- Email: [email protected]
Meta-Analysis of MMP2 –1306T Allele as a Protective Factor in Digestive Cancer
Archives of Medical Research 42 (2011) 239e243
ORIGINAL ARTICLE
Meta-Analysis of MMP2 e1306T Allele as a Protective Factor in Digestive Cancer Le-yao Zhanga,* and Ke-wei Renb,* a
Department of Gastroenterology, Ninth Peoples Hospital, Wuxi, Jiangsu, China Department of Orthopedics, First Affiliated Hospital of Nanjing Medical University, Nanjing, Jiangsu, China
b
Received for publication February 7, 2011; accepted April 13, 2011 (ARCMED-D-11-00068).
Background and Aims. Matrix metalloproteinase-2 (MMP-2) is an enzyme with proteolytic activity on matrix proteins and abolishes a Sp-1 binding site and consequently decreases its activity. Many studies have been carried out on the association between MMP2 e1306C/T polymorphism and digestive cancer risk, but results were somewhat controversial and underpowered. Methods. To examine the risk of digestive cancer associated with e1306C/T polymorphism of MMP2, we performed a meta-analysis of ten case-control studies. Eligible studies were identified by searching the electronic literature using Pubmed and Embase. Odds ratios (ORs) with 95% confidence intervals (CIs) were estimated to assess the association. Results. Overall, we found that e1306T allele can decrease digestive cancer risk in three different genotype models (T-allele vs. C-allele, OR 5 0.73, 95% CI: 0.54e0.98, p 5 0.034; TC vs. CC, OR 5 0.64, 95% CI: 0.53e0.78, p !0.001; TTþTC vs. CC, OR 5 0.68, 95% CI: 0.53e0.86, p 5 0.002). Similarly, in the stratified analysis by cancer type, ethnicity and source of control, significantly decreased cancer risk was indicated. Moreover, in the subgroup of smokers, e1306T allele may protect people against digestive cancer risk (TTþTC vs. CC, OR 5 0.71, 95% CI: 0.51e0.98, p 5 0.037). Conclusions. Our meta-analysis showed evidence that MMP2 e1306T allele may be a protective factor for digestive cancer risk as well as a low-penetrance susceptibility digestive cancer biomarker. Ó 2011 IMSS. Published by Elsevier Inc. Key Words: MMP2, Polymorphism, Digestive cancer, 1306T allele, Risk Meta-analysis.
Introduction Matrix metalloproteinases (MMPs) is a superfamily of proteolytic enzymes capable of degrading extracellular matrix and basement membrane (1). Recent studies have also indicated that MMPs are involved in early tumorigenesis, modulating proliferation, apoptosis and angiogenesis (2). In addition, MMPs can cleave growth factors in extracellular matrix. MMP2 is a member of MMPs, whose activity is known to be regulated not only by posttranscriptional mechanisms including activation of proenzyme and *
These authors contributed equally to this work. Address reprint requests to: Dr. Le-yao Zhang, Department of Gastroenterology, Ninth Peoples Hospital, 999 Liangxi Road, Wux i214063, Jiangsu, China; Phone: þ86-510-85867878; FAX: þ86-510-85879445; E-mail: [email protected]
inhibition of enzyme activity (2), but also transcriptional regulation is also believed to be pivotally important because of a number of cis-acting regulatory elements and several transcription factors such as p53, Sp1, Sp3 and AP-2, to participate in the control of its constitutive expression (3). Increased expression of MMP2 was found to be important for tumor invasion and metastasis (2). For instance, MMP2 can degrade insulin-like growth factor (IGF) binding proteins and release IGF, which can stimulate cell proliferation and inhibit apoptosis of cells (2,4). MMP2 can also target fibroblast growth factor (FGF) receptor 1, yielding active soluble ectodomain of the angiogenic activities of FGF (5). MMP2 (rs243865) is located on chromosome 16q13 and is a gelatinase A that degradates basal membrane composed mainly by type IV collagen (6,7). Price et al. (6) first
0188-4409/$ - see front matter. Copyright Ó 2011 IMSS. Published by Elsevier Inc. doi: 10.1016/j.arcmed.2011.04.013
240
Zhang and Ren/ Archives of Medical Research 42 (2011) 239e243
identified a e1306C/T polymorphism in the MMP2 promoter. This base transition is located in the CCACC box of the Sp1-binding site and eliminates promoter activity; it is likely that CC genotype may be associated with the high transcription level and enzyme activity of MMP2. The substitution of C by the T allele implies the lower activity of the promoter region, lowering the levels of MMP2. This is the mechanism possible that MMP2 e1306C/T polymorphism is related to protection of the development of cancer, such as digestive cancer. So far, many studies have investigated the role of MMP2 e1306C/T polymorphism and digestive cancer risk; however, this relationship remains controversial. Previously, two meta-analyses were published (8,9), which had some limitations. Two studies (10,11) about colorectal cancer were duplicated, some of extracted data were incorrect, and their study did not include the source of control subgroups. Taking into consideration the extensive role of MMP2 in digestive cancer and to derive a more precise estimation of the association of e1306C/T polymorphism in MMP2 and different types of digestive cancer risk, we performed a meta-analysis of all 10 eligible caseecontrol studies: gastric cancer (4,12e14), esophageal cancer (12,15e16), colorectal cancer (10,17) and oral cancer (18).
