Abscopal regression of antigen disparate tumors by antigen cascade after systemic tumor vaccination in combination with local tumor radiation

Abstract

Radiation is a primary modality in cancer treatment. Radiation can also reduce tumor growth outside the treatment field, often referred to as the abscopal effect. The mechanisms and therapeutic potential of the abscopal effect have not been fully elucidated. We evaluated the role of vaccination directed against a tumor-associated antigen (TAA) in the induction and amplification of radiation induced abscopal effects. Active-specific immunotherapy with a TAA-specific vaccine regimen was used to induce and potentiate T-cell responses against carcinoembryonic antigen (CEA) in combination with local irradiation of subcutaneous tumors. We examined the potential synergy of a poxvirus-based CEA vaccine regimen in CEA-transgenic (Tg) mice in combination with either external beam radiation or brachytherapy of local tumors. The induction of CD8(+) T cells specific for multiple TAAs not encoded by the vaccine was observed after the combination therapy. In two tumor models, the antigen cascade responses induced by vaccine and local irradiation mediated the regression of antigen negative metastases at distal subcutaneous or pulmonary sites. Clinically, local control of the primary tumor is necessary and can sometimes prevent metastases; however, irradiation generally fails to control preexisting metastases. These studies suggest that by coupling tumor irradiation with immunotherapy, the abscopal effect can transcend from anecdotal observation to a defined mechanism that can be exploited for the treatment of systemic disease.

FIG 1.

The death receptor Fas is upregulated on tumor cells in vivo after radiation treatment. C57BL/6 mice were transplanted with MC38-CEA⁺ tumor cells subcutaneously. After 7 days, tumors in a subset of mice were subjected to external beam radiation (8 Gy). Tumors were surgically removed 72 hours after radiation treatment, fixed, sectioned at 5 μm, and immunostained with anti-Fas mAb or an isotype control antibody (A). Photomicrographs are shown at 20X. (B) The entire slide set was digitally scanned, analyzed, and Fas positive cells quantified by pixel analysis. Radiation: 0 Gy, open bars; 8 Gy, hatched bars.

FIG. 2.

Radiation therapy of tumor cells and vaccine therapy resulted in synergistic antitumor activity. CEA-Tg mice were transplanted with MC38-CEA⁺ tumor cells subcutaneously. (A) Therapy schema. (B) Mice received no treatment. (C) Tumors in mice were subjected to single-dose external beam radiation (8 Gy) on day 14. (D) Mice were vaccinated with rV-CEA/TRICOM on day 8 (V1), followed by boosting with rF-CEA/TRICOM on days 15, 22, and 29 (V2-V4). (E) Mice were vaccinated with rV-CEA/TRICOM on day 8. Tumors were subjected to single-dose external beam radiation (8 Gy) in situ on day 14, followed by boosting with rF-CEA/TRICOM on days 15, 22, and 29. All vaccines were coadministered with rF-GM-CSF. Tumor volume was monitored.

FIG. 3.

CEA-Tg mice were transplanted with MC38-CEA⁺ tumor cells. (A, E) Mice received no treatment. (C, G) Mice were vaccinated with rV-CEA/TRICOM on day 8, boosted with rF-CEA/TRICOM on day 15. (B, F) Tumors in mice were subjected to radiation (8 Gy) on day 14. (D, H) Mice were vaccinated with rV-CEA/TRICOM on day 8. Tumors were irradiated (8 Gy) on day 14, followed by boosting with rF-CEA/TRICOM on day 15. All vaccines were coadministered with rF-GM-CSF. Tumors were surgically removed at day 21 and analyzed for microvessel architecture (A–D) or infiltrating CD3⁺ T cells by immunohistochemistry (E–H). Photomicrographs are shown at 20X. (I) Percentage of vessel area as determined by pixel analysis (n=2 mice/treatment group). (J) Percentage of infiltrating T cells as determined by pixel analysis (n=2 mice/treatment group). (K) Population analysis of infiltrating lymphocytes. Depicted are percent infiltrations for all groups.

FIG. 4.

CD8⁺ T-cell responses to multiple tumor antigens after therapy with vaccine and irradiation. CEA-Tg mice were transplanted with MC38-CEA⁺ tumor cells. Mice were vaccinated with rV-CEA/TRICOM on day 8 and boosted with rF-CEA/TRICOM on days 15, 22, and 29. Tumors were subjected to radiation (8 Gy) on day 14. Depicted are IFN-γ responses to CD8-restricted peptides specific for CEA (open bars), p53 (black bars), and GP70 (gray bars) from pooled splenic T cells taken from cured mice 6 months after tumor transplant. *: Statistical significance vs age-matched CEA-Tg control mice.

FIG. 5.

Treatment of a CEA⁺ tumor with vaccine and irradiation mediates abscopal regression of a CEA^- tumor at a distal site. (A) Experiment schema; CEA-Tg mice were transplanted with MC38-CEA⁺ tumors (subcutaneously right flank) on day 0 and MC38 (CEA^-) tumors on the left flank on day 5. Mice were treated with rV-CEA/TRICOM on day 8 and administered 8 Gy external beam radiation to the MC38-CEA⁺ tumor on day 14. Mice were boosted with rF-CEA/TRICOM on days 15, 22, and 29. All vaccines were given with rF-GM-CSF. Tumor volume on each flank was monitored. (B) Tumor volume of MC38-CEA⁺ tumors 28 days post-tumor transplant. (C) Tumor volume of MC38 tumors 28 days post-tumor transplant. (D) Total tumor burden of mice (MC38-CEA⁺ plus MC38 tumors).

FIG. 6.

Treatment of a primary tumor with vaccine and irradiation mediates abscopal regression of pulmonary metastases. (A) Therapy schema; CEA-Tg mice were transplanted with LL2-CEA⁺ tumors (subcutaneously right flank) and LL2-CEA⁺ tumors intravenously on day 0. Mice were vaccinated with rV-GP70/TRICOM on day 7, boosted with rF-GP70/TRICOM on days 14, 21, and 28. All vaccines were coadministered with rF-GM-CSF. (B) Primary tumors in subsets of mice were implanted with a single ¹²⁵I-brachytherapy seed on day 13. Depicted is a radiograph of tumor implanted with brachytherapy seed. Spheres, localizing fiducial markers. (C) Fas upregulation of LL2-CEA⁺ primary tumors after 72 hours exposure to ¹²⁵I-brachytherapy. Inset panel depicts % Fas⁺ cells (mean fluorescent intensity). (D) On day 35, lungs were harvested and pulmonary metastases were enumerated. *: Statistical significance.

FIG. 7.

Model of antigen cascade mediated abscopal effect.