Introduction

Oncology Letters

1792-1074 1792-1082

D.A. Spandidos

10.3892/ol.2025.15173

OL-30-3-15173

Articles

Clinical importance of E-cadherin deficiency in resectable gastric cancer: A nested case-control study

Zheng

Cheng

1 Dai

Min

2 Wang

Dongying

3 Liu

Xingchen

4 Xiang

Zhiyi

3 Xu

Anyi

3 Chen

Ping

1 Wu

Feng

1 Yuan

Yuan

1 Ji

Shengqiang

5 Gu

Lihu

1Department of General Surgery, Ningbo No. 2 Hospital, Ningbo, Zhejiang 315010, P.R. China 2The Second Clinical Medical College, Zhejiang Chinese Medical University, Hangzhou, Zhejiang 310053, P.R. China 3The First Clinical Medical College, Zhejiang Chinese Medical University, Hangzhou, Zhejiang 310053, P.R. China 4Intensive Care Unit, Ningbo No. 2 Hospital, Ningbo, Zhejiang 315010, P.R. China 5Department of Histopathology II, Ningbo Clinical Pathology Diagnosis Center, Ningbo, Zhejiang 315021, P.R. China

Correspondence to: Dr Lihu Gu, Department of General Surgery, Ningbo No. 2 Hospital, 41 Northwest Street, Haishu, Ningbo, Zhejiang 315010, P.R. China, E-mail: gulihuyuanzhi@126.com Dr Shengqiang Ji, Department of Histopathology II, Ningbo Clinical Pathology Diagnosis Center, 685 East Section, Huancheng North Road, Jiangbei, Ningbo, Zhejiang 315021, P.R. China, E-mail: 764647607@qq.com

092025

04072025

30 3

427

19022025 05062025

2025

This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

The prognosis of gastric cancer (GC) remains unsatisfactory despite advancements in diagnosis and treatment. Epithelial-mesenchymal transition-induced decreased cell-cell adhesion, often associated with E-cadherin deficiency, serves a crucial role in tumor invasion and metastasis. However, the relationship between E-cadherin deficiency and GC prognosis is unclear. The present study aimed to explore the association between E-cadherin deficiency and the prognosis of resectable GC. The nested case-control study involved prospective data collection and retrospective analysis. Between January 2013 and December 2022, a total of 1,574 patients treated for GC were eligible for prognostic analysis (104 in the E-cadherin deficiency group and 1,470 in the E-cadherin expression group). Logistic regression analysis was utilized to evaluate the univariate and multivariate associations between clinicopathological factors and E-cadherin deficiency. Propensity score matching analysis at a ratio of 1:4 was performed. Kaplan-Meier curves were employed to analyze the relationship between the prognosis of the E-cadherin deficiency group and the E-cadherin expression group. There were 104 cases of E-cadherin deficiency, accounting for an incidence of 6.6%. The results of the analysis revealed that a family history of GC [odds ratio (OR), 7.60; P<0.001], poorly differentiated tumors (OR, 8.67; P<0.001), perineural invasion (OR, 1.63; P=0.030) and elevated carcinoembryonic antigen (CEA) (OR, 1.83; P=0.034) were independent risk factors for E-cadherin deficiency in all enrolled patients. Following propensity score matching analysis, 86 cases in the E-cadherin deficiency group and 344 cases in the E-cadherin expression group were included. Survival analysis demonstrated no statistically significant difference in 5-year disease-free survival rates between the E-cadherin deficiency and expression groups (P=0.590). Similarly, the 5-year overall survival rates were comparable between the two groups (P=0.863). In summary, E-cadherin deficiency is associated with a family history of GC, poorly differentiated tumors, perineural invasion and elevated CEA levels. However, E-cadherin deficiency does not impact the prognosis of resectable GC.

E-cadherin deficiency resectable GC prognosis

Zhu Xiu Shan Talent Project of Ningbo No. 2 Hospital

2023HMYQ09

Ningbo Clinical Research Center for Digestive System Tumors

2019A21003

Ningbo Medical Key Discipline

2022-B09

The present study was funded by the Zhu Xiu Shan Talent Project of Ningbo No. 2 Hospital (grant no. 2023HMYQ09), the Ningbo Clinical Research Center for Digestive System Tumors (grant no. 2019A21003) and Ningbo Medical Key Discipline (grant no. 2022-B09).

Introduction

The latest global cancer data released by the World Health Organization's International Agency for Research on Cancer in 2020 revealed that gastric cancer (GC) ranks as the fifth most common cancer and the fourth leading cause of cancer-associated mortalities worldwide (1). The occurrence of GC is the result of the long-term effects of a variety of factors, such as Helicobacter pylori infection, diet, smoking and family history. There are obvious regional differences in GC, with the main high-end regions being East Asia, Eastern Europe and some Latin American countries. Notably, China stands out as an area with a high incidence of GC, with projections indicating ~10 million new GC cases and 5.6 million associated deaths between 2021 and 2035 (2). In the United States, the 5-year overall age-standardized relative survival rates of patients with GC were 38.3% (2007–2011), 40.6% (2012–2016) and 42.9% (2017–2021) (3). Despite progress in GC diagnosis and treatment, such as use of the gold-standard gastroscopy treatment and biopsy pathology, the patient prognosis remains unsatisfactory.

GC invasion and metastasis represent the primary causes of mortality. In various malignancies, such as breast cancer and nasopharyngeal carcinoma, epithelial-mesenchymal transition (EMT)-mediated reduction in cell-cell adhesion and enhanced migration properties serve a pivotal role in promoting tumor invasion and migration (3–7). Downregulation of E-cadherin, a key epithelial cell adhesion molecule, has been implicated in enhancing EMT occurrence (8). Decreased E-cadherin expression in breast cancer has been associated with a worse prognosis (9). Notably, while E-cadherin deficiency in patients with GC is relatively low and there are limited related reports, GC is associated with mutations in the E-cadherin gene (CDH1), characterizing hereditary diffuse GC (HDGC), which accounts for 1–2% of GC cases (10). Therefore, the present single-center study aimed to investigate the association between E-cadherin expression and the prognosis of patients with GC.