Materials and Methods Identification and Eligibility of Relevant Studies We conducted searches using Pubmed (http://www.ncbi.nlm. nih.gov/) and Embase (http://www.embase.com/home). The last search was updated on January 10, 2011 using the search terms MMP2 or matrix metalloproteinase-2 or MMP, polymorphism and cancer or tumor, or carcinoma and gastric or esophageal or colorectal or oral. There was no restriction on language or publication years. We also screened reference of the retrieved articles and review articles by a hand search. Inclusion and Exclusion Criteria Eligible studies had to meet the following criteria: (a) the study assessed the correlation between different types of digestive cancer and MMP2 e1306C/T polymorphism, (b) case-control studies, and (c) only full-text manuscripts were included. Major exclusion criteria were (a) no control population, (b) no available genotype frequency and (c) duplication of the previous publications. Data Extraction Two of the authors reviewed results of each of the database searches to verify that published papers were not missed. Data were collected on the first author’s last name, year of publication, country of origin, ethnicity, cancer type, number of cases and controls, subjects of genotype in cases and
controls, genotype methods, age range in cases and controls and HardyeWeinberg equilibrium (HWE) in each study. Statistic Analysis Odds ratios (ORs) with 95% confidence intervals (CIs) were used to measure the strength of the association between MMP2 e1306C/T polymorphism and digestive cancer risk based on genotype frequencies in cases and controls. Subgroup analysis stratified by cancer type was performed first. Ethnicity was categorized as European or Asian. Source of control subgroup analysis was performed on two classifications: population-based and hospital-based. Smoking was categorized as smoker and non-smoker. The fixed-effects model and the random-effects model were used to calculate the pooled OR. The statistical significance of the summary OR was determined with the Z-test. Heterogeneity assumption was evaluated with a c2-based Q test among the studies. A p $0.05 for the Q-test indicated a lack of heterogeneity among the studies. In order to better evaluate the extent of heterogeneity between studies, the I2 test was also used. As a guide, I2 values of !25% may be considered low, value of |50% may be considered moderate and values of O75% may be considered high (19). If p #0.05, or I2 $50%, a random-effects model using the DerSimonian-Laird method (20) was adopted; otherwise, a fixed-effects model using the Mantel-Haenszel method (21) was used. For e1306C/T polymorphism, we investigated the association between genetic variants and digestive cancer risk in dominant genetic model (TTþTC vs. CC), heterozygote comparison (TC vs. CC) and allelic contrast (T-allele vs. C-allele). The funnel plot asymmetry and publication bias were assessed with Eggers test; p !0.05 was considered statistically significant (22). The departure of frequencies of MMP2 e1306C/T polymorphism from expectation under HWE was assessed by c2 test in controls using the Pearson c2 test; p !0.05 was considered significant. All statistical tests for this meta-analysis were performed with Stata software (v.10.0; Stata Corp LP, College Station, TX).
Results Eligible Studies Overall, we identified ten articles (11 case-control studies, nine different first authors) to evaluate the association of MMP2 e1306C/T polymorphism with risk for whole digestive cancer. One study (11) was excluded because the same authors published several reports on the same patients and controls, and only the most recent or complete study was selected. Finally, there were ten studies with 2,380 cases and 4,574 controls concerning the MMP2 e1306C/ T polymorphism and digestive cancer. Characteristics of
MMP2 Polymorphism and Digestive Cancer Risk
241
Table 1. Characteristics of studies of MMP2 e1306C/T polymorphism included in this meta-analysis First author, year, ref. no Li 2010 (12) Li 2010 (12) Yu 2004 (15) Miao 2003 (4) Chen 2009 (16) Zhang 2004 (13) Xu 2006 (10) Elander 2006 (17) Kubben 2006 (14) Lin 2004 (18)
Country (Race)
Cancer type
Case/Control
Source of control
Genotyping method
China (Asian) China (Asian) China (Asian) China (Asian) China (Asian) China (Asian) China (Asian) Sweden (European) USA (European) China-Taiwan (Asian)
Gastric cancer Esophageal cancer Esophageal cancer Gastric cancer Esophageal cancer Gastric cancer Colorectal cancer Colorectal cancer Gastric cancer Oral cancer
257/630 371/630 527/777 356/789 188/324 228/774 126/126 127/208 79/169 121/147
PB PB PB HB HB PB HB PB HB HB
PCR-RFLP PCR-RFLP qRT- PCR PCR-DHPLC PCR-RFLP PCR-RFLP PCR-RFLP/PCR-DHPLC PCR-RFLP/PCR-DHPLC PCR-RFLP PCR-DHPLC
PB, population-based; HB, hospital based; PCR-RFLP, polymerase chain reaction restriction fragment length polymorphism; PCR-DHPLC; PCR-denaturing high-performance liquid chromatography; qRT-PCR, quantitative real-time PCR.
studies of MMP2 e1306C/T polymorphism are summarized in Table 1. Distribution of MMP2 e1306 genotypes and age range between digestive cancer and control are showed in Table 2. Cases in most of the studies were histologically diagnosed and controls were free from cancer. For the MMP2 e1306C/T polymorphism, there were four studies of gastric cancer, three of esophageal cancer, two of colorectal cancer and and only one oral cancer; three were eight studies in the Asian population and two in the European population. Hospital-based cases were used in five studies. Li et al. investigated two types of digestive cancer in one publication. All of publications were consistent with the HWE (Table 2). Diverse genotyping methods were used including polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP), quantitative real-time PCR (qRT-PCR) and PCR-denaturing high-performance liquid chromatography (PCR-DHPLC).
CI 5 0.54e0.98, p 5 0.034), the heterozygote comparison (OR 5 0.64, 95% CI 5 0.53e0.78, p !0.001) and the dominant models (OR 5 0.68, 95% CI 5 0.53e0.86, p 5 0.002) (Table 3). In the stratified analysis by cancer type, a significant association between MMP2 e1306C/T polymorphism and gastric, esophageal and oral cancer was found (gastric cancer: OR 5 0.60, 95% CI 5 0.39e0.95, p 5 0.027 for TC vs. CC; OR 5 0.60, 95% CI 5 0.37e0.97, p 5 0.036 for TTþTC vs. CC. esophageal: OR 5 0.73, 95% CI 5 0.63e0.85, p !0.001 for TC vs. CC; OR 5 0.84, 95% CI 5 0.73e0.96, p 5 0.012 for TTþTC vs. CC. Oral: OR 5 0.49, 95% CI 5 0.28e0.85, p 5 0.011 for T-allele vs. C-allele; OR 5 0.53, 95% CI 5 0.27e0.97, p 5 0.038 for TTþTC vs. CC) (Table 3). Subgroup analysis was stratified by ethnicity status. MMP2 e1306T allele can decrease digestive cancer risk in an Asian population in all three genetic models (T-allele vs. C-allele, OR 5 0.68, 95% CI 5 0.48e0.97, p 5 0.034; TC vs. CC, OR 5 0.60, 95% CI 5 0.49e0.74, p !0.001 and TT þ TC vs. CC, OR 5 0.63, 95% CI 5 0.48e0.84, p 5 0.001). Similarly, in the subgroup of source of control, a significantly decreased association was found in hospital-based or population-based groups in all genetic models (Table 3).
