Association Between Tissue Inhibitor of Metalloproteinase-3 Gene Methylation and Gastric Cancer Risk: A Meta-Analysis

Abstract

Background: The tumor suppressor gene tissue inhibitor of metalloproteinase-3 (TIMP-3) has been reported to be frequently and significantly downregulated in gastric cancer, and its downregulation is correlated with hypermethylation in its promoter region. However, the association between TIMP-3 methylation and gastric cancer risk remains unclear. Aim: In this study, we assessed the relationship between TIMP-3 promoter methylation and gastric cancer risk by performance of a meta-analysis. Methods: Relevant studies were identified in a comprehensive literature search using PubMed, Embase, and Web of Science databases. The strength of the association between TIMP-3 methylation and the risk of gastric cancer was assessed by odds ratio (OR) with the corresponding 95% confidence interval (CI). The heterogeneity among studies was tested using the Q-statistics and I² metric. The publication bias was examined by Begg's funnel plots and Egger's linear regression test. Results: A total of 1096 subjects from eight studies were included in the present meta-analysis. Overall, a significant association between TIMP-3 methylation and gastric cancer risk was observed (OR = 8.65; 95% CI 4.31-17.37; p < 0.001). Stratified analyses by ethnicity, sample materials, and detection methods also revealed increased gastric cancer risk in individuals harboring methylated TIMP-3. Moreover, no publication bias was detected in the present meta-analysis. Conclusions: Our results show a positive correlation between TIMP-3 promoter methylation and gastric cancer risk and indicated that TIMP-3 promoter methylation may be used as a molecular marker for gastric cancer.

Introduction

Gastric cancer is one of the most common malignancies in the world, particularly in Eastern Asia, Central and Eastern Europe, and South America (Siegel et al., 2014). Molecular alterations, especially genetic and epigenetic alterations, are involved in the development and progression of gastric cancer (Toyota et al., 1999; Lee et al., 2002; Kim et al., 2005). Increasing evidence shows that aberrant DNA methylation of CpG islands in gene promoter regions is a primary mechanism for tumor suppressor gene inactivation and plays a significant role in gastric carcinogenesis (Das and Singal, 2004; Qu et al., 2013).

The tissue inhibitors of metalloproteinases (TIMPs) are naturally occurring molecules that can specifically inhibit the activity of matrix metalloproteinases (MMPs) (Murphy, 2011). The local balance between MMPs and TIMPs is believed to play a major role in extracellular matrix (ECM) remodeling during development and in diseases such as cancer. Unlike other members of the TIMP family, TIMP-3 is secreted and remains attached to the ECM after synthesis (Anand-Apte et al., 1996). The gene encoding tissue TIMP-3 is located at 22q12.3 and is now regarded as a tumor suppressor gene based on the findings that increased expression of TIMP-3 results in suppression of tumor growth and angiogenesis and decreased TIMP-3 levels facilitate tumor invasion and metastasis (Anand-Apte et al., 1996; Gomez et al., 1997; Ahonen et al., 1998; Murphy, 2011). Interestingly, TIMP-3 has been reported to be frequently and significantly downregulated in gastric cancer, and its downregulation has been correlated with hypermethylation in the promoter region (Kang et al., 2000, 2003b; Gu et al., 2008).

Although several studies have reported the association of TIMP-3 promoter methylation with gastric cancer risk, the results of these studies are conflicting. Specifically, a number of studies have shown that hypermethylation of TIMP-3 is associated with an increased risk of gastric cancer (Kang et al., 2003b; Gu et al., 2008; Yao et al., 2012; Guan et al., 2013; Yu et al., 2014), whereas other studies found no significant difference in TIMP-3 methylation frequency between gastric cancer cases and healthy controls (Leung et al., 2005; Ksiaa et al., 2009). Therefore, in the present study, we conducted a systemic meta-analysis to better evaluate the relationship between TIMP-3 promoter methylation and the risk of gastric cancer.

Materials and Methods

Search strategy

A comprehensive literature search was carried out using PubMed, Embase, and Web of Science databases. The last search was performed on October 1, 2015, without any language restrictions. The following key words were used: tissue inhibitor of metalloproteinase-3, TIMP-3, methylation, gastric, and cancer. References from the primary literatures and review articles were also manually checked to search for additional relevant publications.

