Advances in Clinical and Experimental Medicine

Title abbreviation: Adv Clin Exp Med
JCR Impact Factor (IF) – 2.1
5-Year Impact Factor – 2.2
Scopus CiteScore – 3.4 (CiteScore Tracker 3.7)
Index Copernicus  – 161.11; MNiSW – 70 pts

ISSN 1899–5276 (print)
ISSN 2451-2680 (online)
Periodicity – monthly

Download original text (EN)

Advances in Clinical and Experimental Medicine

Ahead of print

doi: 10.17219/acem/159799

Publication type: review

Language: English

License: Creative Commons Attribution 3.0 Unported (CC BY 3.0)

Download citation:

  • BIBTEX (JabRef, Mendeley)
  • RIS (Papers, Reference Manager, RefWorks, Zotero)

Cite as:


Bukłaho PA, Kiśluk J, Wasilewska N, Nikliński J. Molecular features as promising biomarkers in ovarian cancer [published online as ahead of print on March 15, 2023]. Adv Clin Exp Med. 2023. doi:10.17219/acem/159799

Molecular features as promising biomarkers in ovarian cancer

Patrycja Aleksandra Bukłaho1,A,B,D, Joanna Kiśluk1,E,F, Natalia Wasilewska1,D, Jacek Nikliński1,E,F

1 Department of Clinical Molecular Biology, Medical University of Bialystok, Poland

Graphical abstract


Graphical abstracts

Abstract

Ovarian cancer (OC) is a global challenge for modern medicine, ranking 7th for incidence and the 8th most common cause of mortality from cancers in women. Ovarian cancer has a poor prognosis, characterized by high morbidity and mortality, with detection occurring more frequently in advanced stages. Further issues lie within the heterogeneous nature of this pathology, as well as in its ability to develop multidrug resistance. Therefore, there is a burgeoning need to introduce effective screening for the general population, especially in high-risk groups such as individuals with a family history of cancer. Achieving this would be greatly assisted by identifying new biomarkers in order to, in turn, develop targeted therapies for patients. Advances in molecular biology techniques that enable cancer genetic characterization offer hope for personalized medicine. This article reviews the current findings on the biology of OC at the molecular level. Such knowledge may prove to be crucial and constitute a starting point for the development of new options for the early detection, prevention and treatment of OC.

Key words: ovarian cancer, molecular markers, early detection, genetic testing, ovarian cancer screening

Introduction: Ovarian cancer

Ovarian cancer (OC) is a global issue. In 2018, 295,414 new cases of OC were reported, alongside 184,799 deaths. In Europe, the number of diagnosed women was 67,771, with a mortality rate of 44,576 (The World Ovarian Cancer Coalition Atlas 2020; World Health Organization (WHO) and GLOBOCAN data).1 A range of risk factors may contribute to OC, including increasing age, obesity, family history, and genetic mutations.

Most cases of OC occur in women aged 50–79 years.1, 2 This may be related to hormonal and reproductive cycle changes – for example, early menarche, estrogen replacement therapy used for over 5 years, late menopause, late pregnancy, endometriosis, and polycystic ovary syndrome.2, 3 A high-fat diet and obesity also contribute to an increased risk of OC.1 This may be in part due to the function of adipose tissue – a metabolically active organ that produces pro-inflammatory factors and regulates cell division.4 Obesity may also result in hormonal changes such as a higher level of endogenous androgens.5 It is postulated that obesity increases the risk of developing endometrial cancer of the ovary due to the increased level of estrogens involved in this cancer pathogenesis.6, 7

Family history is linked to the likelihood of ovarian cancer; women with 2 first-degree relatives who have had breast cancer (with 1 diagnosed before the age of 50) have a higher risk of OC. A higher risk of OC in patients may also be seen if they have: 1) 3 or more first- or second-degree relatives with breast cancer; 2) first- and second-degree relatives with a combination of breast cancer and OC; 3) 1 first-degree relative with bilateral breast cancer; 4) 2 or more first- or second-degree relatives with OC; 5) 1 first- or second-degree relative with a combination of breast cancer and OC; and/or 6) a family history of male breast cancer.2, 3, 4, 8

Specific genetic mutations are also related to an increased risk of developing OC. For example, women who carry the BRCA1 or BRCA2 mutation are at a higher risk of OC, and this prevalence may be seen especially in the Ashkenazi Jewish population, where 2% carry mutations in the BRCA1 (185delAG and 5382insC) and BRCA2 (6174delT) genes.2, 9 Mutations in the DNA repair genes hMSH1, hMSH2, hPMS1, and hPMS2 are seen in those with Lynch syndrome (related to hereditary colon cancer), which may also predispose individuals to OC.

Epithelial OC (EOC) is responsible for the overwhelming majority of OCs (85–90%).10 Epithelial OC is a heterogeneous disease, consisting of several histological subtypes with distinct biological features (Figure 1). With the advancement of molecular biology techniques, emerging evidence suggests that different subtypes have different genetic characteristics which will result in changes in classification and, moreover, advances in treatment.11 Currently, EOCs are classified into 5 main groups, according to the clinicopathological features of the disease: clear cell OC (CCOC), endometrioid OC (EOVC), low-grade serous OC (LGSOC), mucinous OC (MOC), and high-grade serous OC (HGSOC).12, 13 In clinical practice, EOCs are currently divided into 2 groups: HGSOC and non-serous high-grade OC (non-HGSOC) which contains CCOC, EOVC, LGSOC, and MOC.10

Genetic and molecular changes in HGSOC

High-grade serous OC is the most common type of OC, accounting for approx. 70% of cases, and in many instances diagnosed at more advanced stages (III–IV).14, 15 There are currently several drugs with molecular targets for the treatment of HGSOC (e.g., antivascular endothelial growth factor (VEGF) antibodies or poly (adenosine diphosphate (ADP)-ribose) polymerase inhibitors (PARPi)). Unfortunately, due to the lack of specific symptoms and late diagnosis, mortality is still high. New, more effective diagnostic molecular markers may be useful in combating this disease.16, 17

DNA sequence

High-grade serous OC is characterized by acquired or inherited mutations in various DNA repair pathways.18 Data from The Cancer Genome Atlas (TCGA) revealed that a mutation in the TP53 gene occurs in 96% of HGSOC tumors, both in primary and metastatic sites.19 The p53 transcription factor encoded by this gene activates further genes responsible for DNA repair, the cell cycle and apoptosis after irreversible DNA damage.18 The TP53 mutations appear to play a significant role in tumor initiation, with most presenting as missense mutations located predominantly in exons 4–8. However, nonsense mutations and frameshifts also occur.19, 20 The TP53 mutations can lead to loss of function (LOF), thus depriving the cell of anticancer protection. Subsequent mechanism is the dominant negative effect, which either masks the wild-type allele function by the mutant allele or gains the function (gain of function – GOF), allowing the cell to promote tumorigenesis. Godwin et al. and Saretzki et al. revealed that the loss of heterozygosity (LOH) at the TP53 locus is characteristic of OC. The TP53 mutant tumors feature poor differentiation, increased invasiveness and high metastatic potential, leading to an aggressive phenotype.acc.20

The BRCA1 and BRCA2 are referred to as tumor suppressor genes (TSGs),21 as they form multiprotein complexes which are involved in the regulation of transcription of DNA synthesis and recognition and correction of particular DNA damage double-strand breaks (DSBs). Moreover, these genes play an important role in controlling cell growth and maintaining genome integrity.6 The BRCA1/2 inactivation causes cancer cells to be devoid of DNA damage repair by homologous recombination (HR). Germline mutations in the BRCA1 and BRCA2 genes are responsible for the majority of hereditary ovarian tumors.15 However, many HGSOCs occur spontaneously, with altered BRCA functioning caused by somatic mutations in BRCA1/2, or as a result of methylation.22 Both germline and somatic BRCA1 mutations accompanied by a LOH suggest that the loss of functionality of this TSG plays a key role in the development of HGSOC.23

