SINGLE-CELL RNA SEQUENCING ANALYSIS OF PANCREATIC DUCTAL ADENOCARCINOMA REVEALS INTERTUMORAL TRANSCRIPTOMIC HETEROGENEITY
Jon Harrison*1, Jonathan E. Chang2, Michelle Piquet2, Slavica Dimitrieva2, Millicent Gabriel2, Thomas P. Hank1, Keith Lillemoe1, Andrew L. Warshaw1, Mari Mino-Kenudson1, Carlos Fernandez-del Castillo1, Viviana Cremasco2, Dave Ruddy2, Andrew S. Liss1
1Surgery, Massachusetts General Hospital, Boston, MA; 2Novartis Institutes for Biomedical Research, Boston, MA
Background: The desmoplastic microenvironment surrounding pancreatic cancer appears to play a key role in this tumor’s aggressive biology by impeding treatment delivery and immune surveillance. Single-cell RNA sequencing (RNA-seq) analysis can help elucidate the roles different cell types play within in the tumor itself. Additionally, it can be employed to compare transcriptomic differences between untreated, neoadjuvant (NAT) treated, and metastatic specimens, which may yield important insights into malignant behavior or treatment effect.
Methods: Fresh untreated and FOLFIRINOX-based NAT treated PDAC tumors were collected after attempted or successful pancreatic resection. All samples were histologically verified as originating from the tumor mass before processing. These tissue specimens were then dissociated, prepared and loaded onto the 10X Genomics single-cell RNA-seq platform, and sequenced with Next-Generation sequencing technologies. Transcriptomic analysis was performed using the standard workflow provided in the R package Seurat.
Results: Fifteen PDAC tumor samples yielded adequate tissue for single-cell RNA sequencing analysis. These 15 cases included six untreated, six FOLFIRINOX-treated, and three untreated metastatic tumors. On average, 7500 cells from each sample were sequenced, and preliminary transcriptomic analysis was performed on 900 to 1700 genes per cell. Cancer cells were identified in all samples based on common transcriptomic features such as CEACAM5 expression. Additionally, endogenous pancreatic cell types such as stellate (RGS5) and endothelial (VWF) cells were seen across all specimens while islet cells (INS), non-malignant ductal cells (CFTR), and acinar cells (PRSS1) were observed only in a subset of tumors. In terms of the tumor microenvironment, cancer-associated fibroblasts (CAF) were identified in all samples based in part on COL1A1 and CXCL14 expression. Myofibroblasts were distinguished from the CAF population by a separate transcriptomic profile, which included ACTA2 expression. Regarding immune cells, canonical makers for B-cells, T-cells, NK cells, and myeloid-derived cells were also surveyed and identified in varying proportions across the specimens.
Conclusion: Single-cell RNA sequencing analysis of PDAC reveals intertumoral cell type heterogeneity, which can be further studied to identify distinguishing features of the untreated, FOLFIRINOX-treated, and metastatic PDAC tumor microenvironment and to potentially guide the development of more targeted therapies.
Back to 2020 Abstracts