Cabozantinib inhibits tumor growth in mice with ovarian cancer

Abstract

Ovarian cancer is usually detected in the advanced stages. Existing treatments for high grade serous ovarian cancer (HGSOC) are not adequate and approximately fifty percent of patients succumb to this disease and die within five years after diagnosis. We conducted pre-clinical studies in a mouse model of ovarian cancer to evaluate disease outcome in response to treatment with the multi-kinase inhibitor cabozantinib. Cabozantinib is a receptor tyrosine kinase inhibitor with multiple targets including vascular endothelial growth factor receptor-2 (VEGFR-2), associated with immune suppression in ovarian cancer. Mice (C57BL/6) were injected with ID8-RFP ovarian tumor cells and treated with cabozantinib. Studies investigated ascites development, tumor burden and regulation of anti-tumor immunity with treatment. Mice treated with cabozantinib had significantly decreased solid tumor burden and decreased malignant ascites as compared to untreated controls. Improved outcome in cabozantinib treated mice was associated with a significantly higher percentage of CD69 early activated T cells, a higher percentage of granzyme B secreting CD8 T cells, the enhanced release of cytokines and chemokines known to recruit CD8 T cells and amplify T cell function, as well as reduced VEGFR-2. Findings suggest that cabozantinib is an important clinical agent capable of improving ovarian cancer in mice potentially in part by priming the autologous immune system to promote anti-tumor immunity.

Keywords: Ovarian cancer; anti-tumor immunity; cabozantinib treatment; disease improvement; mouse model.

Figure 1

Cabozantinib treatment of mice limits ascites development. Female C57BL/6 (B6) mice at 8-10 weeks old were injected with (ID8-RFP) syngeneic murine ovarian cancer cells (1×10⁶ cells/mouse. I.P.), in groups of 5 mice and assigned to treatment and controls. They were treated with cabozantinib (50 mg/Kg body weight/dose). Cabozantinib was diluted in vehicle (1% DMSO, 30% PEG, 1% tween 80) for oral administration of 4 doses weekly, for a total of 24 doses. Mice were euthanized when those in the vehicle control group showed enlarged abdomens due to ascites formation (A). Mice were monitored frequently and weights recorded. Euthanasia was performed at day 66 post tumor cell administration for all mice studied in Figures 1, 2, 3, 4, 5 and 6. (B) The image shows the weight increases in vehicle treated mice in comparison with cabozantinib treated mice over the course of the experiment. (C) The image shows weight changes in the cabozantinib treated group compared with the vehicle treated group at the time of euthanasia (T-test, P<0.01**). Ascites was recovered from mice immediately after euthanasia. (D) The image shows that vehicle treated mice have significantly more ascites recovery than cabozantinib treated mice (n=5 mice/group, P<0.001***). This finding of less weight increase and lower ascites accumulation with cabozantinib compared with vehicle treatment is representative of 2 separate experiments.

Figure 2

Reduced tumor cell fluorescence in mice treated with cabozantinib. Mice were treated with vehicle or cabozantinib post ID8-RFP tumor cell injection. On euthanasia (day 66) images were captured on the IVIS Spectrum to measure RFP in the exposed abdomens of mice (A and B). The average total emission over ROIs of cabozantinib treated mice and vehicle treated mice is shown in (C) (n=5 mice/group, P≤0.05*).

Figure 3

Decreased tumor foci in mice treated with cabozantinib. Mice were injected with ID8-RFP cells, treated with cabozantinib and euthanized at day 66 post tumor cell injection. (A and B) Images show the pattern for two FFPE tissue array blocks for mouse tissues. The images shown are photographs of the actual section on the slide. They are enlarged for visualization. (Magnification 1×). Tissue was embedded in this manner for all mice in experiments and sections cut for H&E staining. Quantification of tumor was performed using the Olympus U-TV0.35XC-2 camera and cellSens Standard software. Each focus of tumor was assessed using the closed polygonal tool to obtain a measurement of area (mm²). Arrows indicate that tumors in the small intestine were reduced in cabozantinib treated mice in comparison with vehicle treated mice (C and D respectively, Magnification 40×). Measurement of all tumors in the small intestines of mice showed that there were significant decreases in average size (mm²) of tumors in the small intestines (P=0.005**) of cabozantinib treated mice when compared with the larger tumor foci in vehicle treated mice (E). The total tumor foci area observed with cabozantinib and vehicle treatment of mice was 81.1 and 127 mm² respectively. (F) The image shows the total average tumor foci in mice treated with cabozantinib versus control vehicle across all organs studied (n=5 mice/group). Each boxplot represents a 5-point summary of the data. The 3 lines in the box represent the first, second/median and third quartiles from bottom to top. Two vertical lines connect to the minimum and maximum values. Any singular point beyond these are identified as outliers. (F) The image shows that the median (middle line) value for tumor foci in vehicle treated mice is much higher compared to that for cabozantinib treatment of mice. The average total tumor foci area per mouse with cabozantinib treatment was 16.2 mm² and for vehicle treatment 25.4 mm². This parameter is significantly higher for vehicle treated mice in comparison to cabozantinib treated mice (P-value=0.04*).

