The Molecular Landscape Influencing Prognoses of Epithelial Ovarian Cancer.

EOCs represent a group of heterogeneous diseases with different precursor lesions, molecular alterations, and hence should be taken into consideration when choosing adjuvant therapeutic agents based on the current concept of targeted therapy. By integrated genomic analyses including exome sequencing, copy number analysis, mRNA expression, miRNA expression, and DNA methylation analysis, the molecular alterations of HGSC identified by The Cancer Genome Atlas (TCGA) [ 10 ] showed high frequency of TP53 mutation (96%) and frequent alteration of homologous recombination related genes (51%) including BRCA1/2 germline and somatic mutations (22%). Four molecular subtypes (immunoreactive, proliferative, differentiated, mesenchymal) were identified by gene expression analysis, and were demonstraed to have distinct clinical outcomes. LGSCs more frequently harbor BRAF , KRAS , or NRAS mutations [ 11 ]. A recent comprehensive genomic analysis and so far the largest genetic study of LGSCs [ 12 ] using whole exome sequencing and genome-wide copy number analysis demonstrated mutations of these key genes in the RAS / RAF pathway in 47% of cases, and up to 57.7% of cases had alterations in genes linked to this pathway. This study also identified mutations in putative novel driver genes including USP9X , MACF1 , ARID1A , NF2 , DOT1L , and ASH1L . Using Reactome pathway analysis, the top recurrently affected pathways include RAS signaling (22%), FGFR signaling (15%), MAPK signaling (15%), ErbB4 signaling (13%), chromatin organization (10%), and ubiquitination (10%). Molecular profiling by targeted massively parallel sequencing or whole genome sequencing, ECs were found to have frequent CTNNB1 , PIK3CA , PTEN , ARID1A , KRAS mutations [ 13 ]. The molecular lanscape of ovarian EC is similar to that of the endometrial EC, although with mild varation in mutational frequencies. Akin to the molecular classification defined by TCGA for endometrial carcinoma [ 14 ], ovarian EC can also be classified into four molecular subtypes: 3% POLE (ultramutated), 19% MSI (microsatellite instability) (hypermutated), 17% copy-number high (serous-like) and 61% copy-number low (endometrioid). The signaling pathways frequently involved include WNT/β-catenin pathway, PI3K pathway, MAPK pathway, and SWI/SNF complex [ 13 ]. CCCs, similar to ECs with endometriosis as a possible precursor, may have mutations of ARID1A (around 50%), PIK3CA (around 50%), TERT (telomerase reverse transcriptase) promoter (5.7–16%), PTEN (2–13%), KRAS (4.7–20%), or amplifications of AKT2 (8–26%), MET (24–37%), ZNF217 (20–36%) as revealed by next generation sequencing of whole genome, whole exome, or targeted genes [ 15 ]. By whole genome sequencing, RNA sequencing, and copy number analysis, MCs, different from other types, most commonly have copy-number loss of CDKN2A (76%). Other frequently mutated genes include KRAS (64%) and TP53 (64%) [ 16 ]. The distinctive genetic and epigenetic dysfunctions in different histological types of EOC involve kinases and transcription factors-associated pathways which promote the tumorigenesis and progression of EOC. In this section, we describe the major molecular pathways identified in EOC. MAPK cascades are key signaling pathways that regulate many cellular processes, including proliferation, differentiation, apoptosis, and stress responses. The Ras/Raf/MAPK (MEK)/ERK pathway ( Figure 1 ) is the most important and thoroughly studied signaling cascade among all MAPK signaling pathways, and it plays a crucial role in the cell signal transduction network and development of tumor cells [ 17 , 18 ]. ERK/MAPK signaling can be stimulated by various factors, such as cytokines, viruses, and oncogenes, and is activated in the following ways: (i) Ca 2+ activation; (ii) receptor tyrosine kinase Ras activation; (iii) PKC-mediated activation; and (iv) G protein-coupled receptor activation [ 19 ]. Therefore, it is closely related to tumor formation. In EOC tumorigenesis, Hong et al. [ 20 ] found that expression levels of MAPK and ERK in EOC tissues were higher compared with those in adjacent normal tissues, while inhibition of the ERK/MAPK signaling pathway could restore tumor cells to a non-transformed state in vitro and in vivo, suggesting that increased activation of ERK/MAPK signaling may be closely associated with the progression of EOC. In addition, activation of ERK/MAPK signaling was also correlated with the characteristics of OC tumorigenesis, including (i) promoting cell proliferation via vascular endothelial growth factor (VEGF)-caused inhibition of apoptosis; (ii) increasing tumor invasion and metastasis by upregulating expression of matrix metalloproteinases (MMPs) that hydrolyse the extracellular matrix (ECM) in the tumor microenvironment (TME); and (iii) regulating expressions of transcription factors (such as hepatocyte growth factor (HGF), VEGF, and interleukin (IL)-8) that cause cytoskeletal deformation, and enhance tumor cell migration and tumor angiogenesis [ 21 , 22 ]. Hyperactivation of the PI3K/AKT/mTOR pathway occurs in nearly 70% of OCs and has been found to be the most frequently altered pathway in EOC [ 23 ] ( Figure 2 ). This pathway plays multiple oncogenic roles including cell proliferation, survival, transcription regulation, protein synthesis, autophage, and angiogenesis [ 23 ]. In EOC, this pathway can be activated through several mechanisms including activating mutations of PIK3CA in CCC (30–40%), EnCA (12–20%) [ 24 ], mutations or amplifications of PIK3CA in HGSC (18%) [ 10 ]. Other subunits altered include mutation of PI3K p85 ( PIK3R1 ) [ 25 ], mutations or amplifications of AKT1 , AKT2 [ 26 , 27 ], loss of phosphatase and tensin homolog (PTEN) through deletion, loss of function mutation, loss of heterozygosity, or epigenetic silencing [ 28 ], loss of inositol polyphosphate4-phosphatase type II ( INPP4B ) [ 29 ], and mutation of mTOR [ 30 , 31 ]. Among these subunits, PI3K p110α ( PIK3CA ) is the most frequently hyperactivated subunit within the PI3K pathway in EOC [ 23 ]. The increased activation of PI3K in EOC and its central role in several cancer-promoting pathways explain its implications in cancer progression including oncogenic transformation, cell proliferation, adhesion, and apoptosis, as well as multiple metabolic programming [ 23 , 32 ]. Moreover, the AKT/protein kinase B (PKB) family encompasses a group of serine threonine kinases, which act as oncogenes through mutation in the plekstrin homology (PH) domain or amplification, and they can be independently activated irrespective of the upstream PI3K signaling. In EOC, activation of AKT causes growth deregulation and strong resistance to apoptotic stimuli, conferring increased resistance to platinum, cisplatin, and paclitaxel [ 33 ], leading to uncontrolled tumor growth and cell invasion [ 34 ]. Additionally, mTOR is activated in about 50% of all HGSC [ 35 ], and it is known to be involved in cell growth, angiogenesis, evasion of cell death, and is thereby associated with poor prognoses [ 10 , 12 ]. An in silico analysis of PI3K/AKT/mTOR and NF-κB expressions, and correlation analyses of EOC TCGA samples using the Kaplan-Meier plot ( http://kmplot.com/analysis , Accessed on 11 August 2012) [ 36 ] and gene expression profiling interactive analysis (GEPIA) web tools ( http://gepia.cancer-pku.cn/index.html Accessed on 11 August 2012) [ 37 ] revealed that high expression of PI3K subunits are associated with poor survival. The complex crosstalk between the PI3K pathway and NF-κB results in decreased survival rates in EOC patients, enhanced aggressiveness, and chemoresistance, indicating that therapeutic targeting of PI3K/AKT/mTOR/NF-κB may be opportunities for clinical trials [ 38 , 39 , 40 ]. Activation of the STAT3/STAT5 pathway regulates a variety of cellular processes, such as tumor cell growth, survival, invasion, cancer stem cell-like characteristics, angiogenesis, drug-resistance, degradation of extracellular matrix, and epithelial-mesnechymal transition ( Figure 3 ), which have significant correlations with reduced survival of recurrent EOC [ 41 , 42 ]. Activation of STAT3 had been found in ovarian cancer cell lines [ 43 ] and ovarian tumor tissues especially high-grade carcinomas including HGSC and CCC [ 44 ]. Activaton of STAT3/STAT5 is mainly through tyrosine phosphorylation, by various stimuli such as cytokines, growth factors, hormones, and oxidative stress. Upon activatioin, dimerization and nuclear translocation result in transcription of target genes [ 41 , 42 ]. Alpinetin, a natural flavonoid, inhibits cell migration through downregulating matrix metallopeptidase (MMP)-2 and MMP-9 via suppression of STAT3 signaling in EOC [ 45 ]. It was also reported that constitutively activated STAT3 is involved in the epithelial-mesenchymal transition (EMT) of EOC by upregulating vimentin, N-cadherin, and IL-6 in STAT3-activated cells. Moreover, activation of STAT3 has strong correlations with elevation of Bcl-xL, cyclin D1, and c-myc, thereby promoting cell proliferation and survival [ 46 ]. Notably, STAT3 plays a vital role in regulating hypoxia-inducible factor (HIF)-1α which is a key modulator of angiogenesis [ 47 ], thus facilitating EOC angiogenesis. Furthermore, studies also demonstrated that STAT3 is correlated with expressions of stemness markers, including c-myc, Nanog, aldehyde dehydrogenase 1 family, member A1 (ALDH1A1), cluster of differentiation 24 (CD24), and β-catenin, which were found to be increased in OC spheroids [ 48 ] and correlated with chemoresistance [ 49 ]. Dysregulation of the Wnt/β-catenin pathway ( Figure 4 ), which leads to hyperactivation of β-catenin, was reported to promote cancer stem cell (CSC) self-renewal, metastasis, chemoresistance, angiogenesis, and immune suppression in all subtypes of EOC [ 50 ]. Specifically, mutation of β-catenin encoding gene CTNNB1 is frequently found in EC (up to 54%) [ 51 ]. Other rare mutations of the β-catenin destruction complex include mutations of AXIN in EC [ 51 ], and mutation of APC in MC [ 52 ]. Furthermore, higher activity of β-catenin through several mechanisms has been found in EOC, especially HGSC [ 50 ]. Accumulating evidence showed that modulation of β-catenin activity, such as microRNA (miR)-1207, suppression of suppression of secreted frizzled-related protein 1 (SFRP1) and axis inhibition protein 2 (AXIN2) (negative regulators of the Wnt/β-catenin pathway), activation of β-catenin signaling, and promotion of expressions of CSC markers promote stemness and chemoresistance in EOC [ 53 , 54 ]. Moreover, Wnt/β-catenin signaling is involved in the remodeling of the extracellular tumor matrix such as MMP-2, -7, -9, and E-cadherin, suggesting enhancement of the EMT/metastasis and angiogenesis in ovarian tumors [ 55 ]. Additionally, EOC was reported to evade the immune system through multiple mechanisms, including the recruitment of regulatory T cells (Treg) and promotion of T-cell apoptosis via programmed death ligand-1 (PD-L1) [ 56 ]. In addition, IL-10 and indolamine 2,3-dioxygenase (IDO) were reported to promote immune evasion by ovarian tumor-associated macrophages [ 57 ], thereby promoting EOC progression.

Early diagnosis of EOC remains an important unmet medical need. To date, two-stage detection strategies have been utilized using biomarker Cancer Antigen 125 (CA125, also known as MUC16) [ 58 ] and/or human epididymis protein 4 (HE4, also known as WFDC2) levels in blood, and subsequent transvaginal sonography (TVS) examination [ 59 ]. However, their sensitivity and specificity are not sufficient enough to detect EOC at the early stage which has a strong correlation with the patient’s survival rate. To improve the initial stage of clinical screening, a combination of CA125 and HE4 has been used to become multi-parametric assays such as the OVA1 test, risk of ovarian malignancy algorithm (ROMA) test, and OVA2 (or OVERA) which are the clinically approved blood-based biomarkers for EOC risk stratification. OVA1 assay was approved by the FDA in 2009 and evaluates combined data from imaging, menopausal status, and five serum protein biomarkers including second-generation CA125-II, apolipoprotein A1, transthyretin, transferrin, and β2-macroglobulin [ 60 ]. When OVA1 was used along with the physical assessment of the patient, an improvement of sensitivity at 96% and specificity at 35% were observed [ 61 ]. ROMA is a screening test which was approved by the FDA in 2011 for predicting the risk of EOC in patients with pelvic masses [ 62 ]. It is a combination of CA125, HE4 levels in blood and the menopausal status of the subject. ROMA produced a sensitivity of 100% and a specificity of 74.2% in premenopausal patients, while the postmenopausal had a sensitivity of 92.3% and a specificity of 76%, respectively [ 62 , 63 ]. OVERA, the second-generation test for OVA1, is a multi-parametric test combining CA125-II, HE4, apoliprotein A1, transferrin, and follicle-stimulating hormone for screening patients with pelvic masses. Compared to OVA1, OVERA test was reported to improve its sensitivity and specificity of 91% and 69%, respectively, therefore was approved by FDA in 2016 [ 64 ]. Obviously, no single biomarker can accurately detect EOC at an early stage, thereby new molecular techniques which allow identify potential clinical epigenetic signatures are important in detection of OC carcinogenesis. In Table 1 , we summarize the clinical utility of biomarkers and list their applications as well as their sensitivity/specificity for clinical prediction of EOC ( Table 1 ). Recent studies showed steady improvements in diagnostic accuracies using circulating cell-free (cf)DNA resulting from genetic mutations, copy number alerations, allelic imbalance, and promoter methylation in tumors [ 65 ]. Using subgroup analyses and meta-regression analyses of systematic reviews suggested that epigenetic markers (i.e., epigenetic DNA modifications, methylation, and alterations) are particularly more effective over other quantitative detection methods of cell-free (cf)DNA and RNA concentrations or chromosomal instability [ 65 , 66 ]. Many susceptibility genes in hereditary ovarian carcinomas are known to alter homologous recombination (HR)/DNA repair pathways [ 67 , 68 ], and recent genomic/epigenomic studies indicated that EOC represents a genetically heterogeneous and complex group of diseases [ 67 , 68 , 69 ]. The genetic heterogeneity may reflect exerted clonal selection in the progression (i.e., chemoresistance and metastatic capacity) of EOCs. This review discusses genomic advances in EOC, and focuses on translating these genomic (both somatic and germline) alterations (as useful biomarkers) into clinical practice. Approximately 15~20% of EOCs occur in a familial context with a high penetrant autosomal dominant genetic predisposition. To increase EOC patient survival, application of biomarkers for early diagnosis/detection and risk factor prediction, including genomic/epigenomic variants, copy number aberrations (CNAs), and DNA methylation in EOC surveillance are promising [ 66 , 70 ]. Moreover, due to rapid developments in DNA sequencing technology, various novel germline mutations have also been identified in familial EOC cases and in patients with early-onset EOC. These selected biomarkers and their clinical significance in EOCs are described in Table 2 .

