Ovarian cancer (OC) remains a critical public health concern due to its poor prognosis and high mortality rate (1,2). According to 2022 GLOBOCAN data, OC is the eighth most common cancer and the eighth leading cause of cancer-related death among women worldwide, accounting for 206,956 deaths (3). This high mortality may be attributed to the lack of symptoms in early-stage disease and the absence of effective screening methods (4,5). Consequently, OC is frequently diagnosed at an advanced stage, contributing to its poor prognosis (6,7). Additionally, the disease is marked by a high recurrence rate, with over 80% of cases developing resistance to treatment, leading to a reduced 5-year survival rate (8).

Epithelial ovarian carcinoma (EOC) is the most common subtype of OC, comprising ~90% of all ovarian malignancies (4,9). EOCs are classified into Type I and II based on their histopathological features and clinical behavior (10-12). Low-grade serous ovarian carcinoma (LGSOC), a Type I tumor, is typically indolent, associated with prolonged overall survival, and presents a relatively favorable prognosis (11,13,14). It is often linked to serous borderline tumors, which are considered common precursor lesions (15,16). LGSOC is frequently associated with mutations in the MAPK signaling pathway, particularly involving the KRAS and BRAF genes (17,18). By contrast, high-grade serous ovarian carcinoma (HGSOC) is a Type II tumor characterized by aggressive growth, rapid progression, and poor clinical outcomes. It is frequently diagnosed at an advanced stage (10,11,19,20). Serous tubal intraepithelial carcinoma, arising from the distal fimbriae of the fallopian tube, has been identified as a probable precursor lesion of HGSOC (21-23). Genomic analyses have revealed that ~96% of HGSOC cases harbor mutations in the TP53 gene (24-27).

Astrocyte Elevated Gene-1 (AEG-1), also known as metadherin, is an oncogene that plays a crucial role in cancer development and progression (28,29). AEG-1 interacts with various proteins and activates key oncogenic signaling pathways, including NF-κB, PI3K/Akt, MEK/ERK and Wnt/β-catenin (28,30). This activation promotes multiple cancer hallmarks, such as increased proliferation, migration, invasion, angiogenesis and metastasis (30-32). Furthermore, AEG-1 expression has been associated with chemotherapy resistance in several cancer types, including OC (33,34). It has been previously shown that elevated AEG-1 expression in EOC correlates with clinicopathological features such as disease stage, tumor grade, residual tumor size, lymph node metastasis and poor prognosis (30).

HGSOC and LGSOC differ in their clinical characteristics and prognoses (14). They also exhibit distinct responses to chemotherapy (19). As such, understanding the molecular differences between these two subtypes of EOC is essential for guiding appropriate treatment strategies. Given AEG-1's involvement in cancer progression and its potential influence on chemotherapy response (28,34), further investigation is warranted. However, the specific role of AEG-1 in EOC, particularly in HGSOC and LGSOC, remains unclear. Therefore, the present study aimed to examine differences in AEG-1 expression between HGSOC and LGSOC and to determine whether AEG-1 expression can be used to distinguish between these two subtypes.

A total of 74 patients were diagnosed with either HGSOC or LGSOC. Of these, 24 patients met the exclusion criteria, including 13 patients for whom no paraffin blocks were available, six patients whose paraffin blocks contained no identifiable tumor cells, four patients who had received NAC, and one patient with incomplete data. Based on the inclusion and exclusion criteria, a final sample of 50 patients was included in the study. The cohort consisted of 27 patients with HGSOC and 23 with LGSOC (Fig. 1).

AEG-1 expression was evaluated by immunohistochemistry. AEG-1 was predominantly localized in the cytoplasm, nuclear membrane, or cell junctions. Staining was assessed across 10 randomly selected high-power fields (magnification, x400) representative of the average tumor size (Fig. 2).

LGSOC and HGSOC are distinct tumor types with differing morphological features, pathogenesis and molecular profiles (17). LGSOC generally has a more favorable prognosis than HGSOC (11,37). In the present study, the initial cohort included 74 patients diagnosed with either HGSOC or LGSOC, comprising 45 patients with HGSOC and 29 with LGSOC. This distribution aligns with existing literature indicating that HGSOC is more prevalent than LGSOC (13,38). HGSOC is the most common subtype of EOC, accounting for ~2/3 of cases and representing ~70% of the total incidence (21,39). By contrast, LGSOC comprises ~5% to 10% of all serous ovarian carcinoma cases (40,41).

The present study investigated several demographic characteristics, including age, parity, menarche and menopausal status. While research exploring the relationship between these variables and AEG-1 expression is limited, numerous studies suggest these factors may influence the risk of developing OC (42). The current findings revealed that younger patients were more frequently diagnosed with LGSOC than with HGSOC, which is consistent with studies that LGSOC tends to occur at a younger age than HGSOC (9,19,41,43).

Infertility has been cited as a risk factor for OC in several studies (1,2,42). Higher parity is associated with fewer ovulatory cycles, which may reduce the opportunity for tumor development due to ovulation-induced injury of the ovarian epithelium (5,44). Additionally, elevated progesterone levels during pregnancy are considered to protect against ovarian carcinogenesis by inhibiting cell proliferation and promoting apoptosis in ovarian epithelial cells (44). A previous study reported that increased parity correlates with a sustained reduction in the risk of EOC (RR=0.81; 95% CI: 0.77-0.86) (5). Specifically, higher parity was linked to a significant decrease in LGSOC incidence (RR=0.84; 95% CI: 0.76-0.93), while no significant association was found with HGSOC (RR=0.97; 95% CI: 0.92-1.02) (heterogeneity: P=0.01) (45). Conversely, another study reported that having more than two pregnancies significantly reduced the risk of HGSOC (RR=0.25; 95% CI: 0.19-0.50) (46).

