Integrated molecular profiling of patient-derived ovarian cancer models identifies clinically relevant signatures and tumor vulnerabilities.

Epithelial ovarian cancer (EOC) is the most lethal gynecological tumor in developed countries and is characterized by high biological and molecular heterogeneity. High-grade serous ovarian carcinoma (HGSOC) is the most frequent and intractable form of the disease, mainly due to its rapid dissemination into the abdominal cavity, a process that is tightly linked to peritoneal ascites. Despite several studies provided insights into the genetic and epigenetic alterations relevant in EOC, the precise molecular alterations involved in tumor onset and progression remain largely unknown. Here we provide an experimental framework to perform a comprehensive investigation of molecular alterations relevant in HGSOC and ovarian cancer stem cells (OCSC) biology. We relied on a well characterized experimental set derived from human HGSOC-associated ascites and consisting of primary tumor cells, OCSC-enriched spheroids and serially propagated patient-derived xenografts, in order to define genetic and transcriptional signatures associated to specific HGSOC evolutionary trajectories. Such signatures exhibited prognostic value in a large cohort of HGSOC patients and allowed to define PI3K signaling as a novel vulnerability in OCSCs. Thus, our approach proved effective for the identification of druggable targets in HGSOC ascites, which is a major player in relapse and poor prognosis. mRNA sequencing was performed on 8 samples derived from a single human ascites. Samples analysed are human ascites (N=1), cultured ovarian cancer cells (N=3) and ascites from patient-derived xenograft (N=4).