Ovarian cancer (OC) is the 7^th most common cancer in women and the 3^rd deadliest gynaecological cancer worldwide (1). In Saudi Arabia, OC affects more than 3% of Saudi women (2–4). This higher mortality rate of OC worldwide seems to be related to the fact that this malignant disease is asymptomatic, especially at early stages (5). In addition, most OC symptoms are nonspecific, misleading and may be confused with other gastrointestinal, urologic, or other diseases (6). Pelvic or abdominal pain and abdominal distension, increased urinary frequency, and some eating disorders such as early satiety are the common OC symptoms in the early stages, while women with advanced stages have a pelvic mass that extends beyond the adnexa (7). OC is classified according to the cellular origin of the malignancy, i.e., epithelial, stromal, or germinal cells. Of note, the vast majority (90%) of OC is of epithelial origin (8). Standard treatment options for OC depend on the type and stage of OC and include surgery along with platinum-based chemotherapy such as carboplatin and paclitaxel, either adjuvant, neoadjuvant, or sometimes both (9). Although 80% of patients diagnosed at an early stage respond to first-line chemotherapy, efficient early diagnosis of OC is still unattainable.

Since most OC patients are not diagnosed until the stage of metastasis, treatment options are not effective enough and are more diverted towards to alleviating symptoms rather than curing the disease. In fact, the 5-year survival rate for OC patients diagnosed with advanced stage disease is about 30% compared to 93% for early stage counterparts (10). Additionally, most OC patients relapse after completion of first-line treatment and require retreatment, mainly with chemotherapy (11). Despite, standard therapies are widely used in the treatment of OC, the prognosis and survival of OC are still poor. In addition, current management and treatment options are challenged by OC heterogeneity, in which a cluster of multiple cells with different genetic and epigenetic features occurs in the same ovarian malignant mass. Furthermore, individuals at the same stage of OC and treated with the same treatment plan have different outcomes. Taken together, these findings highlight the current challenges in optimizing/personalizing current therapeutic strategies for better outcomes (12, 13) and underscore the urgent need for additional effective biomarkers for earlier detection, better prognosis, and more accurate stratification of patients to achieve better individualized treatment options and survival outcomes.

Carbohydrate antigen 125 (CA125) is the first biomarker discovered for the detection of OC. Its level in serum is elevated in most epithelial OC (14). However, the sensitivity of CA125 in OC early stages remains too low, and its level correlates with other diseases such as endometriosis, pregnancy, ovarian cysts, and inflammatory peritoneal diseases. To improve the specificity of OC detection, other biomarkers such as Human Epididymis Protein 4 (HE4) have been developed. HE4 is more sensitive than CA125 and is found in approximately 100% of serous and endometroid subtypes, but its concentration can be influenced by many factors such as body mass index (BMI) (15), smoking (16), and lower HE4 concentration in patients using oral contraceptives (17). Although the combination of CA125 and HE4 has been shown to provide better diagnostic efficacy for risk prediction of OC (18), they are still not accurate and effective enough. More molecular biomarkers are needed to achieve better prognostic, therapeutic and prediction results. Cadherins are important transmembrane glycoproteins that are critical for cell-cell adhesion, especially in epithelial tissues. They were first described as single-pass transmembrane glycoproteins involved in cell–cell adhesion, and are now considered important players in cell polarity and tissue morphology (19). They are also thought to play a direct role in carcinogenesis and metastasis in many cancers (20, 21). In some cases of epithelial carcinoma, epithelial cells lose cell-cell adhesion and polarity and develop migratory and invasive behavior. This process, termed epithelial-mesenchymal transition (EMT), is critical for the development of metastases in cancer progression. A fundamental event in EMT is the “cadherin switch”, defined as loss of E-cadherin expression and increased expression of N-cadherin during cancer progression (22, 23). N-cadherin, also known as CDH2, is a cell-adhesion molecule mapped to 18q11.2 (24). It is a 135 KDa protein that belongs to the family of transmembrane molecules and mediates calcium- dependent intercellular adhesion. It consists of five extracellular cadherin repeats. The cytoplasmic domain of N-cadherin is anchored to the intercellular actin cytoskeleton by interaction with the β-, α-, and γ-catenin complex. CDH2 is expressed in various tissues, including the nervous system, brain, cardiac and skeletal muscles, blood vessels, and hematopoietic function (25, 26). N-cadherin is mainly expressed in the nervous system and promotes intercellular adhesion of neuronal cells, while its expression is low in normal tissues (25–29). However, it has been reported that overexpressed N-cadherin is associated with cell migration, angiogenesis, aggressiveness, and metastasis in many cancers such as breast, lung, bladder, prostate, and hepatocellular carcinomas (25–29). Moreover, the level of soluble N-cadherin in the serum of cancer patients is much higher than that in healthy individuals. As a result, N-cadherin has been suggested as a potential therapeutic target for tumour invasion and metastasis (30). In OC, the role of N-cadherin expression is unclear and there are few studies that have investigated N-cadherin expression in OC (31), especially in the Arabian peninsula. With this background, this study aimed to evaluate N-CAD protein expression patterns as a potential pro-metastatic molecular biomarker that could help improve OC prognosis and management. The associations between N-CAD protein expression patterns with patients’ clinicopathological parameters and its prognostic value in OC were investigated.

