FDA's 2026 CGT Potency Assurance Guidance: Why Product Potency Cannot Be Reduced to a Single Test

LuTra Studio

An analysis of the FDA's final guidance, with practical implications for CAR-T, AAV gene therapy, CMC development, IND submissions, and commercial manufacturing.

Introduction: A Product Can Pass Its Release Tests—But Can It Deliver Its Intended Therapeutic Effect?

Imagine a company preparing a batch of autologous CAR-T cells for a patient. The manufacturing run is complete. Cell viability, CAR expression, and viable cell count meet the predefined specifications. At first glance, everything looks acceptable.

Yet one question remains: can these cells perform the biological function required for the intended treatment?

A viable T cell is not necessarily an effective CAR-T cell. And a vector with an acceptable genome titer is not necessarily capable of delivering functional genetic material. For cell and gene therapy (CGT) products, quality attributes are important—but they do not all measure the same thing, and none should automatically be treated as proof of therapeutic activity.

On November 2, 2026, the U.S. Food and Drug Administration (FDA) issued the final guidance Potency Assurance for Cellular and Gene Therapy Products. It supersedes the Agency's January 2011 guidance, Potency Tests for Cellular and Gene Therapy Products [1, 2].

The CGT Potency Assurance framework is broader than assay development. Potency should be assured through a coordinated, science- and risk-based strategy that combines manufacturing process design, process control, material control, in-process testing, and potency lot-release assays. It is a lifecycle question, not merely a final-release question.

In this article, I examine what the guidance actually says, how its recommendations differ from a narrow testing-centered mindset, and what they mean for CAR-T, AAV, technology transfer, and CMC teams preparing for IND and BLA milestones.

1. What Does Potency Mean in Cell and Gene Therapy?

Under 21 CFR 600.3(s), potency refers to a product's specific ability or capacity to effect a given result, as indicated by appropriate laboratory tests or adequately controlled clinical data obtained through administration as intended [1].

The distinction between what is present and what the product can do is particularly important for CGT.

Consider a CAR-T product. A development team might characterize cell identity, viability, cell count, CAR expression, phenotype, cytokine secretion, and target-cell killing. Each contributes useful information, but none is inherently interchangeable with another. Cell viability addresses survival. CAR expression addresses an engineered feature. A cytotoxicity assay measures a particular biological activity. Their relevance to potency depends on the product's mechanism of action (MOA), intended indication, and supporting evidence.

The same distinction applies to viral-vector gene therapy. Vector genome titer quantifies genomes; it does not by itself establish that the vector retains the ability to enter cells, deliver its payload, or generate a biologically active product.

This does not mean every characterization assay should become a release test. The first task is to determine which attributes are genuinely related to potency and which risks are not already sufficiently controlled by other elements of the manufacturing strategy.

2. CGT Potency Assurance: What Changed from the 2011 Guidance?

The 2011 document focused on potency tests for cellular and gene therapy products. The November 2026 final guidance retains important recommendations on potency assays but places those assays within a comprehensive Potency Assurance Strategy [1].

FDA identifies five mutually supporting elements: manufacturing process design; manufacturing process control; material control; in-process testing; and potency lot-release assays.

For example, if extended cell culture is shown to impair a relevant functional attribute, a sensible control strategy might include a justified upper limit on culture duration, in-process monitoring, and a suitable release assay. Relying only on the final assay may leave avoidable risks uncontrolled.

There are two qualifications worth making. First, the guidance does not impose a universal list of potency assays across different CGT modalities. Second, FDA guidance generally contains nonbinding recommendations, unless it cites applicable statutory or regulatory requirements. The document itself says its integrated approach does not increase sponsor requirements [1].

The practical shift is toward explaining why a set of controls is adequate for a particular product and development stage—not simply demonstrating that a laboratory test produces a numerical result.

3. Building a Potency Assurance Strategy: From MOA to CQAs and CPPs

The central framework starts with product and process understanding, supported by quality risk management.