Overall Effects for Meta-analysis In the overall analysis, a significantly decreased association could be observed between digestive cancer risk and the variant genotypes of MMP2 e1306 in different genetic models: in the allelic contrast (OR 5 0.73, 95%
Table 2. Distribution of MMP2 e1306 genotypes and age range between digestive cancer and control included in the meta-analysis Age range (mean SD), year Author Li Li Yu Miao Chen Zhang Xu Selander Kubben Lin
Case NA NA NA 41e72
(60.5 (60.1 (58.7 (58.4 NA 28e82 (57.7 26e78 (58.4 26e93 (74 35e91 (66 NA (51.3
Case
Control
8.3) 9.33) 9.7) NA)
10.2) NA) NA) NA) 0.9)
NA NA NA 45e76
(60.4 (60.4 (57.6 (57.6 NA 24e83 (57.5 NA NA 18e73 (33 NA
8.42) 8.42) 7.6) NA)
7.7) NA)
Control
TT
TC
CC
TT
TC
CC
PHWE
T (%)
4 39 6 0 3 3 1 9 5 0
46 48 112 44 36 45 19 49 24 20
207 284 409 312 149 180 106 69 50 101
6 6 18 18 6 18 2 10 14 6
137 137 220 229 73 220 32 89 53 34
487 487 539 542 245 536 92 109 102 107
0.283 0.283 0.421 0.279 0.836 0.409 0.678 0.125 0.070 0.133
0.118 0.118 0.165 0.168 0.131 0.165 0.143 0.262 0.240 0.156
NA, not available; PHWE, HardyeWeinberg equilibrium of controls; T (%), frequency of T allele in controls.
242
Zhang and Ren/ Archives of Medical Research 42 (2011) 239e243
Table 3. Stratified analyses of MMP2 e1306 polymorphism on digestive cancer risk T-allele vs. C-allele
TC vs. CC
TTþTC vs. CC
Variables
N
Cases/Controls
OR (95% CI) Ph/P/I2 (%)
OR (95% CI) Ph/P/I2 (%)
OR (95% CI) Ph/P/I2 (%)
Total Cancer type Gastric cancer Esophageal cancer Colorectal cancer Oral cancer Ethnicity Asian European Source of control Hospital-based Population-based Smoking Smoker Nonsmoker
10
2380/4574
0.73 (0.54e0.98) 0.000/0.034/85.7
0.64 (0.53e0.78) 0.022/0.000/53.7
0.68 (0.53e0.86) 0.000/0.002/73.3
4 3 2 1
920/2362 1086/1731 253/334 121/147
0.63 0.95 0.88 0.49
0.60 0.73 0.81 0.62
0.60 0.84 0.84 0.53
8 2
2174/4197 206/377
0.68 (0.48e0.97) 0.000/0.034/88.4 0.96 (0.78e1.19) 0.610/0.739/0.0
0.60 (0.49e0.74) 0.030/0.000/54.9 0.93 (0.75e1.16) 0.913/0.532/0.0
0.63 (0.48e0.84) 0.000/0.001/77.1 0.95 (0.78e1.15) 0.867/0.588/0.0
5 5
870/1555 1510/3019
0.57 (0.37e0.87) 0.001/0.010/77.8 0.90 (0.64e1.26) 0.000/0.532/85.2
0.59 (0.30e0.90) 0.008/0.015/71.1 0.75 (0.67e0.85) 0.422/0.000/0.0
0.56 (0.36e0.88) 0.003/0.011/75.2 0.83 (0.74e0.92) 0.075/0.001/52.9
2 2
259/528 333/720
-
-
0.71 (0.51e0.98) 0.231/0.037/30.4 0.84 (0.64e1.09) 0.548/0.191/0.0
(0.39e1.00) 0.000/0.048/85.6 (0.55e1.64) 0.000/0.862/90.9 (0.70e1.11) 0.060/0.273/71.7 (0.28e0.85)/-/0.011/-
(0.39e0.95) 0.002/0.027/79.5 (0.63e0.85) 0.566/0.000/0.0 (0.64e1.03) 0.120/0.089/58.7 (0.34e1.15)/-/0.132/-
(0.37e0.97) 0.000/0.036/83.6 (0.73e0.96) 0.080/0.012/60.4 (0.68e1.05) 0.075/0.124/68.5 (0.27e0.97)/-/0.038/-
Ph, heterogeneity assumption was evaluated with a c2-based Q test. p, statistical significance of the summary OR was determined with the Z-test.
In one publication (10), there were two types of digestive cancer and the author classified subjects into smokers and nonsmokers. In our current meta-analysis, we stratified smoking subgroup and, surprisingly, pooled decreased digestive cancer risk was found in subjects who were smokers (TT þ TC vs. CC, OR 5 0.71, 95% CI 5 0.51e0.98, p 5 0.001) (Table 3). Test of Heterogeneity There was significant heterogeneity in the entire group for T-allele vs. C-allele (Pheterogeneity !0.001, I2 5 85.7%), TC vs. CC (Pheterogeneity 5 0.022, I2 5 53.7%) and TTþTC vs. CC (Pheterogeneity !0.001, I2 5 73.3%). Sensitivity Analysis and Publication Bias Sensitivity analysis was performed to assess the influence of each individual study on the pooled OR by sequential removal of individual studies. The results suggested that no individual study significantly affected the overall OR dominantly. Beggs funnel plot and Eggers test were performed to assess publication bias. The shapes of the funnel plots did not reveal any obvious asymmetry in any of the comparison models. Eggers test was then used to provide statistical evidence of funnel plot symmetry. The results still did not suggest any evidence of publication bias (T-allele vs. C-allele, t 5 0.82, p 5 0.438; TC vs. CC, t 5 0.52, p 5 0.618 and TT þ TC vs. CC, t 5 0.13, p 5 0.899). Discussion Genetic susceptibility to cancers has led to growing attention to the studies of polymorphism genes involved
in carcinogenesis. If large and unbiased, these can provide insight into the relationships between candidate genes and cancer risk. However, there has been a continuing debate over the association between MMP2 e1306C/T polymorphism and digestive cancer risk. Previous studies have generally been underpowered to draw a convincing conclusion. The meta-analysis is a method that can solve the problem created by low statistical power in a single study to draw a more robust conclusion than the body of evidence. The present meta-analysis, based on ten case-control studies, explored the association between MMP2 e1306C/T polymorphism and digestive cancer risk, involving about 2,380 cases and 4,574 controls. The main message of this meta-analysis is that significantly decreased associations were found between whole digestive cancer and MMP2 e1306C/T polymorphism. The association between MMP2 e1306C/T polymorphism and digestive cancer risk is modified by cancer type: MMP2 e1306T allele represents a protective factor in gastric, esophageal, and oral cancer. We all know that MMP2 is overexpressed in a wide variety of human cancers and contributes not only to cancer progression but also to cancer development (2). Price et al. (6) manifested that gene expression driven by the C allele in MMP2 e1306 promoter was significantly greater than reporter gene expression driven by the T allelic counterpart both in epithelial cells and in macrophages using transient transfection experiments, indicating that the T-allele may play a protective role against cancer risk. This hypothesis was confirmed by our data. Genotype frequency of various polymorphic loci may manifest racial differences. For this reason, we divided the worldwide population into two broad racial groups: Asian and European. In our study, there was also
MMP2 Polymorphism and Digestive Cancer Risk