Patients and methods <sec> <title>Clinical data

A total of 1,574 patients with GC who underwent radical gastrectomy at Ningbo No. 2 Hospital (Ningbo, China) between January 2013 and December 2022 were included in the present study. Among them, the number of men was 1,103 and the number of women was 471. The age ranged from 23 to 92 years, with a median age of 66 years. It is notable that clinical subjects were opportunistically recruited in the present study, with no sample size restrictions and inclusion based only on recruitment time. Due to the rarity of E-cadherin deficiency (~8% of GC), this was the largest sample size available to the research team. Clinical databases were prospectively extracted and retrospectively analyzed. Inclusion criteria comprised histologically confirmed primary gastric adenocarcinoma, no chemotherapy or radiotherapy before surgery, radical gastrectomy, no history of gastrectomy or other malignant tumors, and complete clinical and pathological data. Patient's refusing to participate were excluded. The acquisition and utilization of tissue specimens were approved by the Ethics Committee of Ningbo No. 2 Hospital (approval nos. PJ-NBEY-KY-2019-153-01 and PJ-NBEY-KY-2012-681-08), and patients provided written informed consent.

Immunohistochemical staining and evaluation

Rabbit anti-human E-cadherin antibody (clone number, EP6; Beijing Zhongshan Jinqiao Biotechnology Co., Ltd.) was utilized for immunohistochemical staining. The initial specimens were fixed in 40 g/l formaldehyde for 7 h at 65°C and embedded in paraffin. Specimens from the paraffin embedded block were cut into 4-µm sections. The first section was routinely stained with hematoxylin-eosin for histological diagnosis and subsequent sections were stained for immunohistochemical analysis. E-cadherin immunohistochemical staining was performed using ABC detection system (Vector Laboratories, Inc.). The sections were incubated with a primary antibody against E-Cadherin (clone number, EP6; diluted at 1:200; cat. no. ZA-0565; Beijing Zhongshan Jinqiao Biotechnology Co., Ltd.) for 1.5 h at room temperature. Sections were dewaxed, and rehydrated. Antigen retrieval was performed in 0.1 mM citric buffer with a microwave at 700 W for 5 min. Endogenous peroxidase activity and non-specific bindings were blocked using a 30-min incubation with 3% hydrogen peroxide/methanol followed by a 30-min incubation with 5% goat serum, respectively. The sections were covered with HRP-conjugated secondary antibody at room temperature for 30 min. Thereafter, specimens were stained with 3′,3-diaminobenzidine tetrahydrochloride. The slides were then counterstained with hematoxylin, dehydrated, and evaluated by light microscopy. The results were interpreted via double-blind reading by three pathologists. Any tan-colored staining, indicative of a notable increase in cell membrane expression, was considered a positive signal (Fig. 1A), regardless of nuclear or cytoplasmic staining. E-cadherin-negative was defined as complete loss of membranous expression (Fig. 1B) (11,12).

Follow-up

Patients underwent follow-up every 3–6 months for the first 2 years postoperatively and annually thereafter until mortality or ≥5 years after radical surgery. Adjuvant chemotherapy based on 5-fluorouracil or platinum was recommended for all postoperative patients with stage II–III GC (American Joint Committee on Cancer/International Union against Cancer Tumor-Node-Metastasis classification system, 8th Edition) (13). Follow-up assessments included physical examinations, laboratory tests (complete blood count, liver function tests, tumor markers) and imaging tests [chest X-ray, B-scan ultrasound, gastroscopy, computed tomography (CT), magnetic resonance imaging or positron emission tomography-CT] to detect recurrence or metastasis. Disease-free survival (DFS) was defined as the time from surgery to locoregional recurrence, distant recurrence or mortality, while overall survival (OS) was defined as the time from the start of randomization to mortality from any cause. The median follow-up for the entire cohort was 52 months (range, 1–122 months), with follow-up of all patients included in the present study closing in October 2023.

Statistical analysis

Demographic characteristics of the E-cadherin-negative group (E-cadherin deficiency) and the E-cadherin-positive group were compared and analyzed. Continuous variables were compared using independent-samples t-test or Wilcoxon's rank-sum test, while categorical variables were compared using Pearson's χ² test or Fisher's exact test as appropriate. Logistic regression analysis was employed to evaluate the univariate and multivariate associations between clinicopathological factors and E-cadherin deficiency. Odds ratios (ORs) and 95% confidence intervals were calculated. To mitigate selection bias, 1:4 propensity score matching analysis was performed for the E-cadherin-negative and E-cadherin-positive groups. Kaplan-Meier curves were utilized to estimate DFS and OS, with the log-rank test determining significance. Propensity scores were generated using a logistic regression model incorporating covariates such as age, sex, tumor location, tumor size, perineural invasion, lymphovascular invasion, pathological tumor (pT) category, pathological lymph node (pN) category, number of lymph node dissections, carcinoembryonic antigen (CEA) and adjuvant chemotherapy. Individuals in the E-cadherin-negative group were matched to those in the E-cadherin-positive group based on these propensity scores. All statistical tests were two-tailed, with P<0.05 considered to indicate a statistically significant difference. Statistical analysis was performed using SPSS software (version 25.0; IBM Corp.).

Results <sec> <title>Patient recruitment

Between January 2013 and December 2022, a total of 2,297 gastrectomies were performed at Ningbo No. 2 Hospital. Following application of the inclusion criteria, 723 patients were excluded, resulting in the analysis of 1,574 patients (104 in the E-cadherin-negative group and 1,470 in the E-cadherin-positive group). E-cadherin negativity was observed in 6.6% of cases (Fig. 2).

