Abstract
Objective
Integrin alpha-3 (ITGA3) has been implicated in tumor metastasis in various cancers, but its role in epithelial ovarian cancer (EOC)-associated liver metastasis (LM) remains unclear. This study aimed to investigate its role in LM in primary EOC patients.
Methods
It was a retrospective study with a sample size of n = 235 receiving surgical resection at Puren Hospital Affiliated to Wuhan University of Science and Technology between January 2020 and December 2021, including 98 LM (LM group) and 137 non-LM cases (N-LM group). ITGA3 expression was assessed by immunohistochemistry. ROC curves were used for predictive performance analysis, Kaplan-Meier curves for survival analysis, and Cox regression analysis for identification of risk factors.
Results
Markedly elevated ITGA3 expression in tumor tissues was found in the LM group (P < 0.001), which demonstrated strong predictive value for LM in EOC patients (area under the curve (AUC) = 0.881, sensitivity = 70.41%, specificity = 87.59%, P < 0.001), and strongly correlated with tumor size and postoperative residual lesions (both P < 0.05). Compared with the L-ITGA3 group, the H-ITGA3 group had a higher incidence of postoperative LM (P < 0.001) and showed a left-shifted curve in Kaplan-Meier analysis (P < 0.001). ITGA3 expression in tumor tissues (HR = 5.977), tumor grade (HR = 1.441), and postoperative residual lesions (HR = 1.697) were identified as independent risk factors for postoperative LM.
Conclusions
ITGA3 expression in tumor tissue significantly aids in predicting LM in EOC patients and is independently and closely related to adverse clinicopathological outcomes.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12920-025-02268-4.
Keywords: Integrin alpha-3, Epithelial ovarian cancer, Liver metastasis, Tumor grade, Postoperative residual lesion, Biomarker, Prognosis, Retrospective study
Introduction
Worldwide statistics show ovarian cancer (OC) as the eighth most prevalent cancer type (314,000 cases) and equally the eighth most fatal cancer (207,000 deaths) among female malignancies [1, 2] 10.32604/biocell.2024.047812. Epithelial ovarian cancer (EOC) origins from malignant transformation of ovarian surface epithelial cells and accounts for 90% of cases [3]. Characterized by insidious onset and rapid progression with nonspecific early symptoms [4], EOC often evades timely diagnosis, resulting in 12–33% of patients presenting with International Federation of Gynecology and Obstetrics (FIGO) stage IV disease and synchronous metastases at initial diagnosis [5]. Liver metastasis (LM) is a common and aggressive feature of advanced EOC, associated with poor prognosis and limited therapeutic options [5, 6]. Its high prevalence and contribution to mortality underscore the urgent need for improved management strategies.
Metastasectomy has demonstrated survival benefits in eligible EOC patients [7]. However, due to the absence of specific clinical manifestations during early-stage LM in EOC patients, most of them present with multifocal metastases at diagnosis, thereby missing the optimal window for surgical intervention [8]. Importantly, the clinical management of LM in EOC cases remains challenging owing to the lack of targeted preventive/therapeutic approaches and the limited efficacy of systemic chemotherapy [9]. These critical clinical gaps underscore the urgent need to elucidate the molecular mechanisms underlying LM progression in EOC cases and identify reliable early predictive biomarkers, which would enable timely risk stratification and facilitate early clinical interventions to improve the outcomes of EOC patients.
The integrin family, serving as crucial receptors mediating cell-cell and cell-matrix adhesion, takes pivotal part in maintaining normal epithelial homeostasis [10]. Emerging evidence has established that aberrant integrin expression contributes to cancer progression through diverse mechanisms, including promoting invasion in non-small cell lung cancer (NSCLC) [11], driving malignant advancement in hepatocellular carcinoma [12], and mediating chemoresistance in pancreatic cancer [13]. Notably, integrin alpha-3 (ITGA3), as the primary laminin receptor, has emerged as a key player in tumor metastasis [14]. Bioinformatic analyses have revealed that ITGA3 is notably overexpressed in EOC cases, showing strong associations with advanced-stage disease and poor patient prognosis [15]. A mechanistic study has confirmed that exosomal ITGA3 strongly promotes EOC cell migration and tumor progression via the S100A7/ERK axis [16]. Clinical correlation analyses have further confirmed the association between high ITGA3 expression and poor prognosis in colorectal cancer [17] and the suppression of ITGA3 knockdown in the invasion in head and neck squamous cell carcinoma [18]. The landscape of cancer treatment has evolved from non-specific cytotoxic drugs to complex strategies encompassing targeted therapy, immunotherapy, and precision medicine. The central debate in contemporary oncology centers on the most effective therapeutic focus: targeting intrinsic oncogenic drivers within tumors or disrupting the supportive tumor microenvironment. Emerging evidence suggests dual targeting may yield more durable therapeutic benefits [19]. The future of cancer care involves merging different techniques to formulate personalized and diverse treatment strategies [20]. Against this backdrop, the integrin family of cell adhesion receptors mediates critical cell-cell and cell-matrix interactions, representing a unique class of molecules bridging the gap between intrinsic tumor and microenvironment targeting [21]. Beyond its role in adhesion, ITGA3 affects extracellular matrix remodeling via matrix metalloproteinases and manages immune cell infiltration through its interactions with galectin-3 [22, 23]. ITGA3’s dual roles make it an excellent candidate for influencing both tumor growth and the surrounding environment, positioning it as a key target for therapeutic testing.
