Why You Should Measure Comparability Debt Before Ever Starting A Study

By Siddharth Krishnan

Demonstrating product comparability in the face of manufacturing changes remains one of the key areas for obtaining and maintaining authorization for advanced therapy medicinal products (ATMPs) like cell and gene therapies (CGT).^1,2 However, the ability to robustly demonstrate it in a regulatory context is determined long before process or analytical changes are implemented. Development programs could reach the threshold of a site transfer, scale-up, process optimization, or material change with considerably more uncertainty than is immediately perceptible. While well established in the development of biologics and reinforced by the concepts laid out in International Council for Harmonisation (ICH) Quality Guideline 5E, the execution of comparability exercises for CGT is challenging as they are not stable molecular entities.^3,4

Comparability debt is not a regulatory term. It is simply a way of crystalizing the technical uncertainty that could accumulate when analytical capability, process understanding, and reference materials do not develop at the same pace as the program. While not an entirely new concept, it is an attempt to holistically integrate the underlying ideas of life cycle product understanding, phase-appropriate development, risk-based change management, and the importance of analytical and process knowledge.

The Debt Arises And Accumulates

The development of CGT products is complex, and teams are faced with situations where decisions need to be made in the face of incomplete information and constrained resources. This situation is not uncommon in small therapy developers during early development where product and process knowledge is evolving, some decisions are provisional, and speed is of the essence. Assays to demonstrate potency are particularly good examples as they tend to be required before the mechanism of action is fully elucidated and can become part of broader development decisions, for example, to assess if later process or product iterations remain meaningfully comparable to previous versions.⁵ Here, analytical debt arises and raises the question of assay informativeness wherein the assay is developed with current understanding in the face of evolving science, manufacturing, and regulations and is applied in settings beyond merely potency.

Similarly, as CGT programs mature, manufacturing processes change when there are site changes, optimizations, changes to raw materials, implementation of automation, and so on. As development progresses, the value of evidence from the established process and generated material increases and makes subsequent changes progressively more consequential. From a regulatory angle, once clinical development is built around a process, subsequent knowledge must be related back to that historical process and its product.^1,2,6 Now, retained samples from starting material, historical batches, clinical material, and representative material take center stage as the debt can be further exacerbated by the inability to retrospectively create reference material that was never preserved.

The issues highlighted increase the complexity of the comparability study needed and the range of samples needed, especially when split studies might be required for autologous therapies to account for donor variability. The absence of a quality target product profile framework to support comparability exercises can also decrease the regulatory flexibility available to developers at later stages of development given how they leverage critical quality attributes (CQAs) and their relationship with critical process parameters (CPPs). An important caveat is that industry working groups have published risk assessment-based quantitative ranking tools to identify CQAs for monoclonal antibodies⁷ and vaccines⁸ in conjunction with ICH Q9⁹; such comprehensive guidelines for CGTs are nascent but growing.¹⁰

The Debt Becomes Visible

Debt visibility increases when manufacturing or analytical changes prompt new relationships not previously required.

Kymriah (tisagenlecleucel) provides an illustration. Following a manufacturing site change, CAR-positive T-cell products manufactured at two facilities met lot-release specifications, but additional characterization identified differences in cell growth and transduction efficiency, and the products were not considered comparable in the FDA review.

As comparability between the manufacturing sites had not been completed, the primary efficacy analysis was restricted to patients treated with product manufactured at the established site.¹¹ The manufacturing change exposed the limitations of the available evidence to demonstrate that the change had not altered the product in a clinically meaningful way. In this setting, the contribution of reference materials and historical product knowledge were key as comparability required meaningful reference controls against which the post-change material could be evaluated, using analytical methods capable of distinguishing relevant changes from biological and analytical variability.^12,13

Debt visibility also increases when comparability study design becomes a point of contention. Comparability studies do not create risk and instead reveal where it has accumulated, like the retrospective need to create samples that were not preserved earlier. Equally, frameworks around CQAs and CPPs, while helpful, do not guarantee comparability evidence as analytical methods will need to demonstrate their ability to detect meaningful change. Here, the debt becomes visible when analytical methods have changed or lack the qualification and bridging data sets, which bring into question their assay attributes such as sensitivity, specificity, precision, robustness, and so on. To this point, Zolgensma (onasemnogene abeparvovec) was initially granted conditional marketing authorization by the European Medicines Agency, with its regulatory history highlighting the consequences of unresolved manufacturing and comparability questions between early and later manufacturing processes.^5,14

Managing The Debt

Early programs struggle to eliminate uncertainty given the plethora of constraints they operate under. It is rarely feasible for studies to extensively characterize samples, validate assays, retain all samples, or predict future manufacturing changes. As a result, regulatory agencies acknowledge that products pre- and post-change will not be identical and instead prioritize similarity, efficacy, and safety.^1,2 Thus, comparability debt is not necessarily failure, as the objective is to manage risk using a fit-for-purpose and phase-appropriate approach to potentially reduce the cost of goods (CoGs) and improve access to a wider patient population.^15-17

CGTs by their very nature are complex and expensive to manufacture. Risky and complex comparability studies can increase the cost of manufacturing a successful batch though batch failures, re-manufacturing, out of specification investigations, delays, and additional studies. As such, comparability studies tend to become more manageable, with cost tied more directly to value, when comparability debt is risk-managed up front.

