The -93G>A (rs1800734) polymorphism located in the promoter of mismatch repair gene, MLH1, has been identified as a low-penetrance variant for cancer risk. Many published studies have evaluated the association between the MLH1 -93G>A polymorphism and colorectal cancer (CRC) risk. However, the results remain conflicting rather than conclusive.
The aim of this study was to assess the association between the MLH1 -93G>A polymorphism and the risk of CRC.
To derive a more precise estimation of the association, a meta-analysis of six studies (17,791 cases and 13,782 controls) was performed. Odds ratios (ORs) and 95% confidence intervals (CIs) were used to evaluate the strength of the association. Four of these published studies were performed on subjects of known microsatellite instability (MSI) status. An additional analysis including 742 cases and 10,895 controls was used to assess the association between the MLH1 -93G>A polymorphism and the risk of MSI-CRC.
The overall results indicated that the variant genotypes were associated with a significantly increased risk of CRC (AG versus GG: OR = 1.06, 95% CI = 1.01–1.11; AA/AG versus GG: OR = 1.06, 95% CI = 1.01–1.11). This increased risk was also found during stratified analysis of MSI status (AA versus GG: OR = 2.52, 95% CI = 1.94–3.28; AG versus GG: OR = 1.29, 95% CI = 1.10–1.52; AA/AG versus GG: OR = 1.45, 95% CI = 1.24–1.68; AA versus AG/GG: OR = 2.29, 95% CI = 1.78–2.96). Egger’s test did not show any evidence of publication bias.
Our results suggest that the MLH1 -93G>A polymorphism may contribute to individual susceptibility to CRC and act as a risk factor for MSI-CRC.
Citation: Wang T, Liu Y, Sima L, Shi L, Wang Z, Ni C, et al. (2012) Association between MLH1 -93G>A Polymorphism and Risk of Colorectal Cancer. PLoS ONE 7(11): e50449. doi:10.1371/journal.pone.0050449
Editor: Amanda Ewart Toland, Ohio State University Medical Center, United States of America
Received: August 9, 2012; Accepted: October 22, 2012; Published: November 30, 2012
Copyright: © 2012 Wang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was partially supported by National Natural Science Foundation of China (81102089), National Natural Science Foundation of Jiangsu Province (BK2011773), the Key Program for Basic Research of Jiangsu Provincial Department of Education (11KJB330002), the Practice Innovation Training Program Projects for the Jiangsu College Students (2011), and the Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (Public Health and Preventive Medicine). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Colorectal cancer (CRC) is the third most common cancer worldwide. There were over 1.2 million new cases and an estimated 608,700 deaths in 2008 alone . Accumulating evidence suggests that CRC is caused by a set complex of interactions between environmental and genetic factors . Deficiency in DNA mismatch repair (MMR) plays several important roles in the etiology of CRC. The MMR genes encode a family of highly conserved proteins, including MLH1, MSH2, MSH6, and PMS2 , . MMR systems promote genetic stability by repairing DNA replication errors, inhibiting recombination between non-identical DNA sequences, and participating in responses to DNA damage . DNA replication errors and mispairings cause microsatellite instability (MSI), a phenomenon frequently observed in sporadic CRC . Rare constitutional mutations and methylation of MLH1 and other MMR genes are the primary causes of the autosomal dominant disorder hereditary non-polyposis colorectal cancer (HNPCC) , . MMR genes also contain common single nucleotide polymorphisms (SNPs) which can predispose individuals to sporadic CRC with low to moderate penetrance .
The -93G>A (rs1800734) polymorphism is located in the promoter region of MLH1, which is responsible for maximal transcriptional activity of this gene , . Although the association between the MLH1 -93G>A polymorphism and CRC risk has been demonstrated in several studies, results remain inconsistent. This may be partially due to the relatively small sample size evaluated in each study. To estimate the overall risk of the MLH1 -93G>A polymorphism associated with CRC risk and to quantify potential inter-study heterogeneity, we conducted a meta-analysis on six published case-control studies with a total of 17,791 CRC cases and 13,782 controls.