Based on TCGA data for HGSOC, mutations in NF1 and CDK12 are also frequently observed, as supported by research from Dugo et al., who found additional mutations in RB1, along with CSMD1, NOTCH4 and TMEM132D.24, 25 Other mutations in HGSOC present in the EMSY, RAD51, ATM, ATR, BARD1, BRIP1, PALB2, RB1, NF1, and CDKN2A genes, potentially resulting in a homologous recombination deficiency (HRD).18

Structural and copy number changes

A characteristic feature of HGSOC tumors is the loss of genome integrity, which leads to extreme genome instability.26 The HGSOCs show a relatively high number of somatic copy number changes (copy number variation (CNV)) and structural changes, with >100 repeated amplifications and deletions identified.27 Genomic instability is manifested by common structural and numerical aberrations in chromosomes 3, 8, 11, 17, and 21.28

One of the most common CNVs in OC is the amplification of the 19q12 locus, where CCNE1 resides. The CCNE1 encodes cyclin E1, which is amplified in many solid tumors, as well as in about 20% of cases with HGSOC.17 The overexpression of cyclin E1 increases the rate of passage of cells through the restriction point of the G1/S phase, leading to genomic instability, as an abnormal expression of cyclin E1 triggers unplanned DNA replication, centrosome amplification and chromosome instability.17, 29 The CCNE1 amplification and BRCA mutations are mutually exclusive in HGSOC.17 Copy number amplification of cyclin E1 has only been reported in wild-type BRCA1/2 tumors and is associated with early primary treatment failure and reduced survival in patients suffering from OC. Etemadmoghadam et al. showed that CCNE1-amplified ovarian tumors require the presence of a functional BRCA1 protein and may respond to bortezomib, a proteasome inhibitor.acc.30 Cases of CCNE1 amplification are distinct from those with the BRCA mutation, which suggests that there may be 2 different pathways driving the pathogenesis of HGSOC with heterogeneity among patients.31

According to TCGA data, the most common focal amplifications, apart from CCNE1, concerned MYC and MECOM, each with strong amplification in more than 20% of the analyzed cases.32 Zeng et al. focused on the amplification of the MYC oncogene at the 8q24 locus, and proved that MYC copy number is significantly correlated with the level of expression, and OC is characterized by the highest frequency of MYC amplification compared to many other cancers.33 Data suggest that MYC triggers the selective amplification of gene expression in order to promote cell growth and proliferation as the protein coordinates nutrient sourcing for the production of ATP and, as a result, DNA replication along with cell division occur and cells increase in mass.34 The MECOM is located at the 3q26.2 locus and is essential in early cell development and differentiation (including neurogenesis and craniofacial development), as well as in the regulation of the transforming growth factor β (TGF-β) signaling pathway (allowing for hematopoietic proliferation, differentiation and/or cell death).24, 35 The MECOM somatic mutations are characteristic of leukemias, while the germline variants in RECQL4 occur in osteosarcoma and skin cancer.

Ballabio et al. identified 2 focal and minimal common regions (FMCRs) of amplification in the cytoband 3q26.2 (193 kb α region) and 8q24.3 (495 kb β region). The MECOM gene, located in the α region, is associated with favorable prognosis and relapse-free time in OC.24 The researchers reported that the most frequent was the amplification of the cytoband 8q24 (48%), and the deletion concerned fragments: 5q, 6q, 8p, 13q, 16q, and 18q (from 40% to 48%), as well as 1p, 4p and 4q (approx. 30%).24 Comparatively, Dugo et al. showed that in more than 80% of the HGSOC patients tested, the amplification of the 3q26.2, 17q11.2 and 19p13.3 region, as well as the 4q34.3 deletion was present.25 Another locus, 6q24.2e26, was found to exhibit specific copy number loss, where a reduction in the expression level of 76% of the genes in the lost region was significantly associated with better survival. In a data-driven high-throughput study by Kamieniak et al., regional loss of 6q15eq27 was defined among the 8 copy number losses that significantly influenced the chemoreactive candidate pathways in EOC.36 Amplification of genes such as EMSY, FANC family, RAD51C, and PTEN is able to disrupt the HR pathway in HGSOC tumors.37

Gene expression

Changes to gene expression in HGSOC may reveal underlying mechanisms in which the pathology persists. Differences in gene expression for human HGSOC and normal ovarian surface epithelium (OSE) samples were assessed by Si et al., and the researchers identified 103 differentially expressed genes (DEGs). These included 28 upregulated genes that appear to stimulate cell division and proliferation, and 75 downregulated genes that assist in a variety of processes such as the metabolism of retinol, tyrosine and drug/cytochrome P450 pathways, as well as Wnt signaling. Many studies suggest that these metabolic disorders play an important role in carcinogenesis. Ten genes that may play an important role in the pathogenesis of HGSOC were identified, with 8 being upregulated (EPCAM, ZWINT, BUB1B, NEK2, DLGAP5, MELK, CEP55, CKS2) and 2 downregulated (ALDH1A1, KDR). These genes may be useful in the early diagnosis of patients with HGSOC. Additionally, survival analyses showed that the expression levels of MELK, CEP55 and KDR were significantly correlated with the overall survival (OS) rates of HGSOC patients. The overexpression of MELK was associated with the prolongation of OS, while increased levels of CEP55 and KDR expression in cancerous tissues were significantly associated with the shortening of OS. Moreover, these findings suggest that MELK, CEP55 and KDR are potential predictors of patient prognosis. Therefore, the genes listed above may be viable biomarkers to facilitate early diagnosis and treatment, and predict the prognosis of patients with HGSOC.38

Millstein et al. reported an association between OS and expression of 276 genes, and identified the 5 most significant genes (TAP1, CXCL9, FBN1, PTGER3, and ZFHX4). The TAP1 codes a protein involved in the antigen presentation pathway and had the highest prognostic value in this study. This gene exhibited a reduced expression in metastatic HGSOC and was positively associated with OS as well as tumor regression in response to treatment. The CXCL9, a chemokine that mediates the recruitment of T cells into solid tumors, is a strong prognostic factor in HGSOC and a characteristic feature of the immunoreactive molecular subtype of HGSOC. High expression of intratumor CXCL9 was associated with longer OS and more intensive lymphocyte infiltration. The FBN1 is an extracellular matrix protein, recognized as a biomarker associated with an early relapse in OC patients initially sensitive to chemotherapy and strongly correlated with desmoplasia in HGSOC. The expression of the prostaglandin E2 PTGER3 receptor in ovarian tumor cells is associated with relapse-free survival. The ZFHX4 was newly identified in that study as significant in HGSOC.39

Li et al. sought to underpin molecular features distinguishing HGSOC from serous borderline tumors (SBTs) and revealed significant differences in the expression of 11 genes. High-grade serous OC was characterized by: 1) lower expression of SLC7A2, PIFO, AFF2, HES2, BBS12, RPL12, RPL7A, RPS15, and RPS12; and 2) higher expression of MAFB and CRABP2. Currently, AFF2, MAFB, HES2, RPL12, RPL7A, RPS12, and RPS15 are believed to be involved in pre-transcriptional and post-transcriptional regulatory processes. The expression levels of the transcriptome between stage II and III of HGSOC were also compared. The researchers identified 17 DEGs: TSPAN1, ANK1, SERPINE2, SYNPO, PDGFC, NQO1, CA12, PPP1R14C, TMEM30B, PGK1, DNAJB9, CLIC1, ECE1, ENPP4, ZDHHC7, GCLC, and CHST15. Stage III was characterized by downregulation of all the genes mentioned. The PDGFC, SERPINE2, TSPAN1, ANK1, and SYNPO are involved in cell migration and cytoskeleton organization.40

