Skip Navigation


Carcinogenesis Advance Access originally published online on September 29, 2005
Carcinogenesis 2006 27(3):525-532; doi:10.1093/carcin/bgi227
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
27/3/525    most recent
bgi227v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (13)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Larsen, J. E.
Right arrow Articles by Fong, K. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Larsen, J. E.
Right arrow Articles by Fong, K. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Carcinogenesis vol.27 no.3 © Oxford University Press 2005; all rights reserved.

CYP1A1 Ile462Val and MPO G-463A interact to increase risk of adenocarcinoma but not squamous cell carcinoma of the lung

Jill Everland Larsen 1, 2, Maree Louise Colosimo 1, Ian Anthony Yang 1, 2, Rayleen Bowman 2, Paul Victor Zimmerman 1, 2 and Kwun Meng Fong 1, 2, *

1 Department of Thoracic Medicine, The Prince Charles Hospital, Brisbane, Australia and 2 School of Medicine, University of Queensland, Brisbane, Australia

* To whom correspondence and reprint requests should be addressed. Tel: +61 733 508111; Fax: +61 733 508957; Email: kwun_fong{at}health.qld.gov.au


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
Common polymorphisms in genes encoding phase I and phase II enzymes are considered to modify lung cancer risk due to changes in enzyme activity. Candidates include genetic variants of glutathione S-transferases (GSTM1, GSTT1 and GSTP1) and myeloperoxidase (MPO). We performed a large case–control study of these candidate genes in 1103 patients with non-small cell lung cancer (NSCLC) and 627 controls without NSCLC. Associations between deletion genotypes of GSTM1 and GSTT1 and between single nucleotide polymorphisms (SNPs) of GSTP1 Ile105Val and MPO G-463A were first tested by adjusted logistic regression. Then we analysed gene–gene interactions, also incorporating our published data on the Ile462Val SNP in the phase I enzyme, cytochrome P450 CYP1A1. The homozygous GSTP1 105Val genotype was significantly under-represented in NSCLC compared with controls (OR = 0.73; 95%CI 0.53–1.00; P = 0.050), especially in females (OR = 0.57; 95%CI 0.34–0.98; P = 0.04). The GSTT1-null genotype was significantly over-represented in adenocarcinomas (OR = 1.41; 95%CI 1.06–1.90; P = 0.02) but not in squamous cell carcinomas (OR = 1.03; 95%CI 0.76–1.41; P = 0.84). There was weak risk reduction associated with GSTM1 null in heavy smokers (OR = 0.71; 95%CI 0.54–0.94; P = 0.02), but neither GSTM1 nor MPO genotypes affected the overall risk of NSCLC. The MPO and CYP1A1 risk genotypes interacted to increase the overall risk of NSCLC (OR = 2.88; 95%CI 1.70–5.00; P < 0.001). The data are consistent with the concept that multiple genes of modest effect interact to confer genomic-based susceptibility to lung cancer.

Abbreviations: GST, glutathione S-transferase; MPO, myeloperoxidase; NSCLC, non-small cell lung cancer; SNPs, single nucleotide polymorphisms


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
Smoking is the principal cause of lung cancer, yet <20% of smokers develop the disease, suggesting a possible genetic predisposition. Polymorphic genes encoding enzymes that activate or detoxify harmful chemicals are of particular interest as allelic differences may account for wide inter-individual differences in sensitivity to cancer-inducing or cancer-promoting compounds (1).

Phase I enzymes such as myeloperoxidase (MPO) and cytochrome P450 metabolically activate carcinogens into reactive intermediates which can then covalently bind to DNA forming DNA adducts, thereby initiating the carcinogenic process (2). In turn, tobacco metabolites can be detoxified by phase II enzymes (such as glutathione S-transferases) via conjugation with endogenous molecules to form hydrophilic conjugates that are more readily excreted by the cell (3). The accumulated level of reactive metabolic intermediates is partly dependent upon the metabolic balance of phase I and phase II enzymes. Inter-individual genetic differences (due to common polymorphisms that cause variation in enzymatic activity) in the elements of this metabolic balance may thus affect an individual smoker's susceptibility to lung cancer—especially if gene–gene interactions of phase I and II enzymes occur.

