Next-generation sequencing-based genomic profiling analysis reveals novel mutations for clinical diagnosis in Chinese primary epithelial ovarian cancer patients

Ovarian cancer (OC) is one of the most malignant gynecological tumors worldwide [1]. Although abdominal pain, abdominal distension, gastrointestinal discomfort, and irregular menstruation appear in the initial stage of OC [2, 3], these signs and symptoms are relatively non-specific and difficult to distinguish from other diseases. In addition, the deep anatomy location of the ovary in the pelvic cavity and the absence of effective screening tools prevent the detection of this disease at early - stage. Indeed, among newly diagnosed OC cases, 70% of patients are found at stage III and IV OC (Federation of Gynecology and Obstetrics (FIGO), III-IV period) [1, 4, 5], directly leading to the failure in timely treatment. Therefore, OC is considered as a “silent killer.” Despite the inhibition of tumor growth can be achieved using initial aggressive treatment, both high recurrence rate (70%) and extremely poor overall survival (OS) rate are still observed in patients [6], and the average 5-year OS rate for all stages is 45.6% [1, 4‐9]. According to reports from the National Cancer Institute (NCI), about 140,000 people die each year from OC worldwide [10]. For the past 10 years in China, the incidence and mortality of OC have increased by 30 and 18%, respectively evidenced by ~ 15,000 deaths yearly [11].

Epithelial ovarian cancer (EOC) is the predominant type (about 90%) of OC, and categorized into four types according to histological features: serous, mucinous, endometrioid, and clear cell [12]. Recently, EOC has also been considered as two broad, simplified groups: type I including low-grade serous, endometrioid, clear cell, mucinous, and transitional cell carcinomas, and type II consisting of high-grade serous carcinomas, undifferentiated carcinomas, and carcinosarcomas [12, 13]. Particularly, EOC contributes to 50–70% of primary ovarian tumors and 85–90% of ovarian malignant tumors in China [11]. Due to the genetic heterogeneity of EOC with different pathological characteristics and molecular genotypes, there is a desperate and urgent demand for figuring out the molecular pathogenesis of EOC and identifying novel therapeutic targets and biomarkers, allowing “precision medicine” to be achieved in clinical practice.

Recently, microarray technologies have been used to elucidate the complexity of genomic alterations of OC and identify biomarkers and potential therapeutic targets for developing comparative medicine model [14‐18]. A 126-gene expression profile analysis has been performed on primary tumor tissues and successfully employed as genetic signature to predict OS in high-grade serous OC [19]. Accordingly, a recent study also demonstrated the correlation between BRCAness profile and the survival of EOC patients [20]. Somatic mutations of ARID1A (the AT rich interactive domain 1A (SWI-like)) [21] and PIK3CA (phosphatidylinositol-4, 5-bisphosphate 3-kinase, catalytic subunit alpha) genes [22] have been frequently detected in EOC, which was further supported by the emergence of 45 somatic mutations in 34 genes, including PIK3CA and ARID1A mutations in the most independent Ovarian clear cell carcinoma (OCCCs) [18]. It has been reported that plasma metabolites can be utilized to predict OS and show difference between short-term mortality and long-term survival in EOC patients [16].In addition to tumor biopsy samples, circulating cell-free DNA (cf-DNA) and circulating tumor cells (CTCs) [23] have emerged as “liquid biopsies” for non-invasive biomarkers in both the early and advanced diagnosis, prognosis, and in the identification of resistance mutations in OC.

In the past 5 years, the next-generation sequencing (NGS) technology has become widely available to determine a patient’s precise genetic profiling and identify novel mutations for new drug targets and individualized treatment schemes.

This study identified novel differentially expressed gene (DEG) mutations through comparing the gene expression profiles between EOC and normal healthy tissues from 31 EOC patients in Yunnan Province of China along with the analysis of the Gene Ontology (GO) functions and pathways of the candidate genes involved in EOC progression. The results of the present study provide the better understanding of the molecular mechanisms of EOC as well as potential biomarkers for prognosis and effective therapeutic targets.

