Mutagenesis, Vol. 15, No. 5, 399-404,
September 2000
© 2000 UK Environmental Mutagen Society/Oxford University Press
Glutathione S-transferase µ1 (GSTM1) status and bladder cancer risk: a meta-analysis
Section of Cancer Genetics, Institute of Cancer Research, Sutton SM2 5NG, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Inter-individual differences in bladder cancer susceptibility may be mediated in part through polymorphic variability in the bioactivation and detoxification of procarcinogens. Glutathione S-transferase µ1 (GSTM1) status has been extensively studied as a risk factor in this context. To clarify the impact of GSTM1 deficiency on bladder cancer risk a meta-analysis of 15 case–control studies from the literature has been carried out using a random effects model. The principal outcome measure was the odds ratio for the risk of bladder cancer. Pooling the studies the odds ratio of bladder cancer risk associated with GSTM1 deficiency was 1.53 (95% confidence limits 1.28–1.84). The relationship between GSTM1 status and bladder cancer risk was not confined to a specific population. This meta-analysis supports the hypothesis that GSTM1 deficiency is a determinant of bladder cancer susceptibility. A review of studies does, however, indicate that greater attention should therefore be paid to the design of future studies. The interaction between GSTM1 and other polymorphisms on the risk of bladder cancer and their interaction with environmental risk factors will only be addressed by well-designed studies based on sample sizes commensurate with the detection of small genotypic risks.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Bladder cancer accounts for approximately 6% of all cancers (with an incidence in men of 30 per 100 000 and 10 per 100 000 in women in the UK), is the fifth most frequent cancer in men and has a peak prevalence in the seventh decade (Office of National Statistics, 1992
). Exposure to carcinogens such as 2-napthylamine and other amine compounds is well established as a risk factor for bladder cancer, as is tobacco smoke, where the risk in smokers is between two and six times greater than in non-smokers. Other recognized risk factors include schistosomiasis infection, common in many parts of Africa, and exstrophy of the bladder (Horwich, 1995).
There is a growing realization that the development of most cancers results from a complex interaction of both environmental and genetic factors. Epidemiological studies have shown that relatives of bladder cancer cases are at a 2-fold elevated risk of developing the disease (Houlston and Peto, 1996). It is likely that part of the susceptibility to bladder cancer may be determined by inter-individual differences in the bioactivation of procarcinogens and detoxification of carcinogens. Glutathione S-transferase µ1 (GSTM1) has been of considerable interest as a bladder cancer susceptibility gene in this context. The biochemical basis for a possible association is that GSTM1 is one of a family of glutathione S-transferases capable of detoxifying reactive electrophiles that can act as mutagens. Hence, GSTM1 might be involved in the inactivation of this class of procarcinogens. The GSTM1 gene is polymorphic and at least four alleles exist (Smith et al., 1995). The GSTM1*0 allele represents a deletion and individuals homozygous for this null allele are less efficient at conjugating and detoxifying specific substrate intermediates of carcinogens (Smith et al., 1995).
Fifteen studies have appeared in the literature suggesting or refuting an association between GSTM1 deficiency and bladder cancer risk (Table I). To clarify the effect of GSTM1 status on the risk of bladder cancer, a meta-analysis of published studies has been undertaken.
|
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Identification of studies
A search of the literature was made using two electronic databases, MEDLINE and BIDS EMBASE, in order to identify articles in which GSTM1 status was determined in bladder cancer patients and controls. Additional articles were ascertained through references cited in these publications. Articles included for analyses were primary references and were of case–control design.