In 2021, more than 21,000 new cases were diagnosed with OC worldwide, and about 13,770 patients have died from this deadly disease (36). Several factors are believed to contribute to this increasing incidence and higher mortality rates. Apart from the difficulty of detecting the disease at early stages, the OC treatment options are not very effective at advanced stages, mainly because of the heterogeneity and complexity of this malignant disease. The higher complexity of OC is the result of various intertwined genetic and epigenetic factors that lead to aberrant gene expression and inconsistent treatment outcomes (37). In conjunction with clinicopathological factors such as age, grade, stage, and lymph node invasion, OMICs tools have provided an unprecedented understanding of the molecular complexity and disease progression of the diseases. In particular for OC, many genes have been reported as mutated, including BRCA1, BRCA2, BRIP1, RAD51C, RAD51D, MSH2, MLH1, PMS2, and MSH6. These and other candidate genes have been associated with a higher risk of OC (38, 39). Despite numerous efforts to identify reliable OC biomarkers, early detection strategies still rely mainly on CA125 and HE4, which have not been shown to be specific and sensitive enough (40, 41). Therefore, additional efforts are needed to develop new theranostic tools that can alleviate the suffering of OC patients and improve the treatment of the disease. Currently, the focus is on identifying more effective and clinically useful prognostic markers at the genomic and proteomic levels to detect OC at an early, curable stage and potentially support therapeutic decision making. In this regard, N-CAD has been reported to be expressed in several cancer types and has been associated with several clinicopathological parameters as well as survival outcomes. However, the clinical and prognostic significance of N-CAD in OC has not been well studied, especially in the Arabic Peninsula. Therefore, we conducted this study to investigate the N-CAD expression patterns and evaluate its prognostic value in our cohort of OC patients.

Our study showed that the protein N-CAD was expressed mainly in the cytoplasm of 58% of our patients’ tumor cells, with a recorded expression also in the cell membrane. Similarly, Quattrocchi et al. reported that 99% (158 cases) of their OC cohort expressed N-CAD protein in the cytoplasm (42). However, other studies reported membranous N-CAD expression in 32% of their cohort (43). These discrepancies could be due to cohort size, ethnicity, proportion of histological subtypes, and the complicated molecular heterogeneity of OC within each subtype (44).

The results of this study showed also that the expression of N-CAD protein was significantly associated with some clinicopathological characteristics including histological subtype, grade, tumor necrosis and age of menarche (p< 0.05) ( Table 1 ). These findings are consistent with many studies that reported significant correlations of N-CAD expression in OC with histologic subtypes (45, 46) and grade (45). On the other hand, they found a significant association between N-CAD protein expression and tumor stage, which is not confirmed by our results. Furthermore, and in agreement with our results, other studies on OC reported no significant correlation between the expression of N-CAD and other clinicopathological parameters such as tumor stage, patient’s age, BMI, and tumor size (43, 47). Our results showed that 57% of our patients’ cohort were below 50 years ( Table 1 ). There is a noticeable early onset of OC in the Saudi population compared to the United Kingdom for example where, according to Cancer Research UK, 53% of OC cases were diagnosed at 65 and over. Possible reasons associated to genomic, environmental and lifestyle factors deserve to be investigated to explain this early onset phenomenon.

