Abstract C057: An integrative analysis of pancreatic cancer cells and cancer-associated fibroblasts using single cell transcriptomic analysis

Abstract Background: Despite the advances in diagnosis and treatment of cancer patients over the past decades, the 5-year survival rate of pancreatic cancer (PC) patients still remains lower than 10%. This is because more than 80% of PC patients are diagnosed with unresectable or metastatic disease, and receive systemic chemotherapy as a first-line therapy, but the currently preferred regimens do not extend the patients’ survival by more than 12 months. To improve the clinical outcomes, a comprehensive understanding of molecular biology of pancreatic cancer cells, along with extensive stromal cells including cancer-associated fibroblasts (CAFs), is required. Several studies have performed single-cell RNA sequencing (scRNA-seq) to dissect the heterogeneous tumor microenvironment in PC tissues, and aimed to define the molecular subtypes in PC. However, due to the high inter-patient heterogeneity of tumor microenvironment and low cellularity of cancer cells, it has not been clearly elucidated yet. Methods: We performed scRNA-seq using enriched non-immune cell populations from 13 pancreatic ductal adenocarcinoma (PDAC) tumor tissues and 4 intraductal papillary mucinous neoplasm (IPMN) with high grade dysplasia or invasive carcinoma tissues to investigate the landscape the epithelial cells and fibroblasts precisely, and analyze each subpopulation in an integrated manner. Results: We effectively merged and clustered epithelial cells and fibroblasts using tandem batch correction method, and defined epithelial cells and fibroblasts as 9 different subpopulations, respectively. We further analyzed transcription factor activities and cell-to-cell interactions in each subpopulation and suggested a model for transcriptional evolution of pancreatic cancer cells at single-cell resolution. Interestingly, we also identified a new epithelial cell subtype, “VGLL1+ subtype”, which is a transitional type distinguished from both classical and basal-like subtypes. We demonstrated that VGLL1+ subtype represents a transitional cell population linking classical and basal subtypes and their marker genes are closely related to poor prognosis. Conclusion: This study provides an integrative analysis of epithelial cell and fibroblasts in PC tumor tissues, and suggests a novel model showing the dynamic states of pancreatic epithelial cell subpopulations. Citation Format: Galam Leem, Seongryong Kim, Ho Kyoung Hwang, Jung Hyun Jo, Hee Seung Lee, Moon Jae Chung, Jeong Youp Park, Seung Woo Park, Si Young Song, Eui-Cheol Shin, Chang Moo Kang, Jong-Eun Park, Seungmin Bang. An integrative analysis of pancreatic cancer cells and cancer-associated fibroblasts using single cell transcriptomic analysis [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer; 2022 Sep 13-16; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2022;82(22 Suppl):Abstract nr C057.