Fractionated but not single-dose radiotherapy induces an immune-mediated abscopal effect when combined with anti-CTLA-4 antibody

Abstract

Purpose: This study tested the hypothesis that the type of dose fractionation regimen determines the ability of radiotherapy to synergize with anti-CTLA-4 antibody.

Experimental design: TSA mouse breast carcinoma cells were injected s.c. into syngeneic mice at two separate sites, defined as a "primary" site that was irradiated and a "secondary" site outside the radiotherapy field. When both tumors were palpable, mice were randomly assigned to eight groups receiving no radiotherapy or three distinct regimens of radiotherapy (20 Gy x 1, 8 Gy x 3, or 6 Gy x 5 fractions in consecutive days) in combination or not with 9H10 monoclonal antibody against CTLA-4. Mice were followed for tumor growth/regression. Similar experiments were conducted in the MCA38 mouse colon carcinoma model.

Results: In either of the two models tested, treatment with 9H10 alone had no detectable effect. Each of the radiotherapy regimens caused comparable growth delay of the primary tumors but had no effect on the secondary tumors outside the radiation field. Conversely, the combination of 9H10 and either fractionated radiotherapy regimens achieved enhanced tumor response at the primary site (P < 0.0001). Moreover, an abscopal effect, defined as a significant growth inhibition of the tumor outside the field, occurred only in mice treated with the combination of 9H10 and fractionated radiotherapy (P < 0.01). The frequency of CD8+ T cells showing tumor-specific IFN-gamma production was proportional to the inhibition of the secondary tumor.

Conclusions: Fractionated but not single-dose radiotherapy induces an abscopal effect when in combination with anti-CTLA-4 antibody in two preclinical carcinoma models.

Figure 1. Tumor model and treatment schedule

Immunocompetent mice were injected s.c. with syngeneic TSA cells (1 × 10⁵) into the right (defined as “primary” tumor) and left (defined as “secondary” tumor) flank on day 0 and 2, respectively. Ionizing radiation (IR) was administered locally exclusively to the primary tumor with the rest of the body shielded, in a single dose or multiple fractions given in consecutive days starting on day 12. CTLA-4 blocking mAb 9H10 was given i.p every three days for three times starting on day 12, 14 or 16, as indicated. Primary and secondary tumor volumes were measured until day 35, at which time mice were sacrificed and tumors weighed.

Figure 2. The abscopal effect is induced in TSA tumor-bearing mice by fractionated radiation in combination with anti-CTL-4 antibody

(A) Tumor growth delay of primary irradiated tumors (left panel) and secondary non-irradiated tumor (right panel) in mice treated with PBS (closed circles), 9H10 (open circles), 20 Gy × 1 + PBS (closed diamonds), 20 Gy × 1 + 9H10 (open diamonds), 8 Gy × 3 + PBS (closed squares), 8 Gy × 3 + 9H10 (open squares), 6 Gy × 5 + PBS (closed triangles), or 6 Gy × 5 + 9H10 (open triangles). 9H10 was given on days 14, 17, and 20. Data are the mean ± SE of 5 mice/group. (B) Tumor weight of primary (left panel) and secondary (right panel) tumors at day 35. Data are the mean ± SE. The number of mice with complete tumor regression over the total number of mice per group is indicated. Data shown are from one of two independent experiments with similar results.

Figure 3. Effect of time of administration of anti-CTLA-4 antibody on the abscopal effect induced by radiotherapy in TSA tumor-bearing mice

(A) Tumor growth delay of primary irradiated tumors (left panel) and secondary non-irradiated tumor (right panel) in mice treated with PBS (closed circles), 9H10 givenatday 12, 15, 18 (open circles), 20 Gy × 1 + PBS (closed diamonds), 20 Gy × 1 + 9H10 givenatday 12, 15, 18 (open triangles), or 20 Gy × 1 + 9H10 given at day 14, 17, 20 (open diamonds). Data are the mean ± SE of 5 mice/group. No complete regression of either primary or secondary tumors was observed in any of the treatment arms. (B) Tumor growth delay of primary irradiated tumors (left panel) and secondary non-irradiated tumor (right panel) in mice treated with PBS (closed circles), 8 Gy × 3 + PBS (closed squares), 8 Gy × 3 + 9H10 given at day 12, 15, 18 (open triangles), 8Gy × 3 + 9H10 given at day 14, 17, 20 (open squares), and 8 Gy × 3 + 9H10 given at day 16, 18, 21 (open diamonds). Data are the mean ± SE of 6 mice/group. Complete regression was seen in 3 of 6 primary and 1 of 6 secondary tumors in mice treated with 8 Gy × 3 + 9H10 given at day 12, 15, 18; in 5 of 6 primary and 1 of 6 secondary tumors in mice treated with 8 Gy × 3 + 9H10 given at day 14, 17, 20; and in 1 of 6 primary and 1 of 6 secondary tumors in mice treated with 8 Gy × 3 + 9H10 given at day 16, 18, 21.

Figure 4. Fractionated radiotherapy given to TSA tumor-bearing mice in 3 doses of 8 Gy is more effective than 5 doses of 6 Gy in synergizing with anti-CTLA-4 antibody

(A) Tumor growth delay of primary irradiated tumors (left panel) and secondary non-irradiated tumor (right panel) in mice treated with PBS (closed circles), 8 Gy × 3 + 9H10 (open squares), or 6 Gy × 5 + 9H10 (open triangles). 9H10 was given on days 14, 17, and 20. Data are the mean ± SE of 5 mice/group. (B) Tumor weight of primary (C) and secondary (D) tumors at day 35. Data are the mean ± SE. The number of mice with complete tumor regression over the total number of mice per group is indicated.

