Background

Lung cancer has been the most frequently diagnosed cancer and leading cause of cancer mortality in the USA and other developed countries [1]. Non-small cell lung carcinoma (NSCLC) consists of squamous cell carcinoma (SCC), adenocarcinoma (AC), large cell carcinoma, and others. NSCLC accounts for approximately 85 % of all lung cancers, and the prognosis remains poor [2]. Therefore, investigation on the mechanism of initiation and progression and identification of cancer risk marker are still needed for individualized treatment and better prognosis of NSCLC patients. Epigenetic regulation of tumor suppressor gene expression plays an important role in carcinogenesis. Aberrant methylation of tumor suppressor genes is a commonly observed epigenetic regulation in human tumors including NSCLC [3–5]. Patterns of DNA methylation can classify NSCLC into two phenotypically distinct subtypes of tumors and provide proof of principle that differences in DNA methylation can be used as a platform for predictive biomarker discovery and development [6]. Thus, measure of aberrant gene promoter methylation as a tool for diagnosis of tumors has been widely utilized for many different tumors including NSCLC [7].

Human epithelial cadherin (E-cadherin), a member of transmembrane glycoprotein family, also known as Cadherin-1 (CDH1), CAM 120/80, or uvomorulin, is encoded by the E-cadherin gene [8]. E-cadherin plays a key role in cell-cell adhesion, adherent junction in normal epithelial tissues, contributing to tissue differentiation and homeostasis [9, 10]. Reduced E-cadherin expression was often detected and associated with cancer invasion and metastasis in a variety of epithelial tumors [11–13]. E-cadherin methylation which is associated with the low and absent E-cadherin expression was detected in several kinds of carcinoma including breast cancer, gastric cancer, and NSCLC [14–16]. Although previous studies indicated that inactivation of the E-cadherin is mainly induced by hypermethylation of E-cadherin, the reported E-cadherin hypermethylation rates in NSCLC were remarkably diverse. Evidence concerning E-cadherin hypermethylation in the carcinogenesis and development of NSCLC remains controversial. Numerous studies published in this field examined a small number of patients. In addition, its roles in NSCLC and clinicopathological significance have not been thoroughly investigated. Hence, we conducted a meta-analysis to quantitatively evaluate the effects of E-cadherin hypermethylation on the incidence and clinicopathological characteristics of NSCLC.

Methods

Publication selection

A systematical literature searching was performed by PubMed, Embase, and Web of Science up to October 15, 2014. We used the following search terms: “lung” and “cancer or tumor or neoplasm or carcinoma,” “methylation,” and “E-cadherin or CDH1 or CAM 120/80.” We also searched manually for the reference lists of the retrieved articles and reviews for additional articles. After exclusion of non-relevant and/or redundant publications from different databases, the remaining papers were evaluated in the full-text version for inclusion and exclusion criteria and for relevant articles in the reference lists. All searched data were retrieved. Authors’ bibliographies and references of selected studies were also searched for other relevant studies. The most complete study was chosen to avoid duplication if same patient populations were reported in several publications.

Inclusion and exclusion criteria

Criteria that an eligible study has to meet were as follows: (1) E-cadherin hypermethylation evaluated in the primary NSCLC tissues, (2) researches revealed the relationship between E-cadherin hypermethylation and NSCLC clinicopathological parameters and prognosis, (3) E-cadherin hypermethylation examined by methylation-specific PCR (MSP) or quantitative MSP (QMSP), and (4) studies provided sufficient information to estimate hazard ratio (HR) about overall survival (OS) and 95 % confidence interval (CI). The exclusion criteria included the following: (1) letters, reviews, case reports, conference abstracts, editorials, and expert opinion and (2) all publications regarding in vitro/ex vivo studies, cell lines, and human xenografts were also excluded.

Data extraction and quality assessment

Two researchers (KZ and WC) independently collected the information and extracted the data regarding the authors, year, source of publication, inclusion criteria, E-cadherin methylation frequencies, sexual status, smoking history, pathological types, clinical staging, differentiation degree, lymph node metastasis, and prognostic conditions in patients and control groups. Any discrepancy was adjusted by discussion until they reach an agreement. Methodological evaluation was assessed by two independent researchers (NX and JZ) according to REMARK guidelines and ELCWP quality scale [17, 18].