Heterogeneity

The clonal composition of individual clinical samples and the coexistence of larger and smaller subclones were revealed through deep sequencing of epithelial tumors.41 Next-generation sequencing (NGS) reveals significant genomic heterogeneity of breast, pancreatic, kidney, and HGSOC tumors.17, 41

Changes in the BRCA and CCNE1 pathways represent 2 different genotypes displaying unrepeatable DNA repair susceptibilities.17 The use of PARPi in women who have BRCA1/2 mutations is an example of a shift towards precision medicine in HGSOC.42 Therapeutic approaches for tumors with CCNE1 amplification are being developed and will likely utilize their dependence on HR and replication fork protection routes.17 Four partially overlapping transcriptional subtypes of HGSOC have been identified through the profiling of mRNA expression in HGSOC tumors: 1) C1, mesenchymal; 2) C2, immunoreactive; 3) C4, differentiated; and 4) C5, proliferative. The C1 and C5 subtypes showed the least favorable OS, whereas the C2 subtype exhibited better survival outcomes. Single-cell technologies have allowed for the exploration of intratumor heterogeneity (ITH), and multi-omics profiling has assisted in better understanding of the molecular changes underlying HGSOC. Molecular heterogeneity within the HGSOC tumor is demonstrated both spatially and temporally, and in the case of extensive intraperitoneal dissemination, multi-region molecular profiling of primary tumors and metastases may be useful in identifying the biology underlying HGSOC progression.42

The complex relationship between cancer cells and the tumor microenvironment (TME) also contributes to the success of cancer treatment (Figure 2). In HGSOC, a relatively low burden of point somatic mutations, high levels of aneuploidy and large changes in copy number are associated with low immunogenicity. The T-cell infiltration (CD3+/CD8+) has been shown to play a key role in the prediction of HGSOC survival in primary disease. Research to determine the interaction between mutational ITH and T-cell interactions, as well as the potential impact of chemotherapy on T-cell infiltration in HGSOC is ongoing.43

Jiménez-Sánchez et al. found that the heterogeneity of the transcriptomic pathway is mainly due to the presence or absence of immune and stromal cells. The degree of variability in the patient’s immune signature was comparable to that of the metastatic HGSOC case study, in which different tumor immune microenvironments could be linked to certain clinical results. In this study, all patients had at least 1 tumor with low immune infiltration, which suggests that HGSOC is characterized by distinct microenvironmental niches that underpin both primary and acquired resistance to therapies. Integration of these data revealed, in the case of samples with excluded immune cells but high cellularity, that target genes are overexpressed in samples as well as mutations in the negative Wnt signaling regulators. As Wnt (and MYC) signaling exerts immunosuppressive effects, this could have the capacity to mask immune activation induced by cytotoxic chemotherapy treatment.43

Despite a high response rate to initial treatment, HGSOC is characterized by a high relapse rate likely due to its heterogeneous and adaptive nature described above. Data on significant heterogeneity in solid tumors may be crucial in the development of personalized medicine, which will contribute to increased quality of life and an extended lifespan of patients.41, 42 A more recent discovery of the chemotherapy-induced expansion of natural killer (NK) cells may also provide a translational pathway for novel treatment strategies connecting chemotherapy with immunotherapy.43

Molecular changes in acquired therapy resistance in HGSOC

Acquired resistance to chemotherapy is responsible for the majority of deaths in HGSOC.44 The inborn genetic instability of a neoplastic cell allows for fast adaptation to changes in the local molecular microenvironment. This provides the cancer with an opportunity to use many innate and acquired resistance mechanisms to overcome both chemotherapy and targeted therapies. For HGSOC, acquired resistance research has focused on mechanisms such as drug efflux and increased DNA damage repair ability.45

Currently, platinum-based chemotherapy is the primary form of pharmacological treatment in all cases of advanced EOC. Carboplatin in combination with paclitaxel is the standard treatment, regardless of the histotype.46 The PARPi are used in approx. 50% of patients with HGSOC; however, platinum resistance mechanisms often result in PARPi resistance. The BRCA1/2-deficient HGSOCs are platinum- and PARPi-responsive, yet chemoresistance may well develop through a restored function of the HR repair pathway. Most often, this is achieved through the acquisition of secondary somatic mutations in BRCA1/2 carriers of the germline mutations that reverse mutations to restore the open reading frame (and subsequent BRCA1/2 function), as well as decreased BRCA1 promoter methylation on relapse, resulting in increased BRCA1 expression. Reverse mutations of RAD51C, RAD51D and other genes involved in the HR pathway have been reported in women who progressed after the administration of PARPi. The resistance to paclitaxel, a one-component second-line treatment in women with platinum-resistant HGSOC, arises through alterations in the activity or expression of these apoptosis-controlling proteins. In HGSOC, increased expression of antiapoptotic proteins such as Bcl-2, Bcl-XL, Mcl-1, and survivin (BIRC5) is associated with shorter progression-free survival and resistance to taxanes and platinum in preclinical studies. In addition, the downregulation of pro-apoptotic factors such as Bax and caspases is associated with acquired chemoresistance. One of the best-studied mechanisms underlying multi-drug resistance is the hyperactivity of drug efflux through membrane-bound transporters where cancer cells can lower the intracellular concentration of drugs due to ATP binding cassette (ABC) transporters.44 In the case of OC chemoresistance, the ABCB protein subfamily, with the multi-drug resistance protein 1 (MDR1, other names: P-glycoprotein (P-gp) and ABCB1) has been most intensively studied.45 The overexpression of the ATP-dependent P-gp efflux pump causes resistance to taxanes (paclitaxel and docetaxel) by allowing the expulsion of these substrates from cells.44

Epigenetic and microRNA dysregulation in HGSOC

Epigenetic modifications are mechanisms that affect gene expression without altering the DNA sequence. These modifications alter the packaging of DNA on the histones, which in turn regulates the functioning of the genome. Errant epigenetic changes may lead to pathological over- or underexpression of genes, resulting in various diseases, including cancer. The propensity for alteration makes them an additional site for cancer cell chemoresistance. The most widely known histone modifications include: 1) methylation; 2) acetylation; and 3) phosphorylation. Modifications to cell function may also be driven by microRNAs (miRNAs) – molecules that participate in the regulation of processes such as the cell cycle, cell differentiation, proliferation, apoptosis, and metabolism. The dysregulation of miRNA levels occurs in many types of cancer.47 Unfortunately, there are sparse data describing the role of the noncoding genome in the development of EOC, suggesting a gap in the literature in the area of understanding and developing novel drugs based on targeting epigenetic modifications.48

The DNA methylation plays a key role in the epigenetic regulation of gene expression and takes place mainly on cytosine followed by guanine in CpG dinucleotides. The CpG islands are often located in the regulatory region of genes. Hypermethylation of cytosines located on the CpG islands in the genes promoter region results in decreased gene expression, unlike hypomethylation, which results in increased gene expression. Abnormal methylation patterns are common in cancer and can usually be characterized by global hypomethylation and hypermethylation of TSGs.48 Significant differences in methylation were found, both between the histological subtypes of OC and between the tumor tissue and the normal tissue.46 The CpG island hypermethylation in OC is frequently observed in TSGs such as BRCA1, p16, MLH1, RASSF1, and DARK, suggesting reduced expression and, in turn, inability to repress tumor growth.49

