Dual effects of a targeted small-molecule inhibitor (cabozantinib) on immune-mediated killing of tumor cells and immune tumor microenvironment permissiveness when combined with a cancer vaccine

Abstract

Background: Growing awareness of the complexity of carcinogenesis has made multimodal therapies for cancer increasingly compelling and relevant. In recent years, immunotherapy has gained acceptance as an active therapeutic approach to cancer treatment, even though cancer is widely considered an immunosuppressive disease. Combining immunotherapy with targeted agents that have immunomodulatory capabilities could significantly improve its efficacy.

Methods: We evaluated the ability of cabozantinib, a receptor tyrosine kinase inhibitor, to modulate the immune system in vivo as well as alter the phenotype of tumor cells in vitro in order to determine if this inhibitor could act synergistically with a cancer vaccine.

Results: Our studies indicated that cabozantinib altered the phenotype of MC38-CEA murine tumor cells, rendering them more sensitive to immune-mediated killing. Cabozantinib also altered the frequency of immune sub-populations in the periphery as well as in the tumor microenvironment, which generated a more permissive immune environment. When cabozantinib was combined with a poxviral-based cancer vaccine targeting a self-antigen, the combination significantly reduced the function of regulatory T cells and increased cytokine production from effector T cells in response to the antigen. These alterations to the immune landscape, along with direct modification of tumor cells, led to markedly improved antitumor efficacy.

Conclusions: These studies support the clinical combination of cabozantinib with immunotherapy for the treatment of cancer.

Figures

Figure 1

Cabozantinib inhibits the growth, alters the phenotype and increases the sensitivity of MC38-CEA cells to T cell-mediated killing. (A) MC38-CEA cells were treated with 2.5 μg/mL cabozantinib or vehicle for 1, 3, and 5 days then assayed for growth and viability. Inset panel: MET and VEGFR2 expression of MC38-CEA cells. (B) MC38-CEA cells were exposed to 2.5 μg/mL cabozantinib or vehicle for 24 h, then analyzed by flow cytometry for surface expression of CEA, MHC-I (H-2Kb, H-2Db), ICAM-1, Fas, and calreticulin. Percent positivity and mean fluorescence intensity, in parentheses, are shown. Values in bold denote an increase of >40% relative to vehicle-treated cells. (C, D) MC38-CEA cells treated with cabozantinib (black bars) or vehicle (open bars) or left untreated (gray bars) for 24 h were labeled with 111In, then coincubated with CTLs specific for CEA (C) or gp70 (D) for 18 h at a ratio of 30:1. Bars indicate mean ± SEM. from quadruplicate measurements. Statistical analyses were done by Student’s t test. * = P <0.01. Data are representative of 3 independent experiments.

Figure 2

Cabozantinib alters the immune-cell repertoire of C57BL/6 mice. (A) Serum concentration of cabozantinib in mice fed control diet or diet containing cabozantinib for 10 days (n =5/group). Dashed line indicates the plasma concentration achieved at the maximum tolerated dose in a phase 1 human clinical trial. (B) Number of viable cells/spleen in mice fed control diet or cabozantinib diet for 35 days. The frequency of (C) CD4+ T cells (CD3+ CD4+), (D) CD8+ T cells (CD3+ CD8+), (E) Tregs (CD3+CD4+CD25+FoxP3+), and (F) MDSCs (CD11b+GR-1+) in spleens of mice fed control diet or cabozantinib diet for 35 days was determined by flow cytometry. Error bars indicate mean ± SEM. Statistical analyses were done by Student’s t test. * = P <0.05. Data are representative of 4 independent experiments.

Figure 3

Combining cabozantinib with a cancer vaccine improves antigen-specific immune responses in CEA-Tg C57BL/6 mice. (A) Schema of combination therapy study (n =9). (B) To evaluate Treg function, Tregs (CD3+CD4+CD25+FoxP3+) were isolated from spleens and cocultured with CD4+ T cells from naïve mice, APCs (allogeneic splenocytes irradiated with 30 Gy), and soluble anti-CD3 for 72 h. Control wells containing Tregs, APCs, and anti-CD3 without CD4+ T cells were used to determine the background level of Treg proliferation. Control wells containing CD4+ T cells, APCs, and anti-CD3 without Tregs were used to determine the background level of CD4+ T-cell proliferation. Error bars indicate mean ± SEM. Statistical analyses were done by Student’s t test. * = P <0.01 relative to CD4+ T-cell proliferation in the presence of Tregs from untreated mice; n.s. indicates no significant difference between the proliferation of CD4+ T cells in the absence of Tregs or in the presence of Tregs from mice treated with combination therapy. (C, D) To evaluate CD8+ T-cell responses, harvested splenocytes were incubated with CEA peptide (1 μg/mL) for 7 days. Lymphocytes were then restimulated with fresh, irradiated, naïve splenocytes and 1 μg/mL of either CEA or HIV-gag peptide for 24 h. Supernatants were collected and analyzed for murine IFN-γ (C) and TNF-α (D) using a cytometric bead array. Nonspecific cytokine production in response to HIV-gag was subtracted from that induced by the CEA peptide. Error bars indicate mean ± SEM. Statistical analyses were done by Student’s t test. * = P <0.05 relative to control and single-agent treatments. Data are representative of 2 independent experiments.

Figure 4

Cabozantinib reduces tumor vascularity and improves T cell infiltration when combined with a cancer vaccine. (A) Schema of combination therapy study (n =2/group). (B) Representative histological staining for indicated markers (top) on tumor sections from mice given indicated treatments (left). Inset panels: isotype staining. (C–F) Quantification of (C) blood vessels/tumor section; (D) number of CD3+ T-cells/tumor section; (E) number of CD4+ T-cells /tumor section; and (F) number of CD8+ T-cells /tumor section. Inset panels: number of indicated cells/mm2 of tumor. Marker staining was identified and enumerated using Aperio ImageScope image analysis software. Error bars indicate mean ± SEM. Statistical analyses were done by Student’s t test. * = P <0.05 relative to control.

Figure 5

Cabozantinib alters the tumor infiltration of negative regulatory cell subsets. A portion of the tumors depicted in Figure 4 were homogenized into a single-cell suspension and analyzed by flow cytometry for the presence of (A) Tregs (CD3+CD4+CD25+FoxP3+), (B) MDSCs (CD11b+Gr-1+), and (C) TAMs (CD11b+Gr-1−). Statistical analyses were done by Kolmogorov-Smirnov test. * = P <0.01 upregulation relative to control. ** = P <0.01 downregulation relative to control.

Figure 6

The combination of cabozantinib and immunotherapy significantly reduced tumor growth in a T cell-dependent manner. (A) Schema of combination therapy study (n =10/group). Tumor growth in mice treated with (B) control, (C) vaccine, (D) cabozantinib, and (E) combination therapy, including the number of tumor-free mice at day 35. (F) To determine the role of T cells in the efficacy of combination therapy, we measured tumor growth in control mice with no depletion (open circles), mice receiving combination therapy with no depletion (closed circles), and mice receiving combination therapy in the absence of CD4+ T cells (open squares) or CD8+ T cells (open triangles) (n =8). Depletion antibodies were administered on days 1–3, 10, 17, 24, and 31 (arrows). Tumor dimensions were measured weekly. Error bars indicate mean ± SEM. Statistical analyses were done by Student’s t test. * = P <0.0001 relative to control.