Therefore, this study aimed to probe into the correlation between ITGA3 expression and the occurrence of LM in EOC cases, so as to offer new insights to early prediction of LM risk and lay a theoretical foundation for the development of targeted therapeutic strategies.
Materials and methods
Study population
Totally 284 patients with primary EOC who underwent surgery at Puren Hospital Affiliated to Wuhan University of Science and Technology between January 2020 and December 2021 were retrospectively enrolled. According to the inclusion and exclusion criteria, 235 EOC patients were ultimately enrolled in the study. A 3-year follow-up period is a commonly used and highly valuable timeframe for assessing the prognosis of EOC [24, 25]. Based on the occurrence of LM within 3 years postoperatively, the patients were further categorized into the LM group (n = 98) and the N-LM group (n = 137). This study complied with the ethical principles of the World Medical Association Declaration of Helsinki, relevant clinical research regulations, and the Enhancing the QUAlity and Transparency Of health Research (EQUATOR) network guidelines. The study protocol was approved by the Ethics Committee of Puren Hospital Affiliated to Wuhan University of Science and Technology.
Inclusion and exclusion criteria
Inclusion criteria
(1) patients pathologically confirmed with primary EOC; (2) patients > 18 years old; (3) patients with complete clinical records available; (4) patients eligible for surgical treatment.
Exclusion criteria
(1) patients with preoperative imaging-confirmed distant metastases (All patients underwent preoperative contrast-enhanced computed tomography or magnetic resonance imaging of the chest, abdomen, and pelvis. Patients who had confirmed metastases in distant organs such as the liver, lungs, or bones were excluded); (2) patients with other distant metastases (e.g., pulmonary, osseous, or cerebral) within 3 years post-operation; (3) patients with severe hepatic or renal dysfunction; (4) patients with a history of other malignant tumors; (5) patients with major organ dysfunction; (6) patients with non-EOC subtypes; (7) patients with borderline ovarian tumors.
LM was confirmed by characteristic imaging findings on hepatic ultrasonography, abdominal computed tomography (CT), magnetic resonance imaging (MRI), or positron emission tomography-computed tomography (PET-CT), supported by comprehensive clinical evaluation including medical history, gastrointestinal symptoms, and elevated serum tumor markers. Pathological confirmation was obtained through LM lesion puncture biopsy with histopathological verification of metastatic EOC origin [5].
Sample and data collection
Comprehensive baseline and pathological data were collected for all participants, including age, body mass index (BMI), surgical type, type of initial chemotherapy, laboratory indices [(cancer antigen 125 (CA125), human epididymis protein 4 (HE4), carcinoembryonic antigen (CEA) alpha-fetoprotein (AFP)], and clinicopathological features (FIGO stage, tumor grade, tumor location, histological type, tumor size, regional lymph node status, and postoperative residual lesions).
Immunohistochemistry
As previously described [26], immunohistochemical analysis was performed to detect ITGA3 expression in cancer tissues. Briefly, after routine dewaxing and rehydration of cancer tissue sections, followed by antigen retrieval and inactivation of endogenous enzymes, the sections were incubated overnight at 4 °C with the primary antibody ITGA3 (ab131055, 1:1000, Abcam, UK). After washing, sections were incubated at room temperature with secondary antibody Goat anti-Rabbit IgG (H + L) (31460, 1:1000, ThermoFisher) for 30 min. Stained sections were observed using an Olympus BX53 microscope (Olympus, Center Valley, PA, USA). To ensure experimental specificity, each staining procedure included a positive control (tissue section known to overexpress ITGA3) and a negative control (PBS substituted for the primary antibody). ITGA3 expression was quantified using the H-score method. Scores range from 0 to 12 points, with 0–5 being negative (-), 6–11 being positive (+), and 7–12 being strongly positive (++). Scoring was independently performed by two pathologists with over five years of experience who were unaware of the patients’ clinical and pathological information. Agreement was assessed by calculating the Kappa value (Kappa = 0.85, P < 0.001), demonstrating excellent consistency. Cases with inconsistent scores were adjudicated by a third pathologist with over ten years of experience. The number of ITGA3-positive cells and staining intensity were scored using ImageJ software (National Institutes of Health, USA).