Clinicopathological factors and E-cadherin

E-cadherin deficiency was potentially associated with a family history of GC, tumor size, histological type, perineural invasion, pT category and pN category in GC (all P<0.05; Table I). Univariate logistic regression indicated potential associations between E-cadherin deficiency and a family history of GC, histological type, perineural invasion, pT category, pN category and CEA levels in GC (all P<0.05; Table II). Further multivariate logistic regression identified a family history of GC (OR, 7.60; P<0.001), poorly differentiated histology (OR, 8.67; P<0.001), perineural invasion (OR, 1.63; P=0.030) and elevated CEA levels (OR, 1.83; P=0.034) as independent risk factors for E-cadherin deficiency in all enrolled patients (Table III).

Propensity score matching analysis

In a 1:4 match analysis, age, sex, tumor location, tumor size, perineural invasion, lymphovascular invasion, pT category, pN category, number of lymph node dissections, CEA levels and adjuvant chemotherapy were matched. After matching, 86 patients in the E-cadherin deficiency group and 344 patients in the E-cadherin-positive expression group were analyzed. No significant differences were observed in the covariates of age, sex, tumor location, tumor size, perineural invasion, lymphovascular invasion, pT category, pN category, number of lymph node dissections, CEA levels and adjuvant chemotherapy (all P>0.05; Table IV). Kaplan-Meier analysis revealed no statistically significant differences in 5-year DFS rates between the E-cadherin deficiency group and the E-cadherin-positive expression group (P=0.590; Fig. 3). Similarly, no significant differences were observed in 5-year OS rates between the two groups (P=0.863; Fig. 4).

Discussion

GC histology samples from the Human Protein Atlas demonstrated that E-cadherin deficiency accounted for 8% of GC cases (14). The present study, encompassing 1,574 patients with resectable GC, revealed a proportion of E-cadherin deficiency at 6.6%, which is consistent with the results of the literature. However, another study from China reported that E-cadherin deficiency accounted for 34.5% of patients with GC (15). This result potentially differs from the present study due to inclusion of a limited sample size of 84 patients who had already undergone GC surgery. E-cadherin deficiency is a molecular marker event of GC, and its clinical value is reflected in genetic counseling (CDH1 mutation detection), pathological classification, prognostic stratification and individualized treatment decisions.

The prognosis of resectable GC may be associated with factors such as age, sex, tumor size, depth of invasion, lymph node metastasis and histological type (16). Older age (≥70 years) is an independent risk factor for postoperative complications (17); as some older patients do not receive systemic therapy, they may not experience a survival benefit from adjuvant chemotherapy or radiation therapy (18). However, Hoffman et al (19) reported that postoperative radiotherapy and chemotherapy did not markedly improve survival in older patients with gastric adenocarcinoma resection. It has also been suggested that younger patients have a higher mortality rate compared with older patients, as GC in younger patients exhibits more aggressive features and may be more advanced (20). In summary, the difference in prognosis of GC with age is controversial. A study by Suh et al (21) demonstrated no notable difference between male and female patient outcomes after resection for early-stage GC in younger patients. However, older men have worse OS rate compared with older women, as they generally have more comorbidities compared with women of the same age (22,23). Currently, there are no previous studies that suggest that sex is an independent factor in the prognosis of GC.

Del Rio et al (24) showed that, in patients with GC, tumor size is associated with the degree of malignancy of the tumor and it is a notable predictor of survival. There is a positive association between tumor size and lymph node metastasis, with larger tumors likely to lead to a worse prognosis by promoting lymph node metastasis. A study by Lee et al (25) found that undifferentiated GC was associated with lymph node metastasis, which may also be associated with GC prognosis. T, N and metastasis (M) stages are recognized as the most important independent prognostic factors for GC (26,27). TNM staging for GC remains the most important basis for the selection of clinical treatment strategies and the judgment of clinical prognosis (28).

The expression of E-cadherin has an effect on the prognosis of malignant tumors (1). E-cadherin downregulation leads to E-cadherin-mediated signaling pathway dysfunction, thereby changing cell polarity, increasing cell viability and promoting the EMT process, as well as cell invasion and migration (29). Inactivation of E-cadherin contributes to EMT and tumor progression through a number of signaling pathways, including Wnt signaling, Rho GTPases and epidermal growth factor receptor (EGFR) (30). E-cadherin also has a growth-inhibiting function, inhibiting cell-cell contact and inducing cell cycle arrest by upregulating cyclin-dependent kinase inhibitor 1B (31). E-cadherin deficiency leads to the loss of this growth-inhibiting function, and tumors are more likely to invade and metastasize (3). Chronic inflammation caused by bacterial infection erodes the gastric mucosa, resulting in its eventual regeneration by bone marrow-derived cells, leading to GC through metaplasia-dysplasia sequences. However, this phenomenon may be limited to cancer types that arise after extensive destruction of inflammatory tissue, such as stomach ulcers. It has been established that Helicobacter pylori infection leads to NF-κB activation in the gastric epithelium. NF-κB regulates the phenotype of epithelial cells during inflammation and contributes to inflammation-associated carcinogenesis (32). E-cadherin has the ability to downregulate NF-κB activity, thereby reducing NF-κB-mediated carcinogenic effects (32,33). Thus, E-cadherin deficiency leads to NF-κB activation (4). E-cadherin acts as a tumor suppressor, and downregulation of E-cadherin has been observed in various cancer types, such as breast cancer and cutaneous melanoma (34). Somatic mutations of the the E-cadherin gene CDH1 have been found in sporadic diffuse GC (35). Gene mutations are a major mechanism of silencing tumor suppressor genes. In addition to the downregulation of E-cadherin induced by CDH1 gene mutations, epigenetic factors can also regulate the expression of E-cadherin. DNA methylation is a major type of epigenetic alteration, and promoter hypermethylation often serves as a secondary mechanism for silencing genes (36). This may also be associated with the prognosis of GC.