Although HGSOC exhibits some hypermethylation of genes, it appears more hypomethylated than EOVC or CCOC. For example, HNF1B is overexpressed in CCOC while it is methylated in about 50% of HGSOC cases. The HNF1 homeobox B (HNF1B) plays a key role in the epithelial–mesenchymal transition (EMT), in which cells lose cell adhesion and polarization, acquiring an invasive phenotype. The altered expression of HNF1B has been associated with an increased cancer risk, and decreased expression of HNF1B is involved in cancer development.50 A variant of HNF1B was identified as a HGSOC susceptibility locus, and Ross-Adams et al. reported that the susceptibility allele was associated with the methylation of the HNF1B promoter. Unmethylated HNF1B likely acts as an oncogene in CCOC, but, when hypermethylated, it acts as a tumor suppressor in the more aggressive histotype of HGSOC.acc.46 Sanchez-Vega et al. compared DNA methylation between HGSOC and EOVC and normal tubal tissue. Twelve CpG loci differed in methylation for both HGSOC and EOVC compared to normal tissue, and of these, 11 had reduced methylation in tumors in comparison with normal tissue.acc.46 Millstein et al. demonstrated that TAP1 hypomethylation was associated with a reduction in the time to relapse.39

Histone acetylation has been explored thoroughly in OC.48 As the vast majority of cases in TCGA are HGSOC, further studies are needed to ensure whether these findings can be applied to other histotypes.46 Chapman-Rothe et al. identified gene sets associated with the H3K27me3 (active) and H3K4me3 (repressive) tags at the transcription initiation sites in HGSOC, and investigated their association with epigenetic silencing and malignant progression.51 Researchers examined sets of histone-labeled genes in 8 benign ovarian lesions and 499 HGSOCs. A lower gene expression for H3K27me3 and bivalent gene sets have been demonstrated in neoplastic tissue.46, 51

The miRNAs are post-transcriptional regulators of gene expression which reduce the expression of their target mRNAs by inhibiting translation or promoting degradation. This allows for the regulation of key biological processes such as development, differentiation, apoptosis, and proliferation.28 The miRNAs have also been shown to be involved in cancer tumorigenesis and metastasis.52 Dong et al. revealed that the overexpression of miR-182 in HGSOC was shown to confer strong oncogenic properties by targeting BRCA1 and MTSS1. The overexpression of miR-145 was found in vitro to inhibit proliferation, migration and invasion of OC cells, and in vivo to inhibit tumor growth and metastasis. The miRNA-145 was also found to directly target metadherin, an oncogene strongly overexpressed in breast cancer and OC.52 The members of the miR-106 family are involved in stem cell self-renewal and are highly induced during the early stages of cell reprogramming. In many solid cancers, miR-106a is known to act in tumor-initiating cells and regulate tumor differentiation through the retinoblastoma (Rb) pathway, the cell cycle and FoxO signaling. Liu et al. found significantly increased expression of miR-106a and its family members in the early and late stages of HGSOC. The tumor suppressor gene RBL2, which is mostly downregulated in HGSOC, is a specific target for miR-106a. These findings suggest that miR-106a may specifically inhibit protein expression of a member of the Rb family, and the overexpression of miR-106a results in rapid tumor growth and poor differentiation.53

As previously mentioned, an initiating event of neoplastic metastases in epithelial tumors is EMT. This manifests through increased gene expression and protein levels that are preferentially present in mesenchymal cells. Sun et al. integrated mRNA and miRNA data and subsequently identified a mesenchymal subtype associated with poor survival in HGSOC patients. They revealed a network consisting of 8 major miRNAs and 214 mRNAs. Of these, miR-101, miR-200c, miR-141, and miR-506 are EMT regulators but the role of the remaining 4 miRNAs (miR-25, miR-29c, miR-182, and miR-128) is less clear, with potential involvement in cell migration, invasion and metastasis.47

Genetic and molecular changes in non-HGSOC

Tumors other than HGSOC are often only defined as unclassified, atypical, non-serous, or indistinguishable between EOVC and CCOC. Moreover, non-HGSOC subtypes often show a poor response to chemotherapy; thus, the search for representative molecular features is essential for the development of new targeted therapies.54

EOVC

Endometrioid OC accounts for 10–20% of all EOC cases. Compared to HGSOC, this cancer occurs in younger women (mean age of 56) and is often associated with endometriosis and synchronous endometrial carcinoma.55, 56 The EOVC is characterized by abnormal PI3K signaling and mutations in CTNNB1, which is the major effector of the Wnt pathway.57 Mutations in PTEN and microsatellite instability are also common. Both the histological and molecular profiles of EOVC appear to be more similar to endometrioid endometrial carcinoma (EEC) than to HGSOC.55 Endometrioid carcinomas of the ovary and endometrium have been reported to contain mutations in PTEN, PIK3CA, ARID1A, PPP2R1A, and CTNNB1 (β-catenin), but the frequency varies between these types of tumors. McConechy et al. showed that PTEN is more often mutated in EEC than in EOVC, but CTNNB1 is more frequently mutated in EOVC than in EEC.58

Pierson et al. sequenced the entire exome of 26 EOVC samples, and their findings suggest that PTEN, CTNNB1, PIK3CA, KMT2D, KMT2B, PIK3R1, ARID1A, and TP53 are significantly mutated in EOVC, which also occurred with a similar incidence in uterine corpus endometrial carcinoma (UCEC). Hypermutation due to mismatch repair deficiency (MMRD), as well as POLE mutation, were also observed in EOVC, again with a frequency comparable to UCEC. Common features of EOVC with HGSOC include 1) TP53 mutations, 2) mutations leading to HRD and 3) widespread CNV.55 The new EEC molecular classification suggests EOVC be grouped based on the following characteristics: POLE mutant, abnormal MMR, abnormal p53, and non-special type (NST, individuals without POLE mutation, without abnormal MMR or p53 protein expression). Such classification may have an important prognostic value and enable the selection of the most effective therapy.59

CCOC

Clear cell OC accounts for approx. 10% of EOC in Europe and is more common in younger women.54 It is considered a high-grade neoplasm.59 The most common mutations of CCOC are ARID1A, PIK3CA, TP53, and KRAS.54 Endometriosis is associated with 33–37% cases of CCOC.59 Clear cell tumors are characterized by a transcriptional profile that occurs in clear cell carcinoma of the ovary, endometrium and kidney.54 Clear cell EOC, like renal clear cell tumors (clear cell renal cell carcinoma (ccRCC)), often have inactivating mutations in the chromatin remodeling factor SWI/SNF.57, 60 High similarity of CCOC to ccRCC indicates that the mTOR pathway and angiogenesis may be therapeutic targets. One of the molecular features of CCOC is the overexpression of hypoxia-inducible factor 1α (HIF-1α) and 2α (HIF-2α) with activation of this pathway.57 One characteristic of CCOC is a lower incidence of TP53 mutations than in other histological types. In addition, according to in vivo studies, the inactivation of ARID1A is not efficient in tumor initiation, and further mutations (such as in PIK3CA) are necessary for tumor evolution.59

The amplification of primary copy numbers in CCOC has been revealed in several regions. In 40% of women with CCOC, the amplification of 17q21-24 was present and associated with a worsened prognosis. The PPM1D oncogene and other genes on chromosome 17 were amplified in 10% of this population, with microRNA-21 overexpressed in 14% (resulting in loss of phosphatase tensin homolog (PTEN)). The amplification has also been demonstrated in the 8p, 20q, 17q, and 3p chromosomes that encode genes for CCOC-related proteins such as MYC, ZNF217, HNF1b, and PIK3CA, respectively. The CCNE1 locus amplification occurred in 26% of patients with CCOC. The increase in the copy number resulted in increased protein expression and was associated with a worse prognosis.60

