Production of interferon-gamma by in vivo tumor-sensitized T cells: association with active antitumor immunity

Preoperative chemoradiotherapy induces multiple pathways related to anti-tumour immunity in rectal cancer

BACKGROUND

Rectal cancer treated with preoperative radiotherapy (RT) provides an interesting model to study changes induced on cancer cell immuno-phenotype that could be exploited by immunotherapy interventions to improve prognosis.

MATERIALS AND METHODS

We assessed the expression of HLA-class-I, β2-microglobulin, TAP1, PD-L1 and STING/IFNβ in preoperative biopsies and respective post-RT surgical specimens from patients with rectal cancer (n = 27). The effect of radiation was further investigated in colorectal adenocarcinoma cell lines HT-29 and Caco-2.

RESULTS

Rectal carcinomas exhibited extensive loss of expression of HLA-Class-I related molecules, which was restored in post-irradiation surgical specimens (P < 0.0001). RT induced the expression of IFNβ and STING in cancer cells and tumour-infiltrating lymphocytes (P < 0.0001). In in vitro experiments, irradiation with 4 Gy or 10 Gy induced the expression of HLA-class-I protein (P < 0.001). PD-L1 levels were transiently induced for two days (P < 0.001). cGAS, STING, IFNβ and the downstream genes (MX1, MX2, UBE2L6v2, IFI6v2 and IFI44) mRNA levels significantly increased after 3 × 8 Gy or 1 × 20 Gy irradiation (P < 0.001). TREX1 mRNA levels remained unaltered.

CONCLUSIONS

RT induces the IFN-type-I pathway and the expression of HLA-class-I molecules on rectal carcinoma. The transient induction of PD-L1 expression suggests that long-course daily RT may sustain increased PD-L1 levels. Anti-PD-L1/PD-1 immunotherapy could block this immunosuppressive pathway.

The molecular basis of immuno-radiotherapy

PURPOSE

Radiotherapy (RT) and immunotherapy are powerful anti-tumor treatment modalities. Experimental research has demonstrated an important interplay between the cytotoxic effects of RT and the immune system. This systematic review provides an overview of the basics of anti-tumor immunity and focuses on the mechanisms underlying the interplay between RT and immune anti-tumor response that set the molecular basis of immuno-RT.

CONCLUSIONS

An 'immunity acquired equilibrium' mimicking tumor dormancy can be achieved post-irradiation treatment, with the balance shifted toward tumor eradication or regrowth when immune cells' cytotoxic effects or cancer proliferation rate prevail, respectively. RT has both immunosuppressive and immune-enhancing properties. The latter effect is also known as radio-vaccination. Its mechanisms involve up- or down-regulation of membrane molecules, such as PD-L1, HLA-class-I, CD80/86, CD47, and Fas/CD95, that play a vital role in immune checkpoint pathways and increased cytokine expression (e.g. INFα,β,γ, IL1,2, and TNFα) by cancer or immune cells. Moreover, the interactions of radiation with the tumor microenvironment (fibroblasts, tumor-infiltrating lymphocytes, monocytes, and dendritic cells are also an important component of radio-vaccination. Thus, RT may have anti-tumor vaccine properties, whose sequels can be exploited by immunotherapy agents to treat different cancer subtypes effectively.

Immunomodulation in an apparently non-immunogenic murine tumor

Madison lung carcinoma (M109), a murine tumor of spontaneous origin, appears to be non-immunogenic, according to 2 commonly employed tests for tumor immunogenicity. However, C.parvum-induced immunopotentiation during the growth of M109 tumor results in post-excision anti-tumor immunity to M109 tumor implants. The C.parvum-potentiated post-excision immunity to M109 is tumor-specific and T-cell-dependent. T cells from mice whose progressive M109 tumors have been excised are capable, on passive transfer, of inhibiting adoptive immunotherapy of T-cell-deficient recipients by spleen cells from mice immunized with an admixture of M109 cells and C.parvum. The data are interpreted as evidence supporting the hypothesis that the apparent lack of anti-tumor immunity in this tumor model is not due to the absence of tumor-associated antigens. We suggest that, instead, in this model the balance between the effector and suppressor arms of the immune response favors tumor-induced immunosuppression, resulting in a magnitude of anti-tumor immunity insufficient for detection by commonly employed tests for tumor immunogenicity. Our study shows that shifting the balance in favor of the effector arm by means of immunopotentiation results in a measurable immune response to an apparently non-immunogenic tumor.