The present study is the preliminary research of the complexity of genomic alterations in EOC based on NGS. Particularly, the excessive mutation frequency was identified in TP53 gene, which is consistent with previous reports [16, 18]. Ross et al. (2013) identified TP53 mutations in 79% of OC patients [16], which were more common in papillary serous carcinomas (83%) than in non-papillary serous tumors (50%). Maru et al. (2016) reported multiple TP53 mutations in at least two independent OCCC samples [18]. In agreement with the results of Ross et al. (2013), in which 8 BRCA1 mutations were reported in 44 EOC samples [16], we also identified at least two BRCA1 mutations in the present study. BRCA1 or TP53-gene mutation are predisposed for the increased susceptibility to a variety of cancers, including EOC, and both tumor suppressor genes have been implicated in DNA damage response pathways. Although function in separate pathways to suppress tumorigenesis, BRCA1 can serve as a coactivator to physically interact with TP53 [24, 25]. TP53-encoding gene is positioned the short arm of chromosome 17 [26] and responsible for the expression of the tumor suppressor protein p53 [27] in multicellular organisms. Especially, p53 plays a critical role in determining the repair of the damaged or mutated DNA. The NGS results in the present study confirmed a high degree of TP53 mutations, which is in agreement with previous NGS and further supported by the DNA sequencing, and DNA copy number assessments, as well as mRNA, microRNA and DNA promoter methylation status assays [16, 18, 28, 29]. The tumor suppressor gene, BRCA1, is positioned at the long arm of chromosome 17 [30] and responsible for the expression of breast cancer type 1 susceptibility protein, a functional component of BRCA1-associated genome surveillance complex (BASC) [31]. BASC plays a role in gene transcription, repair of DNA double-strand breaks, protein ubiquitination, and post-transcriptional regulation. Somatic BRCA1 mutations have been identified as a significant feature of high grade serous ovarian carcinoma [32]. Loss of BRCA1 activity, either by germ-line mutations or by down-regulation of gene expression, makes DNA damaging susceptible to therapeutic chemical agents, including platinum, or possibly agents that inhibit DNA repair pathways (i.e. poly ADP ribose polymerase inhibitors), increasing the risk of developing OC by 55% [16, 32]. The beyond molecular mechanisms, through which alternations in TP53 and BRCA1 sensitize cancer cells to cytotoxic or targeted- therapies, remain unclear. However, more detailed understanding of aberrant gene expression could eventually produce beneficial effects on cancer therapy.

The additional mutated genes - TTN, MUC16, OR4N2, CAD, CCDC129, INSR, NAV3, NELL2, NRAS, OBSCN, PGLYRP4, RBM15B and TRPC7 - have been previously reported, which revealed the highly association with cancer development [33]. Among these oncogenic genes, MUC16 encodes a repeating peptide epitope of mucin [34] that promotes cancer cell proliferation and inhibits anti-cancer immune responses [33‐35]. MUC16 is well-known as a clinically reliable diagnostic and therapeutic marker of EOC and is served as differentially diagnose pelvic masse [35]. MUC16 participates in the several signaling pathways, including defective C1GALT1C1, which causes Tn polyagglutination syndrome (TNPS) and CLEC7A (Dectin-1) signaling. It noteworthy that somatic mutations of other genes have not been reported in EOCs cases. TTN and CAD were identified as mutated oncogenes; TTN ranks at the top of 518 potential protein kinase cancer genes [33] and it encodes the largest polypeptide that [36] is expressed in many functional cell types in oncogenesis [37]. INSR encodes a protein that acts as an extracellular pH sensor in the regulation of acid-base balance in humans. INSR expression shows a correlation with degree of apoptosis and dedifferentiation in human neuroblastomas, and is co-expressed with insulin-like growth factor 1 receptor [38]. CAD encodes a trifunctional protein, which is regulated by the MAPK cascade [39]. NAV3 and OBSCN-encoding genes show a highly mutant frequency in human tumors. NAV3 encodes the protein, which contain both-coiled-coil domains and a conserved AAA domain, a featured ATPases functional motif regulating a variety of cellular activities including epithelial migration and invasion [40]. OBSCN has been implicated in cancer biogenesis and development through regulating cell survival [41]. NRAS is a member of the Ras family oncogenes and encodes the GTPase proteins that activate more than 20 signaling pathways, and these signaling cascades are involved in the regulation of essential cellular functions, such as proliferation, survival, and migration, cell division, cell differentiation, and the self-destruction of cells (apoptosis). Upon mutation, oncogenes can convert normal cells into cancerous cells. NRAS mutations were reported to function as reliable predictors of resistance to anti-EGFR (epidermal growth factor receptor) monoclonal antibody therapy [42]. Thus, further analysis is needed to identify the functional role of these mutations in EOC carcinogenesis to identify new prognostic biomarkers and develop novel therapeutic targets. However, it is important to remark that the clinico-pathological features of EOCs and the prognostic impact of these oncogenic mutations remains unclear.