The present study demonstrated that E-cadherin deficiency in patients with resectable GC is associated with a family history of GC, poorly differentiated tumors, perineural invasion and elevated CEA levels. Germline mutations in CDH1 are closely associated with HDGC, leading to a higher incidence of E-cadherin deficiency in patients with a family history of GC (37). Apart from CDH1 gene mutations, factors such as loss of heterozygosity at E-cadherin chromosomal loci, CDH1 promoter hypermethylation, transcriptional repression and post-translational modifications (such as abnormal glycosylation) contribute to E-cadherin deficiency (38). For example, the study by Tedaldi et al (39) utilized next-generation sequencing (NGS) analysis to demonstrate different methylation patterns in the regulatory region of the CDH1 enhancer, serving a crucial role in gene expression regulation and subsequent E-cadherin deficiency. NGS analysis holds promise for diagnosing family members of patients with familial GC, enabling them to undergo preventive total gastrectomy (39,40).

A study by Chen et al (15) indicated a notable association between E-cadherin expression and GC differentiation. E-cadherin expression is more frequently lost in signet ring cell and mucinous adenocarcinomas, while it is often maintained in GC with glandular or nodular formation. Previous studies have identified E-cadherin gene mutations in signet ring cancer, disrupting the Adhesome and facilitating interactions between human EGFR2 and mucin 4, leading to the formation of signet ring cell carcinoma (41,42). The present study further demonstrates the association between poorly differentiated tumors and the prognosis of GC. Studies across different cancer types, including breast, endometrial and prostate cancer, have demonstrated that the reduction or loss of E-cadherin expression is associated with poor differentiation (43–45). This may be attributed to E-cadherin deficiency contributing to tumor dedifferentiation, thereby enhancing invasiveness. Key regulatory transcription factors of EMT, such as Snail, Slug, transcription factor 3 and Smad interacting protein 1/zinc finger E-box binding homeobox 2, are actively involved in inhibiting junction proteins and desmosomes, which are responsible for epithelial cell differentiation (46,47). Previous studies have not extensively explored the associations between E-cadherin deficiency and perineural invasion in GC. We hypothesize that E-cadherin deficiency may enhance the invasiveness of GC, facilitating perineural invasion. Notably, prior studies have not established a direct association between CEA and E-cadherin deficiency. CEA tends to enhance tumor aggressiveness (48), and concurrently, the aggressiveness of E-cadherin deficiency is higher. Therefore, based on the present findings, it is plausible that the two factors may exhibit a synergistic effect. In conclusion, the present study is the first to report the association of CEA elevation with E-cadherin deficiency in GC.

A meta-analysis of 26 studies demonstrated a notable association between decreased E-cadherin expression and poor prognosis in patients with GC (49). E-cadherin serves a multifaceted role in cancer biology, not only influencing invasion and metastasis but also functioning as a tumor suppressor in various cancer types, including GC (50,51). However, the present results indicated no significant association between E-cadherin deficiency and the prognosis of resectable GC. This disparity may stem from differences in study focus, with the present study investigation centering on E-cadherin deficiency rather than decreased expression levels. Additionally, the heterogeneity in thresholds for decreased E-cadherin expression across the original studies in the meta-analysis, ranging from 5 to 90%, coupled with predominantly univariate analyses, contributes to heterogeneity. Consequently, while the meta-analysis suggested a worse prognosis in patients with GC with reduced E-cadherin expression, the confidence in these findings remains tentative. Thus, the association between E-cadherin expression and prognosis in patients with GC remains controversial.

The present study highlights the contentious nature of the associations between four factors associated with E-cadherin deficiency and the prognosis of GC, namely family history of GC, poorly differentiated tumors, perineural invasion and elevated CEA levels. While hereditary cases associated with known cancer susceptibility syndromes or genetic causes account for a small proportion of all GC cases (52), their prognostic implications compared with sporadic GC remain unproven due to limited case numbers. The prognostic importance of histological type in GC has been debated, with numerous studies suggesting it lacks independent prognostic value (26,53). Although perineural invasion is a critical pathway for cancer metastasis, its status as an independent prognostic factor for GC remains unclear (54). Similarly, the association between GC prognosis and elevated CEA levels is disputed. While CEA serves as a vital tumor marker, elevated serum levels may also occur in benign conditions. Numerous patients with advanced metastatic GC may not exhibit elevated CEA levels (55,56). In conclusion, the present study found no significant association between E-cadherin deficiency and the prognosis of resectable GC. If there are more studies on E-cadherin deficiency in the future, a relevant meta-analysis will be performed.

While the present study provides valuable insights into E-cadherin deficiency in patients with GC in China, it has several limitations. Firstly, the low prevalence of E-cadherin deficiency in clinical practice limits the sample size. Additionally, the study lacks sufficient follow-up time. There are five internationally recognized patterns of GC recurrence: Local recurrence, lymph node metastases, peritoneal metastases, hematogenous metastases and mixed recurrence (57). The present study classified the prognosis and a total of 163 out of 430 patients had relapsed. According to the five recognized patterns of GC recurrence, 5 of the cases were local recurrence, 18 were lymph node metastases, 128 were peritoneal metastases, 87 were hematogenous metastases and the remaining patients were mixed recurrences. The primary outcome of the present study was DFS. Due to the sample size, it was not possible to further classify specific recurrence patterns for statistical analysis. Moreover, the cases included were all of patients with resectable GC in stages I–III. After propensity matching, there were 86 patients with E-cadherin deficiency in the study. Among them, there were 14 patients with stage I disease, and none of them relapsed. There were 18 patients with stage II, of whom 6 had recurrence, and there were 54 in stage III, of whom 26 were recurrent. Due to sample size limitations, further stratification analysis could not be performed (especially in patients with stage I). If more samples become available in the future the results should be updated. Finally, E-cadherin deficiency was defined solely based on immunohistochemistry, overlooking other mechanisms such as CDH1 gene mutations, promoter hypermethylation, transcriptional repression and post-translational modifications. Although NGS analysis is recommended for comprehensive gene analysis associated with E-cadherin deficiency, financial and technical constraints precluded further analysis in the present study.