Statistical analyses
Statistical analyses and graphing were performed using SPSS Statistics (version 21.0, SPSS, Inc., Chicago, IL, USA) and GraphPad Prism software (version 8.0.1, GraphPad Software Inc., San Diego, CA, USA). Data normality was evaluated using the Kolmogorov-Smirnov test. Normally distributed measurement data were described as mean ± standard deviation and compared using the independent samples t-test between groups, while non-normally distributed data were described as median (minimum, maximum) and compared with the Mann-Whitney U test between groups. Counting data were described as counts (percentages) and analyzed via the Chi-square test/Chi-square goodness-of-fit test. ROC curve analysis was employed to evaluate the predictive value of ITGA3 expression in cancer tissue and conventional serum biomarkers (CA125, HE4) for the occurrence of LM in EOC patients. Area under the curve (AUC) comparisons were performed using the MEDCALC-Delong test. The impact of the ITGA3 expression on postoperative LM in EOC patients was analyzed by Kaplan-Meier (KM) survival analysis, with the Log rank test used to examine inter-group differences in KM curves. After adjusting for confounding factors including age, type of initial chemotherapy, and tumor stage via multivariate Cox regression, the independent association between ITGA3 expression in cancer tissue and tumor grade (0 = I-II, 1 = III-IV), postoperative residual disease (0 = none or ≤ 1 cm, 1 = > 1 cm) with the occurrence of postoperative LM in EOC patients was analyzed. A two-sided significance level of α = 0.05 was adopted, with P < 0.05 considered notably significant.
Results
Baseline characteristics of enrolled patients
Clinical data were collected and compared between the LM (n = 98) and N-LM (n = 137) groups. The results (Table 1) revealed no notable inter-group difference regarding age, BMI, surgical type, and type of initial chemotherapy (all P > 0.05), but revealed markedly higher levels of CA125, HE4, CEA, and AFP in the LM group than the N-LM group (all P < 0.001).
Table 1.
Baseline characteristics of enrolled patients
| LM (n = 98) | N-LM (n = 137) | z/t/x2 | P | |
|---|---|---|---|---|
| Age (years) | 55 (40, 68) | 56 (46, 65) | 0.405 | 0.686 |
| BMI (kg/m2) | 23.54 (18.23, 27.95) | 23.73 (18.07, 27.81) | 0.022 | 0.982 |
| Surgical type (n, %) | 0.108 | 0.742 | ||
| Hysterectomy or adnexectomy | 40 (40.82) | 53 (38.69) | ||
| Cytoreductive surgery | 58 (59.18) | 84 (61.31) | ||
|
Type of initial chemotherapy (n, %) |
1.030 | 0.794 | ||
| Paclitaxel + Carboplatin | 47 (47.96) | 60 (43.80) | ||
| Carboplatin | 10 (10.20) | 15 (10.95) | ||
| Ifosfamide | 9 (9.18) | 18 (13.14) | ||
| None | 32 (32.65) | 44 (32.11) | ||
| CA125 (×102 U/mL) | 9.55 ± 1.05 | 7.46 ± 0.98 | 14.306 | < 0.001 |
| HE4 (×102 poml/mL) | 4.08 ± 0.35 | 3.17 ± 0.28 | 15.676 | < 0.001 |
| CEA (ng/mL) | 46.23 ± 7.70 | 40.89 ± 7.55 | 19.249 | < 0.001 |
| AFP (ng/mL) | 33.31 ± 5.96 | 19.80 ± 4.78 | 11.687 | < 0.001 |
Data normality was evaluated using the Kolmogorov-Smirnov test. Normally distributed measurement data were described as mean ± standard deviation and compared using the independent samples t-test between groups, while non-normally distributed data were described as median (minimum, maximum) and compared with the Mann-Whitney U test between groups. Counting data were expressed as counts (percentages) and analyzed using Chi-square test/Chi-square goodness-of-fit test
LM Liver metastasis, BMI Body mass index, CA125 Cancer antigen 125, HE4 Human epididymis protein 4, CEA Carcinoembryonic antigen, AFP Alpha-fetoprotein
EOC-LM patients have notably higher ITGA3 expression in tumor tissues than EOC patients
OC tissue sections from EOC patients were subjected to immunohistochemical (IHC) staining to analyze the ITGA3 expression in tumor tissues, with representative pathological images shown in Supplementary Fig. 1. The H scoring results (Fig. 1) demonstrated notably higher ITGA3 expression in tumor tissues in the LM group than in the N-LM group (P < 0.001).