Study selection and data extraction

Studies were selected for the present meta-analysis if they met the following criteria: (1) the study focused on the relationship between TIMP-3 methylation and the risk of gastric cancer; (2) the subjects in the study comprised gastric cancer patients and healthy controls; and (3) the frequency of TIMP-3 methylation was reported or could be calculated from the data presented. If two or more studies had overlapping data, only the study with the largest case number was included in the meta-analysis.

For each eligible study, the following information was extracted with a standard data extraction form: last name of the first author, publication year, country of the subjects, ethnicity, number of cancer cases and controls, DNA methylation detection method, and sample materials. Two investigators independently collected data and reached a consensus on all items.

Statistical analyses

The strength of the association between TIMP-3 methylation and gastric cancer risk was assessed by odds ratio (OR) with the corresponding 95% confidence interval (CI). The heterogeneity between the studies was tested using chi-square-based Q-statistics and was also quantified using the I² metric (I² < 25%, no heterogeneity; I² = 25-50%, moderate heterogeneity; and I² > 50%, strong heterogeneity) (Higgins et al., 2003). If heterogeneity was present, then the random-effects model (the DerSimonian and Laird method) was used to calculate the pooled OR (DerSimonian and Laird, 1986). Otherwise, the fixed-effects model (the Mantel-Haenszel method) was used (Mantel and Haenszel, 1959). Stratified analysis was performed based on ethnicities, sample materials, and detection methods. Moreover, a sensitivity analysis, by which a single study in the meta-analysis was deleted each time to determine the influence of the individual data set on the overall pooled OR, was performed to assess the stability of the results. Publication bias was examined by Begg's funnel plots and Egger's linear regression test (Egger et al., 1997). All statistical analyses were performed using Stata software (version 10.1). All p-values were two sided, and p < 0.05 was considered as statistically significant.

Results

Study characteristics

Using our previously described literature search strategy, we identified 13 studies investigating the relationship between TIMP-3 methylation and gastric cancer. Upon further review, three studies were excluded on the basis of inadequate data for acquiring TIMP-3 methylation frequency (Kang et al., 2000; Kim et al., 2005; Bernal et al., 2008) and two were excluded on the basis of overlapping data (Kang et al., 2001, 2003a). Ultimately, a total of 8 studies comprising 1096 subjects (608 cases and 488 controls) were included in the present meta-analysis (Kang et al., 2003b; Leung et al., 2005; Gu et al., 2008; Kim et al., 2009; Ksiaa et al., 2009; Yao et al., 2012; Guan et al., 2013; Yu et al., 2014). Study characteristics are listed in Table 1. The studies were conducted in four countries (China, Germany, Korea, and Tunisia) and published between 2003 and 2014. All studies were written in English. The methylation status was detected using methylation-specific PCR (MSP), quantitative methylation-specific PCR (QMSP), or a MethyLight assay. TIMP-3 methylation status was detected in serum in one study; gastric tissues were used in the remaining seven out of eight studies.

Table 1.

Characteristics of the Studies Included in the Present Meta-Analysis

Author (Year)	Country	Ethnic	Cases (n)	Controls (n)	Methods	Materials
Kang et al. (2003b)	Korea	Asian	80	131	MSP	Tissue
Leung et al. (2005)	China	Asian	60	22	MethyLight assay	Serum
Gu et al. (2008)	Germany	Caucasian	15	14	MethyLight assay	Tissue
Kim et al. (2009)	Korea	Asian	74	63	MSP	Tissue
Ksiaa et al. (2009)	Tunisia	Caucasian	68	53	MSP	Tissue
Yao et al. (2012)	China	Asian	141	35	Q-MSP	Tissue
Guan et al. (2013)	China	Asian	78	78	MSP	Tissue
Yu et al. (2014)	China	Asian	92	92	Q-MSP	Tissue

MSP, methylation-specific PCR; QMSP, quantitative methylation-specific PCR.

Quantitative data analysis

Strong heterogeneity was found among the eight studies included in the present meta-analysis (p = 0.017, I² = 59.2%, Fig. 1 and Table 2). Therefore, we chose the random-effects model to assess the association of TIMP-3 methylation with gastric cancer risk. Overall, TIMP-3 methylation was significantly associated with an increased risk of gastric cancer (OR = 8.65; 95% CI 4.31-17.37; p < 0.001; Fig. 1 and Table 2). We then performed sensitivity analysis to assess the influence of individual studies on the overall effect, and we found that omission of a single study changed the overall OR from 6.29 (95% CI 3.77-10.47) to 10.96 (95% CI 5.56-21.59), suggesting that there was no singly sensitive study. These results indicated a credibly strong association between TIMP-3 methylation and the risk of gastric cancer.