The areas of the squares reﬂect the study-specific weight (inverse of the variance). The diamonds represent the summary OR and 95% CI. The unbroken vertical line is at the null value (OR = 1.0).
Materials and Methods
Identification and Eligibility of Relevant Studies
We searched the PubMed and EMBASE databases for all relevant articles. The last search update was June 1, 2012, using the search terms “MLH1,” “polymorphism,” and “colorectal cancer.” The search was limited to English-language articles. Additional studies were identified by a manual search of the references of the original studies. In the cases of multiple studies with the same or overlapping data published by the same investigators, we selected the most recent study with the largest number of subjects. Studies included in our meta-analysis met the following criteria: (a) evaluation of the MLH1 -93G.>A polymorphism and CRC risk, (b) case-control design, and (c) sufficient published data for evaluation of the frequencies of various genotypes in cases and controls.
This figure shows the influence of individual studies on the summary OR. The middle vertical axis indicates the overall OR and the two vertical axes indicate the pooled OR when the left study is omitted from the meta-analysis. The two ends of the dotted lines represent the 95% CI.
Each point represents a separate study of the indicated association. Log [or] is the natural logarithm of OR. Horizontal lines indicate the magnitude of the mean effect.
The following variables were extracted by two of the authors of the present paper (Ting Wang and Yang Liu). In the cases of conﬂicting evaluations, agreement was reached after a discussion. For each study, the following data were extracted: the first author’s surname, year of publication, country of origin, ethnicity of study subjects, source of controls, matching criteria, and sample size. Subjects were categorized as European, Asian, or mixed ethnicity. For studies that included subjects from different countries, data were extracted separately for each country group whenever possible.
Hardy-Weinberg equilibrium (HWE) was evaluated for each study using a goodness-of-fit chi-square test. Odds ratios (ORs) with 95% confidence intervals (CIs) were used to assess the strength of association between the MLH1 -93G>A polymorphism and CRC risk. The pooled ORs were performed for co-dominant model (AA versus GG, or AG versus GG), dominant model (AA/AG versus GG), and recessive model (AA versus AG/GG). To assess the heterogeneity between the studies, a statistical test for heterogeneity was performed based on the Q statistic . Where the studies were shown to be homogeneous with a P>0.10 for the Q test, the summary of OR estimate of each study was calculated using a fixed-effects model (the Mantel–Haenszel method) . If there was significant heterogeneity, the random-effects model (the DerSimonian and Laird method) was used . The stability of the results was assessed using sensitivity analyses, by randomly deleting combinations of cases and controls from different studies in the meta-analysis. Funnel plots and Egger’s linear regression test were used to assess publication bias . The same measures were taken to estimate the degree of association between the MLH1 -93G>A polymorphism and MSI-CRC risk. All analyses were performed with Stata software (version 8.2; StataCorp LP, College Station, TX), using two-sided P values.
There were 312 published articles relevant to the search terms (Figure 1). By choosing additional filters, 285 of these papers were excluded (20 were not in English, 10 were not performed in humans, 246 were not performed on CRC, and 9 did not involve polymorphisms). By screening the titles and the abstracts, 21 of these studies were excluded. Only six articles were left for full publication review, and an additional two studies were included by manual searching of the reference lists of retrieved studies. Among these eligible full-text articles, one study was excluded for its small sample size in HNPCC . Another was excluded because the controls had a history of CRC . Finally, a total of six eligible studies involving 17,791 cases and 13,782 controls were included in the pooled analyses , –. The characteristics of selected studies are summarized in Table 1. The CRC cases were confirmed histologically or pathologically in most studies. Controls were usually matched with respect to age and sex. The distribution of genotypes in the controls was consistent with HWE in all studies.
