Engineered Tumor-Targeted T Cells Mediate Enhanced Anti-Tumor Efficacy Both Directly and through Activation of the Endogenous Immune System

Abstract

Chimeric antigen receptor (CAR) T cell therapy has proven clinically beneficial against B cell acute lymphoblastic leukemia and non-Hodgkin's lymphoma. However, suboptimal clinical outcomes have been associated with decreased expansion and persistence of adoptively transferred CAR T cells, antigen-negative relapses, and impairment by an immunosuppressive tumor microenvironment. Improvements in CAR T cell design are required to enhance clinical efficacy, as well as broaden the applicability of this technology. Here, we demonstrate that interleukin-18 (IL-18)-secreting CAR T cells exhibit enhanced in vivo expansion and persistence and significantly increase long-term survival in syngeneic mouse models of both hematological and solid malignancies. In addition, we demonstrate that IL-18-secreting CAR T cells are capable of modulating the tumor microenvironment, as well as enhancing an effective endogenous anti-tumor immune response. IL-18-secreting CAR T cells represent a promising strategy to enhance the clinical outcomes of adoptive T cell therapy.

Keywords: B-ALL; CAR T cell; IL-18; MUC16; adoptive transfer; armored CAR; chimeric antigen receptor; interleukin-18; ovarian cancer.

Figure 1. Human 1928z-hIL18 CAR T Cells Enhance Survival of CD19⁺ Tumor-Bearing Scid-Beige Xenograft Mice

(A) Representation of the human CAR constructs. (B) In vitro cytokine secretion of IL-18, IFN-γ, and IL-2 after coculture with NALM6 tumor cells (E:T, 1:1). (C) In vitro proliferation assay with 1928z and 1928z-hIL18 CAR T cells co-cultured with NALM6 tumor cells (E:T 1:1). CAR T cells were stimulated on day 0 and day 7. (D) In vitro cytotoxicity utilizing standard 4-hr ⁵¹Cr release assay comparing 1928z and 1928z-hIL18 CAR T cells. (E) In vitro cytotoxicity over the course of 72 hr. 1928z versus 1928z-hIL18, *p < 0.05. (F and G) Statistically significant (p < 0.05) differential gene expression between 1928z and 1928z-hIL18 CAR T cells in vitro at day 1 (F) and day 8 after NALM6 tumor stimulation (G). (H) Survival curve comparing 1928z and 1928z-hIL18 CAR T cells in a NALM6-GFP⁺/Luc⁺-tumor bearing xenograft Scid-Beige mouse model. (I) Bioluminescence imaging comparing 1928z and 1928z-hIL18 CAR T cells in Scid-Beige mice inoculated with NALM6-GFP⁺/Luc⁺ tumor cells. Survival curves analyses were performed using Mantel-Cox test. Other analyses were determined with Student’s t test. All results were pooled from at least two independent experiments. Data in (B)–(E) represent means ± SEM. *p < 0.05, **p < 0.01, and ***p < 0.001 for indicated comparison. ns, non-significant.

Figure 2. Murine 19m28mz-mIL18 CAR T Cells Enhance Long-Term Survival of CD19⁺ Tumor-Bearing Syngeneic Mice

(A) Representation of murine CAR T cell constructs. (B) In vitro murine IL-18 secretion after stimulation with EL4hCD19⁺ tumor cells (E:T, 1:1). (C and D) Survival of EL4hCD19⁺ tumor-bearing mice treated on day 1 with (C) full dose (2.5 × 10⁶) or (D) half dose (1.2 × 10⁶) of CAR T cells/mouse. (E) Survival of mice treated with 19m28mz-mIL18 CAR T cells in a delayed tumor model. Mice were inoculated with 1 × 10⁶ EL4hCD19⁺ tumor cells on day 0 and treated with 2.5 × 10⁶ CAR T cells on day 7. Dashed line indicates day of CAR T cell infusion. (F) Survival of mice inoculated with 1 × 10⁶ EL4hCD19⁺ tumorcellson day0, treated with 2.5 × 10⁶ 19m28mz-mIL18 CAR T cells on day 1, and then rechallenged with 1 × 10⁶ EL4hCD19⁺ tumor cells on day 40. Control untreated mice were inoculated with 1 × 10⁶ EL4hCD19⁺ tumor cells on either day 0 or day 40. Data in (B) represent means ± SEM. Survival curves analyseswere made using Mantel-Coxtest. *p < 0.05, **p < 0.01, and ***p < 0.001 for indicated comparison. All results were pooled from at least two independent experiments.