FIG. 1.

Meta-analysis of the association between TIMP-3 promoter methylation and gastric cancer risk. The squares and the horizontal lines represent the study-specific OR and 95% CI, and the diamond represents the pooled OR and 95% CI. CI, confidence interval; OR, odds ratio.

Table 2.

Summary of TIMP-3 Methylation and Gastric Cancer Risk

Variables	Study no.	Cases/controls (n)	OR (95% CI)	p	p_H^a	I² (%)
Total	8	608/488	8.65 (4.31-17.37)	<0.001	0.017	59.2
Ethnicities
Asian	6	525/421	9.71 (4.10-23.00)	<0.001	0.006	69.5
Caucasian	2	83/67	5.58 (1.69-18.36)	0.005	0.459	0
Sample materials
Tissue	7	548/466	8.64 (4.11-18.13)	<0.001	0.009	64.9
Serum	1	60/22	9.36 (0.53-166.71)	0.128
Detection methods
MSP	4	300/325	5.67 (2.78-11.56)	<0.001	0.098	52.3
Q-MSP	2	233/127	30.28 (11.62-78.90)	<0.001	0.414	0
MethyLight assay	2	75/36	9.1 (1.96-42.36)	0.005	0.982	0

p-Value of Q test for heterogeneity.

Next, we stratified the association by subject ethnicity, sample material, and detection method. The ethnic-specific OR showed an increased risk for individuals carrying methylated TIMP-3 compared with those without methylated TIMP-3 in Asian populations (OR = 9.71; 95% CI 4.10-23.00; p < 0.001) and Caucasian populations (OR = 5.58; 95% CI 1.69-18.36; p = 0.005). In the subgroup analysis stratified by sample material, significantly increased risk was found in tissue samples (OR = 8.64; 95% CI 4.11-18.13; p < 0.001), but not in the serum samples. When stratified by the methylation detection method, a remarkably increased risk was detected in all of the detection method groups (Table 2).

Publication bias

As shown in Figure 2, visual assessment of the Begg's funnel plot did not reveal any evidence of obvious asymmetry in the present meta-analysis. Further evaluation using the Egger's linear regression test did not show evidence of publication bias in our study (t = 0.69, p = 0.513).

FIG. 2.

Funnel plot for publication bias test. Each point represents a separate study. The natural logarithm of the OR and its standard error were used for generating the funnel plot.

Discussion

To the best of our knowledge, this is the first meta-analysis to investigate whether DNA methylation of the TIMP-3 gene has an impact on the risk of gastric cancer. In this study, we combined eight independent original studies comprising a total of 1096 samples and revealed that TIMP-3 promoter methylation increased the risk of gastric cancer. The overall OR for TIMP-3 methylation in gastric cancer versus normal samples was 8.65 (95% CI 4.31-17.37; p < 0.001), suggesting a strong positive correlation between TIMP-3 promoter methylation and gastric cancer susceptibility. We also conducted subgroup analyses to identify the relationships between TIMP-3 methylation and gastric cancer risk. The results of subgroup analysis based on ethnicity showed that TIMP-3 promoter methylation was significantly associated with the risk of gastric cancer in both Asian and Caucasian populations, suggesting that there was no ethnic-specific effect of TIMP-3 promoter methylation on gastric cancer susceptibility.

Assessments of the between-study heterogeneity and potential publication bias are regarded as essential components of meta-analyses (Egger et al., 1997; Lau et al., 1998). In this study, we used the Q-test and I² test to detect heterogeneity within studies and implemented Begg's funnel plots and Egger's linear regression test to detect possible publication bias. No evidence of potential publication bias was found. Notably, we also carried out sensitivity analysis to determine the effects of individual studies on the overall effect and found that there was no singly sensitive study in our meta-analysis. These results demonstrate a reliable association between TIMP-3 promoter methylation and the risk of gastric cancer.