Fig. 1.
ITGA3 expression is significantly higher in tumor tissues from EOC-LM patients. Notes: H scores of ITGA3 in tumor tissues. Data normality was evaluated using the Kolmogorov-Smirnov test. Non-normally measurement data were described as median (minimum, maximum) and compared with the Mann-Whitney U test between groups. ***P < 0.001
ITGA3 expression in tumor tissues demonstrates high predictive value for LM in EOC patients
Subsequently, ROC curve analysis was employed to evaluate the predictive value of ITGA3 expression in tumor tissue and conventional serum biomarkers (CA125, HE4) for LM occurrence in EOC patients. ITGA3 expression in tumor tissue (AUC = 0.881, Cut-off value = 6, P < 0.001), CA125 (AUC = 0.914, Cut-off value = 8.73, P < 0.001), and HE4 (AUC = 0.930, Cut-off value = 3.48, P < 0.001) all demonstrated predictive value for LM occurrence in EOC patients. Furthermore, the predictive efficacy of ITGA3 showed no significant difference compared to CA125 and HE4 (both P > 0.05), further indicating the high clinical utility of ITGA3 expression in tumor tissue for predicting LM in EOC patients (Fig. 2, supplementary Table 1).
Fig. 2.
ROC curve analysis of the predictive value of ITGA3 expression, CA125, and HE4 in tumor tissues for LM in EOC patients. Note: AUC, area under the ROC curve
ITGA3 expression in tumor tissues correlates with clinicopathological characteristics in EOC-LM patients
Using the ROC curve cut-off value (6) as threshold, EOC patients were stratified into H-ITGA3 group (> 6, n = 118) and L-ITGA3 group (≤ 6, n = 117). Subsequent analysis was conducted to evaluate the association between ITGA3 expression in tumor tissues and clinicopathological characteristics (FIGO stage, tumor grade, tumor location, histological type, tumor size, regional lymph node status, as well as postoperative residual lesions) in EOC-LM patients. Results (Table 2) revealed no strong associations of ITGA3 expression in tumor tissues with FIGO stage, tumor location, histological type, and tumor grade in EOC patients (all P > 0.05), but revealed strong correlations of ITGA3 expression with tumor size and postoperative residual lesions (both P < 0.05).
Table 2.
Correlations between ITGA3 expression in tumor tissues and clinicopathological characteristics in EOC-LM patients
| H-ITGA3 (n = 118) | L-ITGA3 (n = 117) | x2 | P | |
|---|---|---|---|---|
| FIGO stage (n, %) | 0.551 | 0.508 | ||
| T1-T2 | 51 (43.22) | 45 (38.46) | ||
| T3-T4 | 67 (56.78) | 72 (61.54) | ||
| Tumor grade (n, %) | 0.158 | 0.230 | ||
| I-II | 41 (34.75) | 50 (42.74) | ||
| III-IV | 77 (65.25) | 67 (57.26) | ||
| Tumor location (n, %) | 0.540 | 0.569 | ||
| Unilateral | 100 (84.75) | 103 (88.03) | ||
| Bilateral | 18 (15.25) | 14 (11.97) | ||
| Histologic type (n, %) | 1.642 | 0.648 | ||
| Serous | 63 (53.39) | 67 (57.26) | ||
| Endometrioid | 26 (22.03) | 29 (24.79) | ||
| Mucinous | 15 (12.71) | 10 (8.55) | ||
| Clear cell | 14 (11.86) | 11 (9.40) | ||
| Tumor size (n, %) | 4.631 | 0.037 | ||
| ≤ 8 cm | 48 (40.68) | 64 (54.70) | ||
| > 8 cm | 70 (59.32) | 53 (45.30) | ||
| Regional lymph nodes (n, %) | 0.762 | 0.411 | ||
| N0 | 37 (31.36) | 43 (36.75) | ||
| N1 | 81 (68.64) | 74 (63.25) | ||
| Postoperative residue lesion | 14.161 | < 0.001 | ||
| No | 64 (54.24) | 79 (67.52) | ||
| ≤ 1 cm | 30 (25.42) | 33 (28.21) | ||
| > 1 cm | 24 (20.34) | 5 (4.27) | ||
Counting data were expressed as counts (percentages) and analyzed using Chi-square test/Chi-square goodness-of-fit test
ITGA3 Integrin alpha-3
High ITGA3 expression in tumor tissue substantially increases the risk of postoperative LM occurrence in EOC patients
Subsequent analysis was performed to analyze the association between ITGA3 expression in tumor tissues and postoperative LM in EOC patients. The results (Table 3) revealed a notably higher incidence of postoperative LM in the H-ITGA3 group than the L-ITGA3 group (P < 0.001). Further KM survival analysis (Fig. 3) showed a significant left-shifting KM curve for the H-ITGA3 group versus the L-ITGA3 group (P < 0.001).
