Pre-existing immunity boosts therapeutic efficacy of oncolytic influenza a virus in the presence and absence of neutralizing antibodies

The study of oncolytic viruses that naturally target humans requires models that elucidate the impact of pre-existing immunity on therapeutic efficacy. Recent studies have shown that the presence of anti-viral T cells in the tumor can be associated with potentiation of oncolytic viruses [1,2]. Conversely, the presence of neutralizing antibodies has also been shown to abolish response [1]. Thus, we hypothesized that pre-existing immunity to Influenza A virus (IAV) would lead to abrogation of oncolytic IAV therapy. To study the effect of neutralizing antibodies and IAV-specific T cells on oncolytic virus treatment, we established models of both homosubtypic and heterosubtypic immunity. Mice were infected with either PR8 virus (H1N1) or X-31 virus (H3N2.) 6 weeks post infection, mice were given bilateral CT26 tumors, and one tumor was treated with PR8 deltaNS1. Both tumors were monitored for tumor growth and overall survival. In a single tumor model, flow cytometry was used to define changes in the immune landscape unique to immunized mice. Additionally, anti-viral TILs were analyzed and characterized. Interestingly treated tumors displayed significant delays in tumor growth in PR8 and X-31 mice compared to naïve mice. Additionally, an increased abscopal effect in the contralateral tumor was only observed in X-31 mice. Treatment of X-31 mice resulted in a 20% increase in overall survival compared to PR8 and naïve mice. Rechallenged mice rejected tumor engraftment, indicating a durable anti-tumor memory response.There was an increase in both tumor infiltrating CD8+ and CD4+ T cells in the single tumor model. Amongst CD4s, there was a significant decrease in the presence of T regulatory cells (Treg); corresponding with an increase in CD8:Treg ratio. Immunized tumor bearing mice showed infiltration of anti-viral CD8s prior to treatment with oncolytic virus. X-31 mice had a higher prevalence of anti-IAV CD8s in the tumor and draining lymph node. In both immunization models, CD44+ anti-IAV CD8s were CD127 hi CD69 hi indicating a central memory phenotype. Further phenotyping TILs in X31-immunized, tumor bearing mice displayed a trends to less T-cell exhaustion and increased cytotoxicity. In models of both homosubtypic and heterosubtypic immunity, therapeutic efficacy is strengthened. This indicates that the presence of neutralizing antibodies does not hinder the therapeutic efficacy of the virus and that the presence of anti-IAV T cells in both models may be responsible for this potentiation. Future studies aim to understand how anti-viral T cells in the TME contribute to anti-tumor response via interaction with anti-tumor T cells and overall myeloid cells.