Bioanalytical Community

 View Only

  • 1.  FDA Guidance for Industry: Bioanalytical Method Validation for Biomarkers

    Community Leadership
    Posted 01-22-2025 08:32

    Hello everyone,

    The FDA have just published this guidance for industry: Bioanalytical Method Validation for Biomarkers Guidance

    I leave you all to read and digest.

    Personally, I am disappointed that after all the discussions involving industry and regulatory colleagues on the best scientific practices for biomarker assay establishment and validation, this guidance refers to the ICH M10 PK method validation document and does not include any discussion on the key starting point for biomarker method establishent: understanding the biology of the biomarker and anticipated changes with treatment...



    ------------------------------
    Robert Nelson
    Scientific Officer, Senior Director
    BioAgilytix Europe GmbH
    Hamburg Germany
    [email protected]

    Disclaimer: Opinions expressed are solely my own and do not express the views or opinions of my employer or other entities to which I am affiliated.
    ------------------------------


  • 2.  RE: FDA Guidance for Industry: Bioanalytical Method Validation for Biomarkers

    Posted 01-23-2025 09:19

    Thanks for posting, Robert!

    A few thoughts that come to mind on my end:

    (1) Potential gray area for PD biomarker assays supporting secondary endpoints, especially for gene therapies or cell therapies. Presumably it will depend if any claims are made from that data but this could impact the inclusion of data if there are questions about whether the assays are "fully M10". 

    (2) Endogenous considerations in M10 may not fully handle the questions that arise for LBA biomarker assays that involve recombinant protein in a surrogate matrix. For recombinant standards, spike-recovery may be less informative for selectivity than parallelism (with value of parallelism discussed in Critical Path paper; spiking into healthy matrix with high levels of endogenous material may not be helpful or feasible) and stability of recombinant protein may not be scientifically relevant or useful in relation to stability of endogenous material (although it could support stability of spiked QC samples). Additionally, even A & P may and is likely to differ in terms of mix of QCs between recombinant and endogenous QCs. All these elements do make the considerations for biomarkers unique. 

    (3) Going further from point 2, stability definitely seems to be a sticking point that has come up in recent case studies. There was an EBF presentation on an example this year, and the public documents for Kisunla showed that this was something that was flagged with additional plots made of samples over time to add justification. In the EBF case, I believe there was a request to test benchtop stability at two levels of sample concentration. If I remember correctly, in the Kisunla case, claims were being made based on the data (feel free to correct me if I am wrong on any of the above). 

    I see this as a particular challenge because samples at different levels of analyte may indeed not be available in validation. Patient samples may be needed. This is especially true for gene therapies where low concentration samples may not be available. I suppose it is possible to immunodeplete material then return it back to matrix pools, although this is definitely well beyond the norm for a LBA PK assay. So this assessment may need some clarification as to what is needed. There have also been conference case study presentations looking at assessment of stability by other means, such as statistical analysis of samples measured over time--would want to clarify if this guidance precludes this (possibly acceptable for exploratory but not pivotal?)

    (4) Finally, the key point that Robert makes about fit for purpose and what is needed to discern changes is critical (and of course has been the subject of significant work and discussions). For assays like enzymatic assays or tissue-based measurements, if precision is not that of a standard PK assay but the changes being measured are measurable with the assay (and this can be statistically justified), how does this guidance impact the use of these assays? Does this preclude the use of more variable biomarker measurements for a pivotal decision or dosing even if there is statistical justification, and how is physiological variation incorporated vs just analytical variation? Also, I was curious--is there a good paper that looks at the precision of typical clinical endpoints vs expected variability and expected detectable changes, and compares this with fluid-based biomarkers?



    ------------------------------
    Cathy Vrentas
    [email protected]

    Disclaimer: Opinions expressed are solely my own and do not express the views or opinions of my employer.
    ------------------------------

    Original Message


  • 3.  RE: FDA Guidance for Industry: Bioanalytical Method Validation for Biomarkers

    Community Leadership
    Posted 01-23-2025 18:46

    Thanks for posting, Rob. 

    For our community, I'd like to offer some perspective.  Those of us engaged in biomarker assay development and validation are well aware that the varied contexts of use for biomarker assays cannot be captured in a one size fits all guidance. I want to caution those who may be viewing this guidance as a mandate to use the specific experimental approaches described for PK assays in M10, which itself clearly puts biomarker assays out of scope. Instead, I would view this as indicating that the parameters of interest for quantitative/relative quantitative assays are similar and we should scientifically stand behind our approaches.  This is what we have been doing all along for our biomarker assays - and quite successfully.  I have never failed to gain successful regulator alignment when using a context of use driven approach for biomarkers.

    I'd also like to highlight that this Guidance simply indicates that nothing has changed from what FDA put in the 2018 Guidance: 

    In the 2018 Guidance, FDA stated:

    "Method validation for biomarker assays should address the same questions as method validation for drug assays. The accuracy, precision, sensitivity, selectivity, parallelism, range, reproducibility, and stability of a biomarker assay are important characteristics that define the method. The approach used for drug assays should be the starting point for validation of biomarker assays, although the FDA realizes that some characteristics may not apply or that different considerations may need to be addressed."

    In the current Biomarker Guidance, FDA states: 

    "Method validation for biomarker assays should address the same questions as method validation for drug assays. The accuracy, precision, sensitivity, selectivity, parallelism, range, reproducibility, and stability of a biomarker assay are important characteristics that define the method. The approach described in the guidance for industry M10 Bioanalytical Method Validation and Study Sample Analysis (November 2022) for drug assays should be the starting point for validation of biomarker assays, especially chromatography and ligand-binding based assays."

    All that has changed is that M10 has been adopted for drug concentration assays, so now FDA points there instead of to its own BMV guidance.

    As an industry, we have made great progress in implementing COU-driven biomarker assay validation, even while the 2018 FDA Guidance was in effect.  This new guidance changes nothing and should not alter our trajectory.



    ------------------------------
    Lauren Stevenson Ph.D.
    Chief Scientific Officer
    Immunologix Laboratories
    Tampa FL
    [email protected]

    Disclaimer: Opinions expressed are solely my own and do not express the views or opinions of my employer.
    ------------------------------

    Original Message