Table 3.
Correlation between differential ITGA3 expression in tumor tissues and postoperative LM in EOC patients
| H-ITGA3 (n = 118) | L-ITGA3 (n = 117) | x2 | P | |
|---|---|---|---|---|
| LM (n = 98) | 80 (67.80) | 18 (15.38) | 66.382 | < 0.001 |
| N-LM (n = 137) | 38 (32.20) | 99 (84.62) |
Counting data were expressed as counts (percentages) and analyzed using Chi-square test/Chi-square goodness-of-fit test
Fig. 3.
KM analysis of the impact of ITGA3 expression in tumor tissues on the risk of postoperative LM in EOC patients. Note: The log-rank test was used to assess differences between groups in KM curves
High ITGA3 expression in tumor tissue is an independent risk factor for the occurrence of postoperative LM in EOC patients
After adjusting for potential confounders and performing screening via multiple linear regression analysis, Cox regression analysis was conducted to identify risk factors for postoperative LM in EOC patients. The results (Tables 4 and 5) identified ITGA3 expression in tumor tissues (HR = 5.977), tumor grade (HR = 1.441), and postoperative residual lesion (HR = 1.697) as independent risk factors for postoperative LM in EOC patients.
Table 4.
Variable definitions and assignment
| Factors | Definitions | Assignment |
|---|---|---|
| y | Occurrence of LM | 0 = yes, 1 = no |
| x1 | ITGA3 expression in cancer tissue | 0 = H-ITGA3, 1 = L-ITGA3 |
| x2 | Tumor grade | 0 = I-II, 1 = III-IV |
| x3 | Postoperative residue lesion | 0 = no/≤1 cm, 1 = > 1 cm |
Table 5.
Risk factors for postoperative LM in EOC patients
| Factors | B | S.E. | Wals | P | HR | 95%CI |
|---|---|---|---|---|---|---|
| ITGA3 expression in tumor tissue | 1.788 | 0.268 | 44.506 | < 0.001 | 5.977 | 3.535–10.106 |
| Tumor grade | 0.365 | 0.223 | 2.669 | 0.102 | 1.441 | 0.930–2.232 |
| Postoperative residue lesion | 0.529 | 0.246 | 4.630 | 0.031 | 1.697 | 1.048–2.746 |
Discussion
This study elucidated the critical role of ITGA3 in LM of EOC patients, providing both a novel predictive marker and potential therapeutic target for clinical management. ITGA3, a crucial basement membrane receptor, participates in various physiological and pathological processes by regulating cell adhesion, migration, and cytoskeletal reorganization [27]. Mounting evidence demonstrates a strong association between aberrant ITGA3 expression and the pathogenesis and progression of multiple malignancies [28, 29]. Similarly, Ke et al. [30] have similarly detected ITGA3 overexpression in OC. In this study, the tumor-specific ITGA3 upregulation suggests its potential involvement in regulating the LM of EOC, highlighting its value as a predictive biomarker. Subsequent analysis confirmed ITGA3 as an independent risk factor for LM in EOC patients and its robust predictive efficacy. Li et al. [31] demonstrated that ITGA3 promotes pancreatic cancer LM through glycolytic pathway activation, while Zhang et al. [32] established ITGA3 as a predictor of recurrence and lymph node metastasis in papillary thyroid carcinoma. Our research reveals that ITGA3 serves as a prognostic biomarker for LM in EOC patients, yet its function extends far beyond that of a static membrane receptor. ITGA3 may not just be a static membrane receptor but also be trafficked via exosomes, influencing metastatic tropism (10.20517/2394-4722.2024.78). Furthermore, ITGA3’s role in cell motility must be considered within the broader context of the non-classical migration regulator family [33]. Collectively, ITGA3 can be regarded as a hub responsive to the tumor microenvironment, integrating extracellular signals via exosomes and facilitating intracellular reconnections through non-canonical interactions to advance the metastatic process.
