Abstract A88: RUNX1-deficiency in luminal mammary epithelial cells leads to development of a unique type of immune “hot” ER+ mammary tumor

Abstract

Abstract Genetic mutations in cancer cells play key roles in shaping the tumor immune microenvironment. RUNX1 is a transcription factor (TF) essential for estrogen receptor (ER)+ luminal mammary epithelial cells (MECs) and is mutated in a significant portion of human ER+ luminal breast cancers. It is also an important regulator of immune cells, raising an intriguing question as to whether its loss-of-function mutations found in ER+ breast cancers affect both MECs and immune cells. To address this, we established an in vivo mouse model by inducing loss of both RUNX1 and p53 in luminal MECs, upon intraductal injection of a Cre-expressing adenovirus under the control of the Keratin 8 promoter (Ad-K8-Cre) to mammary glands of female mice carrying their conditional knockout alleles. Compared to similarly injected wild-type (WT) control mice or mice with induced loss of p53 alone, induced loss of both led to hyperproliferation of Runx1/Trp53-null MECs and eventually development of mammary tumors after a long latency (a tumor-free period of ~9-13 months after Ad-K8-Cre injection). Characterization of these tumors revealed that they are ER+ mammary tumors with extensive leukocyte infiltration. CD45+ leukocytes in them constituted ~32-84% of total live tumor cells, which is unusual for a mouse cancer model. Among CD45+ cells, ~45% were macrophages and ~15% were CD3+ T cells. Among tumor-infiltrating myeloid cells (CD45+CD11b+), many were myeloid-derived suppressor cells (MDSCs). In addition, significant differential expression of PD-1 on CD3+ T cells (38.5% vs. 7.3%) and PD-L1 on CD45- cells (2.7% vs. 0.3%) was observed in these Runx1/Trp53-null tumors, when compared to WT controls. This immune profile suggests an immunosuppressive microenvironment in these tumors. The tumor immune phenotype induced by RUNX1-loss in our model is supported by genomic data from human breast cancers analyzed by the TIMER online tool (Li et al., Cancer Res 77(21):e108-e110). The TIMER analysis showed that in human breast cancers, levels of most infiltrating immune cell subtypes (e.g., macrophages, T cells) in RUNX1-mutated cases are higher than those of RUNX1-WT cases; in contrast, a similar analysis for GATA3-mutated cases exhibited the opposite. As both RUNX1 and GATA3 are key TFs for luminal MECs and their mutations are both found in luminal breast cancers, the data from both human cases and our mouse model suggest that RUNX1-mutated breast cancers may represent a unique, immune “hot” subtype of ER+ luminal breast tumors. ER+ breast cancers are typically immune “cold” and do not respond to immune checkpoint blockade (ICB) well. Our data suggest that RUNX1-mutated breast cancers may represent a subset of ER+ breast cancers that may respond to ICB-based therapy. Collectively, our data suggest that loss of RUNX1 promotes development of ER+ breast cancer and modulates the immune microenvironment; a better understanding of how these two processes are interconnected may provide novel strategies to treat breast cancer with RUNX1-mutation. Citation Format: Dongxi Xiang, Zhe Li. RUNX1-deficiency in luminal mammary epithelial cells leads to development of a unique type of immune “hot” ER+ mammary tumor [abstract]. In: Proceedings of the AACR Special Conference on Tumor Immunology and Immunotherapy; 2019 Nov 17-20; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2020;8(3 Suppl):Abstract nr A88.