This work presents a comprehensive gene expression profile of EOC. In the present study, NGS was used to assess differential gene expression between EOC and healthy samples, and the further craft of a pathway signature of EOC was performed using GO and KEGG enrichment analyses. A majority of the 117 candidate genes were classified into “pathways in cancer” and “axon guidance”, upon the activation of “cGMP-PKG signaling pathway”, “oxytocin signaling pathway”, and “adrenergic signaling in cardiomyocytes”. Some of the common pathways were identified as significant in both KEGG and Reactome analyses. “Pathways in cancer” is a comprehensive pathway of multiple cellular processes and crosstalk during cancer development, including the p53 signaling pathway and MAPK signaling pathway. Alternations in TP53 gene expression occur in more than 50% of human tumors [27, 28], leading to the severe reduction of tumor suppression and increasing the possibility that a cell will perform the uncontrolled division. Axon guidance is a process by which axons stretch to their correct targets, and axon guidance pathway genes have been implicated in cancer cell growth, survival, invasion, and angiogenesis [43]. Unfortunately, the incidence of aberrations in these genes in cancer is largely unknown. Several axon guidance genes, including TP53, have been implicated in human cancers including pancreatic carcinogenesis [43]. The cGMP/PKG signaling pathway is involved in cell cycle progression, cellular proliferation, and chromosomal instability [44], playing as an important role in regulating the proliferation and survival of human renal carcinoma cells. In tumors, oxytocin acts as a growth regulator via the activation of the oxytocin receptor, a specific G-coupled transmembrane protein. Adrenergic signaling has been found to regulate multiple cellular processes that contribute to the initiation and progression of cancer, including inflammation, angiogenesis, apoptosis/anoikis, cell motility and trafficking, activation of tumor-associated viruses, DNA damage repair, cellular immune response, and epithelial–mesenchymal transition.

Recent studies have revealed that both pathway signature and mutation signature are important to delineate carcinogenesis in individual cancers [18, 33]. The current study used NGS to deep-sequence hundreds of cancer-related genes from clinical grade tumor samples and elucidated an unexpectedly high frequency of actionable genomic events that may inform targeted treatment decisions. Further work includes the selection and validation of individual genes as biomarkers in a well-designed research with appropriate molecular diagnostic gating to promote precision therapies that specifically target molecules of tumor-driving signaling pathways and predict the response of an individual patient on such a targeted therapy.

In conclusion, our research identified several novel genetic alterations and pathway of OC by NGS, which provided abundant information for understanding molecular mechanisms of the OC, and even contributed to early clinical diagnosis. However, further studies with selective candidate genes and larger sample size are needed to define their role in occurrence and development of OC.