Data analysis

Meta-analysis was performed by Reviewer Manager 5 (Cochrane Collaboration, Oxford, UK). The pooled odds ratios (ORs) and confidence intervals (CIs) were calculated to assess the correlation between E-cadherin methylation and NSCLC. Cochran’s Q test and I 2 were adopted to assess heterogeneity among studies [19]. If Q test showed a P < 0.05 or I 2 test was >50 %, it indicated significant heterogeneity and a fixed effects model was used to calculate the parameters. Otherwise, a random-effects model was used to pool data and attempt to identify potential sources of heterogeneity based on subgroup analyses [20, 21]. Publication bias was detected by Begge’s test and funnel plots [22]. The analysis of meta-regression and publication bias was performed by using STATA version 10.0.

Results

Identification of relevant studies

We identified 97 publications by the search method as described above. Seventy-nine of those were excluded due to non-original articles (review), laboratory studies, or studies irrelevant to the current analysis. Finally, there were 18 studies included in the meta-analysis as shown in Fig. 1.

Fig. 1
figure 1

Flow diagram of the literature search strategy and assessment of studies identified for meta-analysis

Full size image

Study characteristics

Eighteen studies published from 2001 to 2012 were eligible for meta-analysis. A total of 1467 NSCLC patients from China, Singapore, South Korea, Japan, France, Chile, and USA were enrolled. Their basic characteristics are summarized in Table 1.

Table 1 Basic characteristics of the included studies
Full size table

The correlation of E-cadherin hypermethylation with clinicopathological features

The inactivation of E-cadherin through hypermethylation in NSCLC

We first determined that E-cadherin hypermethylation was significantly higher in NSCLC than in normal lung tissues. The pooled OR from 13 studies including 731 NSCLC and 613 normal lung tissues is shown in Fig. 2 (OR = 3.55, 95 % CI = 1.98–6.36, p < 0.0001), indicating that E-cadherin hypermethylation in NSCLC was significantly higher than that in normal lung tissues.

Fig. 2
figure 2

The pooled OR from 13 studies including 731 NSCLC and 613 normal lung tissues, OR = 3.55, 95 % CI = 1.98–6.36, p < 0.0001

Full size image

Relationship between the frequency of E-cadherin hypermethylation and sex status

Next, we determined whether or not E-cadherin hypermethylation rate was correlated with sex status. The pooled OR from eight studies included 723 males and 281 females with NSCLC, as shown in Fig. 3 (OR = 0.81, 95 % CI = 0.59–1.12, p = 0.21), which indicates that E-cadherin hypermethylation was not significantly correlated with sex status.

Fig. 3
figure 3

The pooled OR from eight studies included 723 males and 281 females with NSCLC, OR = 0.81, 95 % CI = 0.59–1.12, p = 0.21, which indicates that E-cadherin hypermethylation was not significantly correlated with sex status in NSCLC patients

Full size image

Relationship between the frequency of E-cadherin hypermethylation and smoking status

Then, we determined whether or not E-cadherin hypermethylation rate was correlated with smoking status. The pooled OR from eight studies including 657 and 225 NSCLCs with and without smoking history is shown in Fig. 4 (OR = 0.95, 95 % CI = 0.65–1.38, p = 0.79), which indicates that E-cadherin hypermethylation was not significantly correlated with smoking status in NSCLC patients.

Fig. 4
figure 4

Nine hundred eighty-two NSCLC patients with the smoking status pooled in eight studies. Aberrant E-cadherin hypermethylation was not significantly correlated with the smoking status in NSCLC patients, OR = 0.95, 95 % CI = 0.65–1.38, p = 0.79

Full size image

Relationship between the frequency of E-cadherin hypermethylation and pathological types

We also determined whether or not E-cadherin hypermethylation was correlated with pathological types. The pooled OR from ten studies including 417 squamous cell carcinoma (SCC) and 474 adenocarcinoma (AD) is shown in Fig. 5 (OR = 0.91, 95 % CI = 0.65–1.25, p = 0.55), which indicates that E-cadherin hypermethylation was not significantly correlated with pathological types.

