Mesothelin-targeted CAR-T cells secreting NKG2D-BiTEs exhibit potent efficacy against triple-negative breast cancer

Abstract

Triple-negative breast cancer (TNBC) is an aggressive subtype with poor prognosis and limited treatment options. Chimeric antigen receptor (CAR)-T cell therapy holds promise, but its efficacy is hindered by tumor antigen escape and heterogeneity. To address these challenges, we developed a novel bispecific T cell engagers CAR-T (BiTEs CAR-T) targeting Mesothelin (MSLN) and secreting NKG2D-Bispecific T cell Engagers (BiTEs) to engage NKG2D ligands (NKG2DL). Analysis of TNBC tissues using The Cancer Genome Atlas and tumor microarrays revealed high but weakly correlated expression of MSLN and NKG2DL, making them ideal targets for dual engagement. To reduce immunogenicity and enhance stability, we used a nanobody and the natural receptor NKG2D as antigen-binding domains instead of traditional scFvs in the CAR construct. The secreted BiTEs could promote the cytotoxicity of untransduced T cells against NKG2DL + tumor cells. In vitro, BiTEs CAR-T cells exhibited superior cytotoxicity, T cell activation, and cytokines production against heterogeneous target cells compared to MSLN CAR-T. In vivo, BiTEs CAR-T cells demonstrated potent antitumor activity in zebrafish and murine TNBC models, significantly reducing tumor burden and prolonging survival without detectable toxicity. These findings suggest that BiTE CAR-T cells offer a highly promising therapeutic strategy for TNBC by addressing antigen heterogeneity and immune escape mechanisms, with promising translational potential for clinical application.

Keywords: BiTEs; CAR-T therapy; Mesothelin; NKG2D ligands; Triple-negative breast cancer.

Fig. 1

BiTEs CAR-T cells are efficacious against TNBC cells with heterogeneous antigens in vitro. (A) Expression (H-score) of Mesothelin (MSLN) and MICA/B in TNBC tissues and proportion of MSLN/MICA/B-positive patients in clinical cohort (n = 38). Subgroup analysis showed that the percentages of MICA/B^low/MSLN^high, MICA/B^highMSLN^low, MICA/B^high/MSLN^high, and MICA/B^low/MSLN^low patients were 28.95%, 13.16%, 42.11%, and 15.79% respectively. (B) Schematic diagrams of MSLN CAR and BiTEs CAR constructs. (C) MSLN and BiTEs CAR expression of T cells on day 7 using FACS. (D) Western blot analysis for concentrated BiTEs from culture supernatants of T cells transduced with BiTEs CAR or control. (E) Flow cytometry analysis demonstrating his-tag detection of concentrated BiTEs binding to NKG2DL on MDA-MB-231^MICA and CD3 on T cells after incubation with BiTEs supernatant (SN). (F) The cytotoxic activity of UTD, MSLN CAR-T, and BiTEs CAR-T cells against MDA-MB-231^MICA/MSLN (a mixture of 80% MDA-MB-231^MICA and 20% MDA-MB-231^MSLN) cells at E: T = 2.5, 5 and 10 using RTCA (n = 3). (G) MDA-MB-231^MICA/MSLN cells were cultured with UTD, MSLN CAR-T, or BiTEs CAR-T cells at an E: T ratio of 5:1 for 24 h. IFN-γ, TNF-α, IL-2, granzyme A, granzyme B, perforin and granulysin were measured using a LEGENDplex multi-analyte Flow Assay Kit (n = 3). (H) Intracellular flow cytometry analysis of IFN-γ secreting CD3⁺ T cells after co-culturing UTD, MSLN CAR-T, or BiTEs CAR-T cells and MDA-MB-231^MICA/MSLN cells at an E: T ratio of 5:1 for 24 h (n = 3). (I) Intracellular flow cytometry analysis of TNF-α secreting CD3⁺ T cells after co-culturing UTD, MSLN CAR-T, or BiTEs CAR-T cells and MDA-MB-231^MICA/MSLN cells at an E: T ratio of 5:1 for 24 h (n = 3). (J) Flow cytometry analysis of expression of CD25 on different CD3⁺ T cells after co-culture with MDA-MB-231^MICA/MSLN cells at an E: T ratio of 5:1 for 24 h (n = 3). (K) Flow cytometry analysis of expression of CD69 on different CD3⁺ T cells after co-culture with MDA-MB-231^MICA/MSLN cells at an E: T ratio of 5:1 for 24 h (n = 3). Each experiment was repeated at least twice with similar results. Representative data are shown. Statistical significance was considered as *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001, ns, not significant

Fig. 2

In vivo antitumor activity of BiTEs CAR-T cells against TNBC with heterogeneous antigens in zebrafish and mice models. (A) Schematic diagram of the zebrafish TNBC xenograft model. MDA-MB-231^MICA/MSLN (a mixture of 80% MDA-MB-231^MICA and 20% MDA-MB-231^MSLN) cells were used to establish the TNBC model with heterogeneous antigens. (B) Cell fluorescence was visualized using fluorescent stereomicroscopy 36 h post T cells injection. (C) Statistical analysis of the fluorescence intensity of MDA-MB-231^MICA/MSLN tumor treated with UTD, MSLN CAR-T or BiTEs CAR-T cells in zebrafish model at 36 h post-treatment (n = 10). (D) Schematic diagram of the murine TNBC xenograft model. MDA-MB-231^MICA/MSLN cells were used to establish the TNBC model with heterogeneous antigens. (E) Growth curve of MDA-MB-231^MICA/MSLN xenograft treated with UTD, MSLN CAR-T or BiTEs CAR-T cells. (F) Kaplan-Meier survival curves of MDA-MB-231^MICA/MSLN tumor bearing-mice treated with UTD, MSLN CAR-T or BiTEs CAR-T cells (n = 4). (G) Body weight of MDA-MB-231^MICA/MSLN tumor bearing-mice treated with UTD, MSLN CAR-T or BiTEs CAR-T cells (n = 4). (H) The concentration of IFN-γ in blood samples of MDA-MB-231^MICA/MSLN tumor bearing-mice treated with UTD, MSLN CAR-T or BiTEs CAR-T cells (n = 3). (I) Percentage of transferred CD45⁺ CD3⁺ T cells in peripheral blood samples from tumor bearing-mice treated with UTD, MSLN CAR-T or BiTEs CAR-T cells (n = 3). Each experiment was repeated at least twice with similar results. Representative data are shown. Statistical significance was considered as *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001, ns, not significant