Pancreatic cancer (PDAC) affects 4400 Canadians each year and with 5year survival rates <8%, there is clearly an unmet need for new therapeutic approaches for treating this deadly disease. Herein I report the development of a surgical model of PDAC that recapitulates many of the hallmarks of human disease and has an immune infiltrate consisting of T cells and suppressive regulatory T cells and myeloid derived suppressor cells. This model allows the exploration of new therapeutics that can be used in combination with surgical resection of primary tumours. Furthermore, I propose that the use of neoadjuvant administration of a prime-boost oncolytic vaccine targeting a pancreatic tumour associated antigen (TAA) - mesothelin - could potentiate pancreatic tumour specific immune responses to improve patient prognosis. We demonstrate that immune tolerance to this self antigen can be broken by the complete depletion of circulating Tregs at the time of vaccination, which leads to the activation of a population of CD8+ T cells responsive to mesothelin. We demonstrate that these T cells respond to mesothelin expressing tumour cells ex vivo, and that CD8+ T cells are recruited to the site of tumour challenge. However, despite the generated CD8+ T cell response, oncolytic vaccine strategies targeting mesothelin provide no protection against Pan02 tumours, or against other mesothelin expressing murine tumour lines. I demonstrate that this is not through common tumour escape mechanisms, nor through the upregulation of suppressive immune populations. Any efficacy observed was found to be provided solely by depletion of Tregs, as the depletion of CD8+ T cells did not reduce protection from tumour outgrowth in vaccinated mice. While mesothelin represents a promising target, it is not an ideal target for oncolytic vaccine platforms, potentially due to its nature as a self antigen.