In conclusion, the present study demonstrated that the proportion of E-cadherin deficiency in resectable GC in China may be ~6.6% based on data from one hospital. E-cadherin deficiency was associated with a family history of GC, poorly differentiated tumors, perineural invasion and elevated CEA levels. However, the present study did not confirm a significant association between E-cadherin deficiency and the prognosis of GC.

Acknowledgements

Not applicable.

Availability of data and materials

The data generated in the present study may be requested from the corresponding author.

Authors' contributions

CZ and MD conceptualized the study and designed the experimental methods. DW, XL and ZX collected and analyzed the data. AX and PC performed the follow-up of the patients. FW and YY analyzed data. SJ and LG were responsible for interpretation of the data. CZ was responsible for writing the original draft and editing the figures. MD and DW reviewed and edited the manuscript. LG supervised the study and contributed to the revision of the manuscript. CZ, MD, DW, XL, ZX, AX, PC, FW, YY, SJ and LG confirm the authenticity of all the raw data. All authors read and approved the final manuscript.

Ethics approval and consent to participate

In accordance with the Declaration of Helsinki, the studies involving humans were approved by the Ethics Committee of Ningbo No. 2 Hospital (approval nos. PJ-NBEY-KY-2019-153-01 and PJ-NBEY-KY-2012-681-08). The participants provided their written informed consent to participate in the study.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Abbreviations

EMT

epithelial-mesenchymal transition

gastric cancer

CEA

carcinoembryonic antigen

HDGC

hereditary diffuse gastric cancer

computed tomography

DFS

disease-free survival

overall survival

odds ratio

confidence interval

NGS

next-generation sequencing

References 1

Ferlay

Colombet

Soerjomataram

Parkin

Piñeros

Znaor

Bray

Cancer statistics for the year 2020: An overview

Int J CancerApril52021(Epub ahead of print)

10.1002/ijc.33588

Wang

Han

Xue

Zhao

Chen

Yang

Jiang

Nationwide gastric cancer prevention in China, 2021–2035: A decision analysis on effect, affordability and cost-effectiveness optimisation

Gut71239124002022

10.1136/gutjnl-2021-325948

Feng

Zheng

Chen

Lyu

Recent estimates and predictions of 5-year survival in patients with gastric cancer: A model-based period analysis

Cancer Control29107327482210992272022

10.1177/10732748221099227

Zhou

Xia

Lin

Wang

Oyang

Tan

Tian

Wang

Cao

Liao

Exosomes in nasopharyngeal carcinoma

J Cancer97677772018

10.7150/jca.22505

29581754

Loh

Chai

Tang

Wong

Sethi

Shanmugam

Chong

Looi

The E-Cadherin and N-Cadherin Switch in Epithelial-to-Mesenchymal transition: Signaling, therapeutic implications, and challenges

Cells811182019

10.3390/cells8101118

31547193

Maeda

Johnson

Wheelock

Cadherin switching: Essential for behavioral but not morphological changes during an epithelium-to-mesenchyme transition

J Cell Sci1188738872005

10.1242/jcs.01634

Park

Kim

Yoon

Breast cancer metastasis: Mechanisms and therapeutic implications

Int J Mol Sci2368062022

10.3390/ijms23126806

Sharaireh

Fitzpatrick

Ward

McKay

Unwin

Epithelial cadherin regulates transition between the naïve and primed pluripotent states in mouse embryonic stem cells

Stem Cells38129213062020

10.1002/stem.3249

32621788

Corso

Figueiredo

De Angelis

Corso

Girardi

Pereira

Seruca

Bonanni

Carneiro

Pravettoni

E-cadherin deregulation in breast cancer

J Cell Mol Med24593059362020

10.1111/jcmm.15140

32301282

Fitzgerald

Caldas

E-cadherin mutations and hereditary gastric cancer: Prevention by resection?

Dig Dis2023312002

10.1159/000063162

12145417

Setia

Agoston

Han

Mullen

Duda

Clark

Deshpande

Mino-Kenudson

Srivastava

Lennerz

A protein and mRNA expression-based classification of gastric cancer

Mod Pathol297727842016

10.1038/modpathol.2016.55

Zhao

Feng

Zang

Huang

Zhou

Nie

Protein expression-based classification of gastric cancer by immunohistochemistry of tissue microarray

PLoS One15e02388362020

10.1371/journal.pone.0238836

33095797

Sano

Coit

Kim

Roviello

Kassab

Wittekind

Yamamoto

Ohashi

Proposal of a new stage grouping of gastric cancer for TNM classification: International Gastric Cancer Association staging project

Gastric Cancer202172252017

10.1007/s10120-016-0601-9

26897166

Caldeira

Figueiredo

Brás-Pereira

Carneiro

Moreira

Pinto

Relvas

Carneiro

Barbosa

Casares

E-cadherin-defective gastric cancer cells depend on Laminin to survive and invade

Hum Mol Genet24589159002015

10.1093/hmg/ddv312

Chen

Chu

Hsu

Lai

Tseng

Chou

Huang

Tseng

Hsiao

Loss of E-cadherin expression correlates with poor differentiation and invasion into adjacent organs in gastric adenocarcinomas

Cancer Lett201971062003

10.1016/j.canlet.2003.07.007

Yang

Zhao

Wang

Guo

Liu

Updates on global epidemiology, risk and prognostic factors of gastric cancer

World J Gastroenterol29245224682023

10.3748/wjg.v29.i16.2452

Shin

Suh

Factors related to morbidity in elderly gastric cancer patients undergoing gastrectomies

