Single cell multiomic profiling of high-grade serous carcinoma.

Abstract

e17535 Background: High-grade serous ovarian carcinoma (HGSOC) is the most frequent and deadliest type of ovarian cancer (OC). With the recent advances in transcriptomic and epigenomic profiling of cancer at single cell resolution, it has become clear that intra-tumor heterogeneity driven by genetic and epigenetic factors may be the basis of drug resistance. Current treatments are not homogeneously effective against all the cancer cell subpopulations, thus enabling resistance. With single cell sequencing technology, characterization of cell signatures and regulatory mechanisms may translate to prognostic markers and novel drug targets. Methods: The transcriptomic and epigenomic landscape of HGSOC was mapped for five tumor samples from debulking surgeries and one tissue patient-derived xenograft (PDX) using the 10X genomics sequencing platform. This strategy permits the combination of single cell ATAC-seq and single cell RNA-seq within the same nucleus (Multiome). Isolation of nuclei, library preparation and sequencing were performed following 10X genomics protocols. Sample quality and library quality were assessed by nuclear morphology and fragment size analysis, respectively. Sequenced reads were mapped to the human genome and downstream analysis including copy number variant prediction, pathway enrichment, motif enrichment and gene regulatory network analysis were performed. The sampled patients continue to be followed for clinical outcomes, such as response to therapy. Clinical data including patient age, grade, and stage of cancer, debulking status, CA-125 levels, and neoadjuvant status, were used in conjunction with the genomics data to characterize patient specific molecular regulatory signatures. Results: Cells from the six samples (total N= 26,421) were projected via UMAP. Immune and stromal cells were shown to cluster by cell-type while cancer cells clustered by patient. Cancer cells were identified as cycling cells, ciliated cells, cells with enrichment of the JAK-STAT signaling pathway, and cells exhibiting cancer stemness signatures. Transcription factor motifs and binding-site enrichment in open chromatin regions reveal transcription regulation-based subpopulations. Using the combined gene expression and open chromatin information from these cells, there is potential to uncover the genetic regulatory network(s) that drives treatment resistance. Conclusions: With single cell technology, specific clusters of cancer and tumor microenvironment cells can be classified. Beyond characterizing patient specific signatures, multiome enables the discovery of genomic, transcriptomic and epigenomic signatures that provide insight in tumor progression and treatment resistance.