Enhanced Memory Phenotype and Proliferation Dynamics of IL-15-Armored CAR-T Cells: Implications for CD8+ T Cell Function and Therapy

CAR-T cells have emerged as a significant alternative for treating hematologic neoplasms in patients with refractory/recurrent disease. Despite encouraging results, many patients experience disease recurrence after CAR-T cell therapy. The factors contributing to CAR-T cell failure are still under investigation. Although the medical and scientific communities have yet to elucidate the factors leading to CAR-T cell failure completely, recent studies have focused on identifying metrics of CAR-T cell products that correlate with better post-infusion responses. Notably, the T cell effector/memory phenotype and CD8/CD4 ratios in CAR-T cells have been identified as important predictive factors correlating with improved responses.

In this study, we utilized anti-BCMA CAR-T cells generated with a second-generation CAR construct (BCMA-41BB-CD3) and a fourth-generation construct incorporating the IL-15 cytokine (BCMA-41BB-CD3-IL15) to evaluate whether IL-15 expression affects the phenotype of CAR-T cells maintained in cell culture. We employed multiparametric flow cytometry to immunophenotype CAR-T cells with or without IL-15 expression. A 12-color panel was used to identify naive T cells, stem cell memory T cells (Tscm), central memory T cells (Tcm), effector memory T cells (Tem/emra), and effector T cells (Teff). CAR-T cells were generated from T cells isolated from peripheral blood mononuclear cells (PBMC) of healthy donors transduced with lentiviral vectors. CAR-T cells were cultured for 15 days with IL-2 supplementation and immunophenotyped at days 0, 7, and 14 post-isolation from PBMC. We also performed functional cytotoxicity and cytokine production assays with CAR-T cells co-cultured with multiple myeloma cell lines MM1.S and U266. Finally, we analyzed exhaustion markers using flow cytometry.

We achieved high transduction efficiency of T cells using lentiviral vectors, resulting in 75±10% CAR+ T cells (n=3). These CAR-T cells demonstrated robust proliferation and maintained high cell viability throughout in vitro expansion. By day 7, both CD4+ and CD8+ CAR-T cells exhibited initial expansion with Tcm and Tem phenotypes. By day 14, the proportion of Tcm CAR-T cells was significantly higher in the IL-15-armored CAR-T cells compared to second-generation CAR-T cells (p<0.001, n= 3). Conversely, CAR-T cells expressing IL-15 showed a lower proportion of Tem and Teff cells (p<0.01, n= 3). Further analysis revealed that within the CD8+ subset, but not the CD4+ subset, the proportion of less differentiated phenotypes (Tscm, Tcm) was significantly higher in CAR-T cells with exogenous IL-15 compared to those without IL-15 expression (p<0.05, n= 3). Notably, CD4+ CAR-T with IL-15 exhibited slower proliferation rates compared to CD8+ CAR-T cells of the same population (p<0.05, n= 3) and to CD4+ CAR-T cells without IL-15 expression (p<0.01, n= 3).

Additionally, we also analyzed CAR-T cells by flow cytometry using exhaustion markers (LAG-3, TIM3 and PD-1) and found no signs of T cell exhaustion due to IL-15 expression. Interestingly, when co-cultured with multiple myeloma cell lines, we identified a trend towards enhanced cytotoxicity and higher production of the cytokine IFNy, in a target specific manner. Further investigation into long-term T cell phenotype, T cell function and In vivo mouse model assays are underway to explore the effect of the incorporation of IL-15 into CAR constructs.

Cytokine armored CAR-T cells hold promise for enhancing cytokine signaling and local production while minimizing systemic side effects. The IL-15-armored CAR-T cells, expanded with IL-2 supplementation, exhibited a less differentiated memory phenotype, particularly in CD8+ T cell subsets, consistent with previous studies using IL-15 supplementation. Our data also suggested an increased CD8/CD4 ratio in IL-15 armored CAR-T cells. Both parameters are correlated with better clinical responses. Our findings align with reports indicating that less differentiated memory T cells proliferate slower in vitro compared to more differentiated, effector-biased T cell subsets. Our data support the development of IL-15 enhanced CAR-T cells as an option to achieve local cytokine support to CAR-T cells without the additional costs and safety concerns of systemic cytokine administration.

No relevant conflicts of interest to declare.