Stability Testing: Monitoring Biological Product Quality Over Time

16 Min Read

19-3-FR-Ritter-P1-300x200.jpgMany physical and chemical factors can affect the quality, safety, and efficacy of biopharmaceutical products, particularly after long-term storage in a container–closure system that can be subject to variations in temperature and light, as well as agitation with shipping and handling. Proteins are inherently complex physiochemically, from their primary amino acid sequences to their higher-order structures, and they require specific conditions to maintain their integrity and functionality. Advanced biological therapies can be even more complicated and particular about their environments.

Stability testing is used to detect changes in identity, purity, and potency of a formulated drug product over time. Stability assays are used throughout each product’s life cycle, beginning with development and performance of comprehensive and specific stability protocols during preclinical development and early clinical phases. Under the quality by design (QbD) paradigm, stability is part of a biotherapeutic’s quality target product profile. Stability test results help analysts understand how critical quality attributes (CQAs) of drug substances and drug products respond under different conditions.

Biomanufacturers conduct stability tests to determine degradation pathways and establish shelf lives and storage conditions for their products. Types of tests and bioassays used to monitor purity, identity, potency, quality, and safety are conducted depending on the product type and intended use. Because in-house stability testing takes time and resources, some companies choose to outsource this work to contract testing laboratories. Consultants who specialize in assay development and related strategies can help prevent problems that might arise otherwise.

With master’s and doctoral degrees in cell and molecular biology from Rice University, long-time BPI editorial advisor Nadine Ritter spent 10 years in basic academic research at the University of Texas Health Science Center. She then entered the biopharmaceutical industry, beginning as a protein chemist in analytical R&D at Abbott Laboratories and later moving on to lead the analytical services division at BioReliance.

Since 2002, she has been an international consultant, trainer, speaker, and writer — having contributed to the success of more than 120 global regulatory filings and laboratory inspections. She serves on the chemistry, manufacturing, and controls (CMC) scientific advisory boards of several organizations and was a cofounder of the CASSS CMC Strategy Forum series. Ritter is now president of the CASSS board of directors. And in January 2021, she was kind enough to answer a few questions about stability assay development for this featured report.

Stability Testing Across the Biotherapeutic Lifecycle

All stages of a biopharmaceutical product life cycle include stability testing. Experts have divided these tests into six stages from early development to late-stage follow-up testing. During preclinical and phase 1–2 clinical trials, stability studies monitor the stability of clinical-trial materials and support formulation development. During phase 3, they provide important documentation for marketing applications. Once a product is marketed, further studies monitor its quality over time and provide data to support postapproval changes.

Source:Rios M. Product Stability Testing: Development Methods for New Biologics and Emerging Markets. BioProcess Int. 13(5) 2015.

Putting Stability into Context
Can you describe the relationships between potency, lot-release, and stability testing? How do the results of one affect decisions made in relation to the other(s)? We expect each batch of a pharmaceutical product to be safe and efficacious from the day the first person gets the first dose of it to the day the last person gets the last. Therefore, testing a product batch at time of release is no more or less important to patients than is testing it over time for stability. We must ensure that quality is maintained throughout a drug’s intended shelf life, which for biological products is based on long-term, real-time stability data. That is because large-molecule biological products do not follow simple, linear, first-order degradation kinetics, so it is challenging to estimate the shelf life of such products from basic high-temperature modeling studies, as is done for many small-molecule drugs. Would you want to assign a predicted shelf life to cold milk based on data from boiling it?

Moreover, sometimes the degradants of therapeutic proteins can be biologically active and create problems such as undesirable immunogenicity. So the safety risk of being wrong on product stability can be quite high. This is not to say that accelerated and stress stability studies are not important — they are for reasons described below.
In terms of potency, biological products have legal quality requirements to be both pure and potent (biologically active). Think of an egg. Whether it is fresh or fried, I can confirm analytically that it is an egg — that it contains the necessary chemical components and is not contaminated. But only one of an egg’s physical formats will be functionally capable of ever becoming a chick. Just having the right components present does not guarantee that a batch of biological product will be effective. The quality control tests applied for biological products therefore must include physiochemical and functional methods of analysis.

