Development of Virus-infected Cancer Cell Vaccine

Abstract

Oncolytic viruses can be genetically modified to limit their replication in normal cells rendering them a cancer specific treatment. In addition, they can induce a “danger signal” in the form of pathogen- and damage-associated molecular patterns leading to anti-tumour immunity. Furthermore, they can be armed with various immunomodulatory molecules to further enhance anti-tumour immunity. In this project I aim to exploit these qualities to develop a translatable cancer vaccine. Virus-infected cancer cells were injected subcutaneously in a prime/boost regimen. Dying cancer cells will release the required danger signal leading to dendritic cell activation and cross-presentation of tumour associated antigens to T cells to elicit an anti-tumour immune response. Our results in the murine pancreatic cancer model showed that vaccination with virusinfected DT6606 cells induced tumour specific immunity capable of protecting vaccinated animals against re-challenge with tumour cells. The highest level of interferon gamma production, a surrogate marker of anti-tumour immunity, was achieved when animals were primed with adenovirus-infected cells. There was no significant difference between various boost groups. To enhance the safety of the proposed protocol a secondary treatment was introduced to arrest the proliferation of tumour cells prior to injection. Our results confirmed that secondary treatment with mitomycin does not affect the induction of tumour specific immunity and it does not affect the release of pathogen-associated molecular patterns in the form of viral proteins and DNA. To test our vaccination regimen in head and neck squamous cell carcinoma (HNSCC) we develop a clinically relevant mouse model using SCC7, B4B8 and LY2 cells to replicate various clinical scenarios including locally advancing disease and post excision locoregional recurrence. Vaccinating mice with HNSCC cells pre-infected with our recently developed tumour-targeted triple-deleted adenovirus (AdTD) resulted in a cell-specific antitumour immune response. In addition, it resulted in an increase in effector memory T-cells of both CD4+ and CD8+ phenotypes. Efficacy studies showed our vaccination can significantly slow down the growth rate of tumours in locally advancing disease. This led to increase survival of the vaccinated mice although it did not reach statistical significance. To further enhance the efficacy of our vaccination regimen, we aimed to increase T cell trafficking to the tumour site. CCL25 is a gut homing chemokine. Priming T cells in the presence of CCL25 will lead to upregulation of the surface expression of α4β7 integrin. The latter is a ligand of MAdCAM-1, a cell adhesion molecule highly expressed in the gut and pancreatic tumours. The α4β7/MAdCAM-1 interaction results in preferential homing of activated T cells to these organs. We hypothesised that vaccinating mice with pancreatic tumour cells pre-infected with a CCL25-armed adenovirus will lead to increased T cell trafficking to pancreatic tumours leading to enhanced efficacy. Although we achieved encouraging results in our pilot experiment, we did not detect any significant increase in α4β7 expression once we added a secondary treatment to the vaccination protocol. Similarly, efficacy experiments in the pancreatic cancer transgenic KPC mice did not show any difference in survival between AdTD-CCL25 and the control virus although both groups showed a trend towards increased survival compared to naïve mice. In conclusion, Virus-infected cancer cell vaccine is a potentially promising immunotherapeutic strategy that can be combined with traditional cancer therapies to increase survival of HNSCC and pancreatic cancer patients.