In addition to using biomarkers as diagnostic/detection tools to improve surveillance in EOCs, biomarkers are also helpful as genetic predictors of treatment responses in EOC patients. Recent studies showed that several genetic predictors ( Table 2 ) of treatment outcomes in EOC patients were identified from new high-throughput genomic techniques, including genome-wide association studies (GWASs), whole-exome sequencing (WES), whole-genome sequencing (WGS), DNA CNAs, and epigenetic alterations. For example, in carboplatin- and paclitaxel-based chemotherapy in 1244 patients with serous OC, a GWAS analysis identified that minor alleles of two single-nucleotide polymorphisms (SNPs) (rs7874043 and rs72700653) were associated with poor PFS [ 110 ]. Furthermore, genetic variations in OCs are complex and differ among different histological types. Therefore, comprehensive genetic analysis and pathway annotation may provide better chance for the personalization of therapy, and patients with concurrent alterations in multiple pathways appear to havea high potential for combinations of targeted therapies [ 111 ]. In addition, WES and exome array-captured sequencing can be used to discover low-frequency variants in individuals with familial high-penetrant diseases and those with certain complex quantitative traits. Most patients with HGSC initially benefit from platinum- and taxol-based chemotherapy, but progressive chemoresistance occurs and results in tumor recurrence and eventual metastasis. Little is known about the mechanisms of chemoresistance. One study applied WES and SNP profiling of 31 paired EOC tissues before and after first-line platinum-based chemotherapy [ 112 ]. They found frequent homologous recombination (HR)-deficiencies [i.e., loss-of-heterozygosity (LOH) and mutations in HR genes] in primary tumors, relapsed tumors, and tumors resistant to second-line platinum, implicating persistent HR deficiencies and qualification for second-line poly(adenosine diphosphate [ADP]-ribose) polymerase (PARP) inhibitors (e.g., olaparib and rucaparib) treatment in recurrent/chemoresistant tumors. Increased CNAs of several genes, including MDS1 and EV11 complex locus(MECOM) , G1/S-specific cyclin-E1 ( CCNE1) [ 112 ], and Erb-B2 receptor tyrosine kinase 2 ( ERBB2) , were discovered in recurrent tumors. Additionally, genomic instability is a hallmark of cancer, and genomic aberrations in the form of DNA CNAs are important in EOC tumorigenesis [ 67 , 113 , 114 ], and specific CNAs are potential biomarkers for chemoresistance, tumor relapse, and survival in EOCs ( Table 3 ). The heterogeneous characteristic of EOC and its direct relationship with treatment failure are well documented [ 133 ]; thus, the broad mutational landscape of EOC, coupled with increasing availability of sequencing technologies, implies the application of precision-based biomarkers in EOC diagnostics and surveillance, including the use of circulating tumor DNA [ 134 ]. Therefore, comprehensive molecular profiling of EOC could lead to the development of new personalized or precision medicine in EOC. According to this scenario, current clinical oncology trials across a range of tumors using biomarker-driven designs were undertaken, including the National Cancer Institute’s Molecular Analysis for Therapy Choice (NCI-MATCH) trial [ 135 , 136 ], the American Society of Clinical Oncology (ASCO) Targeted Agent and Profiling Utilization Registry (TAPUR) study, and the European Organization for Research and Treatment of Cancer-Screening Patients for Efficient Clinical Trial Access (EORTC-SPECTA) program. Among these, the novel phase II NCI-MATCH trial was initiated in August 2015 to investigate agents (in addition to approved agents) of matching targeted therapy to molecular/genomic profiles. The trial is running under ClinicalTrials.gov identifier NCT02465060 , where updated information can be obtained. Furthermore, the TAPUR study is an ongoing, nonrandomized, multicenter clinical trial that opened in 2016 [ 137 ]. This trial is testing the use of drugs already approved by the US Food and Drug Administration (FDA) that target a specific tumor mutation in individuals with advanced cancer outside of the drug’s approved indication. Patients are treated according to their molecular profile regardless of the tissue origin or cancer type as shown in Table 4 .