Quality Target Product Profile (QTPP)

A QTPP should identify the potency-related characteristics expected of the product in light of its MOA, intended indication, and route of administration. A CAR-T program may prioritize relevant cellular functions; an AAV program may focus on functional transgene delivery and the biological activity of the expressed product.

Potency-Related Critical Quality Attributes (CQAs)

FDA recommends identifying attributes that matter for the intended therapeutic effect and controlling them within phase-appropriate limits. Evidence may come from product characterization, nonclinical work, and clinical data.

A correlation between an attribute and an outcome can be informative, especially when the MOA is incompletely understood. But correlation alone does not prove a causal link. A stronger mechanistic understanding should be pursued when additional evidence is needed.

Critical Process Parameters (CPPs)

CPPs are process parameters that may influence CQAs. In cell therapy manufacturing, culture duration or process conditions may alter viability, differentiation, or function. The key is to determine which relationships are relevant to the product's potency and to establish justified operating limits.

Quality Risk Management

Potency risk assessment should consider severity, probability, and, where applicable, detectability. It should extend beyond manufacturing to container closures, equipment and devices that contact the product, storage, shipping, handling, thawing, and preparation for administration.

A useful way to frame the strategy is: MOA and clinical use → candidate potency-related CQAs → risk assessment → process and material controls → in-process monitoring → phase-appropriate lot release → ongoing verification.

FDA also encourages sponsors to use prior knowledge and experience, while recognizing that different starting materials, manufacturing processes, or mechanisms of action can create product-specific risks.

4. Potency Assays: The Best Test Is Not Necessarily the Most Elaborate Biological Model

It is tempting to assume that an ideal potency assay must reproduce the entire therapeutic mechanism. FDA's guidance takes a more practical position. The purpose of a bioassay is to quantitatively measure a potency-related CQA that remains at risk. The assay does not have to mimic the product's complete MOA [1, Section V.B].

This is a meaningful distinction for analytical development. A complex cell-based assay may be biologically compelling but too variable, slow, or difficult to transfer for routine release. Conversely, a simpler method may provide valuable risk reduction if it reliably detects a relevant loss of function.

For most CGT products, the release strategy should include at least one phase-appropriate bioassay that measures biological activity related to the intended therapeutic effect. FDA also explains that multiple release assays are typically appropriate when different at-risk potency-related CQAs require independent control. These may include physicochemical assays and bioassays.

The aim is sufficient control, not maximum assay count. If a later biological activity necessarily depends on earlier steps, a suitable downstream assay may help assure those earlier steps. But if different activities pose independent risks, one readout may be insufficient.

Assay selection should balance relevance, precision, accuracy, specificity, robustness, sample availability, turnaround time, and the ability to detect meaningful changes. The guideline also encourages replacing animal-based bioassays with suitable in vitro methods when potency assurance is not compromised.

5. CAR-T Case Study: Designing a Layered Potency Control Strategy

The following is an illustrative development framework based on FDA principles, not a prescribed CAR-T release specification.

For a patient-specific CAR-T product, the core therapeutic concept is that engineered T cells recognize an antigen-bearing target and generate a relevant functional response. A potency assurance plan might therefore evaluate attributes associated with cell health, CAR expression, phenotype, target recognition, and effector function.

Starting material risk. Patient-derived cellular material varies by disease, previous therapy, and donor-specific biology. Those differences can influence manufacturing performance and product functionality. Material qualification and relevant process controls should account for this variability.

Manufacturing risk. Activation, genetic modification, expansion, harvest, cryopreservation, and thawing can potentially affect potency-related CQAs. Which parameters are critical should be determined experimentally and through risk assessment—not assigned automatically.

Release strategy. A functional assay such as target-cell killing may be useful when appropriately designed and supported, but other assays may be needed to control different at-risk CQAs. A release strategy based only on viability or CAR-positive percentage would need strong evidence that it adequately addresses the relevant potency risks.

