The tandem CD33-CLL1 CAR-T as an approach to treat acute myeloid leukemia

Abstract

Background: Acute myeloid leukemia (AML) is characterized by high heterogeneity, poor long-term survival, and a propensity for relapse. Exceptional efficacy in treating recurrent or refractory B-lymphoid malignancies has been demonstrated by Chimeric antigen receptor T cells (CAR-T cells). Given the therapeutic potential of targeting both CD33 and C-type lectin-like molecule-1 (CLL1) in AML, the development of a dual-targeting CD33-CLL1 CAR-T cells assumes significant importance.

Materials and methods: The expressions of CD33 and CLL-1 antigens in peripheral blood cells and bone marrow cells from AML patients was assessed. Subsequently, a Chimeric Antigen Receptor (CAR) incorporating a dual-specific single-chain variable fragment targeting CLL1 and CD33 (CD33-CLL1-CAR-T) was engineered. The anti-tumor efficacy and potential side effects of CD33-CLL1-CAR-T cells were comprehensively investigated in both in vitro and in vivo settings.

Results: The constructed tandem CD33-CLL1 CAR-T exhibited potent cytotoxicity against leukemia cell lines and human primary AML cells in vitro. Co-cultivation of AML blasts with CD33-CLL1-CAR-T cells resulted in effective proliferation and the secretion of substantial quantities of GM-CSF and IFN-γ. Importantly, the impact of CD33-CLL1-CAR-T cells on normal hematopoietic stem cells was minimal, ensuring safety in vivo mouse models. Notably, significant anti-leukemic activity was observed in the mouse model, with CD33-CLL1-CAR-T cells leading to tumor eradication and prolonged survival.

Discussion: The tandem CD33-CLL1 CAR-T cells not only efficiently eliminated AML blasts but also exhibited low cytotoxicity toward normal hematopoietic stem cells (HSCs). These findings underscore the potential clinical applicability of the tandem CD33-CLL1 CAR-T cells as an effective and safe treatment strategy for AML, representing a noteworthy advancement in the field of CAR-T cells therapy.

Figure 1. Constructing the CD33-CLL1 tandem CAR-T

(A) Expression of CD33 and CLL1 on the surface of human leukemia cell lines (HL-60, MV4–11, THP-1, U937, MOLM-13) was detected by flow cytometry. The gray line represents the isotype control. (B) Expression of CD33 and CLL1 in peripheral blood mononuclear cells (PBMCs, No.=22) and bone marrow (BM, No.=12) of Primary AML patients (down panel). A representative flow plot showed the expression of CD33 and CLL1 in tumor cells from two AML patients (up panel). (C) The structure of CD33-CAR-T (targeting CD33), CLL1-CAR-T (targeting CLL1), and CD33-CLL1- CAR-T (targeting both CD33 and CLL1). (D) The expression of CD45RO and CD62L on T cells before (un T)and after transduction (CAR-T) were detected using flow cytometry. CAR-T: chimeric antigen receptor T-cells; unT: untransfected T-cells.

Figure 2. Human leukemia cells were killed by CD33-CLL1 tandem CAR-T

(A) Killing percentages of HL60, U937, THP-1, and MOLM-13 cells by unT, CD33-CLL1-CAR-T, CLL1-CAR-T, and CD33-CAR-T cells after 18 h co-culture in vitro. E: T ratios mean the ratios of the absolute number of CAR-T cells to target cells. (B) Cytotoxic effects of CD33-CAR-T, CLL1-CAR-T, and CD33-CLL1-CAR-T on 3 primary AML blasts (based on the expression of CD33 and CLL1) after 18 h co-culture in vitro (E: T=1:4). The data were represented as means ± SD and analyzed by One-way ANOVA and two-tailed unpaired Student’s t-test. *p<0.05; **p<0.01; ***p<0.001, ns, no significance.

Figure 3. Long-term cytotoxicity of CD33-CLL1-CAR-T to human leukemia cells

(A) The cytotoxicity of the indicated CAR-T on U937 cells (CSFE labeled) after 8 rounds of co-culture at a 1:4 E/T ratio was detected by flow cytometry. (B) GM-CSF and IFN-γ concentration in the supernatant after 48 h co-culture of the indicated CAR-T cells with target cells (E:T=1:4). The data were represented as means ± SD.

Figure 4. Effects of CD33-CLL1-CAR-T cells on human hematopoietic stem cells in vitro and mice in vivo

(A) Expression of CLL1 and CD33 on CD34⁺ HSCs derived from cord blood was detected by flow cytometry. The gray line represents the isotype control. (B) Cytotoxicity of CLL1-CAR-T, CD33-CAR-T, and CD33-CLL1-CAR-T on CD34+ cord blood cells. CD34+ Cord blood cells were incubated with CAR-T-cells at E: T=10:1 for 16 h, the mixture was then plated (500 HSCs/plate) in semi-solid media, and colonies were counted for each condition after 6 days. (C and D) Experimental scheme: NSG mice were injected intravenously with 5×10⁶ CAR-T cells (unT, CD33-CLL1-CAR-T, CD33-CAR-T, CLL1-CAR-T) or phosphate-buffered saline (PBS) and the weight was monitored to assess the effect on the NOD/SCID IL-2RγC^null (NSG) mice. The data in B and D were represented as means ± SD and analyzed by One-way ANOVA and two-tailed unpaired Student’s t-test. *p<0.05; **p<0.01; ns: no significance.

Figure 5. The anti-tumor activities of CAR-T cells in a U937-Luc xenograft mouse model in vivo

(A) Experimental scheme: NSG mice were injected intravenously with 1×10⁶ U937 cells stably expressing luciferase on day -7. After confirmation of engraftment by Bioluminescence imaging (BLI) on day -1, 3×10⁶ CAR-T cells (unT, CD33-CLL1-CAR-T, CD33-CAR-T, CLL1-CAR-T) or PBS were transferred to all the mice via tail vein. Tumor burden was monitored based on the bioluminescence intensity from the IVIS imaging system at the indicated time points. (B) BLI of the tumor burden. (C) Tumor burden was quantified as the average value of the total flux (p/s). No.= 5 mice per group. (D) Kaplan-Meier survival curve of mice bearing U937 cell-derived tumors after indicated treatment. The data in panel C were represented as means ± SD and analyzed by one-way ANOVA with Tukey’s multiple comparisons tests. The data in panel D were analyzed by log-rank Mantel-Cox test. *p<0.05; **p<0.01; ns: no significance.