Syngeneic mouse model of human HER2+ metastatic breast cancer for the evaluation of trastuzumab emtansine combined with oncolytic rhabdovirus

Background Established mouse models of HER2+ cancer are based on the over-expression of rodent Neu/Erbb2 homologues, which are incompatible with human HER2 (huHER2) targeted therapeutics. Additionally, the use of immune-deficient xenograft or transgenic models precludes assessment of native anti-tumour immune responses. These hurdles have been a challenge for our understanding of the immune mechanisms behind huHER2-targeting immunotherapies. Methods To assess the immune impacts of our huHER2-targeted combination strategy, we generated a syngeneic mouse model of huHER2+ breast cancer, using a truncated form of huHER2, HER2T. Following validation of this model, we next treated tumour-bearing with our immunotherapy strategy: oncolytic vesicular stomatitis virus (VSVΔ51) with clinically approved antibody-drug conjugate targeting huHER2, trastuzumab emtansine (T-DM1). We assessed efficacy through tumour control, survival, and immune analyses. Results The generated truncated HER2T construct was non-immunogenic in wildtype BALB/c mice upon expression in murine mammary carcinoma 4T1.2 cells. Treatment of 4T1.2-HER2T tumours with VSVΔ51+T-DM1 yielded robust curative efficacy compared to controls, and broad immunologic memory. Interrogation of anti-tumour immunity revealed tumour infiltration by CD4+ T cells, and activation of B, NK, and dendritic cell responses, as well as tumour-reactive serum IgG. Conclusions The 4T1.2-HER2T model was used to evaluate the anti-tumour immune responses following our complex pharmacoviral treatment strategy. These data demonstrate utility of the syngeneic HER2T model for assessment of huHER2-targeted therapies in an immune-competent in vivo setting. We further demonstrated that HER2T can be implemented in multiple other syngeneic tumour models, including but not limited to colorectal and ovarian models. These data also suggest that the HER2T platform may be used to assess a range of surface-HER2T targeting approaches, such as CAR-T, T-cell engagers, antibodies, or even retargeted oncolytic viruses.