Optimal trispecific CAR construct simultaneously targets BCMA, GPRC5D, and CD38 to treat antigen-heterogeneous multiple myeloma

Abstract Introduction: BCMA, GPRC5D, and CD38 are highly and independently expressed on malignant plasma cells in relapsed/refractory multiple myeloma (R/RMM), with double- or triple-negative populations rarely observed, making them compelling immunotherapy targets(1). While CAR-T therapies have transformed the treatment landscape for R/RMM, many patients eventually develop resistance and relapse, mainly due to heterogeneity in antigen expression and limited CAR-T cell durability(2-3). Optimizing the in vivo persistence and anti-tumor activity of CAR-T cells is essential to improve their clinical efficacy(4-5). In this context, we have developed a trispecific CAR-T therapy targeting BCMA, GPRC5D, and CD38, integrating the JAK/STAT5 signaling pathway to enhance antitumor efficacy in the presence of soluble BCMA (sBCMA), minimize antigen escape, maintain CAR-T cell fitness, and promote expansion. We aim to achieve broad and durable tumor clearance and enhance clinical outcomes in patients with R/RMM. Methods:In this study, we screened an in-house phage display library, identifying lead antibody candidates with high affinity and specificity for BCMA, GPRC5D, and CD38, based on binding profiles and functional performance in T cells, respectively. We engineered a series of multi-specific CAR constructs sharing the same antigen-binding moiety but differing in configuration, and systematically evaluated their in vitro functionality and in vivoefficacy and persistence in preclinical models. We hypothesized that our lead multi-specific CAR-T cells would enhance their engagement and sensitivity to MM cells, counteract antigen escape, and overcome the inhibitory “antigen sink” effect caused by elevated sBCMA levels. Furthermore, they would show no toxicity toward CD38-positive normal cells. Incorporation of a novel JAK/STAT5 signaling module is expected to preserve T-cell fitness and enhance expansion. Results: Upon stimulation with BCMA, GPRC5D, and CD38 antigens, the multi-specific CAR-T cells exhibited enhanced STAT5 phosphorylation signaling, proliferation, cytotoxicity, and cytokine secretion, along with target-dependent expansion upon repeated tumor cells rechallenge, which are hallmarks of potent immune activation. In vitro studies revealed that these cells exhibit high avidity and effectively target MM cells expressing either BCMA or GPRC5D, even in the presence of sBCMA. Importantly, cells solely expressing CD38 are not targeted, indicating high target specificity and minimized off-tumor toxicity. In vivo, multi-specific CAR-T cells demonstrated potent anti-tumor activity, leading to complete tumor eradication in mice co-inoculated with antigen-heterogeneous MM cell lines and significantly prolonged survival. Mechanistically, targeting multiple antigens promotes CAR multimerization, which enhances cytotoxicity, reduces antigen escape to effectively eliminate MM cells, and improves therapeutic response. Conclusions: Multi-specific CAR-T cells were well tolerated and represent a promising and novel approach for targeting antigen-heterogeneous MM. Compared to single or bi-specific CARs, these cells demonstrated robust anti-tumor activity in vitro and in vivo. Furthermore, incorporation of the JAK/STAT5 signaling pathway enhances T-cell fitness and reduces exhaustion. Our findings highlight the potential of multi-specific CAR-T therapy as a next-generation treatment for R/R MM. Ongoing clinical trials will further assess their therapeutic efficacy.