J Gastric Cancer141731792014

10.5230/jgc.2014.14.3.173

25328762

Song

Jiang

Liu

Cao

Guan

Age-specific effects on the prognosis after surgery for gastric cancer: A SEER population-based analysis

Oncotarget748614486242016

10.18632/oncotarget.9548

Hoffman

Neville

Mamon

Kachnic

Katz

Earle

Punglia

Adjuvant therapy for elderly patients with resected gastric adenocarcinoma: Population-based practices and treatment effectiveness

Cancer1182482572012

10.1002/cncr.26248

21692071

Seo

Jin

Yoon

Shin

Park

Kim

Jung

Lee

Clinicopathologic and molecular features associated with patient age in gastric cancer

World J Gastroenterol21690569132015

10.3748/wjg.v21.i22.6905

Suh

Yook

Kim

Differences in the prognosis of early gastric cancer according to sex and age

Therap Adv Gastroenterol102192292017

10.1177/1756283X16681709

Lim

Lee

Shin

Kim

Park

Jung

Song

Clinicopathological features and surgical safety of gastric cancer in elderly patients

J Korean Med Sci29163916452014

10.3346/jkms.2014.29.12.1639

Lee

Kim

Hong

Alshomimi

Cheong

Hyung

Noh

Kim

Clinical significance of the prognostic nutritional index for predicting Short- and Long-term surgical outcomes after Gastrectomy: A retrospective analysis of 7781 gastric cancer patients

Medicine (Baltimore)95e35392016

10.1097/MD.0000000000003539

27149460

Del Rio

Viani

Bertocchi

Iapichino

Luzietti

Dell'Abate

Sianesi

The prognostic role of tumor size in patients with gastric cancer

Ann Ital Chir884784842017

29339593

Lee

Choi

Han

Kim

Ryu

Yoon

Eom

Kim

Lee

Cho

Risk of lymph node metastasis in differentiated type mucosal early gastric cancer mixed with minor undifferentiated type histology

Ann Surg Oncol22181318192015

10.1245/s10434-014-4167-7

Tan

Wang

Yin

Zhang

Hou

Thirty-year trends in clinicopathologic characteristics and prognosis after gastrectomy for gastric cancer: A single institution in Northern China

J Cancer11105610622020

10.7150/jca.36927

31956352

Kulig

Pach

Majewska

Kulig

Clinicopathological prognostic factors determining outcomes of treatment in gastric cancer surgery

In Vivo36292729352022

10.21873/invivo.13035

36309397

Chen

Chi

Lin

Cheng

Xue

Long

Jiang

Dong

Sui

Prognostic and predictive value of a pathomics signature in gastric cancer

Nat Commun1369032022

10.1038/s41467-022-34703-w

36371443

Van Aken

De Wever

Correia da Rocha

Mareel

Defective E-cadherin/catenin complexes in human cancer

Virchows Arch4397257512001

10.1007/s004280100516

11787845

Cavallaro

Christofori

Cell adhesion and signalling by cadherins and Ig-CAMs in cancer

Nat Rev Cancer41181322004

10.1038/nrc1276

14964308

St Croix

Sheehan

Rak

Flørenes

Slingerland

Kerbel

E-Cadherin-dependent growth suppression is mediated by the cyclin-dependent kinase inhibitor p27(KIP1)

J Cell Biol1425575711998

10.1083/jcb.142.2.557

Karin

Greten

NF-kappaB: Linking inflammation and immunity to cancer development and progression

Nat Rev Immunol57497592005

10.1038/nri1703

Cowell

Yan

Eiseler

Leightner

Döppler

Storz

Loss of cell-cell contacts induces NF-kappaB via RhoA-mediated activation of protein kinase D1

J Cell Biochem1067147282009

10.1002/jcb.22067

19173301

Berx

van Roy

Involvement of members of the cadherin superfamily in cancer

Cold Spring Harb Perspect Biol1a0031292009

10.1101/cshperspect.a003129

Becker

Atkinson

Reich

Becker

Nekarda

Siewert

Höfler

E-cadherin gene mutations provide clues to diffuse type gastric carcinomas

Cancer Res54384538521994

8033105

Strathdee

Epigenetic versus genetic alterations in the inactivation of E-cadherin

Semin Cancer Biol123733792002

10.1016/S1044-579X(02)00057-3

Caldas

Carneiro

Lynch

Yokota

Wiesner

Powell

Lewis

Huntsman

Pharoah

Jankowski

Familial gastric cancer: Overview and guidelines for management

J Med Genet368738801999

van Roy

Berx

The cell-cell adhesion molecule E-cadherin

Cell Mol Life Sci65375637882008

10.1007/s00018-008-8281-1

Tedaldi

Molinari

São José

Barbosa-Matos

André

Danesi

Arcangeli

Ravegnani

Saragoni

Morgagni

Genetic and epigenetic alterations of CDH1 regulatory regions in hereditary and sporadic gastric cancer

Pharmaceuticals (Basel)144572021

10.3390/ph14050457

34066170

Gjyshi

Vashi

Seewald

Kohan

Abboud

Fowler

Suppiah

Halabi

Therapeutic and prophylactic gastrectomy in a family with hereditary diffuse gastric cancer secondary to a CDH1 mutation: A case series

World J Surg Oncol161432018

10.1186/s12957-018-1415-5

Fukui

Mechanisms behind signet ring cell carcinoma formation

Biochem Biophys Res Commun450123112332014

10.1016/j.bbrc.2014.07.025

25019985

Hamilton

Jones

Church

Medlicott

Synchronous appendiceal and intramucosal gastric signet ring cell carcinomas in an individual with CDH1-associated hereditary diffuse gastric carcinoma: A case report of a novel association and review of the literature

BMC Gastroenterol131142013

10.1186/1471-230X-13-114

Gamallo

Palacios

Suarez

Pizarro

Navarro

Quintanilla

Cano

Correlation of E-cadherin expression with differentiation grade and histological type in breast carcinoma

