Epithelial ovarian cancer comprises a heterogeneous group of tumors that have distinct clinicopathological and molecular features as well as multiple underlying causative genetic mutations (1). High-grade serous ovarian cancer originates de novo from the fallopian tube fimbriae, while clear cell endometrioid tumors arise from endometriosis (1). Mucinous ovarian cancer accounts for approximately 10% of epithelial ovarian cancer, but its tissue origin remains controversial (2). Primary mucinous cancer frequently presents as a large (>10 cm) clinically unilateral tumor similar to benign cystadenoma and borderline tumors (3). Occasional presentation as <10-cm tumor or clinically bilateral tumor may be features that contribute to metastases from other sites including the appendix, colon, stomach, pancreas and biliary tract (3). At baseline, primary mucinous ovarian tumors progress from benign to borderline to invasive cancer in a stepwise manner, all of which generally have a good prognosis (3). Mucinous tumors are more frequently detected in early-stage disease with lower tumor grading compared with high-grade serous cancer; however, patients with advanced disease have poor clinical outcome, possibly due to resistance to taxane and platinum-based conventional chemotherapy (4). An evolutionarily conserved signaling cascade system, including growth factor pathways [epidermal growth factor receptor (EGFR), ERBB and fibroblast growth factor receptor (FGFR)] and Wingless (WNT) signaling pathways, regulates a variety of cellular functions, including chemoresistance (5). The crosstalk between EGFR/KRAS proto-oncogene/B-Raf proto-oncogene (BRAF)/mitogen-activated protein kinase (MAPK), phosphatidylinositol-3 kinase (PI3K)/Akt (also known as protein kinase B) and WNT signaling pathways sustains PI3K/glycogen synthase kinase-3β (GSK3β)/β-catenin signal activation, which is associated with chemoresistance in cancer (6). The WNT receptors and their downstream components are being investigated as potential targets in the development of novel anticancer therapies (5,6).

The present article aimed to summarize the underlying molecular mechanisms of chemoresistance in mucinous ovarian cancer, focusing on the WNT signaling pathway. Novel therapeutics that may target chemoresistant processes from bench to bedside were also discussed. In this regard, a systematic review of the literature using an electronic search of the PubMed database (http://www.ncbi.nlm.nih.gov/pubmed) was conducted. Relevant literature published between January 2000 and October 2017 was searched. The search strategy screened for full-text original research or reviews in peer-reviewed journals with at least one of the key words ‘mucinous ovarian cancer’, ‘chemoresistance’, ‘WNT/Wingless’, ‘EGFR/epidermal growth factor receptor’, ‘FGFR/fibroblast growth factor receptor’, ‘signaling pathway’, ‘inhibitor’ or ‘antagonist’ in their titles or abstracts. English-language publication search results from PubMed and references within the relevant articles were analyzed. To minimize selection bias, screening of the studies was independently performed by two reviewers following agreement on the selection criteria.

Previous studies have identified potential gene alterations implicated in the carcinogenesis and progression of mucinous ovarian cancer (2,7–10). Mucinous tumors are likely driven by constitutive signaling activation resulting from mutagenic processes (BRAF and KRAS mutations) and growth factor amplifications (EGFR and MYC proto-oncogene amplifications) (2,8–10). The BRAF and KRAS mutations frequently identified in mucinous ovarian cancer have also been observed in low-grade serous ovarian cancer and serous and mucinous borderline tumors (7). One such activating driver mutation is BRAF^V600E, a substitution of glutamic acid for valine in codon 600 in exon 15 (7). BRAF mutations have diagnostic and prognostic value in many tumors including not only mucinous ovarian cancer, but also melanoma (11), colorectal cancer (12), thyroid cancer (13), brain tumors and various other cancers (14). Furthermore, the mutation rate in KRAS for proven pathogenic mutations is 60–70% (7). EGFR triggers cell proliferation through the RAS/RAF/MAPK signaling pathway. Erb-b2 receptor tyrosine kinase 2 (ERBB2; also known as human epidermal growth factor receptor 2, HER2) amplification is relatively common (~20%) in mucinous ovarian cancer and borderline mucinous tumor (2,7–10). Concurrent aberrant ERBB2 and KRAS signaling has been observed in a marked number of cases (~11%), suggesting that acquired ERBB2 amplification is secondary to the emergence of KRAS activating mutation (10). Although oncogenic KRAS driver mutations and ERBB2 amplification are not mutually exclusive (15), KRAS mutations may be mutually exclusive with c-MYC amplification (9).

