PROJECT SUMMARY Immunotherapy has changed the way we think about and treat cancer. While some patients experience durable remissions, the majority however, will not respond. Some known barriers to response are a lack of immune infiltrate or infiltrate that is suppressed by other cell types in the tumor microenvironment. These patients may find immunotherapies that induce a de novo immune response, such as oncolytic viruses, more successful than one that stimulates the existing immune populations. An oncolytic virus selectively replicates in and lyses tumor cells, stimulating an immune response against viral and tumor antigen, while leaving healthy tissue unharmed. Currently there is one oncolytic approved for use in the clinic. Despite this recent approval and the rapid expansion of clinical trials evaluating oncolytics, little is known about the effects of these treatments on tumor resident immune cells. Many of these trials are currently evaluating oncolytic vaccinia virus as it is an ideal candidate for immunotherapy. Vaccinia replicates entirely in the cytosol preventing incorporation into host DNA, has a dsDNA genome that is easily engineered, and stimulates robust immune responses. Using a genetically engineered strain of oncolytic vaccinia virus (vvDD) with deletions of both viral growth factor and thymidine kinase to increase its tumor selectivity, we have shown that a single dose of this therapy can dramatically remodel the tumor infiltrate. Major increases in CD8+ T cell infiltration are observed as expected, however we also observed surprising phenotypic changes. Seven days post-treatment a loss of regulatory CD4+ T cells (Treg) was seen. Using a GFP expressing strain of the virus (vvDD-GFP) we found that surprisingly, one day post-treatment Tregs were selectively infected. As these cells are experiencing hypoxia in the tumor, we hypothesize that hypoxia leads to infection of these subsets which culminates in their death. The resulting microenvironment, now depleted of immune regulatory cells, is more supportive of immune infiltration and anti-tumor activity. We will test this hypothesis by (1) determining if oncolytic viral infection leads to the death of Treg and determine the contribution of Treg loss to therapeutic efficacy and (2) dissecting the contribution of HIF1α to the ability of vvDD to infect Tregs. The conclusions from these studies will provide us with a better understanding of the mechanism of oncolytic vaccinia. With this information it will be possible to design more efficacious oncolytic viruses as well as determine the patient populations most likely to respond to therapy.