Vector control. FDA specifically recommends that, for vector-transduced patient-specific cell-based gene therapy products, each vector lot have demonstrated biological activity before use in drug-product manufacture. The strategy should therefore include a bioassay and quantitative acceptance criterion for each vector lot, in addition to potency controls for the final drug product [1, Section V.B.2].

This matters in autologous manufacturing. Discovering that a vector lot has inadequate activity only after a patient's cells have been processed can delay an already time-sensitive treatment.

6. AAV Gene Therapy: Vector Genome Titer Is Not Functional Potency

AAV programs illustrate why potency assurance must extend beyond a single concentration measurement.

A batch can contain a measurable number of vector genomes while differing in capsid integrity, nucleic-acid integrity, the ability to deliver genetic material, or subsequent expression and function. These are different quality dimensions and should not be conflated.

For viral-vector gene therapies intended for direct administration that express a transgene, FDA generally recommends including a potency release assay—even in early development—that quantifies transgene mRNA expression, protein expression, or protein activity in transduced cells [1, Section V.B.2].

As development progresses, sponsors should assess risks to vector particles and nucleic acids, including risks to structural integrity, delivery, nucleic-acid length, sequence, and activity. Where other process and quality controls do not sufficiently mitigate a relevant potency risk, additional quantitative potency assays may be appropriate.

For CMC teams, the implication is straightforward: comparing process yields and genome titers alone is not enough to establish comparable biological performance across manufacturing changes.

7. Very Short Shelf-Life Products: What If a Bioassay Cannot Finish Before Release?

Some noncryopreserved cell therapies must be administered before a conventional functional bioassay can be completed. FDA recognizes this constraint and describes a risk-based approach rather than a universal requirement to wait for a full bioassay result.

For extremely short shelf-life products, appropriate physicochemical potency assays at release may be supported by sufficiently informative in-process testing and a tightly controlled manufacturing process.

FDA recommends initiating one or more potency bioassays immediately after drug-product manufacture for investigational products in this situation, then evaluating the results after release. Such post-release studies help verify product potency and process reliability. For licensed short shelf-life products, post-release bioassays may contribute to continued process verification when they meaningfully reduce risk [1, Sections IV.F and V.B.2].

This is not a general exemption from potency requirements. It transfers greater responsibility to process understanding, monitoring, release controls, and follow-up verification. Product-specific justification remains essential.

8. Potency Assurance from IND to BLA: Phase-Appropriate Does Not Mean Optional

Initial IND and Early Clinical Development

Before clinical investigations begin, FDA recommends identifying preliminary potency-related CQAs, assessing risks, and developing a strategy for reducing those risks. The initial IND should describe the approach in CTD Module 3, with a summary in Module 2.

The documentation should cover MOA, QTPP, preliminary CQAs, risk assessments, controls, available potency methods and performance characteristics, acceptance criteria where applicable, and a plan to mature the strategy.

Not every element must be fully developed at the first IND. FDA even addresses early-phase situations where release testing for potency is not yet part of the control strategy: sponsors should explain how process design and other controls still provide adequate assurance.

Later Clinical Development and Studies Supporting Effectiveness

For significant-risk clinical investigations and studies intended to provide substantial evidence of effectiveness for marketing approval, the control strategy should include at least one suitable release assay—physicochemical or biological—that quantifies a potency-related CQA, with justified quantitative criteria capable of rejecting sub-potent lots.

The assay should be qualified for its intended purpose. Sponsors should also have evidence that relevant CQAs remain stable during storage and preparation for administration.

Insufficient potency assurance can raise substantial clinical and interpretability concerns. Under applicable IND rules, FDA may place a study on clinical hold when required information is missing or a subject would face an unreasonable and significant risk; inadequately controlled potency can be relevant to that assessment [1, Section III.B].

