Blog

2018 Bioanalytical Method Validation Guidance: Key Changes and Considerations

On Monday 21st May 2018 the US FDA published the final version of the updated Bioanalytical Method Validation (BMV) Guidance for Industry. Considering this is a pivotal guidance for conducting regulated bioanalysis, a finalized document was highly anticipated since the release of the draft version in 2013. As such, I thought I’d summarize my personal first impressions and anticipate how our community may respond and adapt. My intent is only to spur further discussion and dialogue on this important bioanalytical milestone. I don’t doubt that more substantive review and discussion from across the bioanalytical community will follow and I’ll look to delve deeper myself into the pertinent aspects of this document in the coming weeks.

Please note that the following observations are from an initial review and should not in any way be construed as recommendation or consultation. They are primarily a personal view although they have been derived from early discussions with my colleagues. I strongly recommend the reader to review and interpret any impact (if any) of this new guidance to their own practices and SOPs.

A link to the official document can be found here.

Introduction and Scope

Overall this FDA Guidance is a significant improvement in readability over the 2013 draft with a logical flow of sections covering background, method development, method validation, application to study sample analysis and recommended documentation and reporting. While there are still supplementary sections on endogenous compounds, biomarkers, diagnostic kits and new technology that differentiate from other global BMV guidance (the inclusion of which may be questioned), the completely revamped Appendices including Tables 1 and 2 summarizing acceptance criteria and reporting requirements is a welcome addition. Another major departure from the 2013 draft document, and all other BMV guidance, is the merging of the chromatographic and ligand binding assay recommendations. The practicality of this approach originally seemed daunting but this version has managed to corral the nuances of both techniques under the common expectations of a validated bioanalytical method. Considering the adoption and growth of hybrid, immunoaffinity/LC-MS assays over the last 5 years this approach is even more appropriate today.

The draft version of this guidance elicited a significant public response and is referenced in the introduction/pre-amble. It’s clear that commonly voiced feedback from the bioanalytical community has been taken into consideration and in many cases, reflects the changes and edits. Both sections I and II (Introduction and Background) give a good sense of clarifying the scope of the Guidance which was a frequently cited recommendation from bioanalysts. There is clarification in Section II on the fit-for-purpose (FFP) concept providing the contrast against full validation. It goes on to specifically state full validation should apply to “pivotal studies submitted in an NDA, BLA or ANDA that require regulatory decision making for approval, safety or labeling, such as BE or pharmacokinetic studies.” The FFP concept is noted as more appropriate to exploratory methods that would not be used to support regulatory decision making but may apply to “drugs, their metabolites and biomarkers” relative to the state of drug development.

Enhanced Focus on Method Development

The main section of the Guidance (Section III) starts with Guiding Principles which extends the original reference to developing a bioanalytical method ahead of any validation procedures. Overriding this is the statement “bioanalytical method development does not require extensive record keeping or notation.” However, there is also a comprehensive outline of parameters that should be investigated to ensure that the method is suitable to validate. This extends to stipulations around understanding the analyte, confirming the identity of reference standards and considering physicochemical properties such as metabolism and protein binding. The bioanalyst is well advised to conduct a thorough method development and be prepared to defend the bioanalytical strategy that is ultimately validated. This is consistent with dialogue from FDA representatives since release of the draft document. That is, this guidance is not meant to be overly prescriptive thus allowing adaption to today’s assays and those we are likely to encounter over the next decade or so.

Pre-Study and In-Study Validation Requirements

Section III, B covers the bioanalytical parameters of chromatographic (CCs) and ligand binding assays (LBAs) although notable over the draft document is the removal of the acceptance criteria that have been logically relocated to the Appendix Table 1. This has facilitated the merging of the two technique approaches under common validation experiment descriptions. After clarifying requirements and definitions around reference standards, critical reagents, calibration curves and quality control samples, the expectations of demonstrating the five fundamentals of a bioanalytical method validation are addressed. These are accuracy, precision, sensitivity, selectivity, and stability although they are not tackled in this order in the text. While recovery, specificity and dilution effects are also woven into this section it doesn’t overly detract from logical flow of the process. Analyte stability is given due description through the familiar stability stress testing we are accustomed to. That said, a couple of intriguing points jump out including reference to “consider(ing) the stability of the analyte in the presence of other co-medications” and “determination of stability at minus 20 °C would cover stability at colder temperatures.” No doubt, the bioanalytical community will have more to say on these finer points.

Noticeable by their absence are a couple of routine bioanalytical method validation experiments that I still expect we will continue to conduct for LC-MS assays. While the need for matrix related ion suppression/enhancement is stated earlier in the document, the design of this experiment is not described (in contrast to other BMV guidance). Neither is the extracted sample analytical run reproducibility test referenced, which is not the same as the extracted sample stability test that is described. To be fair, it may be argued that both tests primarily apply to chromatographic assays and therefore fall under the general recommendation of “know your assay.”

Section III, C takes us into application of the validated assay to in-study analysis and reporting. There are some highlights worth noting. The requirement to consider the performance of all QCs (passing and failing acceptance criteria) should not be overlooked. Related is the contentious requirement to report batch performance criteria. With the common practice of employing multiple 96-well plates for sample preparation and analysis by LC-MS this will now require some change of procedure. A welcome clarification is the reporting of analyte concentration below the lower level of quantitation (LLOQ). The draft document recommended reporting as zero concentration but this has now been correctly updated to reporting these results as below the LLOQ (BQL).