a decreased association between MMP2 e1306T allele and digestive cancer in an Asian population in all genetic models. The same relationship was also found in the subgroup by source of control subgroup. Tobacco smoking is an established risk factor for many cancers and has a destructive effect on human immune responses (23). In our study we added another subgroup called smoking, which was classified into smoker and nonsmoker, to discover whether smoking can influence the relationship between digestive cancer and MMP2 e1306C/T polymorphism. To our surprise, a pooled decreased digestive cancer risk was found in people who were smokers and had T allele in MMP2 e1306 gene, which indicated that T allele played a protective factor in smokers against digestive cancer. Some limitations in our meta-analysis should be acknowledged. First, the numbers of published studies included in our meta-analysis was not sufficiently large for a comprehensive analysis. There were no data regarding an African population and further research should focus on this race. Second, publication bias may have occurred and our Eggers test results may have a substantial risk of being affected by such bias, although the funnel plots as well as Eggers linear regression tests indicated no remarkable publication biases in the meta-analyses. Third, interactions between gene-gene, gene-environment and even different polymorphic loci of the same gene may modulate digestive cancer. Fourth, in one study (14), a small number of cases were included. Fifth, our meta-analysis was based on unadjusted estimates. A more precise analysis should be conducted if individual information including other covariates such as age, gender and metastasis/differentiation status. In spite of these, our meta-analysis also had three advantages. First, a substantial number of cases and controls were pooled from different studies, which significantly increased statistical power of the analysis. Second, the quality of caseecontrol studies included in the current pooled analysis was satisfactory based on our selection criteria. Third, we did not detect any publication bias, suggesting that the results are relatively stable and publication biases may not have an evident influence on the results of the meta-analysis. In summary, our meta-analysis showed evidence that the MMP2 e1306T allele was associated with a decreased digestive cancer risk. However, further well-designed large studies, particularly referring to gene-gene and gene-environment interactions and African race studies, are warranted. These future studies should lead to better and more comprehensive understanding of the association between the MMP2 e1306C/T polymorphism and digestive cancer.
Conflict of Interest Statement None declared.
243
References 1. Nagase H, Woessner JF. Matrix metalloproteinases. J Biol Chem 1999; 274:21491e21494. 2. Egeblad M, Werb Z. New functions for the matrix metalloproteinases in cancer progression. Nat Rev Caner 2002;2:161e174. 3. Qin H, Sun Y, Benveniste EN. The transcription factors Sp1, Sp3, and AP-2 are required for constitutive matrix metalloproteinase-2 gene expression in astroglioma cells. J Biol Chem 1999;274:29130e29137. 4. Miao X, Yu C, Tan W, et al. A functional polymorphism in the matrix metalloproteinase-2 gene promoter (1306C/T) is associated with risk of development but not metastasis of gastric cardia adenocarcinoma. Cancer Res 2003;63:3987e3990. 5. Levi E, Fridman R, Miao HQ, et al. Matrix metalloproteinase 2 releases active soluble ectodomain of fibroblast growth factor receptor 1. Proc Natl Acad Sci USA 1996;93:7069e7074. 6. Price SJ, Greaves DR, Watkins H. Identification of novel, functional genetic variants in the human matrix metalloproteinase-2 gene: role of Sp1 in allele-specific transcriptional regulation. J Biol Chem 2001;276:7549e7558. 7. Stamenkovic I. Matrix metalloproteinases in tumor invasion and metastasis. Semin Cancer Biol 2000;10:415e433. 8. Peter MC, Pankaj S. Polymorphisms of matrix metalloproteinases 1,2,3 and 9 and susceptibility to lung, breast and colorectal cancer in over 30,000 subjects. Int J Cancer 2009;125:1473e1478. 9. Peng B, Cao L, Ma X, et al. Meta-analysis of association between matrix metalloproteinases 2, 7 and 9 promoter polymorphisms and cancer risk. Mutagenesis 2010;25:371e379. 10. Xu EP, Huang Q, Lu BJ, et al. The correlation between polymorphisms of matrix metalloproteinase-2 and -9 genes and colorectal cancer of Chinese patients. Zhonghua Yi Xue Yi Chuan Xue Za Zhi 2006;23:78e81. 11. Xu E, Lai M, Lv B, et al. A single nucleotide polymorphism in the matrix metalloproteinase-2 promoter is associated with colorectal cancer. Biochem Biophys Res Commun 2004;324:999e1003. 12. Li Y, Sun DL, Duan YN, et al. Association of functional polymorphisms in MMPs genes with gastric cardia adenocarcinoma and esophageal squamous cell carcinoma in high incidence region of North China. Mol Biol Rep 2010;37:197e205. 13. Zhang XM, Miao XP, Xiong P, et al. Association of functional polymorphisms in matrix metalloproteinase-2 (MMP-2) and MMP-9 genes with risk of gastric cancer in a Chinese population. Ai Zheng 2004;23: 1233e1237. 14. Kubben F, Sier C, Meijer M, et al. Clinical impact of MMP and TIMP gene polymorphisms in gastric cancer. Br J Cancer 2006;95:744e751. 15. Yu C, Zhou Y, Miao X, et al. Functional haplotypes in the promoter of matrix metalloproteinase-2 predict risk of the occurrence and metastasis of esophageal cancer. Cancer Res 2004;64:7622e7628. 16. Chen XB, Chen GL, Liu JN, et al. Genetic polymorphisms in STK15 and MMP-2 associated susceptibility to esophageal cancer in Mongolian population. Zhonghua Yu Fang Yi Xue Za Zhi 2009;43:559e564. 17. Elander N, S€oderkvist P, Fransen K. Matrix metalloproteinase (MMP) -1, -2, -3 and -9 promoter polymorphisms in colorectal cancer. Anticancer Res 2006;26:791e795. 18. Lin SC, Lo SS, Liu CJ, et al. Functional genotype in matrix metalloproteinases-2 promoter is a risk factor for oral carcinogenesis. J Oral Pathol Med 2004;33:405e409. 19. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ 2003;327:557e560. 20. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7:177e188. 21. Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 1959;22:719e748. 22. Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629e634. 23. Zeidel A, Beilin B, Yardeni I, et al. Immune response in asymptomatic smokers. Acta Anaesthesiol Scand 2002;46:959e964.
