Therapeutic effects of anti-GM2 CAR-T cells expressing IL-7 and CCL19 for GM2-positive solid cancer in xenograft model

Abstract

Background: While chimeric antigen receptor (CAR)-T cell therapy has demonstrated excellent efficacy in hematopoietic malignancies, its clinical application in solid cancers has yet to be achieved. One of the reasons for such hurdle is a lack of suitable CAR targets in solid cancers.

Methods: GM2 is one of the gangliosides, a group of glycosphingolipids with sialic acid in the glycan, and overexpressed in various types of solid cancers. In this study, by using interleukin (IL)-7 and chemokine (C-C motif) ligand 19 (CCL19)-producing human CAR-T system which we previously developed, a possibility of GM2 as a solid tumor target for CAR-T cell therapy was explored in a mouse model with human small-cell lung cancer.

Results: Treatment with anti-GM2 IL-7/CCL19-producing CAR-T cells induced complete tumor regression along with an abundant T cell infiltration into the solid tumor tissue and long-term memory responses, without any detectable adverse events. In addition, as measures to control cytokine-release syndrome and neurotoxicity which could occur in association with clinical use of CAR-T cells, we incorporated Herpes simplex virus-thymidine kinase (HSV-TK), a suicide system to trigger apoptosis by administration of ganciclovir (GCV). HSV-TK-expressing anti-GM2 IL-7/CCL19-producing human CAR-T cells were efficiently eliminated by GCV administration in vivo.

Conclusions: Our study revealed the promising therapeutic efficacy of anti-GM2 IL-7/CCL19-producing human CAR-T cells with an enhanced safety for clinical application in the treatment of patients with GM2-positive solid cancers.

Keywords: CAR-T cell; chemokine; cytokine; ganglioside; solid cancers.

FIGURE 1

Generation and characterization of anti‐GM2 CAR‐T cells expressing IL‐7, CCL19, and herpes simplex virus‐thymidine kinase. (A) Schematic representation of Conv. CAR and 7 × 19 CAR against GM2. LS; leader sequence, TM; transmembrane region. (B) Human PBMC transduced with Conv. CAR or 7 × 19 CAR were stained with anti‐idiotype Ab against anti‐GM2 scFv to detect CAR expression, along with anti‐CD8 Ab. UTD T cells were examined as a negative control. The percentage of cells in each quadrant is indicated. (C) The culture supernatants from Conv. CAR‐T and 7 × 19 CAR‐T cells were harvested 4 days after gene transduction, and the concentrations of IL‐7 and CCL19 were measured by ELISA. As a control, the culture supernatant obtained from UTD T cells at the same time point was examined. Data are shown as mean ± standard error of triplicate samples. N.D., not detected.

FIGURE 2

GM2‐specific immune responses of Conv. CAR and 7 × 19 CAR‐T cells in vitro. (A) Surface expression of endogenous GM2 was assessed in human solid cancer cell lines by flow cytometry. Open and filled histograms indicate staining with humanized anti‐GM2 Ab and nonstaining, followed by anti‐human IgG Ab, respectively. (B) Conv. CAR‐T, 7 × 19 CAR‐T, or UTD T cells were co‐cultured with the indicated tumor cells at an effector‐to‐target (E:T) ratio of 1:1 and 1:3 for 2 days. The number of residual tumor cells after co‐culture was analyzed by flow cytometry. Data are shown as mean ± standard error (SE) of triplicate samples. (C) The supernatants of co‐cultured cells as described in (B) were harvested, and the concentration of IFN‐γ was assessed by ELISA. Data are shown as mean ± SE of triplicate samples. ***p < 0.001, N.D., not detected; n.s., not significant.

FIGURE 3

Potent therapeutic effects of 7 × 19 CAR‐T cells in the xenograft model of human small‐cell lung cancer. NOG‐ΔMHC mice were inoculated subcutaneously (s.c.) with 5 × 10⁶ Lu‐135 tumor cells on day 0 and then treated with intravenously injection of 1 × 10⁷ Conv. CAR‐T, 7 × 19 CAR‐T cells, or UTD T cells on day 3, or left untreated. Thereafter, the tumor size (A) and the mouse survival (B) were assessed (n = 5 in the untreated group, n = 6 each in UTD T cell, Conv. CAR‐T cell, and 7 × 19 CAR‐T cell‐treated groups). In (A), each line indicates the tumor volume of an individual mouse. Tumor volumes during the initial 28 days of 7 × 19 CAR‐T cell‐treated mice are shown in the inset. ***p < 0.001.