How does information from early product-development stages inform design of assays for later stages? Several highly diverse elements of stability testing factor into numerous CMC studies during development and postapproval of biopharmaceutical products. Forced-degradation studies are critical for selection and validation of orthogonal stability-indicating methods. Accelerated and stress stability and forced degradation are important elements of comparability studies to support process changes or biosimilar product analytical similarity assessments. Stress studies are used for formulation screening and optimization. And stability testing over time at a target storage temperature is required first to assure us that clinical-trial materials remain stable for use, then to help us establish a proposed shelf life for commercial product.

The single most important aspect of all these studies is to know which study type is needed when and how to design and execute each one properly. I see stability-study design mistakes all the time in which a lot of work is done but the results are not useful for intended objectives.

Some companies choose to keep stability testing in house; others outsource this work. How does that choice affect assay development? And what are some general strategies to prevent technology-transfer problems in outsourced stability testing? Whether for release or stability testing, assay development intended for quality control (QC) applications should target routine, reliable, robust performance under operational constraints that R&D applications never face. The hidden problem I find in method transfers from R&D to QC is that once a transfer is complete, the long-term operational robustness of the method is ignored. Its invalid rate is not tracked or trended to confirm whether the method continues to perform appropriately over time.

The biggest specific gap in transfer of stability-indicating methods is a failure to include known degraded product samples from all potential degradation pathways that are detectable by a given method. Early in product development, it may be acceptable to use only thermally degraded samples in transfer. But for late development and commercial stability methods, the transfer should include degraded samples from all degradation pathways each method is used to detect: e.g., light, heat, agitation, oxidation, and so on. This will help to ensure that the method functions appropriately for its intended stability use.

Addressing Different Product Modalities
Are specific analytical methods and assays applied routinely for particular product modalities? Yes, ICH Q6B defines a “core” analytical profile for biological products (1). It includes testing for product-related substances, product-related impurities (e.g., degradants), process-related impurities, and contaminants (e.g., bioburden, endotoxin, sterility, and leachables). Even though ICH Q6B was issued in 1999 for well-characterized biological products, its exact same principles are relevant for all classes of biological products, regardless of how molecularly simple or complex they are. The central ICH Q6B analytical paradigm is that a biological product could have all the molecular heterogeneity that can be measured with orthogonal analytics, monitored through a suitable analytical control strategy, and maintained over time on stability from point of manufacture to point of use.

Methods for identity, product-related purity and degradants, content and concentration, and potency tend to be product specific, although some developers are exploring platform versions for designated product modalities. Methods for process impurities are usually process specific but might apply across several products that share upstream process components. Methods for environmental contaminants are generally compendial and applicable to all products and processes, with product-specific verification of noninterference. Other compendial general methods are used routinely for product appearance, pH, osmolality, subvisible particulates, and moisture content (if the drug product is lyophilized), with product-specific verification of performance for key method characteristics.

What are the main challenges with stability testing modern biologics? One common product-development issue I see is believing that a product must be stable because you don’t see it degrade. But are you looking with the right methods? When developing pharmaceutical products, we cannot simply assume that our methods will detect degradation. The risks to safety and efficacy are too great. We are required to prove that we have a stability-indicating analytical profile with real data. The technical challenge is to use a complete set of orthogonal analytics that can detect potential physical and functional changes in a product if it should degrade. That’s not to say that it will degrade, but rather that if it does so, then at least one method in our stability tool kit is capable of detecting that.

We’ve heard a lot about cold-chain–related challenges with messenger RNA vaccines for COVID-19. How does that affect stability assay development for such products? The regulatory requirements are exactly the same in that there must be appropriate, orthogonal stability-indicating methods to detect degradation of the product under the biochemical pathways that are possible for nucleic acids and lipids. Recent presentations (e.g., at the CASSS “Well-Characterized Biotechnology Product” meeting) have shown incredible case studies of rapid development, validation, and technology transfer of novel stability methods for mRNA vaccines. It remains to be seen how robust those technologies will be in QC applications over the long term, but given the timelines of the COVID vaccines, the past year has been nothing short of amazing.