Statistical analysis
The odds ratio of bladder cancer associated with GSTM1 deficiency was estimated for each study. These odds ratios and their corresponding 95% confidence intervals were plotted against the number of participants in each of the studies in order to detect any obvious sample size bias. To take into account the possibility of heterogeneity between studies a random effects model was used for the derivation of odds ratios (DerSimonian and Laird, 1986). This model assumes that the studies in question are a random sample of a hypothetical population of studies taking into account within and between study variability. Statistical manipulations were undertaken using the programme Meta-analyst0.989 (obtainable from joseph.lau{at}es.nemc.org). The power of each study was computed as the probability of detecting an association between GSTM1 deficiency and bladder cancer at the 0.05 level of significance, assuming a genotypic risk of 1.5 and 2.0. These estimates of power were performed on the basis of the method published by Fleiss et al. (1980), using the statistical program POWER (Epicenter Software, v.1.30). The aetiological fraction, the proportion of bladder cancer that can be attributed to the GSTM1 status, was calculated from the odds ratio (OR) under the assumption that deficiency can be treated like exposure to a risk factor (Pe). The aetiological fraction is then given by Pe(OR – 1)/[Pe(OR – 1) + 1], under the assumption that Pe (i.e. GSTM1 deficiency) in the control group is similar to that in the target population (Schlesselman, 1982).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Fifteen reports detailing case–control studies of the possible association between GSTM1 status and bladder cancer risk were identified from the literature and judged suitable for analysis (Table I
). Reasons for excluding reports from the analysis were because overlapping data were available in more than one study (Brockmoller et al., 1996a), because no control group of individuals had been ascertained (Brockmoller et al., 1996b; Gabbini et al., 1996; Golka et al., 1997) or because data were presented on urothelial cancer rather than bladder cancer (Katoh et al., 1995).
Of the reports selected for meta-analysis, two assigned GSTM1 status solely by phenotyping (Table I). The ethnicity of cases and controls was detailed in most, but not all, studies (Table I). In four of the 15 studies the controls were age-matched individuals from the general population. Although not universal, smoking histories and industrial exposure to carcinogens had been ascertained from cases and controls in a number of studies (Table I). In some of these studies the relationship between GSTM1 status and bladder cancer risk was analysed in a stratified manner or by logistic regression taking into account smoking information (Table I). In a number of the studies there were no data on the ages of cases or controls. In seven of the studies hospital patients were used as controls (Table I).
Bell et al. (1993) and Lin et al. (1994) reported the frequency of GSTM1 deficiency in cases and controls in a number of different ethnic groups. Since the number of non-Caucasian cases analysed in both studies are tiny, only the data on the Caucasian cases and controls were therefore included in this meta-analysis.
Table I shows the power of individual studies to demonstrate an association between GSTM1 deficiency and bladder cancer risk if the true risk was 1.5 or 2. Given a genotypic risk 
2.0 (
= 0.05), five of the 15 studies had 80% power to demonstrate an association. However, if the genotypic risk was 1.5 or less all studies had <80% power.
Figure 1 shows a plot of odds ratios with 95% confidence limits for the risk of developing bladder cancer associated with GSTM1 deficiency in the 15 case–control studies. The median odds ratio value was greater than unity in 13 of the studies, but was statistically significant (P < 0.05) in only eight. A plot of GSTM1 deficiency and bladder cancer risk showed a trend towards a less significant association between GSTM1 status in the larger of the studies (Figure 1). Pooling all studies the odds ratio of bladder cancer associated with GSTM1 deficiency was 1.53 (95% CI 1.28–1.84).
|
It is conceivable that GSTM1 deficiency may be associated with a specific form of bladder cancer. The vast majority of uroepithelial tumours presenting in the developed countries are transitional cell carcinomas whereas squamous cell carcinomas of the bladder are common in countries such as Egypt where the prevalence of schistomiasis is high. In most studies the GSTM1 status of bladder cancers were not detailed according to histology permitting a pooled analysis to be carried out by subgroup. However, restricting the analysis to the studies of European and US populations, where the risk will be primarily for transitional cancer, the odds ratio associated with GSTM1 deficiency is 1.47 (95% CI 1.21–1.79). Pooling the three studies of Egyptian patients the odds ratio of bladder cancer associated with GSTM1 deficiency is 2.57 (95% CI 1.10–6.02).
