A novel chimeric antigen receptor T-cell therapy targeting CD84 for the treatment of acute myeloid and T-cell lymphoblastic leukemias

Abstract

Despite the remarkable clinical successes of chimeric antigen receptor (CAR) T-cell therapies in treating B-cell malignancies and multiple myeloma, similar outcomes have not been achieved in other indications. For patients with relapsed or refractory (R/R) acute myeloid leukemia (AML) or T-cell acute lymphoblastic leukemia (T-ALL), treatment options are limited, yet CART-cell therapies offer significant potential to address this unmet need. Here, we introduce a first-in-class CART-cell therapy targeting CD84, a novel antigen, for the treatment of R/R AML and T-ALL. CD84 is highly expressed on leukemic blasts, with limited expression on hematopoietic stem progenitor cells (HSPC), and is largely absent in healthy human tissues. Our second-generation CARTs targeting CD84 (CART84) demonstrate potent cytotoxicity against AML and T-ALL cells both in vitro and in vivo in patient-derived xenograft (PDX) models. Furthermore, CART84 eliminated primary leukemic blasts while exhibiting low cytotoxicity against CD34+ HSPC in vitro and in humanized mouse models in vivo, suggesting a low risk of myelotoxicity. These results support CD84 as a promising target for AML and T-ALL and provide the foundation for our upcoming first-in-human phase I/II clinical trial using CD84-directed CAR T cell therapy for patients with R/R AML and T-ALL (EudraCT 2024-519966-31-00).

Fig. 1. CD84 expression on leukemic blasts and healthy bone marrow cells.

A Percentage of CD84, CD123, and CD33-positive cells in primary AML patient samples, with the median indicated (n = 51 AML patient samples). B The normalized mean fluorescence intensity (nMFI) of CD84, CD123, and CD33 in primary AML patient samples, calculated as the ratio of the MFI of cells stained with specific antibodies to the MFI of cells stained with isotype controls (median is shown; n = 51 AML patient samples). See Supplementary Table 1 for AML diagnoses. C The nMFI of CD84 expression was analyzed as the ratio of the MFI of cells stained with the specific anti-CD84 antibody to the MFI of cells stained with the isotype control. Eight samples were analyzed (n = 8: 5 healthy donor bone marrow samples and 3 samples from individuals with monoclonal gammopathy of undetermined significance) (median is represented).

Fig. 2. CD84-targeting CART cells are successfully generated and expanded.

A Scheme of second-generation CAR vector with EF1-α promoter, anti-human CD84 single-chain variable fragment (scFv), CD8α as hinge (H), and transmembrane (TM) domain, 4-1BB as intracellular costimulatory domain, and CD3ζ as signaling domain. B Population doubling of CART84 cells in comparison to UT cells at day 6 of expansion (Min to Max) n = 5–13. C Percentage of CAR-positive cells in CART84 cells at day 6 of expansion (Mean ± SD) n = 9–17. Cytotoxicity of CART84 towards different cell lines: CD84^high MOLM-13-GFP+ (AML) (D); CD84^low U937-GFP+ (AML) (E); CD84^high MOLT-4-GFP+ (T-ALL) (F); CD84^high Ramos-GFP+ (Burkitt lymphoma) (G). The percentage of surviving target cells relative to untreated cells (target cells alone) is shown at different effector-to-target (E:T) ratios at 24 h. Mean of at least 5 independent experiments ± SEM. UT untransduced T cells. Statistical significance was determined with a two-way ANOVA test (multiple comparisons to UT): ***p < 0.001, **p < 0.01, *p < 0.05.

Fig. 3. CART84 selectively kills CD84-positive cells and targets primary AML blasts in vitro.

A Cytotoxicity of CART84 cells towards CD84-negative (left) and CD84-positive (right) HEK-293 cells, at an E:T ratio of 4:1, assessed with the xCELLigence. B Cytotoxicity of CART84 cells against primary AML cells from patient sample #13 (see Suppl. Table 1). Data represent the mean of two CART84 productions derived from T cells of two healthy donors (BC#215 and BC#216), tested against AML cells from patient #13. One representative example out of three is shown. Statistical significance was determined with a two-way ANOVA test (multiple comparisons to UT). C Cytotoxicity of CART84 cells derived from AML patient #52 against autologous AML blasts. Details of AML diagnoses are provided in Supplementary Table 1. Statistical significance was determined with a two-way ANOVA test (multiple comparisons to UT). UT untransduced T cells. ***p < 0.001.

Fig. 4. In vivo efficacy of CART84 in an AML PDX model.

A Experimental design. B Percentage of AML PDX cells in the peripheral blood of mice (n = 6 mice per group). Statistical analysis was performed with a two-way ANOVA with Dunnett’s multiple comparisons vs. control mice (Mean ± SEM). C Kaplan–Meier survival curves for each experimental group. Statistical significance was determined with a log-rank test, with a corrected p value for 3 comparisons (control vs.152.3 / 153.5; 152.3 vs. 153.5). D Percentage of AML PDX cells in the bone marrow or spleen of mice at the end of the experiment (Mean ± SD). Statistical analysis was performed with a one-way ANOVA with Dunnett’s multiple comparisons vs. the control group. ***p < 0.001, **p < 0.01, *p < 0.05.

Fig. 5. In vivo efficacy of CART84 in a T-ALL PDX model.

A Experimental design. B Percentage of T-ALL PDX cells in the peripheral blood of mice (n = 5–6 mice per group). Statistical analysis was performed with a two-way ANOVA model with Dunnett’s multiple comparisons vs. control mice (Mean ± SEM). C Kaplan–Meier survival curves for each experimental group. Statistical significance was determined with a log-rank test, with a corrected p value for three comparisons (control vs.152.3 / 153.5; 152.3 vs. 153.5). D Percentage of T-ALL PDX cells in the bone marrow or spleen of mice at the end of the experiment (Mean ± SD). Statistical analysis was performed with a one-way ANOVA with Dunnett’s multiple comparisons vs. the control group. ***p < 0.001, **p < 0.01, *p < 0.05.

Fig. 6. Myelotoxicity of CART84 cells in vitro.

A Cytotoxicity assay of CART84 cells against G-CSF mobilized peripheral blood CD34⁺ hematopoietic stem/progenitor cells (HSPC) after 24 h (mean of technical duplicates; one representative example out of three is shown). Statistical analysis was performed with a two-way ANOVA with Sidak’s multiple comparisons. B Cytotoxicity of CART84 against CD34+ HSPC from cord blood donors at 24 h (mean of 5 assays using CD34+ cells from five different cord blood units ± SEM). Statistical significance was determined using a two-way ANOVA model with Sidak’s multiple comparisons. C Cytotoxicity of CART84 against donor bone marrow CD34+ HSPC (Mean of 4 assays using CD34+ cells from four different bone marrow samples ± SEM). Statistical significance was determined using an unpaired t-test. *p < 0.05.

Fig. 7. Myelotoxicity of CART84 cells in vivo.

A Experimental design of the humanized NSG mouse model (n = 5–6 mice per group). B–F Percentage of different human subsets in the bone marrow (gated from the human CD45+ population) at day 67 of the experiment. B Percentage of CD45+ human cells gated from the entire bone marrow. C Percentage of human T lymphocytes (hCD3+). D Percentage of human B lymphocytes (hCD19+). E Percentage of human HSPC (hCD34+). F Percentage of human myeloid cells (hCD33+) G Human cell populations per mouse group. Statistical analysis of bone marrow cell populations was performed using a one-way ANOVA with multiple comparisons (all vs. all). UT: untransduced T cells. ****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05.