Although ITGA3 expression levels showed no significant association with tumor grade in univariate analysis, tumor grade itself emerged as an independent risk factor for patient prognosis. Tumor grading reflects the intrinsic biological aggressiveness of tumors, providing robust and independent prognostic value. It also sheds light on the link between tumor grading and prognosis Therefore, while ITGA3 represents a novel prognostic biomarker independent of traditional clinical pathological parameters such as tumor grading, the prognostic information it provides complements rather than overlaps with existing indicators like tumor grading. This adds value to its future clinical translational applications—namely, ITGA3 holds potential for integration with traditional indicators such as tumor grading to construct more precise prognostic prediction models. Our research identified ITGA3 as a novel and independent prognostic biomarker in OC. It is essential to situate this discovery within the rapidly evolving field of OC biomarker discovery, a field increasingly dominated by sophisticated computational and multi-omics approaches [34, 35]. However, the clinical application of such complex detection methods often faces challenges related to cost, technical expertise, computational infrastructure, and turnaround time, particularly in resource-limited settings. In this context, our findings advocate for the enduring value of a simpler, more accessible, and clinically feasible strategy. Future studies may explore integrating ITGA3 immunohistochemistry with these multi-omics features to construct more comprehensive and layered predictive models, potentially achieving both high accuracy and broad clinical applicability.
The plasticity of cancer cells enables them to dynamically switch between epithelial-mesenchymal transition, stem cell-like states, and senescence-associated phenotypes to evade therapeutic pressure and migrate to distant microenvironments (DOI: 10.1002/cso2.1051). Persistent ITGA3 expression in metastatic tumor cells enhances integrin-mediated adhesion to the metastatic microenvironment, promoting phenotypic reprogramming. It also binds to laminin in the liver microenvironment to deliver critical survival signals, thereby shielding cancer cells from extracellular inhibitory effects and conventional therapies. This potential linkage indicates that ITGA3 is not merely a passive marker of invasiveness but an active regulator capable of mediating adaptive survival mechanisms. Concurrently, studies confirm the need for dynamic biomarkers to guide interventions against tumor heterogeneity (DOI:10.1097/ot9.0000000000000039). This further suggests that ITGA3 serves both as a prognostic indicator and a therapeutic vulnerability, enabling patient stratification for targeted treatment against acquired resistance. Future studies could explore whether ITGA3 expression contributes to resistance in EOC to platinum-based chemotherapy or targeted therapies—a highly intriguing and promising avenue for investigation.
This study has certain limitations. First, the retrospective and single-center nature of this study inherently introduces potential selection bias. All enrolled patients were East Asian individuals, limiting representativeness of the broader OC population and affecting the generalizability of findings. Future validation in multicenter, prospective cohorts is therefore necessary to confirm our results. Second, although the 3-year follow-up period was sufficient to capture a substantial number of metastatic events, it may be inadequate for comprehensively assessing the long-term patterns of metastatic disease in OC patients. It is possible that a subset of patients with low-volume metastases may still develop metastatic disease beyond this timeframe, particularly in certain molecular subtypes. Therefore, the predictive capability of this study’s conclusions and the application of ITGA3 for very late-stage metastasis requires further clarification through extended follow-up. Third, our study primarily focused on clinical and correlational aspects. While we have established a strong link between ITGA3 expression and LM, functional experiments are missing. Cell and animal model experiments, such as knocking down ITGA3 in OC cell lines to investigate its impact on in vitro invasion and migration, and establishing a LM xenograft model to confirm whether high ITGA3 expression facilitates liver colonization, represent crucial avenues for in-depth exploration of the ITGA3-lung metastasis mechanism. Similarly, since integrin signaling is typically associated with survival pathways, chemoresistance may further influence ITGA3 expression. The absence of chemoresistance analysis in this study may overlook its potential value as a resistance biomarker or therapeutic target for reversing drug resistance, resulting in an incomplete assessment of ITGA3’s clinical significance. Further investigation and analysis in future studies are warranted.
In summary, this study confirms a close association between high expression of ITGA3 and the clinical pathological characteristics of patients with EOC. ITGA3 is identified as an independent risk factor for LM in EOC patients and holds high predictive value. The detection of ITGA3 can help identify high-risk EOC patients for LM and provide a basis for personalized treatment.
Supplementary Information
Acknowledgements
Not applicable.
Authors’ contributions
KZ, TTC are the guarantors of integrity of the entire study and contributed to the study design; KZ contributed to the study concepts, definition of intellectual content, manuscript review; TTC contributed to the literature research, clinical studies, statistical analysis; YW contributed to the data acquisition, data analysis, manuscript preparation and editing; All authors read and approved the final manuscript.
Funding
No funding was received for this study.
Data availability
All data generated or analysed during this study are included in this article. Further enquiries can be directed to the corresponding author.
Declarations
Ethics approval and consent to participate
This study complied with the ethical principles of the World Medical Association Declaration of Helsinki, relevant clinical research regulations, and the Enhancing the QUAlity and Transparency Of health Research (EQUATOR) network guidelines. The study protocol was approved by the Ethics Committee of Puren Hospital Affiliated to Wuhan University of Science and Technology(Ethical Approval Number: CRWG2024R020). Written informed consent to participate has been obtained from all the participants.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Kai Zhu and Tingting Cheng contributed equally to this work.
