria for accuracy and precision are closely tied to the specific objectives of biomarker measurement. Factors such as biomarker reference ranges, the magnitude and direction of change (increase or decrease) relevant to decision-making, influence the statistical criteria required for the assay. Applying fixed criteria, as we do in drug bioanalysis, is a flawed approach when it comes to biomarkers. This doesn't necessarily mean that more lenient criteria are always appropriate; it depends on the specific biomarker investigation and the subsequent clinical interpretations.
One positive aspect of ICH M10 relevant to this topic is Section 7.1, Methods for Analytes that are also Endogenous Molecules. Despite the reminder that biomarkers are outside the scope of ICH M10, the approaches described for endogenous compounds can be applied to biomarkers. The concise descriptions of surrogate matrix, surrogate analyte, background subtraction, and standard addition are helpful even in the context of biomarker assays and represent an improvement over FDA BMV 2018 language. Likewise, the reference to required parallelism assessments with surrogate matrix and surrogate analyte is applicable to biomarker assays as it is for endogenous compound bioanalysis. This section can aid in formulating discussions early in biomarker assay planning.
The new FDA guidance acknowledges that ICH M10 may not be applicable to some biomarker analyses. There is also the closing statement that M10 Bioanalytical Method Validation and Study Sample Analysis (November 2022) should be a starting point especially for chromatography and ligand-binding assays. With this perspective, we aim to make progress, support our sponsors, and develop bioanalytical strategies that can withstand regulatory scrutiny. At this point, I advise our teams that ICH M10 is a reference point to initiate a COU-driven bioanalytical study plan. This plan will always be tailored to meet the specific objectives of the biomarker analysis, whether it is to enhance our understanding of a disease or to aid in the development of a new therapy.
I’m sure that the interpretation and application of the FDA Guidance for Industry on BMV for Biomarkers will feature heavily in upcoming meetings and conferences. I look forward to continued engagement and discussion on the topic and as always welcome feedback to the opinions shared in this blog.