ORIGINAL ARTICLE
Meta-Analysis of MMP2 e1306T Allele as a Protective Factor in Digestive Cancer Le-yao Zhanga,* and Ke-wei Renb,* a
Department of Gastroenterology, Ninth Peoples Hospital, Wuxi, Jiangsu, China Department of Orthopedics, First Affiliated Hospital of Nanjing Medical University, Nanjing, Jiangsu, China
b
Received for publication February 7, 2011; accepted April 13, 2011 (ARCMED-D-11-00068).
Background and Aims. Matrix metalloproteinase-2 (MMP-2) is an enzyme with proteolytic activity on matrix proteins and abolishes a Sp-1 binding site and consequently decreases its activity. Many studies have been carried out on the association between MMP2 e1306C/T polymorphism and digestive cancer risk, but results were somewhat controversial and underpowered. Methods. To examine the risk of digestive cancer associated with e1306C/T polymorphism of MMP2, we performed a meta-analysis of ten case-control studies. Eligible studies were identified by searching the electronic literature using Pubmed and Embase. Odds ratios (ORs) with 95% confidence intervals (CIs) were estimated to assess the association. Results. Overall, we found that e1306T allele can decrease digestive cancer risk in three different genotype models (T-allele vs. C-allele, OR 5 0.73, 95% CI: 0.54e0.98, p 5 0.034; TC vs. CC, OR 5 0.64, 95% CI: 0.53e0.78, p !0.001; TTþTC vs. CC, OR 5 0.68, 95% CI: 0.53e0.86, p 5 0.002). Similarly, in the stratified analysis by cancer type, ethnicity and source of control, significantly decreased cancer risk was indicated. Moreover, in the subgroup of smokers, e1306T allele may protect people against digestive cancer risk (TTþTC vs. CC, OR 5 0.71, 95% CI: 0.51e0.98, p 5 0.037). Conclusions. Our meta-analysis showed evidence that MMP2 e1306T allele may be a protective factor for digestive cancer risk as well as a low-penetrance susceptibility digestive cancer biomarker. Ó 2011 IMSS. Published by Elsevier Inc. Key Words: MMP2, Polymorphism, Digestive cancer, 1306T allele, Risk Meta-analysis.
Introduction Matrix metalloproteinases (MMPs) is a superfamily of proteolytic enzymes capable of degrading extracellular matrix and basement membrane (1). Recent studies have also indicated that MMPs are involved in early tumorigenesis, modulating proliferation, apoptosis and angiogenesis (2). In addition, MMPs can cleave growth factors in extracellular matrix. MMP2 is a member of MMPs, whose activity is known to be regulated not only by posttranscriptional mechanisms including activation of proenzyme and *
These authors contributed equally to this work. Address reprint requests to: Dr. Le-yao Zhang, Department of Gastroenterology, Ninth Peoples Hospital, 999 Liangxi Road, Wux i214063, Jiangsu, China; Phone: þ86-510-85867878; FAX: þ86-510-85879445; E-mail: [email protected]
inhibition of enzyme activity (2), but also transcriptional regulation is also believed to be pivotally important because of a number of cis-acting regulatory elements and several transcription factors such as p53, Sp1, Sp3 and AP-2, to participate in the control of its constitutive expression (3). Increased expression of MMP2 was found to be important for tumor invasion and metastasis (2). For instance, MMP2 can degrade insulin-like growth factor (IGF) binding proteins and release IGF, which can stimulate cell proliferation and inhibit apoptosis of cells (2,4). MMP2 can also target fibroblast growth factor (FGF) receptor 1, yielding active soluble ectodomain of the angiogenic activities of FGF (5). MMP2 (rs243865) is located on chromosome 16q13 and is a gelatinase A that degradates basal membrane composed mainly by type IV collagen (6,7). Price et al. (6) first
0188-4409/$ - see front matter. Copyright Ó 2011 IMSS. Published by Elsevier Inc. doi: 10.1016/j.arcmed.2011.04.013
240
Zhang and Ren/ Archives of Medical Research 42 (2011) 239e243
identified a e1306C/T polymorphism in the MMP2 promoter. This base transition is located in the CCACC box of the Sp1-binding site and eliminates promoter activity; it is likely that CC genotype may be associated with the high transcription level and enzyme activity of MMP2. The substitution of C by the T allele implies the lower activity of the promoter region, lowering the levels of MMP2. This is the mechanism possible that MMP2 e1306C/T polymorphism is related to protection of the development of cancer, such as digestive cancer. So far, many studies have investigated the role of MMP2 e1306C/T polymorphism and digestive cancer risk; however, this relationship remains controversial. Previously, two meta-analyses were published (8,9), which had some limitations. Two studies (10,11) about colorectal cancer were duplicated, some of extracted data were incorrect, and their study did not include the source of control subgroups. Taking into consideration the extensive role of MMP2 in digestive cancer and to derive a more precise estimation of the association of e1306C/T polymorphism in MMP2 and different types of digestive cancer risk, we performed a meta-analysis of all 10 eligible caseecontrol studies: gastric cancer (4,12e14), esophageal cancer (12,15e16), colorectal cancer (10,17) and oral cancer (18).