In addition, some cases of mucinous ovarian tumors may harbor clinically targetable tumor protein p53 (TP53) mutations; indeed TP53 is the most commonly altered gene in high grade-serous ovarian cancer (2,8). TP53 mutations have been observed to be more frequent in mucinous cancer when compared with borderline tumors (57 vs. 12%, respectively) (10). Mucinous ovarian cancer has been associated with homozygous loss of the cyclin dependent kinase inhibitor 2A (CDKN2A) locus (2,8). Infrequent cases of mucinous ovarian cancer also harbored additional mutations, including in ring finger protein 43 (RNF43), WNT and WNT family members, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha, phosphatase and tensin homolog, cadherin 1, E74 like ETS transcription factor 3 (ELF3), AT-rich interaction domain 1A, GNAS complex locus (GNAS), G protein subunit alpha 11 (GNA11), forkhead box L2, FGFR2, serine/threonine kinase 11 (STK11), β-catenin (also known as catenin beta-1, CTNNB1) and SMAD family member 4 (SMAD4) (2,7,10,16). Mutations of these genes are considered to serve roles in the tumorigenesis, progression, aggressive features and clinical outcome of a subset of mucinous ovarian tumors (16). These mutated genes have been identified in a variety of tumor types, including tumors of the gastrointestinal tract, pancreas and endometrium (16). For example, mutations in RNF43, an E3 ubiquitin-protein ligase, have been observed in pancreatic, colorectal and mucinous ovarian cancers (17). Gene mutation data indicated that mucinous-type tumors from different sites have marked similarities to mucinous ovarian cancer (2). Notably, the majority of these mutated genes, including RNF43, ELF3, GNAS, GNA11, STK11, CTNNB1 and SMAD4, may serve a crucial role in regulating WNT signaling (2,7,10). Mutations in key genes and aberrations in the WNT pathway are typically requisites for mucinous-type cancers and often result in increased nuclear β-catenin (2).

The genetic landscape of a variety of benign, borderline and malignant lesions has been gradually characterized. Genetic alterations of the RAS/RAF/MAPK, PI3K/Akt and WNT/β-catenin signaling pathway members have been reported to increase in a stepwise manner from mucinous borderline tumors to mucinous ovarian cancer (18). The crosstalk between the growth factor and WNT signaling pathways may sustain PI3K/GSK3β/β-catenin signal activation, which is associated with chemoresistance in cancer (18).

Overcoming intrinsic and acquired drug resistance is a challenge in the clinical treatment of patients with ovarian cancer (32). A previous review summarized numerous molecular aspects of chemoresistance, including oncogenes (EGFR/PI3K/Akt and WNT), ATP binding cassette (ABC) transporter pumps, EMT and cancer cell stemness (32). Ovarian cancer subtypes are distinct entities, having different responses to chemotherapy (23). Epithelial ovarian cancer is originally classified into two groups based on chemosensitivity. Serous and endometrioid ovarian cancers exhibit a hallmark of chemosensitivity, with higher response rates for taxane/platinum-based regimens. By contrast, mucinous and clear cell cancers are primarily resistant to chemotherapies (23). Patients with advanced-stage mucinous ovarian cancer exhibit higher rates of chemoresistance and have poorer survival outcomes compared with those with advanced-stage high-grade serous ovarian cancer (33).

Studies have been performed to identify the genes that contribute to chemoresistance in mucinous cancer (34,35). Previous genetic pathway enrichment analyses identified that upregulated transcripts in high-grade serous cancer were enriched for cell cycle signaling pathways, such as the TP53/breast cancer gene (BRCA) driver signaling pathway, while mucinous tumors were associated with upregulation of the WNT signaling pathway (2,35). Furthermore, a number of somatic mutations in proto-oncogenes, including in KRAS (36) and BRAF (2), have been identified in mucinous tumors. Overall, the available data indicate that crosstalk between the EGFR/PI3K/Akt and WNT signal pathways is implicated in the chemoresistance of mucinous ovarian cancer (32,36). Therefore, the WNT receptors and their downstream components may serve as novel therapeutic targets for treatment.