Figure 3. 19m28mz-mIL18 CAR T Cells Display Enhanced Expansion and Persistence and Depend on Autocrine IL-18 Signaling

(A) CAR T cells from peripheral blood quantified by flow cytometry. (B) Peripheral blood B cell aplasia quantified via flow cytometry. (C) Bone marrow PCR analysis of mice treated with 19m28mz-mIL18 CAR T cells. (D) Serum cytokine quantification on day 7, comparing 19m28mz and 19m28mz-mIL18 CAR T cells. (E) Survival of EL4hCD19⁺ tumor-bearing mice treated with 19mz-mIL18 or 19m28mz-mIL18. (F) Quantification of day 7 peripheral blood CAR T cell in vivo. (G) Day 7, 30, and 60 peripheral blood B cell aplasia. (H) Survival of EL4hCD19⁺ tumor-bearing mice treated with IL18R-KO 19m28mz-mIL18 or WT 19m28mz-mIL18 CAR T cells. (I) Day 7 peripheral blood CAR T cell quantification. (J) Day 7 peripheral blood B cell aplasia. (K) Day 7 serum IFN-γ quantification. Survival curves comparisons were made using Mantel-Cox test. All results were pooled from at least two independent experiments. *p < 0.05, **p < 0.01, and ***p < 0.001 for indicated comparison. ns, non-significant.

Figure 4. 19m28mz-mIL18 CAR T Cells Modulate the Tumor Microenvironment and Activate Endogenous Immune Cells

(A) Detection of CAR T cells in the bone marrow. (B) Statistical scaffold map of the bone marrow. Size of unsupervised clusters denotes the relative number of cells in that grouping. Nodes in red denote landmark populations, defined manually. Blue nodes reflect unsupervised clusters of cells from raw dataset. Log₂ fold change of manually gated populations comparing bone marrow of mice treated with 19m28mz or 19m28mz-mIL18 CAR T cells with statistically significant differences noted (p < 0.05). (C) Detection of endogenous CD8 non-CAR T cells and CD8 19m28mz-mIL18 CART cells in the bone marrow of mice treated with 19m28mz-mIL18 CAR T cells. (D–F) Phenotypic analysis of bone marrow endogenous cells of mice treated with 19m28mz or 19m28mz-mIL18 CAR T cells: (D) endogenous CD8 T cells, (E) macrophages, and (F) dendritic cells. (G) Statistical Scaffold map of the bone marrow comparing the phenotype of endogenous non-CAR T cells and 19m28mz-mIL18 CART cells in mice treated with 19m28mz-mIL18 CAR T cells. Results are pooled data of two independent experiments. Adjustment by the Benjamini-Hochberg procedure for multiple testing. Data in (B) and (D)–(F) represent means ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001 for indicated comparison. ns, non-significant.

Figure 5. IL-18-Secreting CAR T Cells Engage Endogenous Anti-tumor Immune Effector Cells

(A) Survival curve of mice inoculated with both 1 × 10⁶ EL4hCD19⁺ and 1 × 10⁶ EL4hCD19⁻ tumor cells and treated on D1 with 19m28mz or 19m28mz-mIL18 CAR T cells. (B) IFN-γ ELISPOT results comparing CD3⁺ CAR-negative splenocytes from mice treated with 19m28mz, 19m28mz-mIL18 CAR T cells, or naive mice after coculture with EL4hCD19⁺ tumor cells. (C) IFN-γ ELISPOT comparing CD3⁺ CAR-negative splenocytes from mice treated with 19m28mz, 19m28mz-mIL18 CAR T cells, or naive mice after coculture with EL4hCD19⁺ tumor cells with/without MHC-I or MHC-II blocking antibodies. (D) IFN-γ cytokine quantification comparing CAR-negative splenocytes after 24 hr of exposure to EL4hCD19⁺ tumor cells (1:1 E:T ratio). (E) IFN-γ cytokine quantification comparing CAR-negative splenocytes after 24 hr of exposure to EL4hCD19⁻ tumor cells (1:1 E:T ratio). (F) Survival of mice treated with 19m28mz-mIL18 CAR T cells with or without macrophage depletion. (G) Survival of mice treated with 19m28mz-mIL18 CAR T cells with or without NK cell depletion. Survival curves analyses performed using Mantel-Cox test. Data in (B)–(E) represent means ± SEM. *p < 0.05, **p < 0.01, and ***p < 0.001 for indicated comparison. ns, non-significant. All results were pooled from at least two independent experiments.