Materials and Methods Identification and Eligibility of Relevant Studies We conducted searches using Pubmed (http://www.ncbi.nlm. nih.gov/) and Embase (http://www.embase.com/home). The last search was updated on January 10, 2011 using the search terms MMP2 or matrix metalloproteinase-2 or MMP, polymorphism and cancer or tumor, or carcinoma and gastric or esophageal or colorectal or oral. There was no restriction on language or publication years. We also screened reference of the retrieved articles and review articles by a hand search. Inclusion and Exclusion Criteria Eligible studies had to meet the following criteria: (a) the study assessed the correlation between different types of digestive cancer and MMP2 e1306C/T polymorphism, (b) case-control studies, and (c) only full-text manuscripts were included. Major exclusion criteria were (a) no control population, (b) no available genotype frequency and (c) duplication of the previous publications. Data Extraction Two of the authors reviewed results of each of the database searches to verify that published papers were not missed. Data were collected on the first author’s last name, year of publication, country of origin, ethnicity, cancer type, number of cases and controls, subjects of genotype in cases and
controls, genotype methods, age range in cases and controls and HardyeWeinberg equilibrium (HWE) in each study. Statistic Analysis Odds ratios (ORs) with 95% confidence intervals (CIs) were used to measure the strength of the association between MMP2 e1306C/T polymorphism and digestive cancer risk based on genotype frequencies in cases and controls. Subgroup analysis stratified by cancer type was performed first. Ethnicity was categorized as European or Asian. Source of control subgroup analysis was performed on two classifications: population-based and hospital-based. Smoking was categorized as smoker and non-smoker. The fixed-effects model and the random-effects model were used to calculate the pooled OR. The statistical significance of the summary OR was determined with the Z-test. Heterogeneity assumption was evaluated with a c2-based Q test among the studies. A p $0.05 for the Q-test indicated a lack of heterogeneity among the studies. In order to better evaluate the extent of heterogeneity between studies, the I2 test was also used. As a guide, I2 values of !25% may be considered low, value of |50% may be considered moderate and values of O75% may be considered high (19). If p #0.05, or I2 $50%, a random-effects model using the DerSimonian-Laird method (20) was adopted; otherwise, a fixed-effects model using the Mantel-Haenszel method (21) was used. For e1306C/T polymorphism, we investigated the association between genetic variants and digestive cancer risk in dominant genetic model (TTþTC vs. CC), heterozygote comparison (TC vs. CC) and allelic contrast (T-allele vs. C-allele). The funnel plot asymmetry and publication bias were assessed with Eggers test; p !0.05 was considered statistically significant (22). The departure of frequencies of MMP2 e1306C/T polymorphism from expectation under HWE was assessed by c2 test in controls using the Pearson c2 test; p !0.05 was considered significant. All statistical tests for this meta-analysis were performed with Stata software (v.10.0; Stata Corp LP, College Station, TX).
Results Eligible Studies Overall, we identified ten articles (11 case-control studies, nine different first authors) to evaluate the association of MMP2 e1306C/T polymorphism with risk for whole digestive cancer. One study (11) was excluded because the same authors published several reports on the same patients and controls, and only the most recent or complete study was selected. Finally, there were ten studies with 2,380 cases and 4,574 controls concerning the MMP2 e1306C/ T polymorphism and digestive cancer. Characteristics of
MMP2 Polymorphism and Digestive Cancer Risk
241
Table 1. Characteristics of studies of MMP2 e1306C/T polymorphism included in this meta-analysis First author, year, ref. no Li 2010 (12) Li 2010 (12) Yu 2004 (15) Miao 2003 (4) Chen 2009 (16) Zhang 2004 (13) Xu 2006 (10) Elander 2006 (17) Kubben 2006 (14) Lin 2004 (18)
Country (Race)
Cancer type
Case/Control
Source of control
Genotyping method
China (Asian) China (Asian) China (Asian) China (Asian) China (Asian) China (Asian) China (Asian) Sweden (European) USA (European) China-Taiwan (Asian)
Gastric cancer Esophageal cancer Esophageal cancer Gastric cancer Esophageal cancer Gastric cancer Colorectal cancer Colorectal cancer Gastric cancer Oral cancer
257/630 371/630 527/777 356/789 188/324 228/774 126/126 127/208 79/169 121/147
PB PB PB HB HB PB HB PB HB HB
PCR-RFLP PCR-RFLP qRT- PCR PCR-DHPLC PCR-RFLP PCR-RFLP PCR-RFLP/PCR-DHPLC PCR-RFLP/PCR-DHPLC PCR-RFLP PCR-DHPLC
PB, population-based; HB, hospital based; PCR-RFLP, polymerase chain reaction restriction fragment length polymorphism; PCR-DHPLC; PCR-denaturing high-performance liquid chromatography; qRT-PCR, quantitative real-time PCR.
studies of MMP2 e1306C/T polymorphism are summarized in Table 1. Distribution of MMP2 e1306 genotypes and age range between digestive cancer and control are showed in Table 2. Cases in most of the studies were histologically diagnosed and controls were free from cancer. For the MMP2 e1306C/T polymorphism, there were four studies of gastric cancer, three of esophageal cancer, two of colorectal cancer and and only one oral cancer; three were eight studies in the Asian population and two in the European population. Hospital-based cases were used in five studies. Li et al. investigated two types of digestive cancer in one publication. All of publications were consistent with the HWE (Table 2). Diverse genotyping methods were used including polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP), quantitative real-time PCR (qRT-PCR) and PCR-denaturing high-performance liquid chromatography (PCR-DHPLC).
CI 5 0.54e0.98, p 5 0.034), the heterozygote comparison (OR 5 0.64, 95% CI 5 0.53e0.78, p !0.001) and the dominant models (OR 5 0.68, 95% CI 5 0.53e0.86, p 5 0.002) (Table 3). In the stratified analysis by cancer type, a significant association between MMP2 e1306C/T polymorphism and gastric, esophageal and oral cancer was found (gastric cancer: OR 5 0.60, 95% CI 5 0.39e0.95, p 5 0.027 for TC vs. CC; OR 5 0.60, 95% CI 5 0.37e0.97, p 5 0.036 for TTþTC vs. CC. esophageal: OR 5 0.73, 95% CI 5 0.63e0.85, p !0.001 for TC vs. CC; OR 5 0.84, 95% CI 5 0.73e0.96, p 5 0.012 for TTþTC vs. CC. Oral: OR 5 0.49, 95% CI 5 0.28e0.85, p 5 0.011 for T-allele vs. C-allele; OR 5 0.53, 95% CI 5 0.27e0.97, p 5 0.038 for TTþTC vs. CC) (Table 3). Subgroup analysis was stratified by ethnicity status. MMP2 e1306T allele can decrease digestive cancer risk in an Asian population in all three genetic models (T-allele vs. C-allele, OR 5 0.68, 95% CI 5 0.48e0.97, p 5 0.034; TC vs. CC, OR 5 0.60, 95% CI 5 0.49e0.74, p !0.001 and TT þ TC vs. CC, OR 5 0.63, 95% CI 5 0.48e0.84, p 5 0.001). Similarly, in the subgroup of source of control, a significantly decreased association was found in hospital-based or population-based groups in all genetic models (Table 3).
