Abstract LB-301: Integrative genomic analysis identifies distinct mutational, epigenetic and immunological patterns among triple-negative breast cancer subtypes

Abstract

Triple-negative breast cancer is an inherently heterogeneous disease defined by the absence of estrogen receptor, progesterone receptor and human epithelial growth factor receptor amplifications. The lack of therapeutically targetable, high-frequency “driver” alterations has impeded development of targeted therapies for TNBC. TNBCs display transcriptional diversity with at least four tumor-intrinsic subtypes that include two basal-like (BL1, BL2), a mesenchymal (M) and a luminal androgen receptor (LAR) subtype. Each subtype displays unique biology and differentially responds to standard-of-care chemotherapy. Therefore, to better understand TNBC genomic alterations within molecular subtypes we have performed an integrative analysis of mutation, copy-number, gene expression, miRNA expression, reverse-phase protein array and methylation of 192 TNBC patients in The Cancer Genome Atlas and drug sensitivity of TNBC cell lines. Gene expression analysis confirmed our previous subtype-specific findings with enrichment of cell cycle, DNA replication response in BL1, Wnt and KRAS signaling BL2, androgen response in LAR and TGF-beta and mesenchymal differentiation in M. Despite having higher mutational burdens and similar copy number alterations, the BL1 and M subtypes display very different patterns of gene expression. These differences likely originate from epigenetic changes, as global methylation analysis demonstrated large hypomethylation across the genomes of M subtype. In contrast, the highly differentiated LAR subtype displayed global patterns of hypermethylation. Mutation analysis revealed enrichment of PIK3CA mutations in LAR tumors, NOTCH1, NOTCH2, NOTCH3 and FBXW7 mutations in M and BL1 tumors and activating MAPK pathway mutations (KRAS, NRAS, BRAF, MAP2K1, MAPK9 and MAP3K13) exclusive to BL2 tumors. Differential copy number analysis identified amplifications specific to the BL1 subtype (AKT3, KDR and KIT), BL2 subtype (TERT, EGFR and KRAS), M subtype (MYCN and WNT5B) and NOTCH pathway (BL1 and M). Reverse-phase protein array analysis confirmed activated MAPK signaling unique to the BL2 subtype and PI3K signaling in LAR. Cell line subtype correlations and published drug sensitivity screens verified LAR sensitivity to AR antagonists and PI3K/AKT/mTORC inhibitors, while BL2 cell models displayed sensitivity to MEK inhibitors. Subtype-specific differences in lymphocyte composition were inferred from gene expression with BL1 subtype displaying the highest CD8 levels. Despite having the greatest level of macrophages, the M subtype had the lowest ratio of M1/M2 polarized macrophages. These data suggest immune checkpoint therapies may be more effective in BL1 subtype and additional strategies targeting the immune suppressive macrophage population necessary for M tumors. Using a comprehensive multi-omics approach we defined the multi-platform genomic landscape of TNBC, identify potential therapeutic strategies and shed light on the evolutionary paths of TNBC.

Citation Format: Brian D. Lehmann, Antonio Colaprico, Jianjiao Chen, Lily Wang, Jennifer Pietenpol, Xi Chen. Integrative genomic analysis identifies distinct mutational, epigenetic and immunological patterns among triple-negative breast cancer subtypes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr LB-301.