Background

Chimeric antigen receptor (CAR)-T cell therapies have led to durable remissions in cancer patients who are at high risk of relapse and death.1 2 However, despite great success, there is still a significant fraction of patients who do not achieve response. It has been a common goal of various research groups to identify which T cell subsets contribute to optimal clinical responses and to characterize the mechanisms that determine their effective antitumor properties.3 4 One of the major challenges is the fact that cells obtained from some patients with hematological malignancies, particularly with chronic lymphocytic leukemia, might display dysfunctional phenotypes. Apheresed cells from these patients are characterized by a lower frequency of naïve, and a higher frequency of effector memory and exhausted T cells, which leads to poor efficacy of their CAR-T cell products.

Preclinical models provide the proof of concept required to translate novel findings into clinical trials and are essential to predict the potency and/or efficacy of a given treatment. It has been previously shown that the concurrent utilization of CD4⁺ and CD8⁺ T cells demonstrates synergistic antitumor effects, both in vitro and in vivo.5 6 Those studies have also reported a higher in vivo potency when using defined CD4:CD8 ratios in the CAR-T products as opposed to unselected T cells. A clinical trial involving 29 adult B-ALL patients used equal proportions of CD4⁺ and CD8⁺ T cells to manufacture anti-CD19 CAR-T cells, which yielded remarkable effects even though these subsets were manufactured separately (NCT01865617).7 Nevertheless, these studies highlight the importance of maintaining CAR-T cells in a less differentiated state and having both subsets present within the product to more likely achieve long-term functionality and avoid T cell dysfunction.

Commentary

One question that Lee et al’s group8 raised was that the separate manufacture of CD4+ and CD8+ CAR T cells adds cost, complexity as well as risks of having a product that does not fulfill the ideal requirements in terms of phenotype and function. They observed that separately cultured CD8+T cells were hypofunctional when challenged with CD19+ target cells, even after they are mixed with CD4+ after infusion. In addition, the preselection of CD4+ and CD8+ T cells takes away one variable which may affect clinical outcome, but this approach will inadvertently remove T cells not expressing either coreceptor. While much remains unclear regarding the nature and function of such cells, they likely to play a role in CAR-T cell response as we recently demonstrated.9 It has been proven possible by Lee et al8 and others that it is possible to manufacture and grow both CD4⁺ and CD8⁺ T cells together, and then formulate the defined ratio that will be infused. However, a commercial product (lisocabtagene maraleucel), which provides equal target doses of CD4⁺ and CD8⁺ T cells that are cultured separately, is an example that this strategy can similarly provide high response rates in patients with relapsed/refractory hematological malignancies. Importantly, on the other hand, some factors that are intrinsic to the patient and vary from one to the other still remain unclear, such as the maintenance of memory and effector function, lack of costimulatory molecules on the surface of some tumor cells, the role of metabolic and epigenetic regulators, and the relationship between lentiviral integration sites and CAR-T function. Another important point raised was regarding the manufacturing conditions of CAR-T cell products that until the moment are not optimized, and vary from one place to the other. During this process, many variables contribute to the function of the final product, such as the T cell selection method, activation with beads versus antibody cocktails (supplemental cytokines) (IL-2, IL-7/15), and culture duration. Insights were provided on how CD8⁺ T cells become hypofunctional if cultured in the absence of CD4⁺ T cells in the ex vivo context. This was explained by the CD4⁺ T cell-mediated cytokine production (elevated IFN-γ, TNF and IL-2 in mixed cultures) and the contact-dependent mechanisms of these cells mediated by the CD40-CD40L and CD27-CD70 axes, that increase CD8⁺ T cell proliferation and memory formation contributing to their overall effector function. In other words, the ligation on CD8+CAR T cells with CD4+CAR T cells by CD4-CD40L or CD27-CD70 (respectively) provides a costimulatory signal that, paired with cytokines produced by CD4+T cells, increases ex vivo expansion and contributes to a more functional phenotype. This corroborates with data from a previous study which investigated the ‘help’ CD4⁺ T cells provide to CD8⁺ T cells through the CD27-CD70 axis.10 Cost and time were factors pointed out by the authors that have to be considered during CAR-T cell production. They modified their protocol considering their findings of improved ex vivo expansion and a similar CD4:CD8 ratio at the end of manufacture compared with culture initiation, in order to design a clinical trial. The expected reduced culture times and improved efficacy of this new product can potentially result in a more cost-effective approach to manufacture CAR-T cells.