DNA methylation of gene regulatory elements, a well-known epigenetic change, acts as an important alternative to gene mutations or deletions as a strategy for gene inactivation (Jones and Takai, 2001). Hypermethylation within the promoter regions of tumor suppressor genes can result in transcriptional silencing of the gene and lead to subsequent downregulation or loss of protein expression, and this phenomenon has been widely reported in human cancers, including gastric cancer (Das and Singal, 2004; Qu et al., 2013). Hypermethylation of the promoter region of TIMP-3 is the main mechanism of TIMP-3 inactivation (Kang et al., 2000, 2003b; Gu et al., 2008). TIMP-3 functions as a tumor suppressor and is capable of inhibiting growth, angiogenesis, migration, invasion, and metastasis of several cancers (Anand-Apte et al., 1996; Ahonen et al., 1998; Anania et al., 2011; Murphy, 2011; Lin et al., 2012). These previous findings support our conclusion from the present study that TIMP-3 methylation increases the risk of gastric cancer.

Molecular-based diagnostic approaches, such as the identification of epigenetic markers, have been extensively investigated in recent years (Elimova et al., 2015; Tahara and Arisawa, 2015). Along with the advancements made in genetics, nucleotide biochemistry, and DNA sequencing technology, various techniques for DNA methylation analysis have been developed. Of these techniques, MSP seems to be the most useful, owing to its sensitivity, quick and simple methodology, and reproducibility (Das and Singal, 2004; Harrison and Parle-McDermott, 2011). In this study, our data showed a strong association between TIMP-3 methylation and the risk of gastric cancer, especially when the analysis was performed in tissue samples, suggesting that TIMP-3 methylation may be used as a molecular marker for early clinical detection and monitoring of gastric cancer. Although no statistically significant difference in TIMP-3 methylation frequency between gastric cancer patients and controls was found when the analysis was conducted on the serum samples (Leung et al., 2005), it is worth mentioning that the frequency of methylated TIMP-3 in cancer patients was 17%, whereas nonmethylated TIMP-3 DNA was detected in the controls, suggesting a potential diagnostic value of TIMP-3 promoter methylation in serum samples. Moreover, TIMP-3 hypermethylation has been reported to be associated with the progression and clinical outcomes of gastric cancer (Gu et al., 2008; Yu et al., 2014). However, due to the small number of such studies, further studies are required to determine the prognostic value of TIMP-3 methylation in gastric cancer.

In summary, our current meta-analysis provides evidence to support a strong association between TIMP-3 promoter methylation and an increased risk of gastric cancer. Although additional studies with larger sample sizes are still needed to provide a more representative and convincing statistical analysis, detection of TIMP-3 methylation may be utilized as a potential biomarker for gastric cancer diagnosis.

Footnotes

Acknowledgment

This study was supported by grants from the Natural Science Foundation of Shandong Province, China (No. ZR2010HL048 to Xiying Luan).

Author Disclosure Statement

No competing financial interests exist.

References

Ahonen

, Baker

, Kahari

(1998) Adenovirus-mediated gene delivery of tissue inhibitor of metalloproteinases-3 inhibits invasion and induces apoptosis in melanoma cells. Cancer Res, 58:2310-2315.

Anand-Apte

, Bao

, Smith

, et al. (1996) A review of tissue inhibitor of metalloproteinases-3 (TIMP-3) and experimental analysis of its effect on primary tumor growth. Biochem Cell Biol, 74:853-862.

Anania

, Sensi

, Radaelli

, et al. (2011) TIMP3 regulates migration, invasion and in vivo tumorigenicity of thyroid tumor cells. Oncogene, 30:3011-3023.

Bernal

, Aguayo

, Villarroel

, et al. (2008) Reprimo as a potential biomarker for early detection in gastric cancer. Clin Cancer Res, 14:6264-6269.

Das

, Singal

(2004) DNA methylation and cancer. J Clin Oncol, 22:4632-4642.

DerSimonian

, Laird

(1986) Meta-analysis in clinical trials. Control Clin Trials, 7:177-188.

Egger

, Davey Smith

, Schneider

, Minder

(1997) Bias in meta-analysis detected by a simple, graphical test. BMJ, 315:629-634.

Elimova

, Wadhwa

, Shiozaki

, et al. (2015) Molecular biomarkers in gastric cancer. J Natl Compr Canc Netw, 13:e19-e29.