As shown in Table 2, the MLH1 -93G>A polymorphism was significantly associated with increased risk of CRC in two genetic models: AG versus GG (OR = 1.06, 95% CI = 1.01–1.11), and AA/AG versus GG (OR = 1.06, 95% CI = 1.01–1.11, Figure 2A) (each study weighting for each study was from 4.1% to 54.2%). In the stratified analysis of MSI status, similar increased risks were found (AA versus GG: OR = 2.52, 95% CI = 1.94–3.28;AG versus GG: OR = 1.29, 95% CI = 1.10–1.52; AA/AG versus GG: OR = 1.45, 95% CI = 1.24–1.68; AA versus AG/GG: OR = 2.29, 95% CI = 1.78–2.96, Figure 2B) (each study was weighed from 5.8% to 50.5%).
Tests for Heterogeneity, Sensitivity Analyses, and Publication Bias
No significant heterogeneity was observed between studies during overall comparisons (Table 2). A single study was removed from meta-analysis each time to determine the inﬂuence of its individual data sets to the pooled ORs, and the corresponding pooled ORs were not materially altered (Figure 3). Begg’s funnel plot and Egger’s test were performed to assess publication bias. The shapes of the funnel plots did not reveal any evidence of obvious asymmetry in any of the models. Then, the Egger’s test was used to provide statistical evidence of funnel plot symmetry. The results did not show any evidence of publication bias for CRC (t = 0.29, P = 0.789, Figure 4A) and MSI-CRC (t = 0.63, P = 0.592, Figure 4B).
MMR genes are among the most important DNA repair genes . Variations in MMR genes may alter predisposition to malignant tumors, especially CRC . Previous studies have shown that the MLH1 -93G>A polymorphism is associated with increased risk of CRC , , . Because two transcription binding sites, NF-IL6 and GT-IIB, exist in this promoter region of the MLH1 gene, the -93G>A polymorphism may reduce MLH1 transcription and expression, thereby reducing overall DNA repair capability .
Several investigators have investigated the association between this polymorphism and risk of CRC, but the results have been inconclusive. In an analysis of 1,518 patients with CRC, Allan et al. demonstrated that the -93A variant was associated with a significantly increased risk of MLH1-deficient CRC detected by immunohistochemistry, though this polymorphism was not correlated with MSI-CRC . Whiffin et al. found a significant difference in -93A allele distribution between a large number of CRC patients and control subjects . However, Campbell et al.  and Koessler et al.  found that the -93G>A polymorphism was not associated with risk of CRC. To explain these conflicting results, a meta-analysis of six studies involving 17,791 cases and 13,782 controls was conducted to derive amore precise estimation of the association. Our results suggested that the -93G>A polymorphism was associated with increased risk of CRC among the studies populations. One biologically plausible mechanism underlying this association may be that the -93G>A substitution alters promoter function, and the -93A allele may be associated with reduced activity . It is also possible that altered transcription factor binding in the MLH1 promoter region causes promoter methylation and gene silencing , .
Although three similar meta-analyses have reported the existence of an association between the MLH1 -93G>A polymorphism and the risk of CRC , , , they showed some different results. For instance, Whiffin et al. reported the pooled effect on the -93G>A polymorphism and CRC based on five case-control studies with a total of 14,121 CRC cases and 10,890 controls . They provided more evidence an association between the -93G>A polymorphism and the risk of CRC. Pan et al. carried out a meta-analysis on all then-published case-control studies to estimate the overall cancer risk of the MLH1 -93G>A polymorphism and showed that this polymorphism was not associated with increased risk of cancer. This null result was inconsistent with our results, possibly because the pooled data evaluated by these teams may have included incorrect information. For example, the controls included in the study by Chen et al. were all CRC patients without MLH1 methylation, not individuals without a history of CRC. Therefore, it would be valuable if Pan et al. could provide a new more accurate estimation of the pooled results after excluding the data reported by Chen et al. . Xu et al. performed a meta-analysis to assess the overall contributions of -93G>A and I219V polymorphisms in the MLH1 gene to cancer susceptibility. They found that the -93G>A polymorphism was associated with increased risk of CRC, but the I219V polymorphism was not. The association between the -93G>A polymorphism and the risk of MSI-CRC could not be estimated. In this way, our meta-analysis provides an overview of the relevant studies and generates more exact pooled results of the associations between the -93G>A polymorphism and risk of CRC.