Figure 5. The combination of fractionated radiotherapy with anti-CTLA-4 antibody enhances TIL in secondary TSA tumors and tumor-specific T cells producing IFNγ

(A, B) Secondary tumors were excised at day 35 and analyzed by fluorescence microscopy for the presence of CD4+ and CD8+ T cells. (A) Representative fields showing CD4+ (top panels) and CD8+ (bottom panels) T cells (white) infiltrating secondary TSA tumors in mice treated as indicated. Nuclei were stained with DAPI (light gray). (B) Mean number ± SE of CD4+ and CD8+ TILs in three mice per group. Both CD4+ and CD8+ TIL were significantly increased in mice treated with the combination of 8 Gy × 3 + 9H10 (p<0.05 compared to all other groups), whereas radiation and 9H10 as single modalities did not have a significant effect. (C, D) Analysis of tumor-specific IFNγ production by spleen cells harvested at day 35 from mice in the various treatment groups. (C) IFNγ concentration in supernatants of total spleen cells isolated from mice treated with 0 Gy + PBS (closed circles), 8 Gy × 3 + 9H10 (open squares), or 6 Gy × 5 + 9H10 (open triangles) and cultured o.n. with irradiated TSA cells were plotted against the volume of the secondary tumor (Top panel). The percentage of CD8+ T cells expressing IFNγ when exposed to TSA cells as determined by intracellular staining (D) following in vitro restimulation with the TSA-derived immunodominant CD8 epitope AH1 was plotted against the volume of the secondary tumor (Bottom panel). Symbols are as above. Each symbol represents one animal. (D) Representative histograms showing the percentage of CD8+ T cells positive for IFNγ by intracellular staining and flow cytometry in response to TSA cells or the irrelevant target RMA-S-Ld. Samples were gated on CD8+ T cells.

Figure 6. The abscopal effect is induced in MCA38 tumor-bearing mice by fractionated radiation in combination with anti-CTL-4 antibody

C57BL/6 mice were injected with syngeneic MCA38 colon carcinoma cells (5 × 10⁵) s.c. into the right and left flank as outlined in Figure 1. (A) Tumor growth delay of primary irradiated tumors (left panel) and secondary non-irradiated tumor (right panel) in mice treated with PBS (closed circles), 9H10 (open circles), 20 Gy × 1 + PBS (closed diamonds), 20 Gy × 1 + 9H10 (open diamonds), 8 Gy × 3 + PBS (closed squares), 8 Gy × 3 + 9H10 (open squares). 9H10 was given on days 14, 17, and 20. Data are the mean ± SE of 5 mice/group. (B) Tumor weight of primary (left panel) and secondary (right panel) tumors at day 35. Data are the mean ± SE. The number of mice with complete tumor regression over the total number of mice per group is indicated. (C) Secondary tumors were excised at day 35 and analyzed by fluorescence microscopy for the presence of CD4+ and CD8+ T cells. Data are the mean number ± SE of CD4+ and CD8+ TILs in three mice per group. Both CD4+ and CD8+ TIL were significantly increased in mice treated with the combination of 8 Gy × 3 + 9H10 (p<0.05 compared to all other groups). (D) Analysis of tumor-specific IFNγ production by spleen cells harvested at day 35 from treated and untreated mice and re-stimulated in vitro with irradiated MCA38 cells. Histograms show the percentage of CD8+ T cells positive for IFNγ by intracellular staining and flow cytometry in response to MCA38 cells or the irrelevant target RMA-S-Ld. Samples were gated on CD8+ T cells. Spleen cells from 3 mice in each treatment group were pooled.

Figure 6. The abscopal effect is induced in MCA38 tumor-bearing mice by fractionated radiation in combination with anti-CTL-4 antibody

C57BL/6 mice were injected with syngeneic MCA38 colon carcinoma cells (5 × 10⁵) s.c. into the right and left flank as outlined in Figure 1. (A) Tumor growth delay of primary irradiated tumors (left panel) and secondary non-irradiated tumor (right panel) in mice treated with PBS (closed circles), 9H10 (open circles), 20 Gy × 1 + PBS (closed diamonds), 20 Gy × 1 + 9H10 (open diamonds), 8 Gy × 3 + PBS (closed squares), 8 Gy × 3 + 9H10 (open squares). 9H10 was given on days 14, 17, and 20. Data are the mean ± SE of 5 mice/group. (B) Tumor weight of primary (left panel) and secondary (right panel) tumors at day 35. Data are the mean ± SE. The number of mice with complete tumor regression over the total number of mice per group is indicated. (C) Secondary tumors were excised at day 35 and analyzed by fluorescence microscopy for the presence of CD4+ and CD8+ T cells. Data are the mean number ± SE of CD4+ and CD8+ TILs in three mice per group. Both CD4+ and CD8+ TIL were significantly increased in mice treated with the combination of 8 Gy × 3 + 9H10 (p<0.05 compared to all other groups). (D) Analysis of tumor-specific IFNγ production by spleen cells harvested at day 35 from treated and untreated mice and re-stimulated in vitro with irradiated MCA38 cells. Histograms show the percentage of CD8+ T cells positive for IFNγ by intracellular staining and flow cytometry in response to MCA38 cells or the irrelevant target RMA-S-Ld. Samples were gated on CD8+ T cells. Spleen cells from 3 mice in each treatment group were pooled.