Augmentation of Antitumor Immunity by Human and Mouse CAR T Cells Secreting IL-18

Abstract

The effects of transgenically encoded human and mouse IL-18 on T cell proliferation and its application in boosting chimeric antigen receptor (CAR) T cells are presented. Robust enhancement of proliferation of IL-18-secreting human T cells occurred in a xenograft model, and this was dependent on TCR and IL-18R signaling. IL-18 augmented IFN-γ secretion and proliferation of T cells activated by the endogenous TCR. TCR-deficient, human IL-18-expressing CD19 CAR T cells exhibited enhanced proliferation and antitumor activity in the xenograft model. Antigen-propelled activation of cytokine helper ensemble (APACHE) CAR T cells displayed inducible expression of IL-18 and enhanced antitumor immunity. In an intact mouse tumor model, CD19-IL-18 CAR T cells induced deeper B cell aplasia, significantly enhanced CAR T cell proliferation, and effectively augmented antitumor effects in mice with B16F10 melanoma. These findings point to a strategy to develop universal CAR T cells for patients with solid tumors.

Keywords: IL-18; T cell; adoptive transfer; cancer; chimeric antigen receptor; immunotherapy.

Figure 1. IL18-CAR T Cells Have Enhanced Proliferation In Vitro and In Vivo

(A) The construct designs. (B) Population doubling and cell volume of SS1 anti-mesothelin or anti-CD19 CAR T cells following the first restimulation with irradiated K562-Meso or K562-CD19 with exogenous IL-2. (C) NSG mice (n = 5) bearing an AsPC1 pancreatic flank tumor received CD19 or SS1 CAR T cells (2e6), and 3 weeks later, peripheral blood was analyzed by Trucount. (D) NSG mice (n = 5) bearing a systemic Nalm6 acute lymphoblastic leukemia (ALL) tumor received 1e6 CD19 or CD19-IL-18 CAR T cells. After 18 days, circulating T cells were assessed. (E) Tumor-free NSG mice (n = 5) were inoculated with 5e6 SS1 or SS1-IL-18 CAR T cells and, 3 weeks later, analyzed for circulating T cells. (F and G) CAR T cells were analyzed on day 9 of ex vivo expansion. In vivo expansion of SS1-IL-18 and CD19-IL-18 CAR T cells was determined by harvesting spleens from the mice described in (C) and (D), respectively. (F) Representative fluorescence-activated cell sorting (FACS) plots of CD8⁺CAR⁺ cells. (G) The percentage of CD4⁺CAR⁺ or CD8⁺CAR⁺ T cells in spleens from (F). All data with error bars are presented as mean ± SEM. Student’s t test: **p < 0.01, ***p < 0.001.

Figure 2. Combined TCR and IL-18 Signaling Promotes CD4⁺ T Cell Expansion in the Absence of CAR Signaling

(A–C) Primary T cells from a healthy donor were activated with anti-CD3 beads and varying concentrations (1.56 doubled up to 200 ng/mL) of rIL-18 added on days 0 and 3. (A) Changes in T cell volume and cell numbers. (B) IL-18-dependent expansion of T cells from two additional donors stimulated with rIL-18, anti-CD3 beads, and anti-CD3 beads plus rIL-18. (C) Luminex analysis of secreted IFN-γ and IL-2 with 12 culture conditions indicated. Additional cytokine data are shown in Figure S2A. (D) Normal donor T cells expanded with anti-CD3/CD28 beads for 10 days were selected by MACS columns to purify CD8⁺ and CD4⁺ T cells. Purified CD4⁺ T cells, CD8⁺T cells, or bulk T cells were restimulated with anti-CD3 beads and 100 ng/mL rIL-18, added on days 0 and 3, as indicated in the legend. (E–H) T cells transduced with lentivectors encoding GFP orGFP-IL-18 were selected using MACS columns to purify CD8⁺ and CD4⁺ T cells. Tumor-free NSG mice (n = 5) were injected with 2e6 T cells as indicated. The data shown are representative of two independent experiments. (E) Circulating CD45⁺ T cells and pictures of spleens from various groups. (F) Circulating CD45⁺CD4⁺ T cells on day 23 following T cell injection and the ratio of CD8⁺ to CD4⁺ T cells. (G) Serum levels of IL-18 and IFN-γ. (H) Serum from mice on day 23 was analyzed by Luminex to assess the cytokine profile. All data with error bars are presented as mean ± SEM. Student’s t test: **p < 0.01, ***p < 0.001.

