WNT signaling inducing activity in ascites predicts poor outcome in ovarian cancer

High grade serous carcinoma of the ovary, fallopian tube, and peritoneum (HGSC) is the deadliest gynecological disease which results in a five-year survival rate of 30% or less. HGSC is characterized by the early and rapid development of metastases accompanied by a high frequency of ascites i.e. the pathological accumulation of fluid in peritoneum. Ascites constitute a complex tumor microenvironment and contribute to disease progression by largely unknown mechanisms. Methods: Malignant ascites obtained from HGSC patients who had undergone cytoreductive surgery were tested for their ability to induce WNT signaling in the Kuramochi cell line, a novel and clinically relevant in vitro model of HGSC. Next, cancer spheroids (the main form of metastatic cancer cells in ascites) were evaluated with respect to WNT signaling. Kuramochi cells were used to determine the role of individual WNT signaling branches in the adoption of metastatic stem cell-like behavior by HGSC cells. Furthermore, we analyzed genomic and transcriptomic data on WNT/Planar Cell Polarity (PCP) components retrieved from public cancer databases and corroborated with primary patient samples and validated antibodies on the protein level. Results: We have shown that ascites are capable of inducing WNT signaling in primary HGSC cells and HGSC cell line, Kuramochi. Importantly, patients whose ascites cannot activate WNT pathway present with less aggressive disease and a considerably better outcome including overall survival (OS). Functionally, the activation of non-canonical WNT/PCP signaling by WNT5A (and not canonical WNT/beta-catenin signaling by WNT3A) promoted the metastatic stem-cell (metSC) like behavior (i.e. self-renewal, migration, and invasion) of HGSC cells. The pharmacological inhibition of casein kinase 1 (CK1) as well as genetic ablation (dishevelled 3 knock out) of the pathway blocked the WNT5A-induced effect. Additionally, WNT/PCP pathway components were differentially expressed between healthy and tumor tissue as well as between the primary tumor and metastases. Additionally, ascites which activated WNT/PCP signaling contained the typical WNT/PCP ligand WNT5A and interestingly, patients with high levels of WNT5A protein in their ascites exhibited poor progression-free survival (PFS) and OS in comparison to patients with low or undetectable ascitic WNT5A. Together, our results suggest the existence of a positive feedback loop between tumor cells producing WNT ligands and ascites that distribute WNT activity to cancer cells in the peritoneum, in order to promote their pro-metastatic features and drive HGSC progression. Conclusions: Our results highlight the role of WNT/PCP signaling in ovarian cancerogenesis, indicate a possible therapeutic potential of CK1 inhibitors for HGSC, and strongly suggest that the detection of WNT pathway inducing activity ascites (or WNT5A levels in ascites as a surrogate marker) could be a novel prognostic tool for HGSC patients.