FDA Details Requirements for Regulatory Submissions of Genomic Biomarkers

FDA’s final guidance for biomarkers related to the development of new drugs or biotechnology products represents the strongest effort made by the agency to date to bring more biomarkers beyond the laboratory to the clinic. The guidance seeks also to help U.S. companies develop biomarkers for the global industry.

The FDA guidance focuses not on the science of biomarkers but instead offers recommendations for the context, structure, and format of regulatory submissions by companies seeking to qualify genomic biomarkers. While the guidance applies to genomic biomarkers, FDA says the principles could apply to other biomarker categories, such as proteomics, imaging, and a genomic-nongenomic combination of biomarkers.  

Efforts to regulate the other biomarker categories via guidance have not been finalized. Last December, FDA’s Center for Veterinary Medicine developed general approaches and analytical techniques for proteomic analysis of complex biological matrices as part of a project to develop techniques to analyze biological specimens for biomarkers.

FDA has even acknowledged how far it still has to go. In a report it issued this month called “In Driving Biomedical Innovation: Initiatives to Improve Products for Patients”, the agency lamented that fully qualified biomarkers are among examples where “much of the applied regulatory science for evaluating the strategies and outcomes for personalized medicine…are still underdeveloped.”

Spelling Out Context and Structure

Under FDA’s new biomarker guidance, descriptions of the context for using a biomarker should include: (1) the general area, (2) specific biomarker use, and (3) critical parameters that define when and how the biomarker should be used. General area includes but isn’t limited to pharmacology, toxicology, efficacy, safety, and disease.

Specific biomarker uses can cover factors ranging from patient selection and dose optimization to drug response monitoring. Critical parameters include specificity of use by class or product, disease diagnosis and phenotypes, sample collection, assay specifications, tissue or physiological/pathological process, species, demographics, and environmental factors.

The context should be supported, the guidance noted, by data available in the biomarker’s initial qualification dossier submission. Should that data prove inconsistent with the proposed context, drug developers could be asked to furnish additional data during the process of qualifying their biomarker.

According to the guidance, the structure of qualification submissions should include an overview of biomarker qualification, with a justification for its proposed context of use, followed by summaries of analytical assay data, nonclinical biomarker data, and clinical biomarker data. Once a biomarker is qualified, researchers need not generate additional data to justify its use.

“These kinds of things help in the determination of whether or not a patient is going to appropriately respond,” David L. Rosen, a partner and co-chair of the life sciences industry team at the law firm Foley & Lardner LLP, told GEN.

“That’s what we’re looking at: How well can the diagnostic agent help people assess whether or not they have particular characteristics that would respond? Can they differentiate between people who have a particular marker or not? How susceptible is the biomarker to false positive readings? I think all those play into that determination,” said Rosen, who spent 14 years at the FDA.

Thinking Globally

In its final biomarker guidance, FDA used the definition of genomic biomarker developed by the International Conference on Harmonisation (ICH) of Technical Requirements for Registration of Pharmaceuticals for Human Use: a measurable DNA and/or RNA characteristic that is an indicator of normal biologic processes, pathogenic processes, and/or response to therapeutic or other interventions.

ICH’s six sponsors comprise a coalition of government and industry groups as well as regulatory bodies in the U.S., Europe, and Japan. Its guidance was endorsed by FDA and its counterparts in the EU and Japan in September 2010. That’s more than a full year after FDA released its draft guidance on the topic and around the time FDA would typically be expected to have issued its own final guidance.

The slower pacing of review compared to typical FDA-only guidance processes reflected the desire of the agency and its counterparts to harmonize global biomarker standards. Until the ICH harmonization process began, FDA’s approach to biomarker review was piecemeal: CDER individually qualified biomarkers like LDL cholesterol and total cholesterol in cardiovascular disease risk as well as the CD4 blood test and viremia in HIV/AIDS.

Fit-for-Purpose Test

The discussion of biomarkers in the guidance is consistent with the “fit-for-purpose” framework for a performance-and-purpose based classification of biomarkers discussed in scientific literature in recent years, Antoinette F. Konski, also a partner with Foley & Lardner, and a member of the firm’s chemical, biotechnology, and pharmaceutical practice, explained to GEN.

“Fit-for-purpose” concludes that for biomarkers that are deemed fit for the purpose of detecting toxicity and monitoring drug safety and therefore qualified for use in a regulatory decision-making context, the underlying science must be sufficiently rigorous to match the critical level of evidence and performance that would be needed. Sufficient qualification data must exist to define the fitness of a biomarker for a regulatory purpose involving safety evaluation and toxicity detection.

“The evolving nature of biomarker qualification further underscores that a significant effort is necessary to cross a scientifically credible threshold from an unqualified or exploratory biomarker, which would not be appropriate for a regulatory data submission filing, to a qualified safety biomarker appropriate for regulatory decision-making,” Konski pointed out.

“For any exploratory safety biomarker to be considered newly qualified as fit-for-purpose, the qualification process is usually incremental. Any additional data required to support qualification for regulatory use will be expected to depend on what data may already be publicly available, may lie within regulatory repositories, with individual sponsors, or available from consortia and institutions.

“Sponsors and regulatory authorities share a responsibility for acknowledging when new safety biomarkers are appropriately qualified for regulatory decision-making,” Konski added. With the final guidance in place, it will now be up to those sponsors and regulators to exercise that responsibility by choosing for review biomarkers with enough data.

While harmonizing global biomarker standards may save time and money for drug developers and others, that savings could be negated if sponsors submit the wrong products for review, especially with the present-day FDA more than willing to sacrifice speed of decisions for what it argues is a better grasp on safety to improve decision-making. Stakeholders will be looking for results that hold the best prospects for approval so that the trickle of biomarkers migrating from the lab to the clinic stands a better chance of building up to a steady stream.