Overall Effects for Meta-analysis In the overall analysis, a significantly decreased association could be observed between digestive cancer risk and the variant genotypes of MMP2 e1306 in different genetic models: in the allelic contrast (OR 5 0.73, 95%
Table 2. Distribution of MMP2 e1306 genotypes and age range between digestive cancer and control included in the meta-analysis Age range (mean SD), year Author Li Li Yu Miao Chen Zhang Xu Selander Kubben Lin
Case NA NA NA 41e72
(60.5 (60.1 (58.7 (58.4 NA 28e82 (57.7 26e78 (58.4 26e93 (74 35e91 (66 NA (51.3
Case
Control
8.3) 9.33) 9.7) NA)
10.2) NA) NA) NA) 0.9)
NA NA NA 45e76
(60.4 (60.4 (57.6 (57.6 NA 24e83 (57.5 NA NA 18e73 (33 NA
8.42) 8.42) 7.6) NA)
7.7) NA)
Control
TT
TC
CC
TT
TC
CC
PHWE
T (%)
4 39 6 0 3 3 1 9 5 0
46 48 112 44 36 45 19 49 24 20
207 284 409 312 149 180 106 69 50 101
6 6 18 18 6 18 2 10 14 6
137 137 220 229 73 220 32 89 53 34
487 487 539 542 245 536 92 109 102 107
0.283 0.283 0.421 0.279 0.836 0.409 0.678 0.125 0.070 0.133
0.118 0.118 0.165 0.168 0.131 0.165 0.143 0.262 0.240 0.156
NA, not available; PHWE, HardyeWeinberg equilibrium of controls; T (%), frequency of T allele in controls.
242
Zhang and Ren/ Archives of Medical Research 42 (2011) 239e243
Table 3. Stratified analyses of MMP2 e1306 polymorphism on digestive cancer risk T-allele vs. C-allele
TC vs. CC
TTþTC vs. CC
Variables
N
Cases/Controls
OR (95% CI) Ph/P/I2 (%)
OR (95% CI) Ph/P/I2 (%)
OR (95% CI) Ph/P/I2 (%)
Total Cancer type Gastric cancer Esophageal cancer Colorectal cancer Oral cancer Ethnicity Asian European Source of control Hospital-based Population-based Smoking Smoker Nonsmoker
10
2380/4574
0.73 (0.54e0.98) 0.000/0.034/85.7
0.64 (0.53e0.78) 0.022/0.000/53.7
0.68 (0.53e0.86) 0.000/0.002/73.3
4 3 2 1
920/2362 1086/1731 253/334 121/147
0.63 0.95 0.88 0.49
0.60 0.73 0.81 0.62
0.60 0.84 0.84 0.53
8 2
2174/4197 206/377
0.68 (0.48e0.97) 0.000/0.034/88.4 0.96 (0.78e1.19) 0.610/0.739/0.0
0.60 (0.49e0.74) 0.030/0.000/54.9 0.93 (0.75e1.16) 0.913/0.532/0.0
0.63 (0.48e0.84) 0.000/0.001/77.1 0.95 (0.78e1.15) 0.867/0.588/0.0
5 5
870/1555 1510/3019
0.57 (0.37e0.87) 0.001/0.010/77.8 0.90 (0.64e1.26) 0.000/0.532/85.2
0.59 (0.30e0.90) 0.008/0.015/71.1 0.75 (0.67e0.85) 0.422/0.000/0.0
0.56 (0.36e0.88) 0.003/0.011/75.2 0.83 (0.74e0.92) 0.075/0.001/52.9
2 2
259/528 333/720
-
-
0.71 (0.51e0.98) 0.231/0.037/30.4 0.84 (0.64e1.09) 0.548/0.191/0.0
(0.39e1.00) 0.000/0.048/85.6 (0.55e1.64) 0.000/0.862/90.9 (0.70e1.11) 0.060/0.273/71.7 (0.28e0.85)/-/0.011/-
(0.39e0.95) 0.002/0.027/79.5 (0.63e0.85) 0.566/0.000/0.0 (0.64e1.03) 0.120/0.089/58.7 (0.34e1.15)/-/0.132/-
(0.37e0.97) 0.000/0.036/83.6 (0.73e0.96) 0.080/0.012/60.4 (0.68e1.05) 0.075/0.124/68.5 (0.27e0.97)/-/0.038/-
Ph, heterogeneity assumption was evaluated with a c2-based Q test. p, statistical significance of the summary OR was determined with the Z-test.
In one publication (10), there were two types of digestive cancer and the author classified subjects into smokers and nonsmokers. In our current meta-analysis, we stratified smoking subgroup and, surprisingly, pooled decreased digestive cancer risk was found in subjects who were smokers (TT þ TC vs. CC, OR 5 0.71, 95% CI 5 0.51e0.98, p 5 0.001) (Table 3). Test of Heterogeneity There was significant heterogeneity in the entire group for T-allele vs. C-allele (Pheterogeneity !0.001, I2 5 85.7%), TC vs. CC (Pheterogeneity 5 0.022, I2 5 53.7%) and TTþTC vs. CC (Pheterogeneity !0.001, I2 5 73.3%). Sensitivity Analysis and Publication Bias Sensitivity analysis was performed to assess the influence of each individual study on the pooled OR by sequential removal of individual studies. The results suggested that no individual study significantly affected the overall OR dominantly. Beggs funnel plot and Eggers test were performed to assess publication bias. The shapes of the funnel plots did not reveal any obvious asymmetry in any of the comparison models. Eggers test was then used to provide statistical evidence of funnel plot symmetry. The results still did not suggest any evidence of publication bias (T-allele vs. C-allele, t 5 0.82, p 5 0.438; TC vs. CC, t 5 0.52, p 5 0.618 and TT þ TC vs. CC, t 5 0.13, p 5 0.899). Discussion Genetic susceptibility to cancers has led to growing attention to the studies of polymorphism genes involved
in carcinogenesis. If large and unbiased, these can provide insight into the relationships between candidate genes and cancer risk. However, there has been a continuing debate over the association between MMP2 e1306C/T polymorphism and digestive cancer risk. Previous studies have generally been underpowered to draw a convincing conclusion. The meta-analysis is a method that can solve the problem created by low statistical power in a single study to draw a more robust conclusion than the body of evidence. The present meta-analysis, based on ten case-control studies, explored the association between MMP2 e1306C/T polymorphism and digestive cancer risk, involving about 2,380 cases and 4,574 controls. The main message of this meta-analysis is that significantly decreased associations were found between whole digestive cancer and MMP2 e1306C/T polymorphism. The association between MMP2 e1306C/T polymorphism and digestive cancer risk is modified by cancer type: MMP2 e1306T allele represents a protective factor in gastric, esophageal, and oral cancer. We all know that MMP2 is overexpressed in a wide variety of human cancers and contributes not only to cancer progression but also to cancer development (2). Price et al. (6) manifested that gene expression driven by the C allele in MMP2 e1306 promoter was significantly greater than reporter gene expression driven by the T allelic counterpart both in epithelial cells and in macrophages using transient transfection experiments, indicating that the T-allele may play a protective role against cancer risk. This hypothesis was confirmed by our data. Genotype frequency of various polymorphic loci may manifest racial differences. For this reason, we divided the worldwide population into two broad racial groups: Asian and European. In our study, there was also
MMP2 Polymorphism and Digestive Cancer Risk