The MMRD occurs only in the histological types associated with endometriosis and is more common in EOVC (18%) than in CCOC (2%).61 Defects in the MMR DNA repair genes cause microsatellite instability. Recent high-case studies have shown that 2.8% of CCOCs are characterized by the loss of at least 1 protein in the MMR pathway.59 Rosen et al. observed a high level of microsatellite instability in the CCOC development, and a strong relationship between changes in the expression of hMLH1 and hMSH2 and microsatellite instability in this type of OC.56

MOC

Mucinous OC is an ovarian tumor of uncertain etiology that accounts for 3–5% of all epithelial OCs. As MOC is morphologically characterized by an epithelium with intestinal differentiation, it is difficult to determine whether the disease is primary OC or secondary metastatic mucinous adenocarcinoma.62

Primary mucinous tumors are classified as benign, borderline or malignant, depending on their histopathological features.56 Genetic analysis of primary MOCs confirms a progressive model of carcinogenesis in which benign cystadenoma develops a KRAS or CDKN2A mutation, resulting in progression to borderline tumors, with the probability of both events and additional CNV, and then further to overt cancer that exhibits a greater frequency of KRAS and TP53 mutations and greater CNV than borderline tumor.54 Mucinous ovarian tumors are characterized by a high frequency of KRAS mutations (46%). The KRAS mutations were found in histologically benign, borderline and malignant regions of the same tumor, suggesting this may be an early (or first) event in carcinogenesis of the ovarian mucosa.56 Comparatively, HER2 amplification or TP53 mutation occurs later in malignant transformation as it is only observed in cancers.63 The comparison of the incidence of lesions between mucinous borderline tumors (MBTs) and MOC showed that TP53 mutations occur in 10–18% of MBTs, and therefore contribute to the progression from MBT to MOC.59

Cheasley et al. indicated that TP53 mutations and copy number aberrations (CNAs), including the noteworthy 9p13 amplicon, are the key factors of cancer progression.62 A high burden of CNA is associated with the progression of MBT to MOC and a worse prognosis in MOC.59, 62 Data from ovarian mucosal tumor sequencing were also compared with records from TCGA and other exome sequencing databases, and MOC revealed different genetic features from high-grade serous tumors of the ovary, endometrium, stomach, and colon, including mucinous colorectal carcinomas and appendicitis. Pancreatic adenocarcinoma was genetically the most similar to the MOC, with common features including the inactivation of CDKN2A, mutation of KRAS and TP53. Contrasting features in primary MOCs were ERBB2 amplification and RNF43 mutations, whilst pancreatic tumors exhibited frequent changes in SMAD4.62

The TP53 mutation is often associated with HGSOC; however, ~25% of MOCs also possess this alteration. The HER2 amplification and overexpression can be observed in 18% of MOC, and high microsatellite instability (MSI-H) may also be seen. Mutations in the CTNNB1 or APC genes were detected, resulting in abnormal signaling in the Wtn pathway.63 Other molecular changes identified in MOCs include RNF43, BRAF, PIK3CA, and ARID1A mutations (8–12%), as well as ERBB2 amplification (26%). On the other hand, CNVs are key carcinogens associated with the increase in the staging and progression of metastases.54 It has also been shown that the copy number changes most strongly enriched in grade 3 MOCs are gains of 1p and 19p, which affect many oncogenes (e.g., JUN, JAK1, MYCL, BRD4).59

LGSOC

Low-grade serous OC accounts for approx. 3–5% of the EOC.54, 57 The LGSOC differs from HGSOC in: 1) age of onset (younger for LGSOC); 2) pathological features; 3) reduced aggressive features; and 4) longer OS.54 However, LGSOCs, unlike HGSOCs, are usually resistant to platinum-based chemotherapy.54, 57 The HGSOC expression profiling revealed an increased expression of genes involved in chromosomal instability and cell proliferation, whilst LGSOC had less overall karyotype instability and a lower mutation rate.57

The LGSOC often contains activating mutations in genes involved in the MAPK signaling pathway, including KRAS (20–35%), BRAF (10–40%), ERBB2 (5%), and NRAS (10%). Mutations in key genes in the MAPK pathway are mutually exclusive.54 The RAS encodes the guanosine triphosphate (GTP) binding protein that is frequently activated in low-grade serous, mucinous and endometrioid OCs. The RAS pathway can also be activated by eliminating regulatory proteins such as Dab2. The Dab2 is known to be highly expressed in normal human tissues, especially in ovarian surface epithelial cells, whereas most OCs were found to lack or inhibit the expression of Dab2 mRNA and protein. Therefore, loss of Dab2 expression may contribute to cell transformation or growth of neoplastic cells.56 The BRAF and KRAS mutations lead to constitutive activation of MAPK/Erk signaling and its downstream pathway, and in turn to increased survival and proliferation of neoplastic cells. These data suggest that the inhibition of MAPK hyperactivation may be a therapeutic target in women with LGSOC who respond very poorly to conventional platinum-based therapy.57

Hunter et al. compared SBT with LGSOC. The mutation frequency in KRAS, BRAF, HRAS, or ERBB2 was significantly higher in SBT than in LGSOC. In contrast, NRAS mutations were recorded in 26% of LGSOC, but not observed in SBT, which indicates a much greater oncogenic potential of this change. Similar results were obtained by Emmanuel et al., who identified a 9% frequency of NRAS mutations that were associated only with cancer, which suggests that it is an oncogenic factor in serous OC.acc.64 The findings of Hunter et al. confirmed that BRAF mutations are less common in LGSOC. Scientists also linked these changes to an earlier stage of cancer, better patient results and a lower likelihood of relapse in both SBT and LGSOC.65 The low frequency of the BRAF mutation in LGSOC challenges the notion that all SBTs and LGSOCs are on a continuum, suggesting that the mutation may be protective. The studies supporting this findings show more favorable results in patients whose tumors have BRAF mutations than in the case of KRAS or BRAF mutations and wild-type KRAS.64 The CNAs are associated with the progression of SBT to LGSOC, but are less common than in HGSOC. Hunter et al. reported that the most significant CNAs in LGSOC was the loss of 9p and homozygous deletions of the CDKN2A/2B locus. They were able to identify markers of progression from SBT to LGSOC, as well as novel LGSOC stimulants. The USP9X and EIF1AX are associated with mTOR regulation, suggesting that mTOR inhibitors may be a crucial adjunct therapy in trials of targeted therapy with MEK and RAF inhibitors.65 Mutations in the TP53 gene are rare in LGSOC, with an incidence of less than 8% in studies, although the absence of TP53 mutation is sometimes used as a criterion for inclusion in LGSOC.54 Finally, when considering CNV, the most common in LGSOC revealed by Van Nieuwenhuysen were loss of 1p, 6q, 9p, 16p/q, 17p, 18p/q, 22q, and gain in 1q, 7p/q and 8q. The most common focal lesion was the loss of 1p36.33 (54.1%). Many human neoplasms are characterized by an alteration in this locus.acc.59

The data discussed above suggest that LGSOC develops gradually from SBT. The formation of SBT appears to be closely related to abnormalities in KRAS signaling, which occurs very early in tumorigenesis. The acquisition of an invasive character, or evolution in LGSOC, appears to be associated with the acquisition of CNVs such as 9p21.3 hemizygous/homozygous deletion and hemizygous deletion of 1p36, which occurs in the final stages of carcinogenesis.66

Conclusions

Ovarian cancer is a challenge of modern medicine. Such tumors are characterized by considerable heterogeneity and their genetic profiles show significant differences (Table 1). The varied molecular characteristics of the individual cases result in differences in the typical stage of presentation (the stage in which the disease gives clinical manifestation), prognosis and success of targeted therapy. Given the heterogeneity between cancers, the individual and the TME, future directions of therapy are based on the concept of personalized medicine. To reduce the mortality rate in OC, there is also a need for an effective strategy to detect the disease at an early stage, which will likely utilize NGS technology.