Although GSTM1 phenotypes and genotypes are highly correlated, concordance is not absolute (coefficient of association 0.85–0.97) (Hirvonen et al., 1993; Nazar-Stewart et al., 1993). Misclassification of GSTM1 status on the basis of phenotypes is therefore a possibility in some studies. Most studies have determined GSTM1 status by genotyping using PCR methods. Restricting the analysis to these studies in Caucasians the odds ratio is 1.49 (95% CI 1.24–1.79).
Information on smoking was ascertained in 11 of the studies and interaction between smoking and GSTM1 deficiency on the risk of bladder cancer examined in some studies. An interaction between smoking and GSTM1 deficiency on bladder cancer risk was seen in the studies of Bell et al. (1993), Lafuente et al. (1993) and Katoh et al. (1998), however, this was not universal and other studies, such as that reported by Brockmoller et al. (1996a), found no evidence for synergism (Table I). Little of the data presented in the studies permits a pooled analysis of smoking interaction to be undertaken.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Given that exposure to carcinogens is recognized to be a risk factor for bladder cancer, modulation of carcinogen metabolism under genetic control is a plausible mechanism for explaining inter-individual susceptibility. GSTM1 deficiency has been evaluated as a risk factor for bladder cancer by a number of researchers. It is not uncommon for the first small published studies to report over-inflated estimates of risk or effects, which subsequent larger studies cannot replicate. The continuing debate about the possible role of GSTM1 deficiency as a bladder cancer risk factor prompted the present meta-analysis in order to derive an estimate of the risk associated with GSTM1 status. It has been argued that it is difficult to pool the results of different epidemiological studies of polymorphisms because of differences in methodology and the amount and type of carcinogen exposure between studies (Smith et al., 1995
). The underlying basis of meta-analysis is, however, not to directly combine results from each study but a relative measure of the observed effect supporting or rejecting a specific hypothesis. An advantage of this statistical procedure is the amalgamation of data collected and analysed by different methods. Providing that the postulated risk factor reflects the same biological phenomenon in each study, as is the case with GSTM1 status, the criticisms advanced by Smith et al. (1995) do not hold. The evidence from this analysis supports the notion that GSTM1 deficiency is associated with an increased risk of bladder cancer. The only caveat to this conclusion is that it is possible for negative findings to go unreported. An effect of GSTM1 status on bladder cancer risk may be mediated either locally, within the bladder urothelium, or systemically by detoxifying foreign compounds in the liver or other tissues (Berendsen et al., 1997; Brockmoller et al., 1996a).
This overview indicates that the design of some of the studies which have evaluated GSTM1 as a risk factor for bladder cancer is less than ideal. Consideration of sample size is crucial to the design of any study of the relationship between common genetic variants and cancer risk. Given that the genotypic risk associated with any metabolic polymorphism is unlikely to be greater than 2 and realistically will be 1.5 or less, many of the studies of GSTM1 and bladder cancer risk are clearly seriously under-powered. The issue of false positive findings in association studies is also a concern. Any stratification within a population sample may lead to spurious evidence for an association between the marker and disease. Avoidance of this problem requires the identification of sub-populations defined in terms of factors influencing disease and marker allele frequencies. These include ethnicity and geographical origin. In a number of the studies the ethnicity of cases and controls were possibly mixed. The frequency of GSTM1 deficiency may vary between ethnic groups (Bell et al., 1993; Katoh et al., 1995; Rothman et al., 1996), therefore, a failure to match cases and controls represents a source of bias. A number of the studies were based on a comparison of cases and hospital patient controls. The use of healthy population controls is preferable as GSTM1 deficiency may confer susceptibility to some non-malignant diseases. Incident and surviving cases are two different groups in terms of any factor that might influence survival. Therefore, the use of prevalent cases represents another potential source of bias, albeit a very minor one.
