Reflections on regulatory updates and industry recommendations for immunogenicity assessments: A CRO’s perspective

LGC are a life sciences company, with sites around the globe providing a variety of services to its customers. Within the Immunogenicity Centre of Excellence at the Drug Development Solutions site in Fordham (Cambridgeshire, UK) the focus is on determining the presence of anti-drug antibodies (ADA) from both non-clinical studies and clinical trials.

Since Drug Development Solutions began performing ADA testing in 2008, the expectations from regulatory authorities have evolved. Drug Development Solutions have implemented changes to ensure our procedures meet these expectations. In addition, as the work we perform can take years from commencement to submission, there is also a need to enact a level of “future-proofing”.

The FDA guidance (2019)¹ covers clinical immunogenicity assessments for submission in North American territories, whereas the EMEA guideline (2017)² covers Europe, the Middle East and Africa. In 2018³ we published an article on our experiences related to the release of the FDA draft 2016 guidance. Since the publication of the current guidance in 2019, we have seen the continued trend towards development of immunogenicity assays that are more sensitive and have higher drug tolerance. This can impact on assay robustness or may result in an increase in the number of samples identified as positives. An example of a challenge when implementing the guidance is sensitivity. Within the FDA 2019 guidance document, there is a recommendation to select a control sample (referred to as a low positive control, or LPC) at a concentration that would lead to a 1% failure rate. While there are statistical analyses that can be performed to estimate this concentration, a difficulty we have faced is ensuring the correct balance is achieved to make sure the concentration selected isn’t too low (resulting in increased number of failures) or too high (resulting in an LPC that may be irrelevant to clinical samples). In addition, there is the issue of what to do if the LPC performance shifts and the fail rate differs over time. As part of the measures introduced to control for this, we developed an in-house tool for trending assay results to ensure proactive implementation of changes to ensure consistent performance.

Another challenge faced is where there are differences between regulatory authorities (e.g. EMA and FDA) and also where there are differences in interpretation between Sponsors. Generally speaking we have noticed an increased level of harmonisation across the industry, however, there are still differences present. In addition, the landscape is ever-changing with current guidance documents being updated or new documents being released. For example, in 2020 the Chinese regulatory authority (National Medical Products Administration (NMPA)) released their draft technical guidelines.

While a simplified approach may be to take the strictest interpretation of the available guidelines, as a CRO we accept that this may not be necessary to meet our Sponsors’ requirements. There is a need to ensure that the work being performed is ‘fit-for-purpose’, and to ensure that as part of a quality by design approach, method validation assessments are not being run without a clear reason. For example, if the target population is not linked with unusual levels of Bilirubin or Rheumatoid Factor, it can be planned form the outset, and scientifically justified, to omit these assessments.

Non-clinical immunogenicity

Historically, there has been little guidance on assay development and validation of ADA assays for non-clinical assessment. The FDA 2019 guidance purposely does not discuss the development or validation of ADA assays for animals studies, however, there is acknowledgement that some concepts discussed within the guidance are relevant to the design of ADA studies for nonclinical testing.

Although there is still a lack of clear industry guidance, the EBF (2021)⁴ recently published White Paper suggests a minimal strategic approach with regards to nonclinical immunogenicity assessments alongside a simplified validation process that is ‘fit-for-purpose’. This involves implementation of screening only, with a lower false positive rate (1% or 0.1%), targeted sensitivity of ≤1000 ng/mL, a 60 data-point cut-point assessment, with just 9 assay runs, as a minimum, to complete the validation.

The Chinese NMPA draft guidance, does discuss nonclinical validations and is in agreement with the EBF White Paper with regards to a smaller cut-point assessment, however, it recommends a sensitivity of 250 to 500 ng/mL, and it is unclear whether a lower false positive rate or single tiered approach should be taken.

As a CRO we support Sponsors who request anything from a full ‘bells and whistles three-tiered’ FDA-style validation and sample analysis, to a more simplified ‘fit-for-purpose’ approach in support of their nonclinical studies. In the opinion of the authors, the issue of the EBF White Paper provides confidence within the industry, that a one-tiered approach is sufficient, thus allowing nonclinical immunogenicity testing to be more efficient, whilst still maintaining the scientific integrity of the data. Additionally, this ‘lean’ approach, leads to faster turn-around times for validations and sample analysis at reduced costs, and supports the 3Rs, with reduced matrix requirements. There are further gains to be had, from employing singlicate analysis, and have been a number of papers and scientific presentations⁵ (e.g. EBF 2017 & 2018 Open Symposium) that demonstrate this would not impact on interpretation of the impact of ADA on PK/PD data. However, there is still industry reluctance, due to fear of regulatory feedback, to change to singlicate analysis.

Conclusion

Drug Development Solutions is aware that evolution of guidelines is likely to cause further challenges in the future and that it is impossible to ‘future proof’ all analytical methods, and is also often unnecessary. Immunogenicity packages are critical to successful regulatory submission, therefore, we recommend a ‘fit-for-purpose’ approach, employing quality by design, for validations that meet the expectations within the guidelines, and follow a logical planned and justified scientific approach. For non-clinical immunogenicity assessments, generally a more ‘lean’ approach is appropriate in order to interpret impacted PK and PD where required.

For more of the latest science news, straight to your inbox, become a member of SelectScience for free today>>

References

Immunogenicity Testing of Therapeutic Protein Products - Developing and Validating Assays for Anti-Drug Antibody Detection, recommendations issued by the U.S. Department of Health and Human Services, Food and Drug Administration, Centre for Drug Evaluation and Research (CDER), Center for Biologics Evaluation and Research (CBER), January 2019

The Guideline on Immunogenicity Assessment of Biotechnology-Derived Therapeutic Proteins, EMEA/CHMP/BMWP/14327/2006 Rev 1, effective 01 December 2017

Regulation and standardisation of immunogenicity assessments: a CRO’s perspective, Hawes A et al (2018), Bioanalysis Zone

A strategic approach to nonclinical immunogenicity assessment: a recommendation from the European Bioanalysis Forum, Lauren et al, (2021) Bionalysis, 17 Mar 2021 

Comparing singlet and duplicate immunogenicity assay in human plasma for pembrolizumab using Gyrolab®, Stanta et al (2021) Bioanalysis, 30 Apr 2021