Physical Poster + E-Poster Presentation 34th Lorne Cancer Conference 2022

Identifying cell of origin in high grade serous ovarian cancer using gene expression data (#232)

Masih Sherafatian 1 2 3 , Luciano Martelotto 4 , Heather Thorne 2 , Matthew Wakefield 5 , Fernando Rossello 4 , Kylie Gorringe 2 3
  1. Sir Peter MacCallum Department of Oncology, The University of Melbourne, Melbourne, VIC, Australia
  2. Cancer Genomics Program, Peter MacCallum Cancer Centre, Melbourne, VIC, Australia
  3. Sir Peter MacCallum Department of Oncology, The University of Melbourne, Melbourne, VIC, Australia
  4. Department of Clinical Pathology and Centre for Cancer Research, University of Melbourne, Melbourne, VIC, Australia
  5. The Walter and Eliza Hall Institute of Medical Research, Melbourne, VIC, Australia

In order to identify the cell of origin in high-grade serous ovarian cancer (HGSOC) we generated scRNA-seq data from the normal tissues of the ovary and fallopian tube (FT) from a BRCA1 mutation carrier. We identified expression markers of different cell types, which were used as references to score publicly available bulk RNA-seq and scRNA-seq datasets of HGSOC, employing various scoring methods to find the most similar profiles. 

Novel rare epithelial cells expressing cancer stem cell markers were found in both FT and OSE. Transcription factor profiles and specific markers for these putative tumour-initiating cells (TICs) were uncovered. One of the PAX8 positive secretory cell types from the FT scored highly against HGSOC bulk RNA-seq samples from TCGA. However, putative TICs from the ovary were identified as the most similar to epithelial cells of an HGSOC scRNA-seq sample. Cancer associated fibroblast (CAF) expression profiles were similar to fibroblasts originating from either the FT or ovary. Surprisingly, some cells previously annotated as CAFs in HGSOC ascites had high correlation to ovarian putative TICs and expressed cancer stem cell markers.  

We discovered novel potential TICs from FT and OSE. Their gene expression profiles were shown to be similar to HGSOC epithelial cells, however there was heterogeneity between individuals, with some HGSOC scoring highly against FT cells, and others against OSE cells. The gene signature of the novel TICs could lead to early disease biomarkers for HGSOC. Moreover, our findings shed new light on CAFs and cancer stem cell gene signatures in the tumour ascites.