One of the more consequential misunderstandings in cell and gene therapy development is treating critical quality attributes as a fixed list — something defined early in a program and carried forward unchanged. In conventional biologics, this assumption has some validity. A monoclonal antibody has a relatively stable characterization profile across its development lifecycle. The CQAs shift in emphasis, but rarely in kind.

In CGT, that assumption is wrong in ways that matter.

Critical quality attributes for cell and gene therapies are not static. They evolve with the program, they expand in scope as the product advances toward regulatory approval, and they interact with manufacturing changes in ways that can require re-characterization of the entire quality profile. Understanding how CQAs change — and why — is essential to building a development strategy that does not generate the wrong data at the wrong time.

What CQAs Are, and Why They Are Different in CGT

The FDA defines a critical quality attribute as a physical, chemical, biological, or microbiological property or characteristic of a product that should be within an appropriate limit, range, or distribution to ensure the desired product quality. In plain terms, a CQA is a measurable attribute of the product that, if it falls outside its acceptable range, affects safety, efficacy, or both.

For cell and gene therapies, the five core CQA categories are safety, identity, sterility, purity, and potency. Each of these applies broadly to biological products, but the specific attributes that fall within each category, and the assays used to measure them, are far more complex for CGT than for conventional biologics.

Safety encompasses tumorigenicity, biodistribution, differentiation and integration behavior, and off-target effects. Identity requires characterization through cell surface markers, expression assays, vector genome analysis, and sequencing. Sterility testing must address a range of microbial and viral contaminants including mycoplasma, adventitious agents, HIV, HBV, HCV, and human TSE. Purity accounts for impurities from the manufacturing process — pyrogenicity, endotoxins, residual manufacturing materials, unintended cell types such as lymphocytes and cytokines. Potency, which the FDA defines as the specific ability or capacity of a product to effect a given result, encompasses the most complex measurement challenge in CGT: mechanism of action confirmation, expression assays, vector copy number, replication competent lentivirus assays, and immune response profiling.

Each of these categories requires customized assays, because potency and identity in CGT are product-specific in a way that has no equivalent in conventional drug development. There is no standard assay for CAR-T potency that applies across programs. There is no single identity test that works for an autologous cell therapy and an AAV gene therapy and a gene-edited stem cell product. The assay development challenge is inseparable from the CQA definition challenge.

How the Development Pathway Shapes CQA Requirements

The CGT development pathway moves through six broad stages — discovery and research, preclinical, Phase 1, Phase 2, Phase 3, and post-approval — and the CQA focus shifts materially at each transition.

In discovery and research, the analytical priority is understanding disease biomarkers, characterizing the tumor microenvironment, and identifying the therapeutic candidate. CQA work at this stage is foundational: understanding what attributes will eventually need to be controlled requires knowing what the product is supposed to do and how it is supposed to do it. This is the stage at which mechanistic hypotheses are formed, and those hypotheses will determine which CQAs matter most later.

At the preclinical stage, the analytical priority shifts to verification. The regulatory questions being asked at this stage focus on proof of concept, toxicology, and post-administration safety. Proof of concept analytics require characterization of the mechanism of action, optimization of route of administration, timing, and dosing schedule, and pharmacological effect assessment. Toxicology requires functional assessments, evaluation of immune response (humoral and cellular), vector biodistribution, tumorigenicity assessment, and evaluation of CRS, HAMA, and immunogenicity. Post-administration monitoring requires tracking vector and gene expression distribution, differentiation and integration, and safety-specific transgene behavior — local versus systemic response, immunogenicity, off-target shedding.

This is also the stage at which the FDA expects an immunogenicity risk assessment as part of the IND submission. That assessment must account for the nature of the drug, the mechanism of action, the route and frequency of administration, patient-specific factors, and the disease indication. Getting this right is not a formality — immunogenicity assessment determines the monitoring strategy for the entire clinical program.

Phase 1 introduces a new set of CQA priorities organized around trial risk, safety, feasibility, preliminary potency, and pharmacological activity. The sponsor focus at this stage includes pharmacological analytics — short-term and long-term response, gene expression levels, biomarker changes, morphological changes, immune function changes — and dose-exploration protocols to establish minimum and maximum effective dose, toxicity levels, dose effectiveness plateau, and dosing schedule. CQA testing at this stage is less about release criteria and more about generating the data that will eventually inform tighter specifications at Phase 3.