Gomez

, Alonso

, Yoshiji

, Thorgeirsson

(1997) Tissue inhibitors of metalloproteinases: structure, regulation and biological functions. Eur J Cell Biol, 74:111-122.

10.

, Xing

, Tanzer

, et al. (2008) Frequent loss of TIMP-3 expression in progression of esophageal and gastric adenocarcinomas. Neoplasia, 10:563-572.

11.

Guan

, Zhang

, Song

, Dai

(2013) Promoter methylation and expression of TIMP3 gene in gastric cancer. Diagn Pathol, 8:110.

12.

Harrison

, Parle-McDermott

(2011) DNA methylation: a timeline of methods and applications. Front Genet, 2:74.

13.

Higgins

, Thompson

, Deeks

, Altman

(2003) Measuring inconsistency in meta-analyses. BMJ, 327:557-560.

14.

Jones

, Takai

(2001) The role of DNA methylation in mammalian epigenetics. Science, 293:1068-1070.

15.

Kang

, Lee

, Kim

, Jung

(2003a) Profile of aberrant CpG island methylation along multistep gastric carcinogenesis. Lab Invest, 83:519-526.

16.

Kang

, Lee

, Kim

, Jung

(2003b) Profile of aberrant CpG island methylation along the multistep pathway of gastric carcinogenesis. Lab Invest, 83:635-641.

17.

Kang

, Shim

, Jung

, et al. (2001) CpG island methylation in premalignant stages of gastric carcinoma. Cancer Res, 61:2847-2851.

18.

Kang

, Choi

, Kim

, et al. (2000) Transcriptional inactivation of the tissue inhibitor of metalloproteinase-3 gene by DNA hypermethylation of the 5′-CpG island in human gastric cancer cell lines. Int J Cancer, 86:632-635.

19.

Kim

, Kim

, Roh

, et al. (2005) Aberrant CpG island methylation in early-onset sporadic gastric carcinoma. J Cancer Res Clin Oncol, 131:733-740.

20.

Kim

, Jung

, Lee

, et al. (2009) Comparative analysis of DNA methylation between primary and metastatic gastric carcinoma. Oncol Rep, 21:1251-1259.

21.

Ksiaa

, Ziadi

, Amara

, et al. (2009) Biological significance of promoter hypermethylation of tumor-related genes in patients with gastric carcinoma. Clin Chim Acta, 404:128-133.

22.

Lau

, Ioannidis

, Schmid

(1998) Summing up evidence: one answer is not always enough. Lancet, 351:123-127.

23.

Lee

, Leung

, Chan

, et al. (2002) Detection of gene promoter hypermethylation in the tumor and serum of patients with gastric carcinoma. Clin Cancer Res, 8:1761-1766.

24.

Leung

, To

, Chu

, et al. (2005) Potential diagnostic and prognostic values of detecting promoter hypermethylation in the serum of patients with gastric cancer. Br J Cancer, 92:2190-2194.

25.

Lin

, Zhang

, Wang

, et al. (2012) Tissue inhibitor of metalloproteinases-3 transfer suppresses malignant behaviors of colorectal cancer cells. Cancer Gene Ther, 19:845-851.

26.

Mantel

, Haenszel

(1959) Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst, 22:719-748.

27.

Murphy

(2011) Tissue inhibitors of metalloproteinases. Genome Biol, 12:233.

28.

, Dang

, Hou

(2013) Gene methylation in gastric cancer. Clin Chim Acta, 424:53-65.

29.

Siegel

, Ma

, Zou

, Jemal

(2014) Cancer statistics, 2014. CA Cancer J Clin, 64:9-29.

30.

Tahara

, Arisawa

(2015) DNA methylation as a molecular biomarker in gastric cancer. Epigenomics, 7:475-486.

31.

Toyota

, Ahuja

, Suzuki

, et al. (1999) Aberrant methylation in gastric cancer associated with the CpG island methylator phenotype. Cancer Res, 59:5438-5442.

32.

Yao

, Shi

, et al. (2012) Quantitative assessment of gene methylation and their impact on clinical outcome in gastric cancer. Clin Chim Acta, 413:787-794.

33.

, Lv

, Han

, et al. (2014) Methylated TIMP-3 DNA in body fluids is an independent prognostic factor for gastric cancer. Arch Pathol Lab Med, 138:1466-1473.