589 Efficacy of oncolytic vaccinia virus requires infection of suppressive immune cells in the tumor microenvironment leading to their reprogramming and deletion. (10th December 2020)

• Record Type: Journal Article
• Authors: DePeaux, Kristin; Vignali, Paolo; Sarkar, Saumendra; Delgoffe, Greg
• Is Part Of: Journal for immunotherapy of cancer. Volume 8(2020)Supplement 3
• Page Start: A352
• Page End: A353
• Publication Date: 2020-12-10
• Subjects: Cancer -- Immunotherapy -- Periodicals; Cancer -- Immunological aspects -- Periodicals; Tumors -- Immunological aspects -- Periodicals; Immunotherapy -- Periodicals; 616.99406105
• DOI: 10.1136/jitc-2020-SITC2020.0589
• Languages: English
• ISSNs: 2051-1426

Abstract

Abstract : Background: Checkpoint blockade immunotherapy has dramatically changed cancer treatment; however, these therapies depend on the presence of a pre-existing immune infiltrate. Unfortunately, some patients have few to no infiltrating immune cells, highlighting the need for therapies that can generate antigenic stimuli. Oncolytic viruses, which infect and lyse tumor cells while leaving healthy tissue unharmed, are an attractive means to provide these signals, although the mechanisms of action of these engineered viral therapies remain incompletely understood. Virally induced immunogenic death causes an influx of tumor- and virus- specific effector CD8+ T cells. Many oncolytic viruses also decrease tumor-infiltrating suppressive immune populations, such as regulatory T cells (Treg), however the mechanism for this is unknown. Here we show that an oncolytic strain of vaccinia virus (VV) infects tumor infiltrating Tregs, in contrast to the prevailing idea that oncolytic viruses only infect tumor cells. Infection leads to viral-mediated Treg depletion that is required for tumor regression. Methods: Using a mouse model of head and neck squamous cell carcinoma (MEER), a VV-resistant line was generated through serial treatment of a VV-sensitive MEER line. At varied time points post-intratumoral treatment with VV, tumor infiltrating lymphocytes (TIL) were isolated from both the VV-resistant and VV-sensitive lines and analyzed by flow cytometry. Results: One day post-treatment of VV-sensitive MEER tumors, tumor isolated Tregs were infected by VV as determined by viral GFP expression. Infection was confirmed in vitro with purified Tregs. Four days post-treatment, tumor infiltrating Treg counts were reduced, and active caspase 3 staining was increased, suggesting that infection lead to Treg death. At 7 days post-treatment, the remaining Tregs in the VV-sensitive tumors acquired a fragile phenotype (IFN?+ Nrp1-). This was not observed in the VV-resistant MEER line. Fragile Tregs are less suppressive and indeed we observed an increase in pro-inflammatory cytokine production from CD8+ and Tconv (CD4+ Foxp3-) T cells in the VV-sensitive tumors compared to VV-resistant. We then engineered oncolytic VV to be susceptible to Cre mediated inactivation. Infection of various murine transgenic Cre lines confirmed the importance of non-tumoral immune infection for therapeutic efficacy, with a particular emphasis on Treg infection. Conclusions: These data reveal a previously unappreciated mechanism of action of oncolytic virus immunotherapy, in which new tumor immunity accompanies the viral mediated loss and phenotypic change of regulatory populations. Importantly, as this treatment is delivered intratumorally the loss of Tregs is tumor specific, resulting in targeted Treg deletion without systemic autoimmunity.