Abstract P6-04-01: Single cell RNA sequencing reveals key expression signatures of primary breast cancer cells and immune infiltrates

Abstract

Abstract Introduction: Cancers display intratumoral heterogeneity which interferes with the precise analyses of the tumor entity, and which may affect therapeutic outcomes of targeted treatments. The aim of this study was to evaluate the feasibility of single cell RNA sequencing on primary breast cancer cells and to demonstrate the key gene expression signatures and transcriptome heterogeneity of breast cancer subtypes. Methods: We performed RNA sequencing on 246 individual cells from 4 primary breast tumors and 2 metastatic lymph nodes from 4 patients, using C1™Single-Cell Auto Prep System (Fluidigm, South San Francisco, CA). RNA sequencing reads were aligned to the human genome reference (hg19) using the 2-pass default mode of STAR_2.4.0d, and gene expression was quantified by RSEM v1.2.18 as the sum of isoform expression. Results: Pathologic characteristics of the 4 patients is summarized in Table 1. Table 1. Pathologic characteristics of patients BC01BC02BC03BC04pathologic stagepT1N0 (IA)pT2N0 (IIA)pT1N3 (IIIC)pT2N1 (IIB)Immunohistochemistry estrogen receptorpositivenegativenegativepositiveprogesterone receptornegativenegativenegativepositiveHER2 (FISH)2+/3 (negative)3+/3 (positive)1+/3 (negative)3+/3 (positive)No. of single cells tumor (lymph node)214947(50)28(51)HER2, human epidermal growth factor receptor 2; FISH, fluorescence in situ hybridization To distinguish tumor cells, we chose epithelial cell adhesion molecule (EpCAM) as an archetypal epithelial tumor marker and performed gene signature enrichment analysis for the EpCAM-positive cells. A total of 673 genes with enrichment score > |0.5| were selected and used as an "epithelial breast cancer" signature. Consensus clustering with the epithelial breast cancer signature separated 246 cells into 145 epithelial tumor and 101 non-tumor cell groups. Gene expression profiling in the tumor cells revealed many co-regulated genes in the estrogen receptor (ER)-positive tumor at a single cell resolution, which represent previously known ER-associated genes including MYC, BCL-2, and GATA3. Individual tumor cells from the two human epidermal growth factor receptor 2 (HER2)-amplified tumors demonstrated drastically different degree of HER2 signaling pathway activation, indicating the necessity of molecular subtyping for the identification of HER2-activated tumors. TNBC tumor cells showed an overall upregulation of activator protein 1 transcriptional pathway and strong activation of epithelial-mesenchymal transition signatures in a small sub-population. Immune cells comprised all the non-epithelial population, with mostly T lymphocytes in the primary tumor samples and B lymphocytes in the lymph nodes. The tumor-infiltrating T cells expressed an activated phenotype and many cytotoxic components. Altogether, the single cell RNA sequencing revealed the true identity of the tumor cells and the tumor-associated immune cells. Conclusion: Single cell RNA sequencing of breast cancer was used to reveal the true gene expression characteristics of the tumor cells and tumor-associated non-tumor compartments. The results showed key gene expression signatures of specific tumor subtypes and a wide range of transcriptome heterogeneity which is shaped by the tumor and microenvironments. Citation Format: Lee H-B, Eum HH, Chung W, Lee H-O, Lee K-M, Kim K-T, Moon H-G, Noh D-Y, Han W, Park W-Y. Single cell RNA sequencing reveals key expression signatures of primary breast cancer cells and immune infiltrates. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P6-04-01.