The strength of the T cell response against a surrogate tumor antigen induced by oncolytic VSV therapy does not correlate with tumor control.

Cancer therapy using oncolytic viruses has gained interest in the last decade. Vesicular stomatitis virus is an attractive candidate for this alternative treatment approach. The importance of the immune response against tumor antigens in virotherapy efficacy is now well recognized, however, its relative contribution versus the intrinsic oncolytic capacity of viruses has been difficult to evaluate. To start addressing this question, we compared glycoprotein and matrix mutants of vesicular stomatitis virus (VSV), showing different oncolytic potentials for B16/B16gp33 melanoma tumor cells in vitro, with the wild-type virus in their ability to induce tumor-specific CD8(+) T cell responses and control tumor progression in vivo. Despite the fact that wild-type and G mutants induced a stronger gp33-specific immune response compared to the MM51R mutant, all VSV strains showed a similar capacity to slow down tumor progression. The effectiveness of the matrix mutant treatment proved to be CD8(+) dependent and directed against tumor antigens other than gp33 since adoptive transfer of isolated CD8(+) T lymphocytes from treated B16gp33-bearing mice resulted in significant protection of naive mice against challenge with the parental tumor. Remarkably, the VSV matrix mutant induced the upregulation of major histocompatibility class-I antigen at the tumor cell surface thus favoring recognition by CD8(+) T cells. These results demonstrate that VSV mutants induce an antitumor immune response using several mechanisms. A better understanding of these mechanisms will prove useful for the rational design of viruses with improved therapeutic efficacy.