Fig. 5
figure 5

The pooled OR from ten studies including 417 squamous cell carcinoma (SCC) and 474 adenocarcinoma (AD), OR = 0.91, 95 % CI = 0.65–1.25, p = 0.55, indicating that E-cadherin hypermethylation was not significantly correlated with pathological types

Full size image

The role of E-cadherin hypermethylation in NSCLC progression

We analyzed 437 NSCLC patients pooled from three studies to assess whether or not the aberrant E-cadherin hypermethylation in NSCLC was associated with the differentiated status. As shown in Fig. 6a, aberrant E-cadherin hypermethylation was not significantly higher in poorly differentiated NSCLC than that in moderately or highly differentiated NSCLC, OR = 0.4, 95 % CI = 0.13–1.28, p = 0.12. Aberrant E-cadherin hypermethylation was also not significantly higher in advanced NSCLC (III and IV) than that in early-stage NSCLC (I and II), OR = 0.89, 95 % CI = 0.63–1.26, p = 0.52 (Fig. 6b). These results suggest that E-cadherin hypermethylation may not play an important role in NSCLC progression and different stages. In addition, aberrant E-cadherin hypermethylation was also not significantly higher in metastatic NSCLC than that in non-metastatic NSCLC, OR = 1.10, 95 % CI = 0.72–1.67, p = 0.67 (Fig. 6c).

Fig. 6
figure 6

Four hundred thirty-seven NSCLC patients pooled from three studies to assess whether or not the aberrant E-cadherin hypermethylation in NSCLC was associated with the differentiated status. Aberrant E-cadherin hypermethylation was not significantly higher in poorly differentiated NSCLC than that in moderately and highly differentiated NSCLCs, OR = 0.4, 95 % CI = 0.13–1.28, p = 0.12 (a). Aberrant E-cadherin hypermethylation was not significantly higher in advanced NSCLC (III and IV) than that in early-stage NSCLC (I and II), OR = 0.89, 95 % CI = 0.63–1.26, p = 0.52 (b). Aberrant E-cadherin hypermethylation was also not significantly higher in metastatic NSCLC than that in non-metastatic NSCLC, OR = 1.10, 95 % CI = 0.72–1.67, p = 0.67 (c)

Full size image

Sensitivity analyses and publication bias

A sensitivity analysis, in which one study was removed at a time, was conducted to assess the result stability. The pooled ORs were not significantly changed, indicating the stability of our analyses. The funnel plots were largely symmetric (Fig. 7a–g), suggesting that there were no publication biases in the meta-analysis of E-cadherin hypermethylation and clinicopathological features.

Fig. 7
figure 7

The funnel plots were largely symmetric, which suggests that there were no publication biases in the meta-analysis of E-cadherin hypermethylation and clinicopathological features. The funnel plot from 13 studies comparing NSCLC and normal lung tissue (a). The funnel plot from eight studies determined the relationship between E-cadherin hypermethylation and the sex status in NSCLC patients (b). The funnel plot from eight studies determined the relationship between E-cadherin hypermethylation and the smoking status in NSCLC patients (c). The funnel plot from ten studies comparing E-cadherin hypermethylation between squamous cell carcinoma (SCC) and adenocarcinoma (AD) (d). The funnel plot from three studies determined E-cadherin hypermethylation in different differentiated NSCLCs (e). The funnel plot from six studies determined E-cadherin hypermethylation in different staged NSCLCs (f). The funnel plot from four studies comparing E-cadherin hypermethylation in metastatic and non-metastatic NSCLCs (g)

