Loop33 × 123 CAR-T targeting CD33 and CD123 against immune escape in acute myeloid leukemia

Abstract

Background: Immunotherapy, such as chimeric antigen receptor T (CAR-T) cells targeting CD33 or CD123, has been well developed over the past decade for the treatment of acute myeloid leukemia (AML). However, the inability to sustain tumor-free survival and the possibility of relapse due to antigen loss have raised concerns. A dual targeting of CD33 and CD123 is needed for better outcomes.

Methods: Based on our previously constructed CD33 and CD123 monovalent CAR-T, Loop33 × 123 and Loop123 × 33 CAR-T were constructed with molecular cloning techniques. All CAR-T cells were generated by lentivirus transduction of T cells from healthy donors. Phenotype detection was evaluated on day 7 concerning activation, exhaustion, and subtype proportions. Coculture killing assays were conducted using various AML cell lines and primary AML cells. Degranulation and cytokine secretion levels were detected by flow cytometry. Cell-derived xenograft models were established using wild-type Molm 13 cell lines, or a mixture of Molm 13-KO33 and Molm 13-KO123 cells as an ideal model of immune escape. By monitoring body weight and survival of tumor-bearing mice, Loop33 × 123 and Loop123 × 33 CAR-T cells were further assessed for their efficacy in vivo.

Results: In vitro study, our results demonstrated that Loop33 × 123 CAR-T cells could efficiently eliminate AML cell lines and primary AML cells with elevated degranulation and cytokine secretion levels. Compared with our previously constructed monovalent CD33 or CD123 CAR-T cells, Loop33 × 123 CAR-T cells showed superior advantages in an immune escape model. In vivo studies further confirmed that Loop33 × 123 CAR-T cells could effectively prolong the survival of mice without significant toxicity. However, Loop123 × 33 CAR-T cells failed to show the same effects. Furthermore, Loop33 × 123 CAR-T cells efficiently circumvented potential immune escape, a challenge where monovalent CAR-T cells failed.

Conclusions: Loop33 × 123 CAR-T targeting CD33 and CD123 could efficiently eliminate AML cells and prolong survival of tumor-bearing mice, while addressing the issue of immune escape.

Keywords: AML; CD123; CD33; Chimeric antigen receptor.

Fig. 1

Construction of Loop33 × 123 and Loop123 × 33 CAR-T cells targeting both CD33 and CD123 a Schematic display of CD33 CAR-T, CD123 CAR-T, Loop33 × 123 CAR-T, and Loop123 × 33 CAR-T. b Representative flow cytometry diagram showed that Loop CAR-T could recognize CD33 and CD123 proteins. c Flow cytometry detection of CAR expression on day 7. d, e Flow cytometry detection of activation markers CD69 and CD25 on day 7. F–h Flow cytometry detection of exhaustion markers PD-1, Tim-3, and LAG-3 on day 7. i Flow cytometry detection pf CD4⁺/CD8⁺ ratio on day 7. j, k Flow cytometry detection of stem cell memory cells (T_SCM), central memory cells (T_CM), effector memory cells (T_EM), and terminal differentiation effector cells (T_EMRA) in CD4⁺ and CD8⁺ cells, respectively. *, p < 0.05; **, p < 0.01; ns, no significance

Fig. 2

Loop CAR-T efficiently eliminated AML cell lines than monovalent CAR-T a FCM shows CD33 and CD123 expression of Molm 13-WT and Molm 13-KO33&123. b–g Residual target cells and CAR-T cells at 24 and 48 h after Vector-T or CAR-T cells were cocultured with Molm 13-WT cells or Molm 13-KO33&123 at various E:T ratios. h, i Degranulation of Vector-T or CAR-T cells defined as CD107a⁺ cells after coculturing with Molm 13-WT cells or Molm 13-KO33&123 for 6 h. j IL-2, IL-6, TNF-α, and IFN-γ secretion of Vector-T or CAR-T cells after coculturing with Molm 13-WT cells for 24 h. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; ns, no significance

Fig. 3

Loop CAR-T cells could effectively eliminate patients’ primary AML cells in vitro a FCM shows CD33 and CD123 expression of six AML patients primary cells. b Summary plot of CD33 and CD123 expression of primary cells. c Percentage of residual target cells at 48 h after coculturing at E:T ratio of 1:1. d Degranulation of Vector-T or CAR-T cells after coculturing with primary cells. e IL-2, IL-6, TNF-α, and IFN-γ secretion of Vector-T or CAR-T cells after coculturing with primary cells for 24 h. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; ns, no significance

Fig. 4

Loop33 × 123 CAR-T cells could significantly prolong survival in tumor bearing NSG mice a The experimental regimen of the B-NDG-CDX model. b Bioluminescence imaging showing tumor burden over time (n = 6). c The body weight of each group was measured after treatment with Vector-T or CAR-T cells. d Statistical analysis of the bioluminescence intensity. e Kaplan–Meier survival curves for each group. (f) Cytokine levels of mice blood serum of each group on day 15. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; ns, no significance

Fig. 5

Loop33 × 123 CAR-T cells were efficient to avoid immune escape in vitro a FCM shows CD33 and CD123 expression on Molm 13-WT, Molm 13-KO33, and Molm 13-KO123. b–d Percentage of residual target cells and CAR-T cells at indicated time points after Vector-T or CAR-T cells were cocultured with Molm 13-KO33 cells at various E:T ratios. e–g Percentage of residual target cells and CAR-T cells at indicated time points after Vector-T or CAR-T cells were cocultured with Molm 13-KO123 cells at various E:T ratios. h–j Percentage of total residual target cells, percentage of residual Molm 13-KO33, and percentage of residual Molm 13-KO123 at 48 h after Vector-T or CAR-T cells were cocultured with a 1:1 mixture of Molm 13-KO33 and Molm 13-KO123 cells at various E:T ratios. k Degranulation of Vector-T or CAR-T cells defined as CD107a⁺ cells after coculturing with Molm 13-KO33, Molm 13-KO123, or a 1:1 mixture of Molm 13-KO33 and Molm 13-KO123 cells for 6 h. l IL-2, IL-6, TNF-α, and IFN-γ secretion of Vector-T or CAR-T cells after coculturing with Molm 13-KO33, Molm 13-KO123, or a 1:1 mixture of Molm 13-KO33 and Molm 13-KO123 cells for 24 h. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; ns, no significance

Fig. 6

Loop33 × 123 CAR-T cells were efficient to avoid immune escape in vivo a The experimental regimen of the B-NDG-CDX model. b Bioluminescence imaging showing tumor burden at indicated time (n = 5). c The body weight of each group was measured after treatment with Vector-T or CAR-T cells. d Kaplan–Meier survival curves for each group. e Statistical analysis of the bioluminescence intensity. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; ns, no significance