Transfer and Bridging of Test Methods
You’ve written with the CASSS CMC Strategy Forum on bridging methods for release and stability testing (2). Have any technical/regulatory developments of note occurred since then in this area? A major US Food and Drug Administration (FDA) guidance was issued in 2015 for analytical method validation, and it contains a method-lifecycle section that covers method bridging and technology transfer (3). It came out at about the same time as the CASSS CMC Strategy Forum proceedings paper (2), so many of the points made in the forum reflect principles from that FDA guidance. I urge anyone working on method bridging and transfer (and validation or verification) to review that excellent FDA guidance document as a primary source of information.

For global regulatory guidance, the World Health Organization (WHO) published in 2011 a guidance that also contains a section on method transfer and includes a table of different method applications with suggested strategies to consider (4). The WHO document is for chemical and biological products, so some suggestions might not be suitable for biopharmaceutical methods. Replication suggestions in the WHO table might surprise some folks, but it is a great overall guidance to use as a resource in planning method transfer activities.

I have not seen other collaborative industry–agency white papers published specifically on analytical method bridging and transfer — at least not for well-characterized biological products. In 2017, the European Federation of Pharmaceutical Industries and Associations (EPFIA) published an industry best-practices white paper on accelerated CMC programs, and it does mention methods (5). And in 2018, the International Society for Pharmacoepidemiology (ISPE) published the third edition of its good practice guidance for technology transfer, which includes a section on analytical methods (6). Both EFPIA and ISPE emphasize the project management elements to consider in these transfers, which is very useful information (7). The Parenteral Drug Association (PDA) put out a 2012 technical report on analytical method validation and transfer (8), and the US Pharmacopeia (USP) has chapter <1224>. Those also contain helpful information on method transfers.

Since 2016, I’ve read a few articles from contract or in-house laboratories on their own method-transfer strategies, including using covalidation in lieu of transfer. However, I urge caution in covalidation to replace formal transfer of methods between laboratories. From my own experience, method validation in one lab is complicated enough!

Moreover, transferring a validated method to a new lab often requires some modification of the procedure for it to work in the recipient lab. So the standard operating procedure (SOP) in the receiving lab will have procedural differences from that in the sending lab.

When it comes to method lifecycle compliance, it is difficult to manage two procedural options of the same method in two labs when all the supportive data reside in one method-validation package that is administratively comanaged between them. All method SOP changes, validation amendments, supportive revalidation data, and so on then must go through two different approval pathways, both of which have to agree. And all data from both sites needs to be shared transparently throughout the lifecycle of the analytical method(s) — something many contract laboratories will not do because of proprietary constraints.

Two questions were posed by audience members at the CMC Strategy Forum but “postponed for future discussion,” and I’m wondering whether you’d like to comment on those. First, is there a possibility of establishing a preapproved comparability protocol for method changes in the same manner as for process changes? Preapproved method comparability (bridging) protocols have continued to be a point of discussion at analytical meetings. Theoretically they should be possible. But here is a serious hitch: To preapprove a comparability protocol (whether for a process or analytical method), regulators expect specifics. Exactly what types of change(s) will be covered by a given protocol? Exactly what experimental design will be used to compare the impact of the change on performance? Exactly what types of samples will be used in the protocol experiments? Exactly what replication scheme will be used to generate sufficient comparative data? Exactly which acceptance criteria will be used for declaring “comparable” performance?

It is enough of a challenge to get technical, quality, and regulatory teams to agree on such specifics for method-bridging protocols when information about the changes is in hand and the change is eminent. You also want to have the chance to do feasibility runs and “sanity check” a protocol before signing off on it. Now imagine committing a detailed comparability protocol in writing to regulators in advance that will be applicable to future method changes — even though you don’t know yet know the reason for change or what you will change to until the time comes. Every method change is different, and even though I’ve probably worked on thousands by now, even when I’ve had all relevant information in front of me, there are still surprises.

Finally, is there a possibility of garnering coordinated regulatory review for global methods changes? ICH Q14 is intended to help promote global regulatory convergence on method quality by design (QbD), which if done and documented appropriately is intended to facilitate postapproval method changes harmonized in ICH Q12 (9, 10). As with everything related to CMC, however, the devils are in the details. Different regulatory agencies that might agree in principle still can disagree on what types and amounts of data will be necessary for their review and approval. Also, some health authorities have regional legal constraints over what regulators are required to do for postapproval changes, some of which include changes to approved specification methods.