Am J Pathol1429879931993

Romero-Pérez

López-García

MÁ

Díaz-Martín

Biscuola

Castilla

MÁ

Tafe

Garg

Oliva

Matias-Guiu

Soslow

Palacios

ZEB1 overexpression associated with E-cadherin and microRNA-200 downregulation is characteristic of undifferentiated endometrial carcinoma

Mod Pathol26151415242013

10.1038/modpathol.2013.93

Umbas

Isaacs

Bringuier

Schaafsma

Karthaus

Oosterhof

Debruyne

Schalken

Decreased E-cadherin expression is associated with poor prognosis in patients with prostate cancer

Cancer Res54392939331994

7518346

Moreno-Bueno

Cubillo

Sarrió

Peinado

Rodríguez-Pinilla

Villa

Bolós

Jordá

Fabra

Portillo

Genetic profiling of epithelial cells expressing E-cadherin repressors reveals a distinct role for Snail, Slug, and E47 factors in epithelial-mesenchymal transition

Cancer Res66954395562006

10.1158/0008-5472.CAN-06-0479

17018611

Vandewalle

Comijn

De Craene

Vermassen

Bruyneel

Andersen

Tulchinsky

Van Roy

Berx

SIP1/ZEB2 induces EMT by repressing genes of different epithelial cell-cell junctions

Nucleic Acids Res33656665782005

10.1093/nar/gki965

16314317

Beauchemin

Arabzadeh

Carcinoembryonic antigen-related cell adhesion molecules (CEACAMs) in cancer progression and metastasis

Cancer Metastasis Rev326436712013

10.1007/s10555-013-9444-6

23903773

Xing

Tang

Yuan

Wang

Yang

Chen

The prognostic value of E-cadherin in gastric cancer: A meta-analysis

Int J Cancer132258925962013

10.1002/ijc.27947

23169395

Jeanes

Gottardi

Yap

Cadherins and cancer: How does cadherin dysfunction promote tumor progression?

Oncogene27692069292008

10.1038/onc.2008.343

19029934

Perl

Wilgenbus

Dahl

Semb

Christofori

A causal role for E-cadherin in the transition from adenoma to carcinoma

Nature3921901931998

10.1038/32433

9515965

Gullo

Grillo

Mastracci

Vanoli

Carneiro

Saragoni

Limarzi

Ferro

Parente

Fassan

Precancerous lesions of the stomach, gastric cancer and hereditary gastric cancer syndromes

Pathologica1121661852020

10.32074/1591-951X-166

Park

Jang

Kim

Park

Kim

Mok

Kim

Gastric cancer histology: Clinicopathologic characteristics and prognostic value

J Surg Oncol985205252008

10.1002/jso.21150

Marchesi

Piemonti

Mantovani

Allavena

Molecular mechanisms of perineural invasion, a forgotten pathway of dissemination and metastasis

Cytokine Growth Factor Rev2177822010

10.1016/j.cytogfr.2009.11.001

20060768

Ruibal Morell

CEA serum levels in non-neoplastic disease

Int J Biol Markers71601661992

10.1177/172460089200700307

1431339

Zhang

Cui

Song

Jiang

The modified Glasgow prognostic score predicts survival in gastric cancer patients with normal CEA and CA19-9

Can J Gastroenterol Hepatol202239530042022

10.1155/2022/3953004

Nakanishi

Ohara

Domen

Shichinohe

Hirano

Ishizaka

Differences in risk factors between patterns of recurrence in patients after curative resection for advanced gastric carcinoma

World J Surg Oncol11982013

10.1186/1477-7819-11-98

Figure 1.

Expression of E-cadherin in resectable gastric cancer tissues (magnification, ×200). (A) E-cadherin-positive group. (B) E-cadherin-negative group.

Figure 1. Expression of E–cadherin in resectable gastric cancer tissues (magnification, ×200). (A) E–cadherin–positive group. (B) E–cadherin–negative group.

Figure 2.

Flow diagram of selection of patients with gastric cancer.

Figure 2. Flow diagram of selection of patients with gastric cancer.

Figure 3.

Comparison of curves of DFS between patients with E-cadherin deficiency and E-cadherin-positive expression in resectable gastric cancer after propensity score matching (P=0.590). DFS, disease-free survival.

Figure 3. Comparison of curves of DFS between patients with E–cadherin deficiency and E–cadherin–positive expression in resectable gastric cancer after propensity score matching (P=0.590). DFS, diseas...

Figure 4.

Comparison of curves of OS between patients with E-cadherin deficiency and E-cadherin-positive expression in resectable gastric cancer after propensity score matching (P=0.863). OS, overall survival.

Figure 4. Comparison of curves of OS between patients with E–cadherin deficiency and E–cadherin–positive expression in resectable gastric cancer after propensity score matching (P=0.863). OS, overall ...

Table I.

Baseline clinicopathological characteristics of the enrolled patients with or without E-cadherin expression.