In Kaplan-Meier survival analysis, N-CAD protein expression was significantly associated with DFS (p=0.03). In fact, patients with higher N-CAD expression have approximately twice the recurrence rate at 5-year follow-up time (42% vs. 78% recurrence at 60 months; p < 0.03, log rank, Figure 2 ). A similar trend was also observed with lower significance for DSS, in which patients with higher N-CAD expression who died more rapidly from the disease compared with their counterparts with low N-CAD expression ( Figure 3 ). In general, the Kaplan-Meier survival curves clearly show that shorter survival and higher recurrence rates were associated with overexpression of the N-CAD protein. These results are consistent with those of Quattrocchi et al., who reported that all patients in their cohort with N-CAD overexpression relapsed by the first year of follow-up time. In the same study, patients with higher E-CAD expression survived shorter than their counterparts with lower N-CAD expression (42). Two important meta-analysis studies using all published data and freely available sources about N-Cadherin showed similar survival outcomes as our results. In fact, they confirmed that N-CAD overexpression is a negative prognosticator of OC (48, 49). When we assessed the freely available KM plotter analysis of TCGA database (https://kmplot.com/analysis/), it appears that this platform did not cover the N-Cadherin protein expression (CDH2) in OC. However, the mRNA data showed that CDH2 gene expression is a good prognosticator of OC (50) (http://kmplot.com/analysis/index.php?p=service). For the protein Atlas database (https://www.proteinatlas.org), the CDH2 protein was not a significant prognosticator in OC possibly due to the heterogeneity of the cohort (51) (https://www.proteinatlas.org/ENSG00000170558-CDH2/pathology).

The survival data indicated that poor disease progression associated with high N-CAD protein expression appears to be either a marker of OC aggressiveness or actively involved in the pathophysiology of disease progression, recurrence, and metastasis. Similar studies in other cancers (breast, lung, bladder, prostate, …) confirmed that overexpression of N-CAD protein was associated with poor treatment outcomes, cell migration, angiogenesis, disease aggressiveness, and metastasis (25–29). Thus, overexpression of N-cadherin in colorectal cancer was significantly associated with poor disease-specific survival and disease-free survival, as well as with many clinicopathological characteristics such as tumor size, lymph node, stage, and grade (52). Similarly, high expression of N-cadherin in bladder cancer was shown to be associated with grade, tumor stage, and poorer recurrence- free survival (53).

Taken together, these results seem to be related to the role of N-CAD in the mesenchymal phenotype, which promotes cell mobility and invasion (31, 54). In fact, several reports have shown that when epithelial tumor cells switch from expressing E-CAD to expressing N-CAD (cadherin switch phenomenon), they acquire the ability to activate Fibroblast Growth Factor Receptor (FGFR) pathways. Our results together with our previous study about E-Cad expression in the same cohort confirmed the cadherin switch (32). In fact, while the E-CAD expression was decreasing at the advanced stages (starting from the EMT and marked by cancer invasion and migration), the N-CAD expression was increasing; and both markers were prognosticators of poor survival outcomes (32). In fact, once N-CAD is overexpressed, it has been shown to affect tumor cell polarity and behavior through its direct interaction with the FGFR, which regulates cancer cell motility and invasion (55, 56). Also, N-CAD was reported to interact with other receptors on tumor cells to promote motility and migration such as Platelet Derived Growth Factor Receptor (PDGFR). This mechanism occurs when the NHERF protein binds the N-CAD with the ß-catenin to the PDGFR to form a complex that drives tumor cells to migrate and motility (56) ( Figure 4 ). These molecular mechanisms of N- CAD protein overexpression, summarized in Figure 4 , played a key role in the phenotypic changes of tumor cells that were actively involved in migration to distant metastases. This pro-metastatic role of N-CAD was also confirmed in vitro with epithelial cells engineered to overexpress N-CAD. These cells have been shown to alter their morphology and behavior, adopting a motile phenotype similar to that observed in cells undergoing EMT (57, 58). This pro-metastatic phenotype depends also on the expression of other interacting proteins in addition to N- CAD, as mentioned previously and summarized in Figure 4 (58, 59).

This study demonstrated a prognostic role of N-CAD in OC, the first to be reported in the Arabic Peninsula. OC patients overexpressing the N-CAD protein had a poor prognosis, as evidenced by higher rates of both OC recurrence and death, as well as its molecular contribution in EMT and distant metastasis; and thus required more frequent and closer follow-up. Further studies with larger patient cohorts are needed to validate these findings, investigate further the role of N-CAD in OC pathophysiology, and explore its role as a potential therapeutic target.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

The study was approved by the Institutional Ethical Review Board of King Abdulaziz University Hospital, Jeddah, Saudi Arabia (Ref. number: KAUH-189-14). The patients/participants provided their written informed consent to participate in this study.

The author confirms being the sole contributor of this work and has approved it for publication.

This work was funded by the Deanship of Scientific Research (DSR), King Abdulaziz University, Jeddah, under grant No. (D-545-117-1443). The author, therefore, gratefully acknowledges the DSR technical and financial support.

The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.