Substantial research has been carried out evaluating the relationship between GSTM1 status and bladder cancer risk. Has this been worthwhile? This meta-analysis indicates that a small but significant increase in risk of bladder cancer is associated with GSTM1 deficiency. Although the risk is modest, the high prevalence of deficiency in the general population means that heritable GSTM1 deficiency is likely to make a significant impact on bladder cancer incidence. In Caucasians complete lack of GSTM1 activity is seen in ~50% of the population. The 1.47-fold increase in bladder cancer risk associated with deficiency translates to GSTM1 deficiency being responsible for around one fifth of cases.
If, as seems probable, genetic susceptibility to bladder cancer is mediated partly by polymorphic variation, it is likely that the risk associated with any one locus will be small. Hence, combinations of certain genotypes may be more discriminating as risk factors than a single locus genotype. For example, N-acetyltransferase slow acetylator status (through reduced detoxification of aromatic amines) has been shown to be a risk factor for bladder cancer (Smith et al., 1995). Slow acetylator status in conjunction with GSTM1 deficiency may therefore define a specific high risk group. The studies that have examined this possibility are inconclusive to date (Brockmoller et al., 1996a; Golka et al., 1997; Gabbani et al., 1996; Peluso et al., 1998) but are all under-powered. A clearer picture of the interaction between different polymorphisms on the risk of bladder cancer and their interaction with environmental risk factors is likely to only be addressed by large studies.
|
Acknowledgments |
|---|
We are grateful to David Phillips for his comments on this manuscript.
|
Notes |
|---|
* To whom correspondence should be addressed. Tel: +44 0181 722 4175; Fax: +44 0181 643 0549; Email: r.houlston{at}icr.ac.uk
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
-
Abdel-Rahman,S.Z., Anwar,W.A., Abdel-Aal,W.E., Mostafa,H.M. and Au,W.W. (1998) GSTM1 and GSTT1 genes are potential risk modifiers for bladder cancer. Cancer Detect. Prev., 22, 129–138.[Web of Science][Medline]
Anwar,W.A., Abdel-Rahman,S.Z., El-Zein,R.A., Mostafa,H.M. and Au,W.W. (1996) Genetic polymorphism of GSTM1, CYP2E1 and CYP2D6 in Egyptian bladder cancer patients. Carcinogenesis, 17, 1923–1929.
Bell,D.A., Taylor,J.A., Paulson,D.F., Robertson,C.N., Mohler,J.L. and Lucier,G.W. (1993) Genetic risk and carcinogen exposure: a common inherited defect of the carcinogen-metabolism gene glutathione S-transferase M1 (GSTM1) that increases susceptibility to bladder cancer. J. Natl Cancer Inst., 85, 1159–1164.
Berendsen,C.L., Peters,W.H., Scheffer,P.G., Bouman,A.A., Boven,E. and Newling,D.W. (1997) Glutathione S-transferase activity and subunit composition in transitional cell cancer and mucosa of the human bladder. Urology, 49, 644–651.[Web of Science][Medline]
Brockmoller,J., Kerb,R., Drakoulis,N., Staffeldt,B. and Roots,I. (1994) Glutathione S-transferase M1 and its variants A and B as host factors of bladder cancer susceptibility: a case-control study. Cancer Res., 54, 4103–4111.
Brockmoller,J., Cascorbi,I., Kerb,R. and Roots,I. (1996a) Combined analysis of inherited polymorphisms in arylamine N-acetyltransferase 2, glutathione S-transferases M1 and T1, microsomal epoxide hydrolase and cytochrome P450 enzymes as modulators of bladder cancer risk. Cancer Res., 56, 3915–3925.
Brockmoller,J., Kaiser,R., Kerb,R., Cascorbi,I., Jaeger,V. and Roots,I. (1996b) Polymorphic enzymes of xenobiotic metabolism as modulators of acquired P53 mutations in bladder cancer. Pharmacogenetics, 6, 535–545.[Web of Science][Medline]
Daly,A.K., Thomas,D.J., Cooper,J., Pearson,W.R., Neal,D.E. and Idle,J.R. (1993) Homozygous deletion of gene for glutathione S-transferase M1 in bladder cancer. Br. Med. J., 307, 481–482.
