Gene Therapy Potency — CBER’s Exacting Expectations for Lot Release Testing

A potency assay for a gene therapy product is not just a release test. It is the primary evidence that your manufacturing process consistently produces a biologically active vector.

That single sentence is the lens through which CBER evaluates your potency strategy at every stage of development — from pre-IND consultation through BLA review. Yet in practice, the gap between what sponsors believe constitutes an adequate potency package and what CBER’s Office of Tissues and Advanced Therapies actually requires is one of the most consequential and most underestimated disconnects in gene therapy CMC development. Sponsors who treat potency assay development as a Phase 3 activity — something to be refined after the clinical signal is established — arrive at BLA submission with the most common and most intractable CMC deficiency in the gene therapy space.

What ‘Potency’ Means for a Gene Therapy Vector: The Regulatory Definition and Why It’s Complex

The regulatory definition of potency is not ambiguous. Under 21 CFR 600.3(s), potency is “the specific ability or capacity of the product, as indicated by appropriate laboratory tests or by adequately controlled clinical data obtained through the administration of the product in the manner intended, to effect a given result.” For a gene therapy vector, this definition carries a specific analytical consequence: potency must be linked to the mechanism of action, not merely to the quantity of vector present. FDA’s guidance “Potency Tests for Cellular and Gene Therapy Products” (January 2011) is unambiguous on this point — the agency expects a functional assay that measures biological activity relevant to the product’s intended therapeutic effect, and it explicitly warns that physicochemical or biological tests used as potency assays are acceptable only when they have been demonstrated to correlate with relevant biological activity.

This is where the vector genome (vg) titer — measured by ddPCR targeting the ITR sequence or a transgene-specific amplicon — fails as a standalone potency test. vg titer is a measure of genome quantity, not biological activity. A preparation with 1 × 1013 vg/mL and a preparation with 1 × 1013 vg/mL but with 70% empty capsids, denatured capsid protein, or a defective promoter driving the transgene cannot be distinguished by vg titer alone. CBER’s position, articulated consistently in pre-IND meeting feedback and in the 2011 guidance, is that the full lot release package for a gene therapy product must include both quantitative evidence (vg titer is acceptable as a surrogate here) and functional evidence derived from a cell-based assay. The potency hierarchy that CBER applies for AAV products places a cell-based transduction assay measuring functional transgene expression at the apex, with an infectious unit (IU) assay on permissive cells such as HeLaRC32 as a secondary functional measure, and vg titer occupying its appropriate role as a quantitative surrogate — not as the primary potency determination.

ICH Q6B, Section 6, reinforces this architecture at the international guidance level: specifications for biological products must include a quantitative measure and defined acceptance criteria for potency, and those criteria must reflect the biological activity of the product. The practical consequence for sponsors seeking global submissions is that neither CBER nor EMA’s Committee for Advanced Therapies will accept a potency specification that relies on titer alone, regardless of how precisely the titer is measured.

Transduction Efficiency, Gene Expression, and the Functional Potency Assay Requirements

Designing a cell-based potency assay for an AAV vector that will satisfy CBER requires three decisions made in sequence, each of which constrains the next. The first is cell line selection. The target cell line must express the physiologically relevant receptor for the serotype in question — and where the transgene encodes a protein with a cell-type-specific mechanism of action, the cell line should permit the transgene product to execute that function. Using HEK293T cells as the transduction target for an AAV9 vector delivering a secreted enzyme because they are convenient to culture is not a scientifically defensible design choice; HEK293T cells are a manufacturing substrate, not a physiologically relevant potency target. CBER reviewers examining your assay design will ask whether the cell line reflects the intended target tissue biology, and the answer must be grounded in the transgene’s mechanism of action.

The second decision is functional endpoint selection. The assay must measure something that reflects biological activity, not merely gene transfer. Three endpoint categories are routinely accepted by CBER: transgene mRNA quantified by RT-qPCR (which confirms transcriptional activity but not necessarily protein function), transgene protein quantified by ELISA or Western blot (which confirms translation), and enzyme activity measured by a substrate conversion assay (which confirms functional protein). For enzyme replacement gene therapies, the enzyme activity endpoint is the strongest regulatory argument because it is mechanistically linked to the therapeutic effect. For gene silencing or gene editing approaches, the endpoint design becomes more complex, but the principle is identical — the assay must measure the functional consequence of the biological event the product is intended to trigger.

The third decision is the relative potency calculation method. CBER expects potency to be expressed relative to a characterized reference standard — not as an absolute unit — using either parallel-line analysis or four-parameter logistic (4PL) curve fitting. The system suitability criteria for the assay must include a requirement that the reference standard response fall within ±2 standard deviations of the historical mean for that reference lot, and intermediate precision must be characterized with an assay CV of no greater than 25%. For parallel-line analysis specifically, parallelism acceptance requires a p-value greater than 0.05. These are not aspirational targets; they are the operational thresholds that distinguish a qualified potency assay from a research-grade transduction experiment, and their absence from the assay control strategy is a primary contributor to the deficiency pattern CBER identifies in pre-Phase 3 review meetings.