The glutathione S-transferase family of five subclasses of enzymes (alpha, mu, pi, theta and zeta) (4,5) catalyse conjugation of reactive chemical intermediates such as the carcinogenic polycyclic aromatic hydrocarbons (PAHs) to soluble glutathione. GSTM1 and GSTT1 are enzymatically inactive from germ line deletion polymorphisms in ~50 and 20% of Caucasian populations, respectively (6,7). Although deficiency of GSTM1 or GSTT1 genes (GSTM1 null and GSTT1 null, respectively) has been proposed to lead to increased susceptibility to tobacco-related cancers, such as lung cancer, reported studies to date suggest that GSTT1 null has no effect upon lung cancer susceptibility (8,9), whilst the effect of GSTM1 null remains ambiguous. Enzymatic studies of the coding sequence single nucleotide polymorphism (SNP) A313G (Ile105Val) in GSTP1 suggest that the codon 105 isoleucine allele displays a higher catalytic activity towards fjord region diol epoxides whilst the valine allele has a higher catalytic activity towards bay region diol epoxides (10–13). Consequently, such substrate specificity could influence the metabolism of different environmental carcinogens leading to altered individual carcinogenic risk depending upon both the type of carcinogen exposure and GSTP1 genotype.

MPO is released locally from neutrophil lysosomes in the lung in response to inflammatory activation and converts tobacco-smoke carcinogens into highly reactive intermediates (14). A single base substitution (G->A) in a promoter Alu repeat 463 bases upstream of MPO down-regulates gene expression through disruption of the SP1 transcription factor binding site (15).

In previous studies, variable study design and limited statistical power due to small sample size have contributed to the continuing debate regarding the true role of these SNPs in modifying lung cancer risk. Here, we postulate that overall lung cancer risk is determined by the combined interaction of multiple genes, each of which may have only a modest effect size individually. We therefore designed a large case–control study in a predominantly Caucasian population (99%) to test four priority candidate SNPs (GSTM1, GSTT1, GSTP1 and MPO) in combination with our previously studied cytochrome P450 CYP1A1 Ile462Val SNP to define individual and combined effects of susceptibility genotypes (16). These genes were selected on the basis of the strength of associations from the candidate gene literature available at the time of conception of this study and the functional importance of SNPs in these genes based on in vitro data.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Sample collection
Cases consisted of patients with cytologically or histologically confirmed primary non-small cell lung cancer (NSCLC) (n = 1103) treated at The Prince Charles Hospital from 1980 to 2003. These consisted of 836 surgically resected cases (accounting for a low percentage of Stage IV cases overall) and 267 cases treated with other modalities. Overall stage of the disease and Tumour Node Metastasis (TNM) stage were classified using the American Joint Committee Staging system (17). Controls consisted of patients with chronic obstructive pulmonary disease (COPD) but without lung cancer (n = 380), treated at the same hospital from 1998 to 2003, or healthy smokers attending a smoking cessation clinic held at the hospital from 2000 to 2003 (n = 247). These two groups were selected as controls because their known risk factors for lung cancer, age and tobacco-smoke exposure, were similar to the cases. Of the invited subjects undergoing surgery ~95% consented to participate in the study and the population from which the subjects were drawn was >99% Caucasian.

The study protocol and access to archived NSCLC paraffin blocks (n = 259) were approved by the Ethics Committee of The Prince Charles Hospital, and subjects (both cases and controls) gave informed written consent for use of resected fresh lung tissue or blood (n = 1471). Demographic characteristics of cases and controls (Table I) were collected by a research nurse or treating physician, and data were checked against patient records and the institutional lung cancer database.