Hormonal and reproductive factors are among the most significant additional risk factors for EOC (44,47). The total number of menstrual cycles throughout a woman's life is positively correlated with an increased risk of EOC, highlighting the potential role of repeated ovulation in ovarian carcinogenesis (47). As a result, early menarche and late menopause, which prolong the ovulatory lifespan, can elevate the risk of developing the disease (2,42). Although specific literature on the role of menarche in LGSOC is currently lacking, it has been suggested that both early menarche and late menopause may increase cumulative estrogen exposure, potentially contributing to a higher risk of HGSOC (48,49).

CA 125 is widely recognized as the most important tumor biomarker for the screening and detection of EOC (50). Elevated serum CA 125 levels are observed in ~50% of early-stage tumors, primarily Type I cancers such as LGSOC and in ~92% of advanced-stage cases, predominantly HGSOC (50,51). Higher CA 125 levels have been associated with increased AEG-1 expression, likely due to AEG-1's role in promoting tumor cell proliferation and cancer progression (28,30,34). However, the present findings did not fully align with previous studies, as elevated CA 125 levels were more frequently observed across both groups in our dataset. As a result, the expected cell counts in the statistical analysis exceeded 20%, introducing uncertainty in interpreting the association between CA 125 levels and AEG-1 expression.

HGSOC is characterized by marked aggressiveness in its proliferation, invasion and metastatic behavior, often resulting in diagnosis at advanced stages and a higher likelihood of residual tumor masses >1 cm (52,53). This understanding supports the current findings, which demonstrated that a significantly greater number of HGSOC cases presented with residual tumor sizes >1 cm compared with LGSOC (OR=7.219; 95% CI: 1.399-37.252). While this result is consistent with existing literature, the wide CI suggests that the sample size may have been insufficient. Historically, achieving a maximum residual tumor diameter of #x003C;2 cm was considered indicative of successful cytoreductive surgery (54). However, more recent evidence identified that reducing the residual tumor burden to #x003C;1 cm provides improved survival outcomes, and it is now widely accepted that complete macroscopic resection (that is, no visible residual disease or R0) offers the greatest survival benefit (53).

AEG-1 contributes to cancer development and progression by activating multiple oncogenic signaling pathways (28,30). Specifically, AEG-1 enhances the PI3K/Akt pathway, promoting the phosphorylation of murine double minute 2 (MDM2) by Akt. This phosphorylation facilitates MDM2's nuclear translocation, leading to the degradation of p53 and enabling continued tumor cell proliferation (30,55,56). Additionally, the PI3K/Akt pathway is involved in angiogenesis, as Akt activation upregulates HIF-1 expression, which in turn increases VEGF transcription (35,57,58). AEG-1 also plays a key role in cancer cell migration, invasion and metastasis through the activation of the NF-κB pathway (30). Within this pathway, AEG-1 interacts with multiple components, serving as a crucial mediator in NF-κB activation and the subsequent induction of inflammation (59,60). Moreover, by inhibiting retinoid X receptor function, AEG-1 has emerged as a significant regulator of lipid metabolism and the tumor microenvironment during cancer development (59).

According to current theoretical frameworks, AEG-1 is associated with the clinicopathological characteristics of various cancers, including OC. Previous studies investigating AEG-1 expression in OC have demonstrated significant associations with several clinical variables, including age over 55 years (P=0.031), advanced FIGO stage (P#x003C;0.001), higher histological grade (P#x003C;0.001), elevated CA 125 levels (>35 U/ml, P#x003C;0.001), residual tumor size >1 cm (P#x003C;0.001) and lymph node metastasis (P=0.027) (30).

By contrast, the present study did not identify a statistically significant association between AEG-1 expression and these clinicopathological characteristics. However, this does not diminish the relevance of the findings. Limitations such as a relatively small sample size and uneven group distribution likely impacted the statistical power, underscoring the need for future studies with larger cohorts or alternative study designs.

Compared with LGSOC, AEG-1 appears to play a role consistent with the more aggressive biological behavior of HGSOC, contributing to increased cancer cell proliferation and invasiveness (28,30,36). HGSOC is clinically recognized for its aggressive progression and poor prognosis (10,11,22). The findings of the present study support this distinction, revealing that AEG-1 expression was significantly higher in HGSOC than in LGSOC (RR=3.228; 95% CI: 1.188-8.776) with the effect size, calculated using Cohen's D, is 0.745. These findings suggest that the aggressive clinical course of HGSOC may be partly driven by elevated levels of AEG-1 expression.

In conclusion, the present study demonstrated a significant difference in AEG-1 expression between HGSOC and LGSOC, with higher levels observed more frequently in HGSOC. These findings contribute to an improved understanding of the molecular distinctions between these two subtypes of EOC and may serve as foundational data for future research. Ultimately, such insights can support more tailored therapeutic approaches and improve patient outcomes and quality of life.