Phase 2 marks the transition from dose exploration to qualification. Gene expression levels, persistence, quality, stability, and biodistribution all require formal study. Biomarker changes must be systematically tracked against clinical endpoints. The CQA testing at Phase 2 is where the correlative relationship between product characteristics and clinical outcomes begins to be established — or where its absence becomes apparent.

By Phase 3, CQA requirements have become substantially more demanding. Lot release testing requires demonstration of potency across multiple dimensions: VCN, RCL assays, and expression assays. Safety testing includes tumorigenicity, biodistribution, differentiation, integration, and off-target effects. Identity requires characterization. Sterility must address mycoplasma, adventitious agents, and the full panel of viral contaminants. Purity testing must address process-related manufacturing impurities.

Post-approval, the CQA focus shifts to monitoring — potency in the context of disease recurrence and long-term immune response, expression stability, and persistence. For some products, post-approval safety monitoring extends years beyond the initial approval, because the long-term behavior of integrated genetic material is not yet fully characterized at the population level.

Table 2. Critical Quality Attribute (CQA) focus by development stage across the CGT pathway. Each cell identifies the primary analytical questions and testing activities required at that stage for each CQA category. Sourced from MKA research, FDA CGT CMC Guidance (January 2020), and MKA CGT Analytical Services research (2022). Gray shading indicates Phase 3/Lot Release, where CQA requirements are most comprehensive.

Product Design and Process Design: The CQA Feedback Loop

Underlying all of this is a dynamic that many CGT developers underestimate: product design and process design do not operate independently. They operate in a close feedback loop that changes and evolves with each stage of development, and every change to one side of that loop has implications for the other.

Product design encompasses the hypothesis driving the therapeutic, the mechanism of action, the effective attributes of the therapy, its safety profile, persistence characteristics, and the specific cellular or genetic target. Process design encompasses starting cell characteristics, stable gene expression, gene editing efficiency, cell fitness for therapeutic benefit, reproducible quality, purity and impurity control, and dose. When a process change is made — a different cell expansion protocol, a modified purification step, a change in vector titer — it does not merely affect the process. It affects the product. And if it affects the product in ways that touch any of the CQAs, it may require re-characterization and, depending on the stage of development, re-submission of data to regulators.

The FDA’s CMC guidance for gene therapy INDs makes this explicit. Changes in manufacturing may affect safety, identity, quality, purity, potency, and stability. Sponsors are expected to understand which process parameters are critical — those that have the potential to influence batch reproducibility, product performance, and product quality — and to demonstrate that their control strategies adequately address the CQAs at each stage.

The practical implication is that CQA strategy cannot be treated as a one-time activity. It must be built as a living framework that evolves with the program, that anticipates process changes and their downstream quality implications, and that maintains alignment between what the FDA expects to see at each stage and what the sponsor’s analytical program is actually measuring.

Building CQA Strategy That Grows With the Program

The most common failure mode in CGT CQA strategy is not a failure to define the CQAs — it is a failure to build the assay development and analytical infrastructure to measure them at the right stage. Potency assays, in particular, present a persistent challenge. Because potency in CGT is product-specific and mechanism-dependent, developing a validated potency assay is itself a multi-year scientific program. Organizations that defer this work until Phase 2 or Phase 3 routinely find themselves without the analytical tools needed to support regulatory submissions on the timelines those stages demand.

The same applies to identity and characterization. The level of molecular, biochemical, immunologic, phenotypic, physical, and biological characterization required for a BLA submission is substantially greater than what is typically in place at IND. Building toward that level of characterization requires a deliberate analytical development roadmap — one that maps each CQA to the assays needed to measure it, the timeline for assay development and qualification, and the regulatory touchpoints at which that data will be required.

For sponsors building these roadmaps, the starting point is a clear articulation of what the product is and how it works — because the CQAs flow from the mechanism of action, not the other way around. The programs that build the most robust analytical foundations are those that take the time to connect their scientific understanding of the product to a systematic map of what needs to be measured, when, and with what level of rigor.

MKA Insights supports CGT developers in building CQA and analytical development strategies that are calibrated to the actual regulatory requirements at each stage of development. If your program is approaching an IND or navigating the transition between clinical phases, we bring the cross-domain perspective to ensure your quality framework grows with your program rather than behind it.