MSI is characterized by widespread changes in the length of short regions of repetitive DNA sequences, called microsatellites. A number of studies have reported associations between the -93G>A polymorphism and susceptibility to MSI-CRC. Our results showed that the -93G>A polymorphism could contribute to the risk of MSI-CRC. This was compatible with previous results. Raptis et al. showed the MLH1 -93G>A variant allele to be associated with a higher risk of developing MSI-CRC than the wild-type allele . Whiffin et al. also showed the effect on the -93G>A polymorphism and CRC risk to be confined to MSI-CRC, acting a marker for a somatic event . There have been relatively few studies of CRC and microsatellite stability (MSS-CRC). In this way, the sample size for the -93G>A polymorphism and MSS-CRC was too small for meta-analysis. However, no significant association between the -93G>A polymorphism and MSS-CRC risk has been demonstrated in any previous study , .
This meta-analysis has limitations that must be acknowledged. First, the eligible studies included only published studies; so any preexisting publication bias will be reflected in our results, although the statistical data may not show it. Second, our lack of access to original data from the reviewed studies limited further evaluation of potential interactions. For example, the smoking status and die can help evaluate gene-gene and gene-environment interactions. Third, misclassifications of disease status and genotypes may also inﬂuence the results; many of the cases in several studies were not confirmed by pathology or other gold-standard methods. This is especially relevant to diversity within MSI status.
In conclusion, our meta-analysis suggests that the MLH1 -93G>A polymorphism is associated with an increased risk of MSI-CRC. Large and well-designed epidemiological studies are warranted to validate our findings.
Conceived and designed the experiments: TW YL MW. Performed the experiments: TW YL MW. Analyzed the data: TW YL L. Sima. Contributed reagents/materials/analysis tools: TW YL L. Sima L. Shi ZW CN ZZ MW. Wrote the paper: TW YL L. Sima.
- 1. Garcia-Albeniz X, Chan AT (2011) Aspirin for the prevention of colorectal cancer. Best Pract Res Clin Gastroenterol 25: 461–72.
- 2. Malekzadeh R, Bishehsari F, Mahdavinia M, Ansari R (2009) Epidemiology and molecular genetics of colorectal cancer in iran: a review. Arch Iran Med 12: 161–9.
- 3. de la Chapelle A (2004) Genetic predisposition to colorectal cancer. Nat Rev Cancer 4: 769–80.
- 4. Kunkel TA, Erie DA (2005) DNA mismatch repair. Annu Rev Biochem 74: 681–710.
- 5. Jean M, Pelletier J, Hilpert M, Belzile F, Kunze R (1999) Isolation and characterization of AtMLH1, a MutL homologue from Arabidopsis thaliana. Mol Gen Genet 262: 633–42.
- 6. Campbell PT, Curtin K, Ulrich CM, Samowitz WS, Bigler J, et al. (2009) Mismatch repair polymorphisms and risk of colon cancer, tumour microsatellite instability and interactions with lifestyle factors. Gut 58: 661–7.
- 7. Bronner CE, Baker SM, Morrison PT, Warren G, Smith LG, et al. (1994) Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with hereditary non-polyposis colon cancer. Nature 368: 258–61.
- 8. Peltomaki P, Vasen H (2004) Mutations associated with HNPCC predisposition – Update of ICG-HNPCC/INSiGHT mutation database. Dis Markers 20: 269–76.
- 9. Tomlinson IP, Houlston RS, Montgomery GW, Sieber OM, Dunlop MG (2012) Investigation of the effects of DNA repair gene polymorphisms on the risk of colorectal cancer. Mutagenesis 27: 219–23.
- 10. Arita M, Zhong X, Min Z, Hemmi H, Shimatake H (2003) Multiple sites required for expression in 5′-flanking region of the hMLH1 gene. Gene 306: 57–65.