References
- 1.Tavares V, Marques IS, Melo IG, Assis J, Pereira D, Medeiros R. Paradigm shift: a comprehensive review of ovarian cancer management in an era of advancements. Int J Mol Sci. 2024. 10.3390/ijms25031845. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Sherazi SAM, Rafique F, Haris M, Arshad A, Qaiser H, Uzair M, et al. Applications of CRISPR Cas-9 in ovarian cancer research. Protein Peptide Lett. 2023;30(8):653–67. [DOI] [PubMed] [Google Scholar]
- 3.Wei YF, Ning L, Xu YL, Ma J, Li DR, Feng ZF, et al. Worldwide patterns and trends in ovarian cancer incidence by histological subtype: a population-based analysis from 1988 to 2017. EClinicalMedicine. 2025;79:102983. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Arnaoutoglou C, Dampala K, Anthoulakis C, Papanikolaou EG, Tentas I, Dragoutsos G, et al. Epithelial ovarian cancer: a five year review. Medicina (B Aires). 2023. 10.3390/medicina59071183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hussain I, Xu J, Deng K, Noor Ul A, Wang C, Huang Y, et al. The prevalence and associated factors for liver metastases, development and prognosis in newly diagnosed epithelial ovarian cancer: a large population-based study from the SEER database. J Cancer. 2020;11(16):4861–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Deng K, Yang C, Tan Q, Song W, Lu M, Zhao W, et al. Sites of distant metastases and overall survival in ovarian cancer: a study of 1481 patients. Gynecol Oncol. 2018;150(3):460–5. [DOI] [PubMed] [Google Scholar]
- 7.Cheng F, Wang J, Chen Q, Zhu J. Pulmonary metastasectomy for recurrent metastatic ovarian cancer. J Obstet Gynaecol Res. 2023;49(12):2905–9. [DOI] [PubMed] [Google Scholar]
- 8.Kuroki L, Guntupalli SR. Treatment of epithelial ovarian cancer. BMJ. 2020;371:m3773. [DOI] [PubMed] [Google Scholar]
- 9.Yuan Y, Wang R, Luo M, Zhang Y, Guo F, Bai G, et al. JingZhao: a machine learning approach using XGBoost predicts lung metastasis in patients with ovarian cancer. Biomed Res Int. 2022;2022:8501819. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pang X, He X, Qiu Z, Zhang H, Xie R, Liu Z, et al. Targeting integrin pathways: mechanisms and advances in therapy. Signal Transduct Target Ther. 2023;8(1):1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Matsumoto Y, Kage H, Morota M, Zokumasu K, Ando T, Maemura K, et al. Integrin alpha 2 is associated with tumor progression and postoperative recurrence in non-small cell lung cancer. Jpn J Clin Oncol. 2023;53(1):63–73. [DOI] [PubMed] [Google Scholar]
- 12.Peng Z, Hao M, Tong H, Yang H, Huang B, Zhang Z, et al. The interactions between integrin alpha(5)beta(1) of liver cancer cells and fibronectin of fibroblasts promote tumor growth and angiogenesis. Int J Biol Sci. 2022;18(13):5019–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Liu M, Zhang Y, Yang J, Cui X, Zhou Z, Zhan H, et al. ZIP4 increases expression of transcription factor ZEB1 to promote integrin alpha3beta1 signaling and inhibit expression of the gemcitabine transporter ENT1 in pancreatic cancer cells. Gastroenterology. 2020;158(3):679–92. e671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hu F, Zhao L, Wang Y, Ye H, Tang H, Zhou J. Integrated bioinformatics analysis shows integrin alpha 3 is a prognostic biomarker for pancreatic cancer. Open Med. 2022;17(1):1935–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wu A, Zhang S, Liu J, Huang Y, Deng W, Shu G, et al. Integrated analysis of prognostic and immune associated integrin family in ovarian cancer. Front Genet. 2020;11:705. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Yin Z, Ma J, Adu-Amankwaah J, Xie G, Wang Y, Tai W, et al. Exosomal integrin alpha 3 promotes epithelial ovarian cancer cell migration via the S100A7/p-ERK signaling pathway. Acta Biochim Biophys Sin. 2025. 10.3724/abbs.2025024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sa KD, Zhang X, Li XF, Gu ZP, Yang AG, Zhang R, et al. A miR-124/ITGA3 axis contributes to colorectal cancer metastasis by regulating anoikis susceptibility. Biochem Biophys Res Commun. 2018;501(3):758–64. [DOI] [PubMed] [Google Scholar]