BLA and Licensed Products

Before BLA submission, sponsors should reevaluate potency assurance using accumulated quality and clinical data. Release and in-process methods must meet applicable validation requirements. FDA also expects sponsors to explain important differences between assays or reference materials used during pivotal clinical investigations and those proposed for the licensed product.

Licensed biologics are subject to lot-release potency testing requirements under 21 CFR 610.1 and 610.10, with possible alternative approaches under 21 CFR 610.9 when the specified showing is made.

The goal is continuity: the licensed product should remain sufficiently consistent in potency with the product administered in clinical investigations that established effectiveness.

9. Acceptance Criteria: Why 'Report Only' Is Not a Release Standard

FDA recommends quantitative acceptance criteria for potency-related CQAs. Criteria such as "report" or "for information only" do not, by themselves, reduce the risk of releasing a sub-potent lot [1, Section V.D].

An appropriate quantitative lower limit should help demonstrate that each lot has adequate capacity to achieve its intended therapeutic effect. Where excessive biological activity may pose safety concerns, or the safety of unusually high potency is unclear, a suitable upper limit should also be considered using available process, nonclinical, and clinical evidence.

Early development criteria may be comparatively permissive for products with high inherent variability when justified in an IND—but they still need to reject lots outside an acceptable potency range.

For licensed products, limits should support consistency with the clinical lots that provided the primary evidence of effectiveness. A specification derived only from the distribution of manufacturing data, without considering clinical relevance and risk, can miss that objective.

10. Assay Qualification, Validation, and Reference Material Management

Biological assays are vulnerable to variability across operators, instruments, cell lines, reagent lots, and test days. FDA recommends evaluating analytical characteristics including precision, accuracy, specificity, reportable range, and robustness.

Qualification establishes whether assay performance is fit for its intended use; validation confirms performance against appropriate predefined criteria for the finalized method, where required. The guidance recommends qualifying potency release methods as soon as feasible and no later than the start of investigations intended to provide substantial evidence of safety and effectiveness for a marketing application.

Reference materials deserve early attention. Many CGT products have no suitable compendial potency standard, making in-house standards necessary. A representative, well-characterized DS or DP lot may be useful, provided material quantity and stability permit it. Replacing a reference lot should involve a defined qualification protocol, independent comparisons to the existing lot, and prespecified statistical procedures.

FDA also recommends considering a control material distinct from the calibrator reference lot for system suitability. Control charts can help detect changes in assay performance over time.

For method transfer to a CDMO or external testing laboratory, a prospective transfer study should examine reproducibility with sufficient statistical power. FDA specifically discusses equivalence testing as one approach for assessing whether results from the new method or laboratory are sufficiently similar to the original [1, Section V.C].

11. Process Scale-Up, Technology Transfer, and Comparability

One of the most consequential questions during CMC development is whether a manufacturing change preserves the properties that matter clinically.

I have encountered versions of this challenge throughout drug-delivery and process-development work: a method performs well at research scale, but a different manufacturing scale, supplier, facility, or analytical platform can alter product attributes in ways that are not obvious from yield alone.

For CGT, changes in cellular starting materials, vector supply, equipment, culture conditions, cryopreservation, filling, handling, or testing laboratories can affect potency-related CQAs. These changes do not necessarily reduce potency; they create risks that need to be assessed using relevant evidence.

The 2026 guidance points out that manufacturing changes may introduce risks not captured by routine release tests. Consequently, a comparability exercise may require additional product characterization beyond the standard lot-release panel [1, Section V.A].

This is why potency assay development, process characterization, and comparability planning should not operate in separate silos. If a program waits until late development to identify potency-related attributes, it may have inadequate historical data to support a future process change.

12. Five Practical Actions for CGT Development Teams

1. Start the potency strategy before the pivotal trial. Identify candidate potency-related CQAs during preclinical and early clinical development, and plan a path toward qualified and validated methods. FDA especially encourages early assay development when the clinical program may move quickly.

2. Link process development to function. Use product and process characterization to establish which materials, steps, and process parameters can affect functional potency—not just productivity or conventional quality measurements.

