Novel CTLA-4 costimulatory domain mitigates cytokine-related toxicity, improving the anti-cancer therapeutic index of CS1-CAR T cells for multiple myeloma

Abstract Multiple myeloma (MM) is a prevalent hematological malignancy in the United States, with approximately 34,000 new cases reported annually. Chimeric antigen receptor (CAR) T cell therapy has demonstrated remarkable response rates and sustained anti-tumor activity in MM. CS1/SLAMF7, a cell surface glycoprotein highly expressed on MM cells, is an attractive CAR target for MM. Our ongoing clinical trial testing CS1-targeted CAR T cells with a 4-1BB costimulatory domain (NCT03710421) has demonstrated potent anti-MM activity. However, this efficacy was accompanied by severe cytokine-related toxicity. Thus, we proposed to refine the CS1 CAR design by replacing the 4-1BB signaling domain with that of cytotoxic T-lymphocyte antigen 4 (CTLA-4), to reduce inflammatory cytokine release while improving the anti-cancer therapeutic index. CTLA-4 is a transmembrane inhibitory receptor expressed on activated T cells that competes with CD28 for binding to CD80/CD86 on antigen-presenting cells and is a negative regulator of T cell activation. The intracellular domain of CTLA-4 modulates T cell activity by recruiting two key phosphatases to the membrane: 1) PTP1B, which attenuates T cell receptor and CAR signaling, thereby dampening activation; and 2) PP2A, which reduces the production of pro-inflammatory cytokines like IFN-γ, IL-2, and TNF-α, which prevents excessive immune activation and tissue damage. To date, CAR T cells that use the CTLA-4 costimulatory domain have not been explored in the context of cancer therapy. We hypothesized that this design would provide favorable kinetics of CAR T cell activation, to preserve anti-tumor activity while reducing the risk of sustained hyperactivation and toxicity. We designed two CS1 CARs with different costimulatory domains—4-1BB or CTLA-4—using the same lentiviral backbone. Lentiviral vectors were produced in-house, and CS1 CAR T cells were generated from the same donor to control donor variability. We performed phenotypic characterization using flow cytometry to assess markers of memory (CD62L, CD28, CCR7, CD45RA), exhaustion (PD-1, TIM-3, LAG-3), and activation (CD25, CD69, CD137). We evaluated effector function following stimulation with MM.1S cells by measuring CD107a expression and intracellular IFN-γ production. We assessed cytolytic activity following co-culture of CS1 CAR T cells with GFP-expressing MM.1S cells at an effector-to-target ratio of 1:1 for 24 hours by measuring the remaining GFP+ tumor cells by flow cytometry. Following co-culture with MM.1S cells, CS1-CTLA-4 CAR T cells exhibited 2–3-fold lower Th1 cytokine secretion (IFN-γ, GM-CSF, IL-2, TNF-α, and granzyme B) and decreased CD137 expression (P<0.05) compared to CS1-4-1BB CAR T cells, indicating lower levels of activation. CS1-CTLA-4 CAR T cells maintained cytotoxic activity and a comparable memory T cell phenotype to that of CS1-4-1BB CAR T cells, suggesting that CTLA-4 signaling suppressed cytokine production without impairing anti-MM function (P>0.05). We generated a mutant CTLA-4 construct (CTLA-4m) with a tyrosine-to-phenylalanine mutation in the cytoplasmic domain. The CS1-CTLA-4m CAR restored T cell overactivation, high intracellular cytokine production, and degranulation compared with wildtype CS1-CTLA-4 CAR (P<0.05), supporting the modulatory function of the CTLA-4 co-stimulatory domain. To test anti-MM activity, we engrafted NSG mice with 2×106 MM.1S cells expressing firefly luciferase by intratibial injection. Five days later, we injected 1x106 CS1-CTLA-4 CAR T cells intravenously into tumor-bearing mice, using untransduced (mock) T cells and untreated mice as controls. We monitored tumor burden weekly using bioluminescent imaging. CS1-CTLA-4 CAR T cells exhibited significantly better anti-tumor activity compared to mock T cells (P<0.05, N=5 different donor-derived CAR T cell products), confirming the potency of these cells. We have expanded the evaluation of the CTLA-4 costimulatory platform to CAR constructs targeting CD19 and IL-13Ra, confirming the improvement in the CAR T cell anti-cancer therapeutic index. The ultimate goal of testing CS1-CTLA-4 CAR T cells is to develop a novel cellular therapy that addresses the urgent need for safer and more effective treatment options for patients with relapsed/refractory MM. Our CS1-CTLA-4 CAR T cells had reduced activation and Th1 cytokine secretion while maintaining cytotoxic activity, supporting its potential as a safer CAR design for MM.