View this table:
[in this window]
[in a new window]
  		Table I. Characteristics of the study subjects

 
Nucleic acid purification
DNA from control subjects was extracted from peripheral blood. DNA from subjects diagnosed with NSCLC was extracted from peripheral blood or resected normal lung tissue, either fresh-frozen or paraffin embedded (18–20). In 166 cases, DNA was extracted from more than one source (94 cases in which normal lung was available both as paraffin-embedded blocks and fresh-frozen tissue, 72 cases where both peripheral blood and fresh-frozen tissue were available). Genotyping results for individuals were completely concordant in all cases where DNA had been extracted from multiple sources, confirming the reproducibility of genotyping methodology. Of the samples 10% were randomly selected and retested for consistency.

Genotyping
PCR-based restriction fragment length polymorphism methods were used to analyse GSTP1 Ile105Val and MPO G-463A genotypes as described previously (21,22). GSTM1 and GSTT1 were genotyped using real-time quantitative PCR with primer and probe sequences from a previously published study (23). Due to the high degree of sequence homology between GSTM1 and GSTM4/GSTM5 an alternative reverse primer was designed (5'-AAACTCTGTCAGATGCAGCTCACT-3') using the Primer Express software (Applied Biosystems, Foster City, CA, USA), and it was paired with the published forward primer and probe in order to selectively amplify GSTM1 alone. GSTM1 and GSTT1 probes were labelled with a 5'-reporter dye (6-carboxyfluorescein; FAM) and a 3'-quencher dye (6-carboxy-N,N,N1,N1-tetrachlorofluorescein; TAMRA). VIC-labelled Taqman Ribosomal RNA reagents (Perkin Elmer) were used as an internal control. Each PCR contained the primers and FAM-labelled probe of GSTM1 or GSTT1 with the control primers and VIC-labelled probe, each at a final concentration of 50 nM. The reaction contained 1x Platinum Quantitative PCR SuperMix-UDG (Invitrogen) (0.375 U Platinum Taq DNA Polymerase, 20 mM Tris–HCl, pH 8.4, 50 mM KCl, 5 mM MgCl2, 200 µM of each dGTP, dATP and dCTP, 400 µM dUTP, 0.25 U UDG), a passive reference ROX dye (Invitrogen) and 10 ng genomic DNA in a final volume of 12.5 µl. Real-time PCR was performed with an Applied Biosystems ABI PRISM 7700 sequence detector.

Statistical analysis
Genotyping was performed blinded to case–control status. Hardy–Weinberg equilibrium was confirmed by {chi}2 analysis. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using logistic regression analysis where log odds of lung cancer were adjusted for smoking (pack-years), age (as a continuous variable) and sex (as a categorical variable). In order to detect important differences in population subgroups, stratification by subgroup analysis of clinically relevant factors (age, pack-years and sex) was performed. All tests were two-sided and a P-value of 0.05 or less was considered significant. Interaction was formally tested by comparing –2 log (likelihood) in models that included an appropriate interaction term. Data were analysed using the SPSS for Windows Version 11.5 (SPSS Inc., Chicago, IL, USA).

Gene–gene interactions of two polymorphic genes were investigated when addition of a second gene into a model identified the second gene as a potential confounder [>10% change to Exp(ß)]. We used a linear effects model to avoid an increase in the chance of type I error associated with more complex interaction models. The two genotypes were merged to form a new variable with three categories—no risk alleles (reference), one risk allele (of either gene) and two risk alleles. Prior to interaction analyses, genotypes were classified into dichotomous variables where the risk alleles for each gene were as follows: GSTP1 Ile (Ile/Ile and Ile/Val genotypes), GSTT1 null, GSTM1 null, MPO G (G//G genotype) and the previously studied CYP1A1 Val (Ile/Val and Val/Val genotypes).