DerSimonian,R. and Laird,N. (1986) Meta-analysis in clinical trials. Control. Clin. Trials, 7, 177–188.[Web of Science][Medline]
Fleiss,J.L., Tytun,A. and Uray,H.K. (1980) A simple approximation for calculating sample sizes for comparing independent proportions. Biometrics, 36, 343–346.[Web of Science]
Gabbani,G., Hou,S.M., Nardini,B., Marchioro,M., Lambert,B. and Clonfero,E. (1996) GSTM1 and NAT2 genotypes and urinary mutagens in coke oven workers. Carcinogenesis, 17, 1677–1681.
Golka,K., Reckwitz,T., Kempkes,M., Cascorbi,I., Blaskewicz,M., Reich,S.E., Roots,I., Soekeland,J., Schulze,H. and Bolt,H.M. (1997) N-Acetyltransferase 2 (NAT2) and glutathione S-transferase mu (GSTM1) in bladder-cancer patients in a highly industrialized area. Int. J. Occup. Environ. Health, 3, 105–110.[Medline]
Houlston,R.S. and Peto,J. (1996) Genetics of common cancers. In Eeles,R.A., Ponder,B., Easton,D.E. and Horwich,A. (eds) Inherited Predisposition to Cancer. Chapman & Hall, London, UK, pp. 208–226.
Horwich,A. (1995) Bladder cancer. In Horwich,A. (ed.) Oncology—A Multidisciplinary Test Book. Chapman & Hall, London, UK, pp. 471–484.
Katoh,T., Inatomi,H., Nagaoka,A. and Sugita,A. (1995) Cytochrome P4501A1 gene polymorphism and homozygous deletion of the glutathione S-transferase M1 gene in urothelial cancer patients. Carcinogenesis, 16, 655–657.
Katoh,T., Inatomi,H., Kim,H., Yang,M., Matsumoto,T. and Kawamoto,T. (1998) Effects of glutathione S-transferase (GST) M1 and GSTT1 genotypes on urothelial cancer risk. Cancer Lett., 132, 147–152.[Web of Science][Medline]
Kempkes,M., Golka,K., Reich,S., Reckwitz,T. and Bolt,H.M. (1996) Glutathione S-transferase GSTM1 and GSTT1 null genotypes as potential risk factors for urothelial cancer of the bladder. Arch. Toxicol., 71, 123–126.[Web of Science][Medline]
Lafuente,A., Pujol,F., Carretero,P., Villa,J.P. and Cuchi,A. (1993) Human glutathione S-transferase mu (GST mu) deficiency as a marker for the susceptibility to bladder and larynx cancer among smokers. Cancer Lett., 68, 49–54.[Web of Science][Medline]
Lafuente,A., Zakahary,M.M., el-Aziz,M.A., Ascaso,C., Lafuente,M.J., Trias,M. and Carretero,P. (1996) Influence of smoking in the glutathione-S-transferase M1 deficiency-associated risk for squamous cell carcinoma of the bladder in schistosomiasis patients in Egypt. Br. J. Cancer, 74, 836–838.[Web of Science][Medline]
Lin,H.J., Han,C.Y., Bernstein,D.A., Hsiao,W., Lin,B.K. and Hardy,S. (1994) Ethnic distribution of the glutathione transferase Mu 1-1 (GSTM1) null genotype in 1473 individuals and application to bladder cancer susceptibility. Carcinogenesis, 15, 1077–1081.
Okkels,H., Sigsgaard,T., Wolf,H. and Autrup,H. (1996) Glutathione S-transferase mu as a risk factor in bladder tumours. Pharmacogenetics, 6, 251–256.[Web of Science][Medline]
Office of National Statistics (1992) Cancer Registrations. HMSO, London, UK.