In particular, novel therapeutics that target chemoresistant processes may be useful from bench to bedside in mucinous ovarian cancer. Key components identified in the WNT signaling pathways are potential candidates of chemoresistance. Table I lists the WNT-related candidates and corresponding targeting agents for overcoming chemoresistance. Fig. 1 depicts the WNT pathway antagonists, inhibitors and modulators of particular interest in preclinical/clinical studies and their different mechanisms of action.

WNT ligand interacts with its cognate co-receptors LRP5/6 and Fzd. A number of approaches have been tested to target this interaction, including the use of natural compounds (37), small molecule inhibitors (38) and antibody-based inhibitors (8), with promising results (37). Natural compounds, including curcumin, 3,3-diindolylmethane, and phytoestrogen, have been tested for their capacity to reduce the activity of WNT signaling in cancer cells. Synthetic/small WNT inhibitors, including rofecoxib (cyclooxygenase-2 inhibitor), PRI-724 (β-catenin antagonist) and CWP232291 (synthetic/small WNT inhibitor), and the monoclonal antibody against Fzd receptors, vanituctumab, have been described (37). According to previous literature, several families of secreted antagonists consist of secreted frizzled-related proteins (sFRPs), the Dickkopf (Dkk) protein family, sclerostin (a soluble WNT antagonist), cerberus and WNT inhibitory factor-1 (39,40). However, the majority of these have not been incorporated into clinical practice for the treatment of patients with mucinous ovarian cancer. These WNT inhibitors have been previously reviewed in detail (35,37,41,42).

Activation of the EGFR or HER/PI3K/Akt cascade is considered to represent a major mechanism of chemoresistance in ovarian cancer (81). Therapeutic antibodies, including cetuximab (EGFR inhibitor), lapatinib (EGFR kinase inhibitor), rituximab (chimeric monoclonal anti-CD20 antibody) and trastuzumab [HER-family receptor tyrosine kinase (HER2) inhibitor] elicit an effective therapeutic response (82). Therefore, addition of EGFR/HER antibodies may improve the therapeutic effect of the conventional cytotoxic agents. Recent developments and the future potential in antibody-based targeting of the EGFR pathway have been previously reviewed in detail (83).

Chemotherapy with taxanes promotes activation of WNT signaling (47). The WNT inhibitors/antagonists, including vantictumab (anti-FZD) and ipafricept (FZD8-Fc), potentiate taxane-mediated cancer cell death. Furthermore, platinum treatment may induce activation of Src kinase (23). Combination therapy of oxaliplatin with dasatinib, a Src inhibitor, has previously exhibited antitumor effects in a mucinous cancer model (23). Additionally, the ABC inhibitors verapamil and elacridar re-sensitized chemoresistant cells to taxanes (84). Therefore, chemotherapeutics may be combined with tyrosine kinase inhibitors or P-glycoprotein inhibitors to enhance cytotoxicity (85).

Advanced mucinous ovarian cancer is established to be resistant to taxane- and platinum-based chemotherapy and is associated with poor clinical outcome, and there remains to be a requirement for effective anti-cancer therapies. The present article reviewed the underlying molecular mechanisms of chemoresistance, focusing on growth factor and WNT signaling pathways. Chemoresistance is mediated by the coordinated action of the WNT signaling axis, together with crosstalk from other growth factor receptor pathways, notably EGFR, ERBB2 and FGFR. Research has focused on molecular mechanisms for antagonizing WNT signaling by directly or indirectly targeting Wnt receptors on the cell surface and their downstream components (36). The present review described the characterization of synthetic small molecule inhibitors, small interfering peptides, antibiotics, organic molecules, proteases, protease inhibitors, and monoclonal antibodies that disrupt WNT signaling. Several WNT antagonists, inhibitors or modulators, including Fzd, LRP5/6, Dvl, CK1, AXIN, GSK3β and β-catenin (Table I and Fig. 1), are being evaluated in preclinical/clinical studies. WNT antagonists or blockades may have synergistic effects with platinum and taxane (47). From the published data it may be hypothesized that key components identified in the WNT signaling pathways are potential candidates of chemoresistance. Thus, the WNT inhibitors may provide novel therapeutic benefit in combination with current chemotherapies for mucinous ovarian cancer.

In conclusion, the present review has summarized promising WNT inhibitors for the targeting of chemoresistant processes in mucinous ovarian cancer from bench to bedside.