Figure 3. IL-18-Enhanced T Cell Proliferation Requires TCR and IL-18R Signaling

(A) Tumor-free NSG mice (n = 5) were injected with 2e6 TCR KO or TCR WT GFP-IL-18 T cells or the indicated controls. After 18 days, T cells in the blood (top) or spleens (bottom) were assessed. (B) Cell-competitive repopulation design. CD19 CAR-IL-18 WT, GFP WT, AmCyan IL-18R KO and DsRed TCR KO T cells were mixed together and injected into tumor-free NSG mice (n = 5). (C) Purity of TCR KO and IL-18R KO cells after CRISPR/Cas9 gene editing and MACS sorting. (D) Initial frequency of each cell population shown in (B). (E–H) The mice were analyzed 3 weeks after T cell injection. (E) Left: total circulating CD45⁺ cells, indicating robust expansion. Right: individual populations of T cells. (F) Expression levels of IL-18Rα on T cells in the blood (left) and spleen (right). (G) Representative FACS plot of each cell population in the blood. (H) The ratios of CD19 CAR-IL-18 WT, IL-18R KO AmCyan, and TCR KO DsRed to GFP WT cells in the blood (left) and spleen (right). All data with error bars are presented as mean ± SEM. Student’s t test: **p < 0.01, ***p < 0.001.

Figure 4. IL-18 Costimulation Significantly Enhances Proliferation and Antitumor Activity of Human TCR-Deficient and Syngeneic Murine CD19 CAR T Cells

(A) Expansion kinetics and volume of TCR KO CAR T cells following the first (with IL-2) and second (without IL-2) restimulation with irradiated K562-CD19cells. (B and C) NSG mice were injected intravenously with 1e6 Nalm6 ALL cells expressing CBG-GFP. On day 7 after tumor injection, mice (n = 5) received the indicated TCR KO CAR T cells (1e6). The data shown are representative of two independent experiments. (B) Live animal imaging of tumor growth. (C) Tumor growth curves of four groups (left) and a magnified plot of TCR KO CD19-GFP and TCR KOCD19-IL-18 CAR T cohorts (right). Additional data include body weight, survival, number of CD45⁺CD3 KO T cells, IL-18 plasma concentration, percentage of CD3 KO cells among total CD45⁺ cells, distribution of CD4⁺ and CD8⁺ T cells within CD19 CAR T cells, and memory phenotype in CD4⁺ and CD8⁺ CAR T cells. (D) Study of murine CD19-IL-18 CAR T cells in the B16F10 melanoma tumor model that was derived from the B6 mouse in 1954. Lymphocyte-replete C57BL/6 (CD45.2 strain) mice were inoculated subcutaneously (s.c.) with 2e6 B16F10 tumor cells expressing murine CD19 (B16-mCD19). One week later, mice were randomized and treated with 5e6 murine CD19-GFP (n = 10) or CD19-IL18 (n = 10) CAR T cells (CD45.1 origin) or left untreated (n = 7). The mice were monitored for tumor growth, B cell frequency, survival, body weight, CD45.1 T cells, and murine IL-18 plasma concentration. All data with error bars are presented as mean ± SEM. In some cases, the magnitude of the SEM is less that the size of the symbol. Student’s t test: *p < 0.05, **p < 0.01, ***p < 0.001.