1 ICH Q6B. Specifications: Test Procedures and Acceptance Criteria for Biotechnological/Biological Products. US Fed. Reg. 64, 1999: 44928;

2 Ritter N, et al. Bridging Analytical Methods for Release and Stability Testing: Technical, Quality, and Regulatory Considerations. BioProcess Int. 14(2) 2016: 12–23;

3 CBER/CDER. Analytical Procedures and Methods Validation for Drugs and Biologics. US Fed. Reg. 80, 44357–44358;

4 TR 961, Annex 7: Transfer of Technology in Pharmaceutical Manufacturing. World Health Organization: Geneva, Switzerland, 2011;

5 White Paper on Expedited CMC Development: Accelerated Access for Medicines of Unmet Medical Need — CMC Challenges and Opportunities. EFPIA-EBE: Brussels, Belgium, December 2017;

6 Good Practice Guide: Technology Transfer, Third Edition. International Society for Pharmaceutical Engineering: North Bethesda, MD, December 2018;

7 TR 57: Analytical Method Validation and Transfer for Biotechnology Products. Parenteral Drug Association: Bethesda, MD, 2012;

8 Martin GP, et al. Lifecycle Management of Analytical Procedures: Method Development, Procedure Performance Qualification, and Procedure Performance Verification. Pharmacopeial Forum 39 (5) 2013;

9 ICH Q12. Technical and Regulatory Considerations for Pharmaceutical Product Lifecycle Management. ICH Steering Committee, September 2014;

10 ICH Q14 (concept paper). Analytical Procedure Development and Revision of Q2(R1) Analytical Validation. International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use: June 2018;

Further Reading
Patel J, et al. Stability Considerations for Biopharmaceuticals: Overview of Protein and Peptide Degradation Pathways. BioProcess Int. 9(1) 2011:

Rieder N, et al. Binding Assays and Bioassays: What Are Their Roles in Lot Release and Stability Testing? BioProcess Int. 8(6) 2010.

Rios M. Biologics Stability: Lifecycle Management of Drug Products. BioProcess Int. eBook December 2019:

Rios M. Outsourcing Stability Testing: Discussions with Contract Laboratories. BioProcess Int. 13(10) 2015: 16–21;

Wang X, et al. Antibody Higher Order Structure Stability: Polymorphism Revealed By Protein Conformational Array. BioProcess Int. 15(11) 2017:

Nadine M. Ritter, PhD, is president and analytical advisor for Global Biotech Experts, LLC, in Germantown, MD; 1-240-372-4898; [email protected]; Cheryl Scott is cofounder and senior technical editor of BioProcess International, part of Informa Connect; [email protected].

2021 Bioassay Events from Informa Connect Life Sciences

Analyzing Bioassays and Immunoassayslive online course premiering in 2021 (dates to be determined)with Melody Janssen (SciPot Consultancy) and Zeban Kolen (Byondis) virtual-classroom course covers basic statistical processes necessary for analysis of bioassay data, setting up meaningful system suitability acceptance criteria, and designing validation plans. Important considerations include choosing appropriate linear/nonlinear regression fit, calculating correct residual weights, computing practical limits for sensitivity and acceptable error, and measuring nonparallelism and potency.Well-Characterized Biologics and Biological Assays26–28 October 2021, Hyattsville, MD features several regulatory speakers along with presentations from Amgen, AstraZeneca, Biogen, Bluebird Bio, Global Biotech Experts, Genentech/Roche, Eli Lilly, Janssen, KBI Biopharma, Sangamo, Sanofi, and more. Tracks cover bioassay development for non-MAb products; bioassay optimization and improvements; control of adventitious agents in raw materials; implications of global pharmacopeial monographs on biosimilar products; next-generation approaches; analytical challenges and considerations for formulations, drug products, biosimilars, and novel modalities; structure and function development; and strategies for accelerated method development and phase-appropriate GMP work.

You May Also Like