Linking Cancer driver mutations to regulatory T cell immunosuppression

Abstract Breast cancer immunotherapy efficacy is still very limited. Thus, there is a pressing need to understand molecular determinants enabling clinically distinct breast cancers to suppress antitumor immune response and harness these mechanisms for novel treatments. Emerging evidence suggests that oncogenic mutations can directly adversely affect tumor immune responses. However, such functions for vast majority of breast cancer mutations have not been explored. In this proposal, we focus on aggressive breast tumors carrying combinations of most frequent breast cancer driver mutations, as a proof of principle of a new and customizable platform of clinically relevant cancer models to understand immunosuppressive mechanisms. MLL3 (also known as KMT2C), encoding a histone methyltransferase, is a novel tumor suppressor in various human cancers. MLL3 is frequently inactivated by gene deletions or truncating point mutations, with especially high rate in breast cancers (up to 24%). Additionally, MLL3 mutant cancers have significantly worse outcomes compared to MLL3 wild type cancers. We have developed a novel method for rapidly generating genetically engineered mouse models (GEMMs) by efficiently expanding and “custom genome editing” mouse mammary stem cells (MaSCs) in culture and then using these MaSCs to regenerate genetically engineered mammary glands in syngeneic immunocompetent mice. Using this model of MLL3 deletion in conjunction with constitutive activation of PI3-kinase (PI3KCA, ~60% of MLL3 mutant tumors in patients are also PI3KCA) and inactivation of p53 (these three mutations altogether account for the most frequent combinations of cancer driver mutations in human breast cancers), we found that the loss of Mll3 promotes early infiltration of Foxp3+ regulatory T (Treg) cells and their further expansion and differentiation to a highly suppressive phenotype, leading to faster tumor immune escape in primary tumors and at metastatic sites. Monoclonal antibody targeting of specific immune receptors highly expressed on tumor-infiltrating Treg cells show remarkable efficacy in inhibiting tumor initiation and growth. Based on these findings, we next propose to investigate how MLL3 loss mechanistically activates HIF1 and harness the understanding to develop therapeutic interventions applicable to aggressive human breast cancers. We will explore underlying mechanisms of Treg cell differentiation into highly suppressive effector Treg cells in the tumors, mediated by both extracellular cues and cell-intrinsic regulators. Thus, in addition to uncovering new mechanisms by which major breast cancer drivers favor early immune escape, the power of our approach can be easily extended to test the tumor promoting effect of any breast cancer mutations, singly or in combination. The mechanistic investigations of these complex tumor-immune system interactions in vivo require the use of similarly complex in vivo models that closely recapitulate tumor development in an immunocompetent host environment in which tumors arise. To our knowledge, mice represent the most cost effective and best tractable mammalian models available to us.