Potency Assay Development and the Regulatory Validation Standard for GT Potency Methods

The most operationally damaging misconception in gene therapy potency development is the belief that assay qualification can be deferred until the clinical program has generated data to anchor the specification. The 2011 CBER guidance on potency testing, read alongside ICH Q2(R2)/Q14 on analytical procedure development and validation, establishes a stage-gated expectation that is consistent with what CBER Office of Therapeutic Products (OTP) communicates in pre-IND meetings under the framework described in MAPP 5015.1: for Phase I, the potency assay must be qualified — meaning the assay has defined system suitability criteria, a preliminary reference standard has been established, and the method has been shown to be repeatable under defined conditions. Full ICH Q2(R2)-compliant validation is expected for BLA submission.

The reference standard is the element sponsors most consistently underinvest in early-stage development. USP <1032> and <1033> define the requirements for reference standard qualification: a primary reference standard should be established from a minimum of six independently manufactured lots, characterized by the full panel of analytical methods in the release specification, assigned a defined potency value, and stored under conditions demonstrated to preserve activity over the intended use period. A single-lot reference standard qualified from a pilot-scale batch — common in Phase I programs — is not disqualifying, but the qualification report must clearly define the lot characterization data, the assigned potency, and the plan for transition to a primary reference standard as clinical development advances. When that qualification report does not exist, or exists as an internal memo rather than a controlled document, the potency assay has no metrological anchor — and CBER will identify this absence precisely.

The potency specification range for gene therapy products is typically set at 50–200% of the reference standard, with the final range justified by the combined weight of manufacturing history and clinical evidence accumulated through Phase 3. The critical regulatory discipline is that the range cannot remain “TBD” at the Phase 2/3 boundary. CBER’s expectation, consistent with ICH Q6B, is that a pre-specified potency range with clinical justification — meaning a demonstrated relationship between assay response and the in vivo efficacy signal — must be in place before the pivotal trial initiates. A program that enters a pivotal study without a pre-specified potency specification cannot retrospectively justify the specification at BLA and will face a review cycle deficiency that, at minimum, delays approval by one complete review cycle.

Building a GT Potency Testing Strategy That Holds Up From Phase 1 Through BLA Submission

The XGene GT Potency Assay Lifecycle Architecture is a stage-gated development framework that maps every potency assay decision — cell line, endpoint, reference standard, system suitability, and specification range — to the specific regulatory deliverable required at each IND, Phase 1, Phase 2/3, and BLA milestone.

Step 1 — Mechanistic Anchoring (Pre-IND): Define the functional endpoint that is directly linked to the transgene’s mechanism of action, not to gene transfer efficiency, and document the scientific rationale in a potency assay design justification memo that can be presented at a CBER pre-IND Type B meeting as evidence that the assay strategy reflects the product’s biological activity.

Step 2 — Assay Qualification and Preliminary Reference Standard Establishment (IND to Phase 1): Execute qualification experiments demonstrating repeatability and intermediate precision (CV ≤25%), establish a preliminary reference standard with a documented characterization report under USP <1032>/<1033>, and define system suitability criteria — including reference standard response within ±2 SD of historical mean — that will govern every future lot release decision.

Step 3 — Specification Anchoring with Clinical Correlation (Phase 2 Transition): Establish the pre-specified potency range (typically 50–200% RS) using manufacturing history from Phase 1/2 clinical lots combined with the earliest available in vivo efficacy data, and build the bridging documentation that connects early clinical lots to the proposed commercial specification before pivotal study initiation.

Step 4 — Full Validation and BLA Readiness Package (Phase 3 to BLA): Execute full ICH Q2(R2)/Q14-compliant method validation, qualify the primary reference standard from ≥6 lots, and compile the potency assay validation report as a controlled BLA-ready document that maps every assay performance parameter to its acceptance criterion and links the specification range to the clinical justification narrative in Module 2.7.

The output of this framework is a BLA-ready potency assay dossier — not a gap list — that gives CBER reviewers a complete, stage-mapped record of every analytical decision made from pre-IND through submission, preempting the most common CMC deficiency in the gene therapy biologics license application review cycle.

Programs that defer potency assay development treat a foundational regulatory requirement as a late-stage analytical exercise — and the cost of that deferral is not measured in assay development hours but in BLA review cycles. A Complete Response Letter citing potency assay deficiencies at BLA is a minimum twelve-month setback for a product that may already have been in clinical development for a decade. The sponsors who arrive at BLA with a fully validated, mechanistically justified potency assay, a qualified primary reference standard, and a pre-specified potency range backed by clinical correlation data did not achieve that position by beginning potency assay development in Phase 3. They began it at the pre-IND stage, they invested in qualification during Phase 1, and they built the clinical-to-analytical bridge during Phase 2 — because that is the stage-gated approach CBER has described in its guidance and communicated in pre-IND meetings since 2011.

For your current GT program, can you confirm today that your potency assay measures biological activity linked to the mechanism of action (not just gene transfer), name the cell line and endpoint used, identify the reference standard lot and its characterization report, and state the specification range in your current lot release certificate?