- 18.Koshizuka K, Hanazawa T, Kikkawa N, Arai T, Okato A, Kurozumi A, et al. Regulation of ITGA3 by the anti-tumor miR-199 family inhibits cancer cell migration and invasion in head and neck cancer. Cancer Sci. 2017;108(8):1681–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Liu H, Dilger JP. Different strategies for cancer treatment: targeting cancer cells or their neighbors? Chin J Cancer Res. 2025;37(2):289–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Sonkin D, Thomas A, Teicher BA. Cancer treatments: Past, present, and future. Cancer Genet. 2024;286–287:18–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Saini P, Mirji G, Islam SMS, Simons LM, Bhat SA, Bonfanti AP, et al. Targeting Siglec-10/alpha3beta1 Integrin Interactions Enhances Macrophage-Mediated Phagocytosis of Pancreatic Cancer. bioRxiv. 2025. [DOI] [PMC free article] [PubMed]
- 22.Wang K, He H. Pancreatic tumor microenvironment. Adv Exp Med Biol. 2020;1296:243–57. [DOI] [PubMed] [Google Scholar]
- 23.Gui J, Yang L, Liu J, Li Y, Zou M, Sun C, et al. Identifying the prognosis implication, immunotherapy response prediction value, and potential targeted compound inhibitors of integrin subunit alpha3 (ITGA3) in human cancers. Heliyon. 2024;10(2):e24236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Gong TT, Guo S, Liu FH, Huo YL, Zhang M, Yan S, et al. Proteomic characterization of epithelial ovarian cancer delineates molecular signatures and therapeutic targets in distinct histological subtypes. Nat Commun. 2023;14(1):7802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Loizzi V, Rossi C, Cormio G, Cazzolla A, Altomare D, Selvaggi L. Clinical features of hepatic metastasis in patients with ovarian cancer. Int J Gynecol Cancer. 2005;15(1):26–31. [DOI] [PubMed] [Google Scholar]
- 26.Uchikura E, Fukuda T, Sengiku T, Noda T, Awazu Y, Wada T, et al. Role of Fyn expression in predicting the sensitivity to platinum–based chemotherapy in patients with ovarian serous carcinoma. Oncol Lett. 2024;28(5):525. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Tan W, Chen G, Ci Q, Deng Z, Gu R, Yang D, et al. Elevated ITGA3 expression serves as a novel prognostic biomarker and regulates tumor progression in cervical cancer. Sci Rep. 2024;14(1):27063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Feng C, Jin X, Han Y, Guo R, Zou J, Li Y, et al. Expression and prognostic analyses of ITGA3, ITGA5, and ITGA6 in head and neck squamous cell carcinoma. Med Sci Monit. 2020;26:e926800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Li Q, Ma W, Chen S, Tian EC, Wei S, Fan RR, et al. High integrin alpha3 expression is associated with poor prognosis in patients with non-small cell lung cancer. Transl Lung Cancer Res. 2020;9(4):1361–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Ke FY, Chen WY, Lin MC, Hwang YC, Kuo KT, Wu HC. Novel monoclonal antibody against integrin alpha3 shows therapeutic potential for ovarian cancer. Cancer Sci. 2020;111(10):3478–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Li R, Ji Q, Fu S, Gu J, Liu D, Wang L, et al. ITGA3 promotes pancreatic cancer progression through HIF1alpha- and c-Myc-driven Glycolysis in a collagen I-dependent autocrine manner. Cancer Gene Ther. 2025;32(2):240–53. [DOI] [PubMed] [Google Scholar]
- 32.Zhang G, Li B, Lin Y. Evaluation of ITGA3 as a biomarker of progression and recurrence in papillary thyroid carcinoma. Front Oncol. 2021;11:614955. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Liu H, Hamaia SW, Dobson L, Weng J, Hernandez FL, Beaudoin CA, et al The voltage-gated sodium channel beta3 subunit modulates C6 glioma cell motility independently of channel activity. Biochim Biophys Acta Mol Basis Dis. 2025;1871(6):167844. [DOI] [PubMed] [Google Scholar]
- 34.Kwon GY, Lee S, Hong J, Kim Y, Choi HJ, Yun J, et al. A pilot study of evaluation of a deep-learning-based homologous recombination deficiency assay in Korean patients with ovarian high-grade serous carcinoma: diagnostic performance and clinical implications. Cancer Genet. 2025;298–299:69–77. [DOI] [PubMed] [Google Scholar]
- 35.Shen D, Wu C, Chen M, Zhou Z, Li H, Tong X, et al. Prognosis prediction and drug guidance of ovarian serous cystadenocarcinoma through mitochondria gene-based model. Cancer Genet. 2025;292–293:1–13. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All data generated or analysed during this study are included in this article. Further enquiries can be directed to the corresponding author.