Humanized mouse model of MAGE-A1-targeted anti-melanoma T cell therapy

Adoptive cellular therapy (ACT) has resulted in durable anti-tumor responses for both solid and hematological malignancies, mediated by the persistence and activation of transferred T cells. Unfortunately most patients, especially those with solid tumors, continue to progress despite persistent T cells. Thus revealing challenges specific to the hostile microenvironment of solid tumors and a critical need to identify factors that limit ACT efficacy. Furthermore, serial patient biopsies are challenging to obtain and accessible peripheral blood does not reflect relevant cell interactions at the tumor site. Xenograft models that lack infiltrating human macrophages, NK cells and lymphocytes cannot faithfully recapitulate critical interactions within the tumor microenvironment. To overcome this limitation, we initiated ACT in the MISTRG mouse that supports hematopoiesis after transplantation of human CD34 + stem cells resulting in functional human innate and adaptive immune cells. Co-transplantation of the melanoma cell line (Me275) results in tumor infiltration by human macrophages (TAMs) supporting tumor progression. The cancer/testis antigen Melanoma associated Antigen Gene (MAGE)-A1 is an attractive ACT target due to broad expression in tumors and limited expression in normal tissues rendering off-tumor tissue toxicities unlikely. We have identified a high-affinity HLA-A*02:01-restricted TCR that recognizes MAGE-A1 278-286 peptide (TCR MA1 ) and confers effector functions in transduced CD8 and CD4 T cells when co-transduced with CD8αβ. Ex vivo analysis of MISTRG T cell cytolysis and proliferation suggest that CD4 + T cell help is required in this model. After two weeks of in vitro expansion, CD4 + and CD8 + T cells from humanized MISTRG spleens were transduced and transferred into MISTRG littermates harboring MAGE-A1 + Me275. Although TCR MA1 treated mice subtly controlled the growth of Me275, the changes were not significant. However, compared to an irrelevant TCR, only TCR MA1 CD8 + T cells localized to tumor and persisted in blood and spleen 8 and 28 days post transfer. TCR MA1 CD8 T cells displayed an effector phenotype, with high levels of PD-1 expression characteristic of recently activated and/or exhausted T cells. These results suggest that TCR MA1 cells recognized the tumor and were consequently activated, but could not overcome suppression in the microenvironment, mirroring human studies. Additional studies are being initiated to overcome the specific factors associated with the TAMs that limit in vivo T cell function. This model provides an ideal environment to test next generation T cell therapies, such as addition of intrinsic co-stimulation, to enhance responses against solid tumors. Humanized MISTRG mice are a powerful tool to study suppression of ACT in vivo, offering streamlined approaches to overcome hostile tumor microenvironments before clinical translation. Citation Format: Megan S. McAfee, Trisha Sippel, Daniel Hunter, Jean Campbell, Thomas Schmitt, Robert Pierce, Anthony Rongvaux, Aude Chapuis. Humanized mouse model of MAGE-A1-targeted anti-melanoma T cell therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3569.