Comparison of tumour-associated antigen-specific immune responses induced by the poxvirus vectors, MVA and ALVAC
Viral cancer vaccines comprise a form of immunotherapy that aims to generate anti-tumour immune responses that will lead to tumour destruction and enhanced survival of the patient. Accordingly, many tumour-associated antigens (TAAs) and viral vectors have been tested to gain insights into how these responses can be elicited most effectively. The poxviruses MVA and ALVAC have been characterized as vectors for a number of TAAs, however, an 'in vivo' comparison of their immunogenicity has not been performed. We compared these vectors in the context of the TAAs, 5T4 (an oncofetal carcinoma antigen) and gp100 (a melanosomal membrane glycoprotein), to illuminate any differences that may exist regarding antigen expression and immunogenicity. To facilitate the comparison, rMVA-gp100M and rALVAC(2)-5T4 were constructed to complement rALVAC(2)-gp100M and rMVA-5T4, respectively, which have been characterized previously. Expression of TAA protein by each recombinant was confirmed by western blot analysis 24 hours following infection of chicken embryo fibroblasts (CEFs). The level of 5T4 expression was approximately equal for both viruses, whereas gp100 stability appeared to be greater when MVA was the vector. Substantial degradation of gp100 at this time point has been previously shown to occur in both CEFs and mammalian MRC5 cells infected with rALVAC(2)-gp100M, suggesting that this unique vector-transgene combination leads to enhanced degradation of the recombinant protein. Transgenic mice expressing human Class I MHC (HLA-A2) molecules were vaccinated with each recombinant virus using a homologous prime boost schedule and IFN-[gamma] ELiSPOT was used to compare CD8+ T cell responses elicited against each TAA. 5T4-specific T cell responses could not be detected using either virus vector, whereas MVA elicited a significantly higher gp100-specific response than ALVAC at 106 PFU (p<0.001). Comparison of TAA antibody titers in CD-1 mice (vaccinated as above) showed that similar 5T4 responses were elicited by each vector, whereas MVA induced gp100 responses at a lower dose than ALVAC (p<0.001). In this study, 'in vitro' TAA protein stability correlated with 'in vivo' immunogenicity. The use of ALVAC as a vector for gp100 led to reduced protein stability in infected cells (' in vitro'), and lower potency of both antibody production and CD8 + T cell activation, relative to MVA. These findings suggest that the effects of novel vector-transgene combinations may not be predictable and should therefore be assessed individually when designing vaccines.