Targeted Focus on ISR


Maybe due to heightened awareness and precedence set by other BMV guidance, the incurred sample reanalysis (ISR) is separated into its own section. The only notable update from the draft, and consistent with what we have already become accustomed to by using the EMA BMV Guidance, is the selection of samples. For the detail, one needs to reference Table 1 where the number of samples required is clarified as 10% of the number of study samples up to 1,000 samples and an additional 5% thereafter. Also noted is selection of samples from the Cmax and elimination phases somewhat discounting recent discussions about true random sampling of ISR samples.

Additional Issues

A significant section on Additional Issues (Section V) remains from the draft document but includes several important updates. The text around Endogenous Compounds section has been consolidated notably including a bullet point on recommending the parallelism test. Based off the definition of ‘parallelism’, now included in the glossary, this test is restricted to dilution linearity as a means of demonstrating potential matrix effects. The reference to parallelism also extends to the next sub-section on biomarkers. It’s worth pointing out that the application of this guidance document to biomarkers remains from the draft version and is likely to continue to be a clear differentiator over other global past, present and future BMV guidance.

Section V, C comprises a significant re-write on use of diagnostic kits. A previous sentence from the draft version stating that these recommendations do not apply to Clinical Laboratory Improvements Amendments (CLIA) activities is now removed. With inclusion of biomarkers it will be interesting as to where this guidance takes us regarding clinical laboratory practices. Otherwise the reader interested in use of diagnostic kits (or their components) for regulated bioanalytical assays is well advised to become familiar with the specifics of this section.

The New Technologies section of the draft has now been split into Bridging Data from Multiple Bioanalytical Technologies and Dried Blood Spots (DBS). It makes sense to have more general reference to new technologies and recommendations for accommodating them in bioanalytical strategy. Contrasting this is the singling out of DBS. Maybe the objective is using it as a case example of a new technology. In which case, what stands out in the DBS section is the requirement for correlative studies with traditional sampling during drug development. This continued reference to DBS will likely draw discussion in the bioanalytical community if it is used as an example of what we should be prepared for in the adoption and adaptation to new technologies.

Documentation and Reporting

The main body of the Guidance concludes with Section VI on documentation. The text around documentation is now significantly consolidated compared to the draft version. With inclusion of Appendix Table 2 this makes sense and helps clarify expectations. As well as in-lab documentation, the text overviews the final reporting but for specific requirements the reader needs to consult Table 2 rather than the rely on Section VI language. Depending on individual lab interpretations and existing report templates, the stipulated requirements for the final report may induce some significant changes to current practice.

The Finer Points and Criteria

Table 1 and Table 2 are new to the Guidance replacing the example report tables of the draft version. Here is where the differentiation between chromatographic and ligand binding assays is spelled out. Both Tables also extend to in-study analysis experiment requirements and the recommended reporting respectively. While there are no radical changes to the acceptance criteria we have become familiar with (e.g. +/- 15% criteria for CCs and +/- 20% for LBA) there are some subtle additions such as the acceptance criteria for any processing batches as well as the whole analytical run. Regarding documentation and reporting requirements, there is a reference to temperature during sample shipment tracking records. Likewise, extended requirements around sample management and handling are proposed for study reports. In practice, these seemingly fine points will likely disproportionally impact bioanalytical laboratory procedures.

Summary

Over the next few weeks and months I will likely refer to this blog article and I will do so around the major themes of the 2018 FDA BMV Guidance. I’ve touched on many of these in this overview but I anticipate they deserve further expansion as our bioanalytical community implements this new guidance. These areas include:

  • Preanalytical requirements
  • Bioanalytical method development
  • Bioanalytical method validation and associated criteria for CCs and LLBAs
  • Application of the validated assay to in-study sample analysis
  • Documentation and reporting
  • Fit-for-purpose bioanalysis and application to endogenous compounds and biomarkers
  • Managing new technologies in the regulated bioanalytical laboratory

Conclusion

The bioanalytical community had been wondering if we would see a finalized BMV Guidance from the FDA. After 5 years of the draft document under review and with the ICH M10 initiative to develop a global harmonized BMV Guidance started, it was at least theorized that a finalized FDA version may not update the 2001 official document. I expect the bioanalytical community will adapt to this 2018 guidance. The fundamentals are quite consistent with the EMA BMV Guidance that has been the reference for most regulated bioanalytical activities since its release (2012). Of course, there are the additional references to biomarkers, endogenous compounds and new technologies, however, these fall more under ‘confirm your bioanalytical method is suitable for the intended use.’ The prescriptive language is primarily reserved for full validations and pivotal PK / BE studies supporting safety, efficacy and labelling claims. It is here we largely find consistency with other global health authority BMV guidance.

There are some subtle differences and updates that will require the bioanalyst supporting regulated assays to become familiar with this document. Most laboratories will need to adapt accordingly. A gap analysis against existing SOPs and practice is probably warranted at most bioanalytical facilities. Such a gap analysis is now initiated at my own laboratory and we will update our sponsors and clients accordingly to ensure their interests are protected and we are aligned with their expectations and those of the regulators.