Target populations for screening may be divided into 2 groups based on average lifetime risk, with general (1.32%) and high-risk (10%) populations. There is a need to modify and standardize the guidelines for screening, as the high-risk population is currently identified through family history and genetic testing only for BRCA1 and BRCA2. Additional SNPs should be included as these contribute to the polygenic risk of OC.67

The current standard of care for women at high risk of OC is risk-reducing salpingo-oophorectomy (RRSO). This procedure is recommended to be performed up to 40 years of age in carriers of the BRCA1 mutation and up to 45 years of age in the case of carriers of the BRCA2 mutation. In Europe and the UK, the RRSO is recommended for high-risk women without screening, while in the USA, the CA125 blood test and transvaginal ultrasound are performed before the surgery. The RRSO reduces the risk of OC and mortality; however, many factors may influence the choice of undergoing surgery. Oophorectomy may result in complications but also increase the risk of cardiovascular disease.68 In their review, Nebgen et al. emphasize the role of the detection of circulating tumor DNA (ctDNA) in the blood, tests based on DNA methylation analysis and miRNA as potential biomarkers in detecting cancer at an early stage, as well as in complementing and improving screening strategies offered to patients before RRSO.69

For patients diagnosed with EOC, germline genetic testing is recommended, and in addition, pathogenic variants (PVs) of genes other than BRCA1 and BRCA2 should be included. After obtaining a negative germline PV result, a tumor tissue test to assess somatic mutations for BRCA1 and BRCA2 (at the minimum) should be performed.70 It is reasonable to use multigene NGS in OC where there are somatic BRCA1/2 mutations associated with increased benefit for PARPi. As recommended by the European Society for Medical Oncology (ESMO), larger panels can only be used under specific contracts with payers who must consider the total cost of such a strategy. However, ESMO stresses the need for multi-gene sequencing by cancer clinical research centers and the development of new treatment strategies.71

To conclude, current research has revealed many changes at the molecular level in OC. However, further, large-scale investigations are required to assess the increased application in diagnostics. In turn, this would allow for both the extension of the gene panels offered to patients with EOC or suspected family predisposition, as well as for tailoring therapeutic intervention.

Tables


Table 1. Main genetics and molecular differences between ovarian carcinomas subtypes

HGSOC

EOVC

CCOC

MOC

LGSOC

mutations in: TP53, BRCA1/2, CDK12, NF1, RB1, CSMD1, NOTCH4, TMEM132D

mutations in: PTEN, PIK3CA, ARID1A, PPP2R1A, CTNNB1, KMT2D, KMT2B, PIK3R1, TP53

mutations in: ARID1A, PIK3CA, TP53, and KRAS; inactivating mutations in the chromatin remodeling factor SWI/SNF

mutations in: KRAS, CDKN2A, TP53

– activating mutations of genes involved in the MAPK signaling pathway, including KRAS, BRAF, ERBB2, and NRAS – mutually exclusive

 TP53 mutations are rare

widespread CNV

widespread CNV

CNV:

 17q21-24 amplification, chromosome 17  PPM1D oncogene

– amplification of 8p  MYC, 20q  ZNF217, 17q  HNF1b, and 3p  PIK3CA (genes for proteins related to CCOC)

 amplification at the CCNE1 locus

CNV:

 HER2 and ERBB2 amplification

 1p and 19p – impact on multiple oncogenes (JUN, JAK1, MYCL, BRD4)

CNV:

– loss of 1p, 6q, 9p, 16p/q, 17p, 18p/q, and 22q

 gain in 1q, 7p/q and 8q

 loss of 1p36.33 – the most common focal change

mutations leading to HRD

mutations leading to HRD

microRNA-21 overexpression lead to loss of PTEN

mutations in the CTNNB1 or APC gene lead to abnormal signaling in the Wtn pathway

less overall karyotype instability and lower mutation rate than HGSOC

very frequent structural and numerical aberrations in chromosomes 3, 8, 11, 17, and 21 lead to genomic instability

MMRD (EOVC – 25%) causes MSI-H

MMRD (CCOC – 2%) causes MSI-H

MSI-H

OC – ovarian cancer; HGSOC – high-grade serous OC; EOVC – endometrioid OC; CCOC – clear cell OC; LGSOC – low-grade serous OC; MOC – mucinous OC; SWI/SNF – switch/sucrose nonfermentable; CNV – copy number variation; HRD – homologous recombination deficiency; MMRD – mismatch repair deficiency; MSI-H – high microsatellite instability; PTEN – phosphatase and tensin homolog.

Figures


Fig. 1. Characteristics of histopathological types of ovarian cancer (Images courtesy by Department of Pathology, Johns Hopkins Hospital (Baltimore, USA))
OC – ovarian cancer; HGSOC – high-grade serous OC; EOVC – endometrioid OC; CCOC – clear cell OC; LGSOC – low-grade serous OC; MOC – mucinous OC.
Fig. 2. Simplified characterization of tumor microenvironment (TME)
IL – interleukin; CTL – cytotoxic T lymphocyte (also called CD8+ T cell); Th cell – T helper cell; Treg – regulatory T cell; M1 – macrophage type 1; M2 – macrophage type 2; TNF – tumor necrosis factor; IFN-γ – interferon gamma; TGF – transforming growth factor; GM-CSF – granulocyte-macrophage colony-stimulating factor.

References (71)

