Aberrant non-canonical WNT pathway as key-driver of high-grade serous ovarian cancer development

Epithelial ovarian cancer (EOC), the deadliest among gynecologic malignancies, is ranked as the fifth leading cause of cancer deaths in females [1]. Based on morphological findings, cellular origins, clinical characteristics, and several molecular genetic/epigenetic alterations, EOC has been subdivided into five main types: high-grade serous (HGSC, 70%), endometrioid (EC, 10%), clear cell (CC, 10%), mucinous (MC, 3%), and low-grade serous carcinomas (LGSC, < 5%) that account for over 95% of cases [2]. HGSCs generally harbor TP53 alterations, a pronounced genomic instability and, also, inherited and somatic BRCA1 and BRCA2 mutations. The other abovementioned cancer types are frequently characterized by mutations in KRAS, BRAF, PTEN, and CTNNB1 (Beta-catenin), and a relatively stable karyotype [2]. In this complex and heterogeneous scenario, the Wnt/beta-catenin signaling pathway, known to regulate stemness, cellular homeostasis, embryonic development, and physiological processes, in a broad spectrum of stem cell niches (also including the ovarian site), seems to play an important role in ovarian cancer [3]. WNT proteins are a large family of secreted glycoproteins activating at least three signaling pathways: the canonical pathway (WNT-beta-catenin), the non-canonical pathway (planar cell polarity), and WNT-Ca2+ pathway [4]. The first operates by stabilizing beta-catenin and translates a WNT signal into the transient transcription of a TCF/LEF target gene program; the second is beta-catenin independent and controls cell movement during morphogenesis [5]. Several WNT-antagonists, distinct in two types, are well known: (a) those ones binding to low-density lipoprotein receptor-related proteins (LRP-5 or LRP-6), including Sclerostin and Dickkopf (DKK) proteins, and (b) those ones interacting directly with WNT proteins, including WIF-1, Cerberus, and secreted Frizzled-related proteins (SFRPs) [6]. Different molecular events may determine the chronic activation of WNT target gene program in human cancers: (a) mutations in APC or Axin1 genes resulting in the production of truncated scaffold proteins being unable to bind beta-catenin [6]; (b) mutation of the conserved Ser/Thr phosphorylation sites at the N-terminus of beta-catenin [7, 8]; and (c) loss of WNT inhibitors through epigenetic silencing [6]. Although dysregulation of the WNT pathway via beta-catenin is a frequent event in several human cancers [7], its potential implications in ovarian cancer are still under investigation. In particular, 16–54% of endometrioid and mucinous histotypes are characterized by mutations of beta-catenin or, with a considerably less frequency, APC, AXIN1, and AXIN2 [3]. However, the altered expression of beta-catenin (overexpression or alterations in subcellular location: nuclear/cytoplasmatic vs membrane) has been observed also in other histotypes (HGSC), where mutations in Wnt-related genes are relatively uncommon. This suggests that the constitutive deregulation of Wnt signaling with the consequent over-activity of normal structured beta-catenin protein may contribute to cancer progression in ovarian HGSCs. Evidence accumulated in the years supports Wnt pathway key role in EOC development, by promoting CSC (cancer stem cell) self-renewal, EMT (epithelial-mesenchimal transition), metastasis, and tumor angiogenesis, and suppressing tumor immunity/immune escape [8]. Data are limited regarding its potential role in predicting chemoresponse or its prognostic relevance [8]. Molecularly, different mechanisms could be involved in Wnt pathway hyperactivation: overexpression of ligands and receptors, under-expression of inhibitors of the Wnt/beta-catenin pathway, and altered expression of proteins that regulate beta-catenin/E-cadherin or beta-catenin/TCF interactions. In addition, the role of many non-coding RNAs (lncRNAs, miRNAs, and circRNAs) in modulating beta-catenin signaling in EOC is emerging [9]. Nevertheless, many of these studies were only conducted in vitro or by using cell lines. Therefore, it is important to confirm these key findings in primary tumor cells collected from patients on FFPE specimens. In order to confirm the contribute of non-canonical pathway in driving ovarian HGSC development, the study of M. Chehover, R. Reich, and B. Davidson, published in this journal [10], has well investigated the expression and clinical role of > 20 Wnt pathway molecules (WNT1, WNT2, WNT3, WNT4, WNT5A, WNT6, WNT7, WNT11, FZD1, FZD4, FZD5, FZD6, FZD7, FZD8, FZD10, LRP5, LRP6, DKK, CCND, RUNX2, YAP) in metastatic HGSC surgical specimens and effusions, by different molecular methods (qPCR, IHC, and WB). The authors observed that Wnt pathway members, with the exception of WNT1 and WNT11, are hyper-activated in HGSC and that levels of multiple mRNAs of the Wnt pathway were inversely related to membrane expression of beta-catenin, suggesting that activation of this pathway is more pronounced in HGSC effusions, where there is reduced cell adhesion. In this way, the authors have highlighted the concept that the expression of Wnt pathway molecules is anatomic site-dependent and in HGSC effusions, it is also informative of chemoresponse and survival. The anatomic site-related variation of Wnt pathway molecules expression in HGSC provides further evidence of tumor heterogeneity and underlines the role of tumoral microenvironment in this cancer. This paper also considers the Wnt pathway potential role in chemoresistance. Significant association between expression of WNT4-7, RUNX2, LRP5-6, FZD6-7-10 and favorable response at diagnosis and/or absence of intrinsic resistance has been reported. As the authors declare, data with respect to OS were less uniform. FZD5 levels in pre-chemotherapy effusions and WNT2 levels in post-chemotherapy effusions were significantly associated with shorter OS, whereas the opposite was true for RUNX2, FZD1, and FZD4 in post-chemotherapy effusions. In conclusion, the study emphasizes the possibility of non-canonical Wnt pathway hyperactivation in HGSCs and supports the rationale of further investigating this signaling in larger series. Advanced-stage HGSC is almost an uniformly lethal pathology. This underscores the urgent need for new therapeutic strategies. Several studies performed on EOC cells have shown that WNT inhibitors strongly reduced tumor growth and metastasis [9, 11]. Up to now, only one clinical study on Wnt targeting drugs has been reported for EOC: in a phase 1b clinical trial, Ipafricept, a fusion protein that antagonizes Wnt signaling, was found to be well tolerated when used with standard chemotherapies [12]. Given the Wnt pathway hyperactivation and its strong tumor-promoting effects in EOC, further studies should be performed to elucidate this complex protein network driving cancer progression and therapy resistance as well as more clinical investigations should explore the strong therapeutical potentials of Wnt inhibitors, as a single agent or in combination with standard chemotherapies.