Figure 4

Cabozantinib enhances immune activity in mice. Spleen cells of ID8-RFP inoculated mice, treated with cabozantinib and euthanized on day 66 were used to prepare a single cell suspension. Cells from 5 mice in each group were monoclonal antibody labeled to study CD3, CD4, CD8 and other antigens. Events were acquired on a flow cytometer and data analyzed by FloJo software. For data analysis a gate was drawn in the live cell region. Within the live cell gate, a region of smaller cells positive for CD3 was selected as the lymphocyte gate. CD3, CD4 and CD8 T cells were analyzed in the lymphocyte gate. (A-C) Results of dot plots summaries show that a higher percentage of CD3, CD4 and CD8 T cells were present in the spleens of mice treated with cabozantinib than in vehicle treated mice. Spleen cells were stimulated with anti-CD3 antibody to investigate the percentage of CD69 activated T cell subsets in resulting cultures. The upper quadrants of dot plots in (D) show that there was significantly increased percentages of CD69 positive CD4 T cells from spleen cells of cabozantinib treated mice (P<0.05*). Dot plots in (E) show that there was also a significantly higher percentage of CD69 positive CD8 T cells from cabozantinib treated mice compared with vehicle control mice (P<0.05*). After surface staining for T cell antigens, anti-CD3 antibody stimulated cells were fixed and permeabilized and stained for granzyme B by intracellular staining. Events were acquired on a flow cytometer and dot plots were analyzed by FloJo analysis. A summary of these plots show that CD8+ T cells from mice treated with cabozantinib secreted significantly more granzyme B than those of vehicle treated mice (F). (G) The image shows the average granzyme B secretion by cabozantinib treated mice and vehicle treated mice (P<0.01**).

Figure 5

Mice treated with cabozantinib are potent secretors of soluble molecules. Mice were inoculated with ID8-RFP tumor cells and treated with cabozantinib. Spleen cells were stimulated overnight with anti-CD3 antibody. Cultures from wells with identical treatments were harvested, pooled and centrifuged to remove cells. The resulting cell culture supernatants were stored at -20°C for assay in a mouse 32-Plex Discovery Assay system. The box plots summarize the most abundant cytokines which were significantly different between cabozantinib treated (n=5 mice) and vehicle treated mice (n=5). In each box plot, the average pg/ml of soluble molecule secretion of 5 mice is indicated by “X”. The lowest line on the whisker shows the minimum cytokine/chemokine secretion in each group. The highest line on the whisker shows the maximum cytokine/chemokine secretion in each group. The inner line in each box indicates the median soluble molecule secretion in each group.

Figure 6

Cabozantinib treatment of mice decreases VEGFR-2 in the liver. Tissue sections from ID8-RFP tumor cell inoculated mice which received treatment of cabozantinib or vehicle (and euthanized at day 66) were stained by IHC for VEGFR-2 (1:1000 dilution). By microscopic evaluation and analysis, (A) shows a low H-score for a liver of a cabozantinib treated mouse and a higher H score for a liver of a vehicle treated mouse is shown in (B). Each bar in (C and D) represents the H-score for 1 mouse. Overall, the data shows that there is no difference in expression of VEGFR-2 in the spleens of mice with cabozantinib or vehicle treatment. (C and D) Images show higher H-scores in livers of mice than in spleens. (D) The image shows significantly lower expression of VEGFR-2 in cabozantinib treated mice livers than in vehicle control mice livers (P<0.01**). This pattern of VEGFR-2 expression in cabozantinib treated livers and spleens was representative of 2 separate experiments.