  1. Reid F. World Ovarian Cancer Coalition Atlas 2020. World Ovarian Cancer Coalition; 2020. https://worldovariancancercoalition.org/wp-content/uploads/2020/10/2020-World-Ovarian-Cancer-Atlas_FINAL.pdf. Accessed May 9, 2022.
  2. Roett MA, Evans P. Ovarian cancer: An overview. Am Fam Physician. 2009;80(6):609–616. PMID:19817326.
  3. Rooth C. Ovarian cancer: Risk factors, treatment and management. Br J Nurs. 2013;22(17):S23–S30. doi:10.12968/bjon.2013.22.Sup17.S23
  4. Kujawa K, Lisowska K. Ovarian can­cer: From biology to clinic [in Polish]. Postepy Hig Med Dosw (online). 2015;69:1275–1290. https://phmd.pl/api/files/view/116449.pdf. Accessed May 9, 2022.
  5. Fairfield K, Willett W, Rosner B, Manson J, Speizer F, Hankinson S. Obesity, weight gain, and ovarian cancer. Obstet Gynecol. 2002;100(2):288–296. doi:10.1016/S0029-7844(02)02053-7
  6. Varol U, Kucukzeybek Y, Alacacioglu A, et al. BRCA genes: BRCA1 and BRCA 2. J BUON. 2018;23(4):862–866. PMID:30358186.
  7. Foong KW, Bolton H. Obesity and ovarian cancer risk: A systematic review. Post Reprod Health. 2017;23(4):183–198. doi:10.1177/2053369117709225
  8. Mavaddat N, Peock S, Frost D, et al. Cancer risks for BRCA1 and BRCA2 mutation carriers: Results from prospective analysis of EMBRACE. J Natl Cancer Inst. 2013;105(11):812–822. doi:10.1093/jnci/djt095
  9. Moslehi R, Chu W, Karlan B, et al. BRCA1 and BRCA2 mutation analysis of 208 Ashkenazi Jewish women with ovarian cancer. Am J Hum Genet. 2000;66(4):1259–1272. doi:10.1086/302853
  10. Kim J, Park E, Kim O, et al. Cell origins of high-grade serous ovarian cancer. Cancers (Basel). 2018;10(11):433. doi:10.3390/cancers10110433
  11. Coetzee SG, Shen HC, Hazelett DJ, et al. Cell-type-specific enrichment of risk-associated regulatory elements at ovarian cancer susceptibility loci. Hum Mol Genet. 2015;24(13):3595–3607. doi:10.1093/hmg/ddv101
  12. Casey L, Singh N. Ovarian high-grade serous carcinoma: Assessing pathology for site of origin, staging and post-neoadjuvant chemotherapy changes. Surg Pathol Clin. 2019;12(2):515–528. doi:10.1016/j.path.2019.01.007
  13. Desai A, Xu J, Aysola K, et al. Epithelial ovarian cancer: An overview. World J Transl Med. 2014;3(1):1–8. doi:10.5528/wjtm.v3.i1.1
  14. Takaya H, Nakai H, Takamatsu S, Mandai M, Matsumura N. Homologous recombination deficiency status-based classification of high-grade serous ovarian carcinoma. Sci Rep. 2020;10(1):2757. doi:10.1038/s41598-020-59671-3
  15. Hollis RL, Meynert AM, Churchman M, et al. Enhanced response rate to pegylated liposomal doxorubicin in high grade serous ovarian carcinomas harbouring BRCA1 and BRCA2 aberrations. BMC Cancer. 2018;18(1):16. doi:10.1186/s12885-017-3981-2
  16. Wada M, Kukita A, Sone K, et al. Epigenetic modifier SETD8 as a therapeutic target for high-grade serous ovarian cancer. Biomolecules. 2020;10(12):1686. doi:10.3390/biom10121686
  17. Kroeger PT, Drapkin R. Pathogenesis and heterogeneity of ovarian cancer. Curr Opin Obstet Gynecol. 2017;29(1):26–34. doi:10.1097/GCO.0000000000000340
  18. Lheureux S, Braunstein M, Oza AM. Epithelial ovarian cancer: Evolution of management in the era of precision medicine. CA Cancer J Clin. 2019;69(4):280–304. doi:10.3322/caac.21559
  19. Lee S, Zhao L, Rojas C, et al. Molecular analysis of clinically defined subsets of high-grade serous ovarian cancer. Cell Rep. 2020;31(2):107502. doi:10.1016/j.celrep.2020.03.066
  20. Silwal-Pandit L, Langerød A, Børresen-Dale AL. TP53 mutations in breast and ovarian cancer. Cold Spring Harb Perspect Med. 2017;7(1):a026252. doi:10.1101/cshperspect.a026252
  21. Mehrgou A, Akouchekian M. The importance of BRCA1 and BRCA2 genes mutations in breast cancer development. Med J Islam Repub Iran. 2016;30:369. PMID:27493913.
  22. Madariaga A, Lheureux S, Oza A. Tailoring ovarian cancer treatment: Implications of BRCA1/2 mutations. Cancers (Basel). 2019;11(3):416. doi:10.3390/cancers11030416
  23. Berchuck A, Heron KA, Carney ME, et al. Frequency of germline and somatic BRCA1 mutations in ovarian cancer. Clin Cancer Res. 1998;4(10):2433–2437. PMID:9796975.
  24. Ballabio S, Craparotta I, Paracchini L, et al. Multisite analysis of high-grade serous epithelial ovarian cancers identifies genomic regions of focal and recurrent copy number alteration in 3q26.2 and 8q24.3. Int J Cancer. 2019;145(10):2670–2681. doi:10.1002/ijc.32288
  25. Dugo M, Devecchi A, De Cecco L, et al. Focal recurrent copy number alterations characterize disease relapse in high grade serous ovarian cancer patients with good clinical prognosis: A pilot study. Genes (Basel). 2019;10(9):678. doi:10.3390/genes10090678
  26. Wu JWY, Dand S, Doig L, et al. T-cell receptor therapy in the treatment of ovarian cancer: A mini review. Front Immunol. 2021;12:672502. doi:10.3389/fimmu.2021.672502
  27. Kossaï M, Leary A, Scoazec JY, Genestie C. Ovarian cancer: A heterogeneous disease. Pathobiology. 2018;85(1–2):41–49. doi:10.1159/000479006
  28. Marí-Alexandre J, Pellín-Carcelén A, Agababyan C, et al. Interplay between microRNAs and oxidative stress in ovarian conditions with a focus on ovarian cancer and endometriosis. Int J Mol Sci. 2019;20(21):5322. doi:10.3390/ijms20215322
  29. Gorski JW, Ueland FR, Kolesar JM. CCNE1 amplification as a predictive biomarker of chemotherapy resistance in epithelial ovarian cancer. Diagnostics (Basel). 2020;10(5):279. doi:10.3390/diagnostics10050279
  30. Stronach EA, Paul J, Timms KM, et al. Biomarker assessment of HR deficiency, tumor BRCA1/2 mutations, and CCNE1 copy number in ovarian cancer: Associations with clinical outcome following platinum monotherapy. Mol Cancer Res. 2018;16(7):1103–1111. doi:10.1158/1541-7786.MCR-18-0034
  31. Lisio MA, Fu L, Goyeneche A, Gao ZH, Telleria C. High-grade serous ovarian cancer: Basic sciences, clinical and therapeutic standpoints. Int J Mol Sci. 2019;20(4):952. doi:10.3390/ijms20040952
  32. The Cancer Genome Atlas Research Network. Integrated genomic analyses of ovarian carcinoma. Nature. 2011;474(7353):609–615. doi:10.1038/nature10166
  33. Zeng M, Kwiatkowski NP, Zhang T, et al. Targeting MYC dependency in ovarian cancer through inhibition of CDK7 and CDK12/13. Elife. 2018;7:e39030. doi:10.7554/eLife.39030
  34. Stine ZE, Walton ZE, Altman BJ, Hsieh AL, Dang CV. MYC, metabolism, and cancer. Cancer Discov. 2015;5(10):1024–1039. doi:10.1158/2159-8290.CD-15-0507
  35. Chen P, Zuo N, Wu C, et al. MECOM promotes supporting cell proliferation and differentiation in cochlea. J Otol. 2022;17(2):59–66. doi:10.1016/j.joto.2021.11.002
  36. Kamieniak MM, Rico D, Milne RL, et al. Deletion at 6q24.2-26 predicts longer survival of high-grade serous epithelial ovarian cancer patients. Mol Oncol. 2015;9(2):422–436. doi:10.1016/j.molonc.2014.09.010
  37. Andrews L, Mutch DG. Hereditary ovarian cancer and risk reduction. Best Pract Res Clin Obstet Gynaecol. 2017;41:31–48. doi:10.1016/j.bpobgyn.2016.10.017
