The potency test ensures that a product meets predefined biological functional characteristics before batch release for clinical use. In the case of CAR-T cells, this test assesses their ability to recognize the CAR’s specific target antigen by measuring one of three induced lymphocyte functional activities: cytotoxic activity, cytokine production, or proliferation. A quantitative evaluation of one of these functional activities helps validate the manufacturing process and ensures the quality of the final CAR-T cell preparation. This test can also be used to validate a freezing/thawing step and to conduct stability studies. In the best-case scenario, although not its primary purpose, the potency assay could also serve as a predictive biomarker of clinical efficacy in treated patients.

Several biological factors associated with favorable clinical outcomes in CAR-T cell therapy have been identified. For instance, the dose of reinfused CAR-T cells has been suggested to influence clinical outcomes [32, 33]. Higher degrees of expansion (Cmax) and/or longer durations of expansion (area under the curve, AUC) of CAR-T cells in patients have been identified as biomarkers associated with positive clinical responses [34, 35]. On the other hand, CAR-T cells with insufficient potency may be ineffective, whereas an overly potent product can lead to toxicity, such as cytokine release syndrome (CRS) or on-target, off-tumor toxicity, including neurological (ICANS) and hematological (ICAHT) toxicities. Moreover, prolonged in vivo persistence of CAR-T cells has been correlated with the depth and durability of clinical responses [32, 36].

Currently, the most performed potency test is the quantification of interferon-gamma (IFN-γ) production after CAR-T cell stimulation. Recently, the percentage of CAR-positive T lymphocytes (transduction efficiency) and cell viability have also been classified as potency assays under certain regulatory frameworks, particularly by the FDA. Indeed, the FDA has recently drafted standardized expectations for CAR-T cell products. In particular, lot release criteria for early-phase Investigational New Drug (IND) applications do not require fully validated potency assays and specifications. However, such assays must be included when generating data in support of a Biologics License Application (BLA) (Center for Biologics Evaluation and Research, 2022a).

Potency test methods for assessing the functional activity of CAR-T cellsAssessment of cytotoxic activity

After several hours of co-culture between CAR-T cells and target antigen-expressing cells, CAR-T cell cytotoxic activity can be assessed in two different ways. The first approach involves measuring target cell lysis. Currently, four techniques can be used for this measurement: Chromium-51 release, bioluminescence (luciferase-luciferin reaction), enzymatic release assays, impedance-based assays, and flow cytometry. Each method has its own advantages and limitations [37].

The second approach provides an indirect assessment of cytotoxic activity by detecting surface expression of CD107a/b on CAR-T cells via flow cytometry. This marker is upregulated following the degranulation of cytotoxic molecules such as perforins and granzymes.

After analyzing these methods, the working group concluded that target cell lysis measurement is the most relevant approach for evaluating CAR-T cell cytotoxic activity. Furthermore, to ensure broad implementation across French cell therapy units using existing equipment, flow cytometry was identified as the preferred technique. However, the feasibility of this choice depends on laboratory access to flow cytometers, which may not be universally available.

Assessment of proliferative capacity

Assessing the proliferative capacity of CAR-T cell preparations is a relevant parameter for CAR-T cell efficacy and toxicity in patients, as studies have shown that higher expansion levels (Cmax) and/or longer expansion durations (area under the curve, AUC) in patients are biomarkers associated with improved clinical responses [38].

After analyzing the advantages and limitations of available techniques for evaluating this biological function, the working group concluded that the most feasible approach would be a proliferation assay based on fluorochrome labeling (e.g., CFSE) and flow cytometry analysis. However, the group does not propose this test as a potency assay, as it requires a minimum of 3 to 4 days of CAR-T cell culture following antigen-specific stimulation. This extended timeframe is incompatible with the release of fresh CAR-T products.

Assessment of cytokine production

The production of one or more cytokines following co-culture of CAR-T cells with their specific antigen can be measured between 4- and 24-hours post-stimulation, depending on cytokine production kinetics. This assessment can be performed using three (or four) major assay types:

Intracellular staining with flow cytometry,

Cytokine quantification in co-culture supernatants via ELISA,

ELISpot assay,

Luminex technology, which combines ELISA with flow cytometry for multiplexed detection of multiple protein and/or nucleic acid targets.

Each of these techniques has advantages and limitations. Currently, commercially available CAR-T cell products with MA are released based on IFN-γ quantification. The working group emphasizes that this test is only relevant if it specifically quantifies CAR-dependent IFN-γ production. To ensure specificity, the assay should include a control mechanism, which can be based on:

effector cells (using either CAR-negative patient T lymphocytes or, ideally, T cells expressing an irrelevant CAR),

and/or target cells (using the same cell line with and without the CAR target antigen).

Ideally, both approaches should be included. Since obtaining additional patient-derived T lymphocytes is incompatible with clinical-scale CAR-T cell production, the working group recommends using target cell lines that are both positive and negative for the CAR antigen. This approach allows for the detection of CAR-specific IFN-γ production while distinguishing specific CAR activity from alloreactivity, which occurs when T cell receptors (TCRs) recognize major and minor histocompatibility complex differences between cell lines.

Thus, the proposed test should include four conditions.

1.

No stimulation (negative control).

2.

Stimulation with a cell line expressing the CAR target antigen.

3.

Stimulation with a cell line lacking the CAR target antigen.

4.

Non-specific polyclonal stimulation (e.g., anti-CD3 antibodies, anti-CD3/CD28 beads, PHA-L, etc.).

Group recommendations

The objective of this workshop is to identify a potency assay that can be implemented across all academic units while meeting the following criteria: simplicity, compatibility with automated equipment, and use of a technique already used in hospital immunology laboratories. Additionally, since most batches will be released fresh, the assay must be rapid. Therefore, in addition to cell counting/viability assessment and determination of the percentage of transduced cells, the group has chosen to develop and validate a potency assay based on IFN-γ quantification using ELISA or ELISpot. This choice is driven by the relative ease of implementation compared to cytotoxicity or proliferation assays, as well as the fact that all currently approved commercial CAR-T products are released based on IFN-γ quantification.

IFN-γ measurement is performed following antigen-specific stimulation of the CAR-T cell preparation. Several types of stimulatory agents can be used, including:

recombinant protein [39],

beads coated with the CAR target antigen,

human or murine [40] cell lines expressing the CAR target antigen.

The advantages and limitations of each of these stimulatory agents are summarized in Tables 4 and 5.

Table 4 Comparison of the advantages and limitations of different stimulatory agents for CAR-T cell preparations.Table 5 Comparison of ELISA and ELISpot techniques for IFN-γ quantification.

Following the analysis of the advantages and limitations of stimulatory agents and the two techniques, the working group will likely recommend the use of coated beads with the antigen or murine cells modified to express the target antigen, which were developed by Jean-Baptiste Latouche’s team [40], based on the results of a preliminary comparative study.

While the ELISpot test provides additional quantitative data (number of cells capable of producing IFN-γ), it is more complex and costlier (not universally available in all labs) compared to the ELISA technique. Therefore, the working group will likely propose conducting an ELISA test with an 18- to 24-hour incubation period to align with the end time of the manufacturing process and the working hours of technical staff.

A first comparative study (ELISA versus ELISpot, with different stimulatory agents) will be conducted by the biotherapy laboratories. Based on what has been done and the results of this study, recommendations will be issued for all biotherapy laboratories in future workshop.