Clinicopathological feature	No E-cadherin expression (n=104)	Positive E-cadherin expression (n=1,470)	P-value
Age, years	64.5±9.3	65.5±9.6	0.315
Sex			0.508
Male	70 (67.3)	1,033 (70.3)
Female	34 (32.7)	437 (29.7)
Family history of gastric cancer			<0.001
Absence	93 (89.4)	1,447 (98.4)
Presence	11 (10.6)	23 (1.6)
Tumor location			0.692
Upper third	14 (13.5)	172 (11.7)
Middle third	18 (17.3)	258 (17.6)
Lower third	68 (65.4)	1,007 (68.5)
Two-thirds or more	4 (3.8)	33 (2.2)
Type of gastrectomy			0.708
Distal subtotal	76 (73.1)	1,091 (74.2)
Total	28 (26.9)	371 (25.2)
Proximal subtotal	0 (0.0)	8 (0.6)
Tumor size, cm	4.5±2.3	4.0±2.3	0.046
Histological type			<0.001
Differentiated	10 (9.6)	759 (51.6)
Undifferentiated	94 (90.4)	711 (48.4)
Perineural invasion			<0.001
Absence	36 (34.6)	815 (55.4)
Presence	68 (65.4)	655 (44.6)
Lymphovascular invasion			0.123
Absence	36 (34.6)	626 (42.6)
Presence	68 (65.4)	844 (57.4)
pT category			<0.001
T1	16 (15.4)	402 (27.3)
T2	6 (5.8)	209 (14.2)
T3	8 (7.7)	198 (13.5)
T4a	70 (67.3)	609 (41.4)
T4b	4 (3.8)	52 (3.6)
pN category			0.005
N0	34 (32.7)	638 (43.4)
N1	16 (15.4)	260 (17.7)
N2	18 (17.3)	255 (17.3)
N3a	22 (21.2)	250 (17.0)
N3b	14 (13.4)	67 (4.6)
Number of lymph node dissections	24.7±10.0	25.1±9.2	0.644
Carcinoembryonic antigen level^a			0.050
Normal	82 (82.0)	1,253 (88.9)
Abnormal	18 (18.0)	156 (11.1)

Data are presented as n (%) or mean ± SD.

4 and 61 patients in the no E-cadherin and positive E-cadherin expression groups, respectively, did not have CEA examination data. pT, pathological tumor stage; pN, pathological lymph node stage.

Table II.

Univariate regression analysis of clinicopathological factors associated with E-cadherin deficiency.

Clinicopathological feature	OR	95% CI	P-value
Age, years
≤60	1.00
>60	0.74	0.49–1.12	0.156
Sex
Male	1.00
Female	1.15	0.75–1.76	0.524
Family history of gastric cancer
Absence	1.00
Presence	7.50	3.30–16.90	<0.001
Tumor location
Upper third	1.00
Middle third	0.86	0.42–1.77	0.677
Lower third	0.83	0.46–1.51	0.540
Two-thirds or more	1.49	0.46–4.81	0.505
Tumor size, cm
≤5	1.00
>5	1.24	0.80–1.92	0.345
Histological type
Differentiated	1.00
Undifferentiated	10.04	5.19–19.41	<0.001
Perineural invasion
Absence	1.00
Presence	2.35	1.55–3.57	<0.001
Lymphovascular invasion
Absence	1.00
Presence	1.40	0.92–2.13	0.113
pT category
T1	1.00
T2	0.72	0.28–1.87	0.502
T3	1.02	0.43–2.41	0.973
T4a	2.89	1.65–5.04	<0.001
T4b	1.93	0.62–6.00	0.254
pN category
N0	1.00
N1	1.16	0.63–2.13	0.645
N2	1.33	0.74–2.39	0.350
N3a	1.65	0.95–2.88	0.077
N3b	3.92	2.00–7.67	<0.001
Carcinoembryonic antigen level
Normal	1.00
Abnormal	1.76	1.03–3.02	0.038

pT, pathological tumor stage; pN, pathological lymph node stage; OR, odds ratio; CI, confidence interval.

Table III.

Multivariate regression analysis of clinicopathological factors associated with E-cadherin deficiency.

Clinicopathological feature	OR	95% CI	P-value
Family history of gastric cancer
Absence	1.00
Presence	7.60	3.30–17.30	<0.001
Histological type
Differentiated	1.00
Undifferentiated	8.67	4.44–16.94	<0.001
Perineural invasion
Absence	1.00
Presence	1.63	1.05–2.53	0.030
Carcinoembryonic antigen level
Normal	1.00
Abnormal	1.83	1.05–3.21	0.034

OR, odds ratio; CI, confidence interval.

Table IV.

Baseline characteristics of patients with or without E-cadherin expression after 1:4 propensity score matching.

Clinicopathological feature	Noe E-cadherin expression (n=86)	Positive E-cadherin expression (n=344)	P-value
Age, years	64.0±9.6	63.9±9.9	0.959
Sex			0.897
Male	58 (67.4)	236 (68.6)
Female	28 (32.6)	108 (31.4)
Tumor location			0.264
Upper third	12 (14.0)	44 (12.8)
Middle third	14 (16.3)	53 (15.4)
Lower third	56 (65.1)	243 (70.6)
Two-thirds or more	4 (4.7)	4 (1.2)
Tumor size, cm	4.5±2.3	4.4±2.2	0.533
Perineural invasion			0.389
Absence	30 (34.9)	139 (40.4)
Presence	56 (65.1)	205 (59.6)
Lymphovascular invasion			0.801
Absence	32 (37.2)	121 (35.2)
Presence	54 (62.8)	223 (64.8)
pT category			0.542
T1	12 (14.0)	47 (13.7)
T2	6 (7.0)	33 (9.6)
T3	4 (4.7)	26 (7.6)
T4a	60 (69.8)	212 (61.6)
T4b	4 (4.7)	26 (7.6)
pN category			0.234
N0	28 (32.6)	83 (24.1)
N1	16 (18.6)	56 (16.3)
N2	14 (16.3)	73 (21.2)
N3a	16 (18.6)	94 (27.3)
N3b	12 (14.0)	38 (11.0)
Number of lymph node dissections	24.2±8.4	25.8±9.4	0.133
Carcinoembryonic antigen level^a			0.285
Normal	70 (83.3)	291 (87.7)
Abnormal	14 (16.7)	41 (12.3)
Adjuvant chemotherapy			0.496
No	20 (23.3)	95 (27.6)
Yes	66 (76.7)	249 (72.4)

Data are presented as n (%) or mean ± SD.

2 and 12 patients in the no E-cadherin and positive E-cadherin expression groups, respectively, did not have CEA examination data. pT, pathological tumor stage; pN, pathological lymph node stage.