  38. Si M, Zhang J, Cao J, et al. Integrated analysis to identify molecular biomarkers of high-grade serous ovarian cancer. Onco Targets Ther. 2019;12:10057–10075. doi:10.2147/OTT.S228678
  39. Millstein J, Budden T, Goode EL, et al. Prognostic gene expression signature for high-grade serous ovarian cancer. Ann Oncol. 2020;31(9):1240–1250. doi:10.1016/j.annonc.2020.05.019
  40. Li Y, Jaiswal SK, Kaur R, et al. Differential gene expression identifies a transcriptional regulatory network involving ER-alpha and PITX1 in invasive epithelial ovarian cancer. BMC Cancer. 2021;21(1):768. doi:10.1186/s12885-021-08276-8
  41. Schwarz RF, Ng CKY, Cooke SL, et al. Spatial and temporal heterogeneity in high-grade serous ovarian cancer: A phylogenetic analysis. PLoS Med. 2015;12(2):e1001789. doi:10.1371/journal.pmed.1001789
  42. Govindarajan M, Wohlmuth C, Waas M, Bernardini MQ, Kislinger T. High-throughput approaches for precision medicine in high-grade serous ovarian cancer. J Hematol Oncol. 2020;13(1):134. doi:10.1186/s13045-020-00971-6
  43. Jiménez-Sánchez A, Cybulska P, Mager KL, et al. Unraveling tumor–immune heterogeneity in advanced ovarian cancer uncovers immunogenic effect of chemotherapy. Nat Genet. 2020;52(6):582–593. doi:10.1038/s41588-020-0630-5
  44. Freimund AE, Beach JA, Christie EL, Bowtell DDL. Mechanisms of drug resistance in high-grade serous ovarian cancer. Hematol Oncol Clin North Am. 2018;32(6):983–996. doi:10.1016/j.hoc.2018.07.007
  45. Cornelison R, Llaneza D, Landen C. Emerging therapeutics to overcome chemoresistance in epithelial ovarian cancer: A mini-review. Int J Mol Sci. 2017;18(10):2171. doi:10.3390/ijms18102171
  46. Natanzon Y, Goode EL, Cunningham JM. Epigenetics in ovarian cancer. Semin Cancer Biol. 2018;51:160–169. doi:10.1016/j.semcancer.2017.08.003
  47. Sun Y, Guo F, Bagnoli M, et al. Key nodes of a microRNA network associated with the integrated mesenchymal subtype of high-grade serous ovarian cancer. Chin J Cancer. 2015;34(1):28–40. doi:10.5732/cjc.014.10284
  48. Matthews BG, Bowden NA, Wong-Brown MW. Epigenetic mechanisms and therapeutic targets in chemoresistant high-grade serous ovarian cancer. Cancers (Basel). 2021;13(23):5993. doi:10.3390/cancers13235993
  49. Ivan C, Hu W, Bottsford-Miller J, et al. Epigenetic analysis of the Notch superfamily in high-grade serous ovarian cancer. Gynecol Oncol. 2013;128(3):506–511. doi:10.1016/j.ygyno.2012.11.029
  50. Li J, Zhang Y, Gao Y, et al. Downregulation of HNF1 homeobox B is associated with drug resistance in ovarian cancer. Oncol Rep. 2014;32(3):979–988. doi:10.3892/or.2014.3297
  51. Chapman-Rothe N, Curry E, Zeller C, et al. Chromatin H3K27me3/H3K4me3 histone marks define gene sets in high-grade serous ovarian cancer that distinguish malignant, tumour-sustaining and chemo-resistant ovarian tumour cells. Oncogene. 2013;32(38):4586–4592. doi:10.1038/onc.2012.477
  52. Dong R, Liu X, Zhang Q, et al. miR-145 inhibits tumor growth and metastasis by targeting metadherin in high-grade serous ovarian carcinoma. Oncotarget. 2014;5(21):10816–10829. doi:10.18632/oncotarget.2522
  53. Liu Z, Gersbach E, Zhang X, et al. miR-106a represses the Rb tumor suppressor p130 to regulate cellular proliferation and differentiation in high-grade serous ovarian carcinoma. Mol Cancer Res. 2013;11(11):1314–1325. doi:10.1158/1541-7786.MCR-13-0131
  54. Barnes BM, Nelson L, Tighe A, et al. Distinct transcriptional programs stratify ovarian cancer cell lines into the five major histological subtypes. Genome Med. 2021;13(1):140. doi:10.1186/s13073-021-00952-5
  55. Pierson WE, Peters PN, Chang MT, et al. An integrated molecular profile of endometrioid ovarian cancer. Gynecol Oncol. 2020;157(1):55–61. doi:10.1016/j.ygyno.2020.02.011
  56. Rosen DG, Yang G, Liu G, Mercado-Uribe I, Chang B, Xiao XS. Ovarian cancer: Pathology, biology, and disease models. Front Biosci (Landmark Ed). 2009;14(6):2089–2102. doi:10.2741/3364
  57. Krzystyniak J, Ceppi L, Dizon DS, Birrer MJ. Epithelial ovarian cancer: The molecular genetics of epithelial ovarian cancer. Ann Oncol. 2016;27(Suppl 1):i4–i10. doi:10.1093/annonc/mdw083
  58. McConechy MK, Ding J, Senz J, et al. Ovarian and endometrial endometrioid carcinomas have distinct CTNNB1 and PTEN mutation profiles. Mod Pathol. 2014;27(1):128–134. doi:10.1038/modpathol.2013.107
  59. Romero I, Leskelä S, Mies BP, Velasco AP, Palacios J. Morphological and molecular heterogeneity of epithelial ovarian cancer: Therapeutic implications. EJC Suppl. 2020;15:1–15. doi:10.1016/j.ejcsup.2020.02.001
  60. Ji JX, Wang YK, Cochrane DR, Huntsman DG. Clear cell carcinomas of the ovary and kidney: Clarity through genomics. J Pathol. 2018;244(5):550–564. doi:10.1002/path.5037
  61. Leskela S, Romero I, Cristobal E, et al. Mismatch repair deficiency in ovarian carcinoma: Frequency, causes, and consequences. Am J Surg Pathol. 2020;44(5):649–656. doi:10.1097/PAS.0000000000001432
  62. Cheasley D, Wakefield MJ, Ryland GL, et al. The molecular origin and taxonomy of mucinous ovarian carcinoma. Nat Commun. 2019;10(1):3935. doi:10.1038/s41467-019-11862-x
  63. Babaier A, Ghatage P. Mucinous cancer of the ovary: Overview and current status. Diagnostics (Basel). 2020;10(1):52. doi:10.3390/diagnostics10010052
  64. Goulding EA, Simcock B, McLachlan J, Griend R, Sykes P. Low-grade serous ovarian carcinoma: A comprehensive literature review. Aust N Z J Obstet Gynaecol. 2020;60(1):27–33. doi:10.1111/ajo.13105
  65. Hunter SM, Anglesio MS, Ryland GL, et al. Molecular profiling of low grade serous ovarian tumours identifies novel candidate driver genes. Oncotarget. 2015;6(35):37663–37677. doi:10.18632/oncotarget.5438
  66. Della Pepa C, Tonini G, Santini D, et al. Low grade serous ovarian carcinoma: From the molecular characterization to the best therapeutic strategy. Cancer Treat Rev. 2015;41(2):136–143. doi:10.1016/j.ctrv.2014.12.003
  67. Nash Z, Menon U. Ovarian cancer screening: Current status and future directions. Best Pract Res Clin Obstet Gynaecol. 2020;65:32–45. doi:10.1016/j.bpobgyn.2020.02.010
  68. Wentzensen N, Berg CD. Population testing for high penetrance genes: Are we there yet? J Natl Cancer Inst. 2018;110(7):687–689. doi:10.1093/jnci/djx282
  69. Nebgen DR, Lu KH, Bast RC. Novel approaches to ovarian cancer screening. Curr Oncol Rep. 2019;21(8):75. doi:10.1007/s11912-019-0816-0
  70. Fostira F, Papadimitriou M, Papadimitriou C. Current practices on genetic testing in ovarian cancer. Ann Transl Med. 2020;8(24):1703. doi:10.21037/atm-20-1422
  71. Mosele F, Remon J, Mateo J, et al. Recommendations for the use of next-generation sequencing (NGS) for patients with metastatic cancers: A report from the ESMO Precision Medicine Working Group. Ann Oncol. 2020;31(11):1491–1505. doi:10.1016/j.annonc.2020.07.014