- 11. Ito EYY, Iwahashi Y, Suzuki Y, Nagasaki H, AkiyamaY, et al (1999) A core promoter and a frequent single-nucleotide polymorphism of the mismatch repair gene hMLH1. Biochem Biophys Res Commun 256: 488–494.
- 12. Handoll HH (2006) Systematic reviews on rehabilitation interventions. Arch Phys Med Rehabil 87: 875.
- 13. Mantel N, Haenszel W (1959) Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 22: 719–48.
- 14. DerSimonian R, Laird N (1986) Meta-analysis in clinical trials. Control Clin Trials 7: 177–88.
- 15. Egger M, Davey Smith G, Schneider M, Minder C (1997) Bias in meta-analysis detected by a simple, graphical test. BMJ 315: 629–34.
- 16. Chen H, Taylor NP, Sotamaa KM, Mutch DG, Powell MA, et al. (2007) Evidence for heritable predisposition to epigenetic silencing of MLH1. Int J Cancer 120: 1684–8.
- 17. Allan JM, Shorto J, Adlard J, Bury J, Coggins R, et al. (2008) MLH1 -93G>A promoter polymorphism and risk of mismatch repair deficient colorectal cancer. Int J Cancer 123: 2456–9.
- 18. Koessler T, Oestergaard MZ, Song H, Tyrer J, Perkins B, et al. (2008) Common variants in mismatch repair genes and risk of colorectal cancer. Gut 57: 1097–101.
- 19. Raptis S, Mrkonjic M, Green RC, Pethe VV, Monga N, et al. (2007) MLH1 -93G>A promoter polymorphism and the risk of microsatellite-unstable colorectal cancer. J Natl Cancer Inst 99: 463–74.
- 20. Samowitz WS, Curtin K, Wolff RK, Albertsen H, Sweeney C, et al. (2008) The MLH1 -93G>A promoter polymorphism and genetic and epigenetic alterations in colon cancer. Genes Chromosomes Cancer 47: 835–44.
- 21. Tulupova E, Kumar R, Hanova M, Slyskova J, Pardini B, et al. (2008) Do polymorphisms and haplotypes of mismatch repair genes modulate risk of sporadic colorectal cancer? Mutat Res 648: 40–5.
- 22. Whiffin N, Broderick P, Lubbe SJ, Pittman AM, Penegar S, et al. (2011) MLH1-93G>A is a risk factor for MSI colorectal cancer. Carcinogenesis 32: 1157–61.
- 23. Mohrenweiser HW, Jones IM (1998) Variation in DNA repair is a factor in cancer susceptibility: a paradigm for the promises and perils of individual and population risk estimation? Mutat Res 400: 15–24.
- 24. Knudson AG (1997) Hereditary predisposition to cancer. Ann N Y Acad Sci 833: 58–67.
- 25. Ito E, Yanagisawa Y, Iwahashi Y, Suzuki Y, Nagasaki H, et al. (1999) A core promoter and a frequent single-nucleotide polymorphism of the mismatch repair gene hMLH1. Biochem Biophys Res Commun 256: 488–94.
- 26. Perera S, Mrkonjic M, Rawson JB, Bapat B (2011) Functional effects of the MLH1-93G>A polymorphism on MLH1/EPM2AIP1 promoter activity. Oncol Rep 25: 809–15.
- 27. Goldsborough AS, Kornberg TB (1996) Reduction of transcription by homologue asynapsis in Drosophila imaginal discs. Nature 381: 807–10.
- 28. Pan XM, Yang WZ, Xu GH, Bai P, Qin HJ, et al. (2011) The association between MLH1 -93G>A polymorphism of DNA mismatch repair and cancer susceptibility: a meta-analysis. Mutagenesis 26: 667–73.
- 29. Xu JL, Yin ZQ, Huang MD, Wang XF, Gao W, et al. (2012) MLH1 polymorphisms and cancer risk: a meta-analysis based on 33 case-control studies. Asian Pac J Cancer Prev 13: 901–7.