Ovarian cancer (OC) is one of the leading causes of cancer deaths in women. According to the Global Cancer Observatory GLOBOCAN 2020 database, OC accounted for 313,959 new cases (1.6% of all cancers) worldwide in 2020 with around 207,252 deaths (2.1% of all cancers) [ 1 , 2 ]. Moreover, according to the International Agency for Research on Cancer (IARC), the estimated number of global OC incidents for 2040 will be 434,184 new cases and 293,039 deaths, with respective increased incidence and death rates of 27.7% and 29.3% [ 3 ]. The majority of OC cases are of epithelial origin, which are named epithelial OC (EOC). According to the 5th edition of World Helath Organization classificaton, there are at least five subtypes of EOCs based on histomorphology, immunoprofile, and molecular pathogenesis: high-grade serous carcioma (HGSC), low-grade serous carcioma (LGSC), endometrioid carcinoma (EC), clear cell carcinoma (CCC), and mucinous carcinoma (MC) [ 4 ]. Among these, HGSC is the most common type (70%) and accounts for most deaths of EOCs. Approximately 75% of all women with EOCs are diagnosed when the cancer has spread into the peritoneal surface, corresponding to International Federation of Gynecology and Obstetrics (FIGO) stages IIIc and IV. The 5-year survival rate for patients with FIGO stage IIIc-IV EOC is <30%, and most deaths occur within the first 2 years after a diagnosis [ 5 ]. With the addition of the antiangiogenic agent bevacizumab and the poly (ADP-ribose) polymerase (PARP) inhibitors, there is only a modest improvement in progression-free survival (PFS) [ 6 , 7 , 8 ], however, there has been no significant improvement in overall survical (OS). A recently published phase 3 trial using maintanance PARPi olaparib therapy in paients with platinum-sensitive relapsed ovarian cancer carrying BRCA1/2 mutation showed a median overall survival benefit of 12.9 months with a median follow-up perioid of about 65 months. However, a statistical significance was not accomplished which may be due to crossover and use of post-progresssion therapies [ 9 ] Thus, a comprehensive understanding of the mechanisms and molecular profiles is needed to improve clinical management and identify new therapeutic targets of EOC. In this review, we provide an updated overview of the different molecular profiles described for EOC and discuss their roles in current and future management of this malignancy.

EOC causes substantial morbidity and mortality in the developed world. A number of clinical features are known to affect PFS and OS rates in EOC, including the disease stage, tumor grade, chemoresistance, and recurrence. Current concepts in treating EOC are focused on new therapies (e.g., PARP inhibitors) as well as molecular testing. At present, only the BRCA status is routinely used clinically, with germline BRCA1 and BRCA2 genetic testing now in place at a number of centers as a biomarker for the use of PARP inhibitor therapy. The benefits of genetic testing for EOC patients and their family members can guide personalized treatment decisions, while also potentially preventing disease in others who carry inherited gene mutations. Therefore, genomic defects in homologous recombination DNA repair (HRR) pathway components remain an area of great interest, which is believed to benefit and improve clinical therapeutic outcomes. Although molecular markers/profiling for diagnosis, prognosis prediction, surveillance, and treatment are becoming more important as knowledge of the molecular mechanisms of EOC increases, molecular testing still faces several challenges, which needs to be addressed before broad implementation into clinical practice can be achieved. There are currently several ongoing clinical trials, which are investigating new targeted treatments for advanced EOCs. This will allow researchers and clinicians to work with useful diagnostic and therapeutic tools to combat advanced EOCs. The keys to success are molecular testing and personalized medicine, to obtain flexibility and fit the treatment for each patient with a unique EOC type.