Peluso,M., Airoldi,L., Armelle,M., Martone,T., Coda,R., Malaveille,C., Giacomelli,G., Terrone,C., Casetta,G. and Vineis,P. (1998) White blood cell DNA adducts, smoking and NAT2 and GSTM1 genotypes in bladder cancer: a case-control study. Cancer Epidemiol. Biomarkers Prev., 7, 341–346.[Abstract]
Rothman,N., Hayes,R.B., Zenser,T.V., DeMarini,D.M., Bi,W., Hirvonen,A., Talaska,G., Bhatnagar,V.K., Caporaso,N.E., Brooks,L.R., Lakshmi,V.M., Feng,P., Kashyap,S.K., You,X., Eischen,B.T., Kashyap,R., Shelton,M.L., Hsu,F.F., Jaeger,M., Parikh,D.J., Davis,B.B., Yin,S. and Bell,D.A. (1996) The glutathione S-transferase M1 (GSTM1) null genotype and benzidine-associated bladder cancer, urine mutagenicity and exfoliated urothelial cell DNA adducts. Cancer Epidemiol. Biomarkers Prev., 5, 979–983.[Abstract]
Salagovic,J., Kalina,I., Stubna,J., Habalova,V., Hrivnak,M., Valansky,L., Kohut,A. and Biros,E. (1998) Genetic polymorphism of glutathione S-transferases M1 and T1 as a risk factor in lung and bladder cancers. Neoplasma, 45, 312–317.[Web of Science][Medline]
Schlesselman,J.J. (1982) Case–Control Studies—Design, Conduct, Analysis. Oxford University Press, Oxford, UK, pp. 220–226.
Smith,G., Stanley,L.A., Sim,E., Strange,R.C. and Wolf,C.R. (1995) Metabolic polymorphisms and cancer susceptibility. Cancer Surv., 25, 27–65.[Web of Science][Medline]
Wolf,C.R., Smith,C.A.D., Bishop,T., Forman,D., Gough,A.C. and Spurr,N.K. (1994) CYP2D6 genotyping and the association with lung cancer susceptibility. Pharmacogenetics, 4, 104–106.[Web of Science][Medline]
Zhong,S., Wyllie,A.H., Barnes,D., Wolf,C.R. and Spurr,N.K. (1993) Relationship between the GSTM1 genetic polymorphism and susceptibility to bladder, breast and colon cancer. Carcinogenesis, 14, 1821–1824.
Received on January 31, 2000; accepted on May 23, 2000.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. P. A. Ioannidis Effect of Formal Statistical Significance on the Credibility of Observational Associations Am. J. Epidemiol., August 15, 2008; 168(4): 374 - 383. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Lai, L. Crevier, and L. Thabane Genetic Polymorphisms of Glutathione S-Transferases and the Risk of Adult Brain Tumors: A Meta-analysis Cancer Epidemiol. Biomarkers Prev., July 1, 2005; 14(7): 1784 - 1790. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Iida, K. Itoh, Y. Kumagai, R. Oyasu, K. Hattori, K. Kawai, T. Shimazui, H. Akaza, and M. Yamamoto Nrf2 Is Essential for the Chemopreventive Efficacy of Oltipraz against Urinary Bladder Carcinogenesis Cancer Res., September 15, 2004; 64(18): 6424 - 6431. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Zhang and G. B. Gordon A strategy for cancer prevention: Stimulation of the Nrf2-ARE signaling pathway Mol. Cancer Ther., July 1, 2004; 3(7): 885 - 893. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Sanyal, F. Festa, S. Sakano, Z. Zhang, G. Steineck, U. Norming, H. Wijkstrom, P. Larsson, R. Kumar, and K. Hemminki Polymorphisms in DNA repair and metabolic genes in bladder cancer Carcinogenesis, May 1, 2004; 25(5): 729 - 734. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. S. Engel, E. Taioli, R. Pfeiffer, M. Garcia-Closas, P. M. Marcus, Q. Lan, P. Boffetta, P. Vineis, H. Autrup, D. A. Bell, et al. Pooled Analysis and Meta-analysis of Glutathione S-Transferase M1 and Bladder Cancer: A HuGE Review Am. J. Epidemiol., July 15, 2002; 156(2): 95 - 109. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||