Full size image

Discussion

The hypermethylation of tumor suppressor gene is an essential component of the molecular mechanism in the gene epigenomic regulation for cancer initiation and progression [41]. Inactivation of E-cadherin by promoter hypermethylation plays an important role in tumorigenesis in several types of tumors including NSCLC [14, 16, 42–44]. Although there have been some studies describing the methylation status of E-cadherin in NSCLC, the roles of inactivation of E-cadherin by hypermethylation in NSCLC and clinicopathological significance have not been thoroughly investigated. Our meta-analysis combining 18 published articles demonstrated that the hypermethylation frequencies in NSCLC were significantly higher than those in normal control tissues, OR = 3.55, 95 % CI = 1.98–6.36, p < 0.0001. Further analysis showed that E-cadherin hypermethylation was not strongly associated with the sex or smoking status in NSCLC patients. In addition, E-cadherin hypermethylation was also not strongly associated with pathological types, differentiated status, clinical stages, or metastatic status in NSCLC patients. The results from the current study indicate that the hypermethylation frequency of E-cadherin in NSCLC is strongly associated with NSCLC incidence; however, E-cadherin hypermethylation may be an early event in carcinogenesis of NSCLC. In support of this conclusion, Ceteci F et al. observed that postnatal inactivation of E-cadherin affected Clara cell differentiation and compromised airway regeneration under injury conditions, and the loss of E-cadherin function leads to tumor formation when additional mutations are sustained [45]. Their results indicate that E-cadherin plays a critical role in the regulation of proliferation and homeostasis of the epithelial cells lining the conducting airways.

Since changes in E-cadherin promoter hypermethylation are reversible, drug treatment through demethylation may be useful to delay carcinogenesis and progression. In fact, treatment of E-cadherin-negative tumor cells with the demethylating agent, 5-aza-2′-deoxycytidine, induced re-expression of E-cadherin mRNA and/or protein in several types of tumor cells including colorectal cancer [46], esophageal cancer [47], prostate cancer [48], and NSCLC [49]. 1α,25(OH)(2)D(3) promoted differentiation of breast cancer MDA-MB-231 cells by inducing de novo E-cadherin expression, an effect that was time- and dose-dependent [50]. Transfection of E-cadherin cDNA into R-HepG2 cells, in which E-cadherin promoter was hypermethylated in drug resistance of a doxorubicin-induced multidrug-resistant hepatocellular carcinoma cell line, led to increased amount of doxorubicin uptake, decreased cell viability, decreased P-glycoprotein expression, and increased apoptotic population of cells exposed to doxorubicin [51]. Interestingly, a combination of histone deacetylase inhibitors and DNA methyltransferase inhibitors suppresses the growth of endometrial cancer, which is likely mediated by upregulation of E-cadherin and downregulation of Bcl-2 [52]. Therefore, the approaches targeting E-cadherin to reverse epigenetic silencing, reactivate gene expression, and, finally, induce a therapeutic effect such as differentiation, growth arrest, or apoptosis may bring new direction and hope for cancer treatment through gene-targeted therapy.

E-cadherin as a tumor suppressor gene functionally keeps cell-cell adhesion and controls epithelial cell arrangement in normal order and layer. A number of studies demonstrate that loss of the expression or function of E-cadherin can initiate the activation of several signaling pathways including the canonical Wnt and Rho family GTPase-mediated modulation of the actin cytoskeleton which are associated with epithelial-mesenchymal transition, finally leading to cancer cell metastasis [53, 54]. To better understand the correlation between E-cadherin methylation and NSCLC, comprehensive evaluation on the methylation markers in NSCLC should be further addressed. Although a large number of studies have demonstrated the potential relationship between E-cadherin methylation and NSCLC, a meta-analysis can summarize the studies and compare different subgroup characters.

A sensitivity analysis was used to assess the result stability, and the pooled ORs were not significantly changed, indicating the stability of our analyses. In addition, the funnel plots were largely symmetric, indicating that there were no publication biases in the meta-analysis of E-cadherin hypermethylation and clinicopathological features. However, there are several potential limitations in this meta-analysis. First, we only selected articles published in English; thus, other articles that were published in other languages were not selected, due to anticipated difficulties in obtaining accurate medical translation. Second, most selected articles are from Asia; hence, cautions should be taken when our findings are interpreted among the general populations. Third, DNA methylation is influenced by several clinicopathological parameters that are not taken into account in the study (i.e., age, ethnics, previous treatments, etc.).

In summary, this meta-analysis shows that E-cadherin hypermethylation is strongly associated with NSCLC incidence; however, E-cadherin hypermethylation is also not significantly associated with pathological types, differentiated status, clinical stages, or metastatic status in NSCLC patients. These results indicate that E-cadherin methylation might be an early biomarker of carcinogenesis of NSCLC, with potential value for predicting the diagnosis of NSCLC patients. In addition, the potential value of E-cadherin as a drug target may bring new direction and hope for cancer treatment through gene-targeted therapy.