FIGURE 4

Massive infiltration of T cells in the tumor tissues by the treatment with 7 × 19 CAR‐T cells. NOG‐ΔMHC mice were inoculated subcutaneously with 1 × 10⁷ Lu‐135 cells on day 0, followed by treatment with intravenously injection of 1 × 10⁷ Conv. CAR‐T or 7 × 19 CAR‐T cells on day 3. Tumor tissues were resected from the mice on day 12, and FFPE slices were prepared from each tumor tissue and stained with H&E and IHC. In IHC staining, rabbit anti‐human CD4 and rabbit anti‐human CD8 mAb were used. Stained cells were visualized and observed by microscopic examinations at ×100 and ×400 magnifications. (A) Representative images are displayed. Scale bar indicates a length of 200 μm (×100) or 50 μm (×400). (B) The number of stained cell per tumor area (mm²) was calculated by BZ‐X analyzer software. Data are shown as mean ± standard error (n = 6 each per group). *p < 0.05.

FIGURE 5

Generation of target‐specific long‐term memory responses by the treatment with 7 × 19 CAR‐T cells. NOG‐ΔMHC mice were inoculated subcutaneously (s.c.) with 1 × 10⁷ Lu‐135 cells on day 0, followed by treatment with intravenously injection of 1 × 10⁷ 7 × 19 CAR‐T cells on day 3. (A) The mice with complete tumor regression were rechallenged s.c. with 1 × 10⁷ Lu‐135 tumor cells at the right flank and 5 × 10⁶ SW480 tumor cells at the left flank on day 28. As a control, naïve NOG‐ΔMHC mice were inoculated s.c. with Lu‐135 and SW480 tumor cells in the same fashion (n = 6 in the tumor‐rejected mice group, n = 4 in the naïve NOG‐ΔMHC mice group). (B) The mice with a complete tumor regression were rechallenged s.c. with 3 × 10⁶ KMS11 tumor cells at the right flank and 5 × 10⁶ SW480 tumor cells at the left flank on day 35. As a control, naïve NOG‐ΔMHC mice were also inoculated s.c. with KMS11 and SW480 tumor cells in the same fashion (n = 4 in the tumor‐rejected mice group, n = 5 in the naïve NOG‐ΔMHC mice group). Tumor size was measured twice a week by a digital caliper. Tumor volumes are shown as mean ± standard error. ***p < 0.001, n.s., not significant.

FIGURE 6

Depletion of 7 × 19 CAR‐T cells equipped with Herpes simplex virus‐thymidine kinase (HSV‐TK) by GCV treatment. (A) Anti‐GM2 7 × 19 CAR‐T cells equipped with or without HSV‐TK suicide gene were cultured for 3 days in the presence of GCV at a concentration of 0, 0.1, 1, and 10 μM. The number of residual CAR‐T cells was analyzed by flow cytometry. Data are shown as mean ± standard error (SE) of triplicate samples. (B, C) NOG‐ΔMHC mice were inoculated subcutaneously with 1 × 10⁷ Lu‐135 tumor cells on day 0, followed by intravenously injection of 1 × 10⁷ 7 × 19 CAR‐T cells expressing HSV‐TK on day 3, and then treated with i.p. administration of GCV (100 mg/kg) on days 7 and 21. (B) The proportion of CAR‐T cells in PBMC was assessed by flow cytometry. Data are shown as mean ± SE (n = 8 in the GCV‐treated mice group, n = 7 in nontreated mice group). **p < 0.01, ***p < 0.001. (C) Tumor size of Lu‐135 in the mice treated with 7 × 19 CAR‐T cells with or without GCV administration was measured twice a week by a digital caliper. Each line indicates the tumor volume of an individual mouse.