The decisions that determine future comparability evidence are distributed across functions in the development organization and a risk-based comparability strategy begins long before the comparability study, with the progressive product and process understanding. Managing comparability debt should ideally focus on maintaining the ability to interrogate changes as product and process knowledge develop. Comparability is therefore a life cycle responsibility, and the objective is not to predict every future manufacturing change but to ensure that the evidence base can evolve with the program. What you choose to understand and measure early in development can determine what can be meaningfully interrogated when the process and analytics change.

Key takeaways

• Comparability debt can accumulate long before a manufacturing change is proposed, as product and process understanding, analytical capability, and reference materials develop at different rates.
• Comparability studies reveal accumulated uncertainty rather than create it, with manufacturing or analytical changes exposing limitations in the existing evidence base.
• Early development decisions influence future comparability, particularly around what is measured, which samples are retained, and how product and process knowledge are built over time.
• Comparability is a life cycle responsibility, requiring a fit-for-purpose, phase-appropriate strategy that preserves the ability to meaningfully assess future changes.

References:

1. European Medicines Agency, “Questions and answers Comparability considerations for Advanced Therapy Medicinal Products (ATMP)” (2019).
2. Food and Drug Administration; Center for Biologics Evaluation and Research, “Manufacturing Changes and Comparability for Human Cellular and Gene Therapy Products; Draft Guidance for Industry” (2023).
3. European Medicines Agency, “International Council for Harmonisation - Q5E Comparability of Biotechnological/Biological Products” (2005).
4. Medicines & Healthcare products Regulatory Agency, “Advanced Therapy Medicinal Products Guidance T Cell and NK Cell Characterisation Assays” (2024).
5. P. Salmikangas, B. Carlsson, C. Klumb, T. Reimer, S. Thirstrup, Potency testing of cell and gene therapy products. Front. Med. (Lausanne). [Preprint] (2023).
6. N. Clément, S. H. Kassim, D. Leblanc, K. Spink, J. Tomtishen, The comparability tales: A phase-appropriate roadmap for CGT drug product development. Mol. Ther. Methods Clin. Dev. [Preprint] (2024).
7. N. Alt, et al., Determination of critical quality attributes for monoclonal antibodies using quality by design principles. Biologicals 44, 291–305 (2016).
8. CMC Vaccine Working Group, “A-VAX: Applying Quality by Design to Vaccines CMC-Vaccines Working Group” (2012).
9. European Medicines Agency, “ICH guideline Q9 (R1) on quality risk management” (2025).
10. Alliance for Regenerative Medicine, The National Institute for Innovation in Manufacturing Biopharmaceuticals, “A - CELL: A case study-based approach to integrating QbD principles in Cell-based Therapy CMC programs” (2022).
11. Food and Drug Administration, “Summary Basis for Regulatory Action (Kymriah)” (2017).
12. A. Cockroft, A. Wilson, Comparability: What We Can Learn from the Review of Advanced Therapy Medicinal Products. Regenerative Med. 16, 655–667 (2021).
13. S. Nahum, et al., From production to bedside: A tiered QC framework for regulatory alignment and analytical comparability in decentralized CGT. Regen. Ther. [Preprint] (2026).
14. Novartis, “Summary of the EU Safety Risk Management Plan v2.2 (Onasemnogene abeparvovec)” (2023).
15. K. R. Poudel, Z. Taraporewala, D. Blumenthal, D. Shah, K. Francissen, Comparability for Cell and Gene Therapy Products: A Challenge and an Opportunity. Bioprocess Int. (2024).
16. The Technology Partnership Plc, When COGs is more than just a number: engineering scalable cell and gene therapy manufacturing. (2026).
17. R. McCoy, S. Ward, N. Gaddum, J. Hasan, The necessity of automated manufacture for cell-based immunotherapies: a cost-based analysis. Cell Gene Ther. Insights 6, 673–690 (2020).

About the Author:

Siddharth Krishnan, Ph.D., is founder and principal consultant at Cellvanta Ltd., a specialist consultancy supporting cell and gene therapy development. Previously, he was a senior scientist in analytical research at Quell Therapeutics and before that an associate senior scientist in analytical development at the Cell and Gene Therapy Catapult. He received his Ph.D. in neuroscience from the University of Manchester.