Summary of FDA Guidance for Industry: Considerations for the Development of Chimeric Antigen Receptor (CAR) T Cell

In their associated clarification webinar, CAR T cell design was highlighted in particular with respect to changes made to the final guidance. As CAR T cell design has evolved to target multiple antigens, that is, when multiple CARs are expressed in a single drug product, FDA has observed that recombination events can occur during manufacturing which reduce product consistency and result in reduced expression of the CARs. These recombination events can adversely impact product development. From a CAR design perspective, sponsors and developers need to reduce if feasible this risk of recombination events. This change to the final guidance reflects the rapidly evolving scientific knowledge and understanding that will continue to occur as these products develop further.

Vector quality directly contributes to CAR T cell quality and consistency, described in the guidance as being ‘a critical component that furnishes a pharmacological activity for the treatment of disease’. From a design and development perspective, a key element of the guide to note is the use of vectors that integrate into the DNA (e.g. retroviral based vectors or transposons). This is due to the potential for long term transgene expression compared to non-integrated vectors. Consequently, long term follow up is recommended, ‘because integrating vectors may increase the risk of delayed adverse events’. In comments submitted by the Biotechnology Innovation Organization in 2022⁹, they requested that as more safety related data becomes available based on real time clinical experience, that FDA remains flexible in its definition of ‘long term follow up’. FDA has issued Guidance for Industry in 2020 – ‘Long Term Follow-Up After Administration of Human Gene Therapy Products’¹

The potential impact to patients who have been previously administered CAR T cells is emphasised in the updated guidance – Particular consideration should be given to patients who have received CAR T cells previously’. Such patients may be considered for different CAR T cell clinical studies due to lack of response. It is not yet clear how previously administered CAR T cells impact the safety and efficacy of the CAR T cells under development. As noted in the guidance, ‘Previously administered CAR-T cells in the starting material may have unexpected effects on CAR T cell manufacturing (e.g., expansion or transduction rates), potency, in vivo expansion safety and efficacy’. This impact can be assessed by performing an evaluation of previously administered CAR T cell levels in the cellular starting material, through the detection of common structural elements, although this monitoring may not necessarily impact patient eligibility.

As vectors are considered a drug substance (in contrast to the EMA position describing the vector as a starting material), all aspects of its manufacture according to phase specific expectations, including full GMP manufacturing for late-stage clinical studies and licensure need to be considered. The current gene therapy CMC Guidance⁸ provides recommendations for manufacturing and testing of the vector. Lot release expectations are phase specific, with full safety testing and appropriate characterisation for early phase studies. FDA recommends that lot release testing includes the measure of vector concentration (i.e. strength), such as titre for viral vectors so the amount of vector can consistently be applied in CAR T cell manufacturing. For biological activity, while transgene expression alone may be sufficient to support early phase IND studies, additional measures of biological activity are needed for clinical study(s) intended to provide evidence of effectiveness to support the market application. As noted above, there are additional safety concerns and testing expectations related to the use of retroviral-based vectors.

While CAR T cells can be manufactured from a variety of cells, this guidance focusses on leukapheresis derived materials as this is the most common sources at this time. Consistent leukapheresis procedures are important to support consistent product manufacturing, with a recommendation that procedures for collection are consistent across different sites. In particular, the maintenance of the Chain of Identity (COI) should be described and validated for licensure.

In the associated FDA Webinar at www.fda.gov, FDA responded to a query regarding controlling the impact of donor variability on drug product variability in late phase CAR T cell development. Donor variability is a significant contributor to lot-to-lot variability of the product. FDA recognizes that even with a consistent manufacturing process, there is going to be an expected level of lot-to-lot variation in the final drug product. For autologous products, it is recommended to include a sufficient number of product lots and patients in clinical studies, which should inform sponsors of the acceptable level of variability that will be apparent in the commercial product. For allogenic, while the incoming material is generally more uniform as it comes from healthy donors, it is recommended that a range of donors are used to manufacture different lots as part of the clinical study enabling the definition of a more robust manufacturing space and to understand the clinical impact of product attribute variability.

This section of the guidance provides details on process control expectation to address the inherent sources of variability within the manufacturing process as a result of use of different donors where applicable. Added to this is the variability associated with a multi-step manufacturing process. Ultimately, expectations for CAR T manufacture align with the normal product development expectations including progressive characterisation and control of the manufacturing process.

Changes to this section mostly include updates on the original text that in most cases, have not changed the original intent but do enhance the clarity of the requirements including. The following changes are points to note:

  The extent of in-process testing and testing of intermediate and final product to be as appropriate for the phase of product development.

  To assure product safety, the aseptic processing must be validated for licensure.

  The updated guidance distinguishes between stability test requirements for fresh or cryopreserved product.

  For stability studies, the final guidance expands on the use of health donor material for both early stability studies and licensure, taking into account the need to enable determination of product shelf life.

The guidance gives direct advice on some assays that are particularly applicable to CAR T cells. It’s notable that the delineation between autologous and allogenic is made with respect to the requirement for additional testing for allogenic products may be required beyond what is currently outlined in the final guidance.

The assays outlined below are addressed in the guidance with additional specific examples for potency expectations.

  Flow cytometry

  Vector copy number

  Identity

  Potency

While the changes to this section are mostly clarification, it’s clear that the agency recognises the challenges for the developers who need to demonstrate appropriate control of the test methods as distinct from having validated methods for Phase 1 clinical studies.

With respect to Vector Copy Number (VCN), over recent years, the FDA had updated its recommendations for monitoring the average number of vector integrations in the final drug product is included in the draft guidance from 2022. FDA in the past accepted vector copy number measurements as a function of total cells or as a function of total number of CAR positive cells. FDA now recommends, as it did in the draft guidance, reporting vector copy number per CAR positive cell on the final certificate of analysis provides a more useful assessment of patient risk and allows for a better comparison of product variability across lots.

For potency testing, the final guidance provides specific examples for different product design potency expectations. This additional detail is to support the need to address where CAR T cells express multiple transgenes.