Anti-melanoma vaccines engineered to simultaneously modulate cytokine priming and silence PD-L1 characterized using ex vivo myeloid-derived suppressor cells as a readout of therapeutic efficacy

Abstract

Efficacious antitumor vaccines strongly stimulate cancer-specific effector T cells and counteract the activity of tumor-infiltrating immunosuppressive cells. We hypothesised that combining cytokine expression with silencing programmed cell death ligand 1 (PD-L1) could potentiate anticancer immune responses of lentivector vaccines. Thus, we engineered a collection of lentivectors that simultaneously co-expressed an antigen, a PD-L1-silencing shRNA, and various T cell-polarising cytokines, including interferon ? (IFN?), transforming growth factor ? (TGF?) or interleukins (IL12, IL15, IL23, IL17A, IL6, IL10, IL4). In a syngeneic B16F0 melanoma model and using tyrosinase related protein 1 (TRP1) as a vaccine antigen, we found that simultaneous delivery of IL12 and a PD-L1-silencing shRNA was the only combination that exhibited therapeutically relevant anti-melanoma activities. Mechanistically, we found that delivery of the PD-L1 silencing construct boosted T cell numbers, inhibited in vivo tumor growth and strongly cooperated with IL12 cytokine priming and antitumor activities. Finally, we tested the capacities of our vaccines to counteract tumor-infiltrating myeloid-derived suppressor cell (MDSC) activities ex vivo. Interestingly, the lentivector co-expressing IL12 and the PD-L1 silencing shRNA was the only one that counteracted MDSC suppressive activities, potentially underlying the observed anti-melanoma therapeutic benefit. We conclude that (1) evaluation of vaccines in healthy mice has no significant predictive value for the selection of anticancer treatments; (2) B16 cells expressing xenoantigens as a tumor model are of limited value; and (3) vaccines which inhibit the suppressive effect of MDSC on T cells in our ex vivo assay show promising and relevant antitumor activities.Efficacious antitumor vaccines strongly stimulate cancer-specific effector T cells and counteract the activity of tumor-infiltrating immunosuppressive cells. We hypothesised that combining cytokine expression with silencing programmed cell death ligand 1 (PD-L1) could potentiate anticancer immune responses of lentivector vaccines. Thus, we engineered a collection of lentivectors that simultaneously co-expressed an antigen, a PD-L1-silencing shRNA, and various T cell-polarising cytokines, including interferon ? (IFN?), transforming growth factor ? (TGF?) or interleukins (IL12, IL15, IL23, IL17A, IL6, IL10, IL4). In a syngeneic B16F0 melanoma model and using tyrosinase related protein 1 (TRP1) as a vaccine antigen, we found that simultaneous delivery of IL12 and a PD-L1-silencing shRNA was the only combination that exhibited therapeutically relevant anti-melanoma activities. Mechanistically, we found that delivery of the PD-L1 silencing construct boosted T cell numbers, inhibited in vivo tumor growth and strongly cooperated with IL12 cytokine priming and antitumor activities. Finally, we tested the capacities of our vaccines to counteract tumor-infiltrating myeloid-derived suppressor cell (MDSC) activities ex vivo. Interestingly, the lentivector co-expressing IL12 and the PD-L1 silencing shRNA was the only one that counteracted MDSC suppressive activities, potentially underlying the observed anti-melanoma therapeutic benefit. We conclude that (1) evaluation of vaccines in healthy mice has no significant predictive value for the selection of anticancer treatments; (2) B16 cells expressing xenoantigens as a tumor model are of limited value; and (3) vaccines which inhibit the suppressive effect of MDSC on T cells in our ex vivo assay show promising and relevant antitumor activities.