3. Treat materials as possible potency risks. Patient starting material, vectors, growth factors, and other biologically active inputs may need specifications or testing tied to their potential effect on potency.

4. Build in analytical lifecycle management. Establish suitable reference materials, system suitability criteria, control strategies, and a plan for method changes or transfer to external laboratories.

5. Engage FDA with specific questions. When seeking feedback, explain the proposed potency-related attribute, its relationship to MOA and intended effect, the assay design and limitations, and the evidence supporting the risk-control strategy. CBER recommends early consultation and targeted questions rather than broad requests for general approval of an assay concept [1, Section IV.H].

13. My Perspective: Potency Assurance Is Ultimately About Understanding the Product

For a research team, the first milestone is often to show that a therapeutic concept works. An engineered T cell kills its target. An AAV vector produces a functional protein. That proof of concept matters—but clinical development asks a more difficult question: can we reproduce the required biological performance across lots, sites, operators, and product lifecycles?

I view the FDA's final guidance as an argument for closer integration of discovery biology, analytical development, process development, quality, and regulatory affairs. Potency is not simply an analytical-department deliverable. It is the thread connecting mechanism, manufacturing variability, process control, release, stability, and patient use.

Startups will not always have the resources to validate every assay at the outset. The guidance recognizes phase-appropriate development. What it does not support is postponing the scientific reasoning until a late regulatory milestone.

The strongest CGT programs will be those that can explain not only what they measure, but why those measurements and controls are enough to provide confidence that every released lot retains its intended therapeutic capacity.

About LuTra Studio

LuTra Studio works at the intersection of biotechnology consulting, scientific communication, career development, and collaboration between the U.S. and Taiwan. We cover translational research, RNA therapeutics, drug delivery, cell and gene therapy, and the CMC questions that arise as technologies advance from laboratory studies toward clinical development.

Through The Biomedical Scientist's Blog, we share evidence-based discussions of emerging science, manufacturing, and regulatory developments. If your team is working on CGT, LNP, biologics process development, or cross-border collaboration, visit LuTra Studio to explore how we can support your next stage.

References

1. U.S. Food and Drug Administration. Potency Assurance for Cellular and Gene Therapy Products: Guidance for Industry. Final guidance, November 2026. https://www.fda.gov/media/196183/download

2. U.S. Food and Drug Administration. FDA Provides Clarity to Support Cell and Gene Therapy Development and Quality. News release, November 2, 2026. https://www.fda.gov/news-events/press-announcements/fda-provides-clarity-support-cell-and-gene-therapy-development-and-quality

3. U.S. Food and Drug Administration. Potency Tests for Cellular and Gene Therapy Products. January 2011 (superseded). https://www.fda.gov/regulatory-information/search-fda-guidance-documents/potency-tests-cellular-and-gene-therapy-products

4. U.S. Food and Drug Administration. Considerations for the Development of Chimeric Antigen Receptor (CAR) T Cell Products. January 2024. https://www.fda.gov/media/156896/download

5. U.S. Food and Drug Administration. Human Gene Therapy Products Incorporating Human Genome Editing. January 2024. https://www.fda.gov/media/156894/download

6. ICH / FDA. Q9(R1) Quality Risk Management. May 2023. https://www.fda.gov/media/167721/download

7. U.S. Food and Drug Administration. Chemistry, Manufacturing, and Controls Flexibilities for Developing Human Cellular and Gene Therapy Products for a Biologics License Application. May 2026. https://www.fda.gov/media/192321/download

Editorial note: The CAR-T, AAV, and process-development scenarios in this article illustrate how the guidance's principles may be applied. They are not FDA-endorsed product-specific specifications. FDA guidance generally provides nonbinding recommendations; applicable statutes and regulations remain controlling.

Image note: The cover is a conceptual scientific illustration created for LuTra Studio; it does not depict a literal AAV genome structure, a real CAR-T micrograph, or experimental findings.