a decreased association between MMP2 e1306T allele and digestive cancer in an Asian population in all genetic models. The same relationship was also found in the subgroup by source of control subgroup. Tobacco smoking is an established risk factor for many cancers and has a destructive effect on human immune responses (23). In our study we added another subgroup called smoking, which was classified into smoker and nonsmoker, to discover whether smoking can influence the relationship between digestive cancer and MMP2 e1306C/T polymorphism. To our surprise, a pooled decreased digestive cancer risk was found in people who were smokers and had T allele in MMP2 e1306 gene, which indicated that T allele played a protective factor in smokers against digestive cancer. Some limitations in our meta-analysis should be acknowledged. First, the numbers of published studies included in our meta-analysis was not sufficiently large for a comprehensive analysis. There were no data regarding an African population and further research should focus on this race. Second, publication bias may have occurred and our Eggers test results may have a substantial risk of being affected by such bias, although the funnel plots as well as Eggers linear regression tests indicated no remarkable publication biases in the meta-analyses. Third, interactions between gene-gene, gene-environment and even different polymorphic loci of the same gene may modulate digestive cancer. Fourth, in one study (14), a small number of cases were included. Fifth, our meta-analysis was based on unadjusted estimates. A more precise analysis should be conducted if individual information including other covariates such as age, gender and metastasis/differentiation status. In spite of these, our meta-analysis also had three advantages. First, a substantial number of cases and controls were pooled from different studies, which significantly increased statistical power of the analysis. Second, the quality of caseecontrol studies included in the current pooled analysis was satisfactory based on our selection criteria. Third, we did not detect any publication bias, suggesting that the results are relatively stable and publication biases may not have an evident influence on the results of the meta-analysis. In summary, our meta-analysis showed evidence that the MMP2 e1306T allele was associated with a decreased digestive cancer risk. However, further well-designed large studies, particularly referring to gene-gene and gene-environment interactions and African race studies, are warranted. These future studies should lead to better and more comprehensive understanding of the association between the MMP2 e1306C/T polymorphism and digestive cancer.
Conflict of Interest Statement None declared.
243
References 1. Nagase H, Woessner JF. Matrix metalloproteinases. J Biol Chem 1999; 274:21491e21494. 2. Egeblad M, Werb Z. New functions for the matrix metalloproteinases in cancer progression. Nat Rev Caner 2002;2:161e174. 3. Qin H, Sun Y, Benveniste EN. The transcription factors Sp1, Sp3, and AP-2 are required for constitutive matrix metalloproteinase-2 gene expression in astroglioma cells. J Biol Chem 1999;274:29130e29137. 4. Miao X, Yu C, Tan W, et al. A functional polymorphism in the matrix metalloproteinase-2 gene promoter (1306C/T) is associated with risk of development but not metastasis of gastric cardia adenocarcinoma. Cancer Res 2003;63:3987e3990. 5. Levi E, Fridman R, Miao HQ, et al. Matrix metalloproteinase 2 releases active soluble ectodomain of fibroblast growth factor receptor 1. Proc Natl Acad Sci USA 1996;93:7069e7074. 6. Price SJ, Greaves DR, Watkins H. Identification of novel, functional genetic variants in the human matrix metalloproteinase-2 gene: role of Sp1 in allele-specific transcriptional regulation. J Biol Chem 2001;276:7549e7558. 7. Stamenkovic I. Matrix metalloproteinases in tumor invasion and metastasis. Semin Cancer Biol 2000;10:415e433. 8. Peter MC, Pankaj S. Polymorphisms of matrix metalloproteinases 1,2,3 and 9 and susceptibility to lung, breast and colorectal cancer in over 30,000 subjects. Int J Cancer 2009;125:1473e1478. 9. Peng B, Cao L, Ma X, et al. Meta-analysis of association between matrix metalloproteinases 2, 7 and 9 promoter polymorphisms and cancer risk. Mutagenesis 2010;25:371e379. 10. Xu EP, Huang Q, Lu BJ, et al. The correlation between polymorphisms of matrix metalloproteinase-2 and -9 genes and colorectal cancer of Chinese patients. Zhonghua Yi Xue Yi Chuan Xue Za Zhi 2006;23:78e81. 11. Xu E, Lai M, Lv B, et al. A single nucleotide polymorphism in the matrix metalloproteinase-2 promoter is associated with colorectal cancer. Biochem Biophys Res Commun 2004;324:999e1003. 12. Li Y, Sun DL, Duan YN, et al. Association of functional polymorphisms in MMPs genes with gastric cardia adenocarcinoma and esophageal squamous cell carcinoma in high incidence region of North China. Mol Biol Rep 2010;37:197e205. 13. Zhang XM, Miao XP, Xiong P, et al. Association of functional polymorphisms in matrix metalloproteinase-2 (MMP-2) and MMP-9 genes with risk of gastric cancer in a Chinese population. Ai Zheng 2004;23: 1233e1237. 14. Kubben F, Sier C, Meijer M, et al. Clinical impact of MMP and TIMP gene polymorphisms in gastric cancer. Br J Cancer 2006;95:744e751. 15. Yu C, Zhou Y, Miao X, et al. Functional haplotypes in the promoter of matrix metalloproteinase-2 predict risk of the occurrence and metastasis of esophageal cancer. Cancer Res 2004;64:7622e7628. 16. Chen XB, Chen GL, Liu JN, et al. Genetic polymorphisms in STK15 and MMP-2 associated susceptibility to esophageal cancer in Mongolian population. Zhonghua Yu Fang Yi Xue Za Zhi 2009;43:559e564. 17. Elander N, S€oderkvist P, Fransen K. Matrix metalloproteinase (MMP) -1, -2, -3 and -9 promoter polymorphisms in colorectal cancer. Anticancer Res 2006;26:791e795. 18. Lin SC, Lo SS, Liu CJ, et al. Functional genotype in matrix metalloproteinases-2 promoter is a risk factor for oral carcinogenesis. J Oral Pathol Med 2004;33:405e409. 19. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ 2003;327:557e560. 20. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7:177e188. 21. Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 1959;22:719e748. 22. Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629e634. 23. Zeidel A, Beilin B, Yardeni I, et al. Immune response in asymptomatic smokers. Acta Anaesthesiol Scand 2002;46:959e964.