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
Demographics and genotype distributions of the 1103 cases and 627 controls are shown in Table II. To confirm the validity of combining the two subgroups of controls, the variant allele frequencies of GSTP1, GSTM1, GSTT1 and MPO were initially compared between COPD patients and healthy smokers. As there was no significant difference in the variant frequencies between these two subgroups of controls (P > 0.05, data not shown) they were combined to form one control group. The genotype frequencies are listed in Table I. All genotypes in the control population were in Hardy–Weinberg equilibrium. The study had >80% statistical power to detect an odds ratio of 1.5 or more for the variant alleles of GSTM1, GSTT1 and GSTP1, and an odds ratio of 0.7 or less for the variant allele of MPO with {alpha} = 0.05. The main hypothesis tested for each gene was the effect upon overall NSCLC risk in relation to the two main risk factors (age and tobacco-smoke exposure). Exploratory subset analyses were performed following histological and sex stratification.


View this table:
[in this window]
[in a new window]
  		Table II. Analysis of GSTM1 null, GSTT1 null, GSTP1 Ile105Val and MPO G-463A polymorphisms in relation to NSCLC risk

 
GSTM1
The null genotype of GSTM1 did not modify the risk of lung cancer overall (OR = 0.83; 95%CI 0.68–1.02; P = 0.071) within either of the two main histological subtypes AC (OR = 0.82; 95%CI 0.65–1.04; P = 0.108) and SCC (OR = 0.83; 95%CI 0.65–1.05; P = 0.124) (Table II). When cases and controls were stratified into subgroups above and below the median cumulative tobacco-smoke exposure of 46 pack-years, no association was identified between GSTM1 genotype and lung cancer risk in smokers below the median (OR = 0.97; 95%CI 0.73–1.79; P = 0.837), but the null genotype significantly reduced lung cancer risk in smokers above the median (OR = 0.71; 95%CI 0.54–0.94; P = 0.019) (Table II). A formal test of interaction between GSTM1 genotype and tobacco-smoke exposure was not significant (P>0.05, data not shown). Nested analyses following stratification by age (median of 64 years) or sex yielded no significant associations (Table II).

GSTT1
The null genotype of GSTT1 did not modify the overall risk of lung cancer (OR = 1.16; 95%CI 0.90–1.49; P = 0.245). GSTT1 null was associated with significantly increased risk in AC (OR = 1.41; 95%CI 1.06–1.90; P = 0.020) but not in SCCs (OR = 1.03; 95%CI 0.76–1.41; P = 0.842) (Table II). Nested analyses following stratification around cumulative tobacco-smoke exposure (median of 46 pack-years), age (median of 64 years) or gender found no significant associations (Table II).

GSTP1
Comparing Ile/Ile with Val/Val genotypes, Val/Val was associated with decreased overall lung cancer risk, but the upper 95% confidence limit for the estimated odds ratio was 1.00 (OR = 0.73; 95%CI 0.53–1.00; P = 0.050). There was no significant alteration of risk associated with the Ile/Val genotype as compared with the Ile/Ile genotype (OR = 0.98; 95%CI 0.80–1.21; P = 0.885) (Table II). No significant effect of the valine allele was observed when Ile/Val and Val/Val genotypes were combined (OR = 0.92; 95%CI 0.76–1.12; P = 0.410). In both histological subgroups, a trend towards the protective effect of the Val/Val genotype was demonstrated, perhaps stronger for ACs than SCCs (OR 0.69 versus 0.82, respectively), but neither odds ratio was statistically significant (Table II). Stratification for cumulative tobacco-smoke exposure (median of 46 pack-years) and age (median of 64 years) followed by nested analysis revealed no significant associations (Table II). The GSTP1 Val/Val genotype was associated with a significantly decreased risk of NSCLC in females (OR = 0.57; 95%CI 0.34–0.98; P = 0.042) but not in males (OR = 0.83; 95%CI 0.55–1.24; P = 0.359) (Table II). A formal test of interaction between the GSTP1 genotype and gender was not significant (P > 0.05, data not shown).

MPO
MPO genotyping of the larger 350 bp PCR amplicon was not possible on paraffin block-derived DNA, leaving a total of 627 cases and 624 controls for this analysis. No protective effect was observed in NSCLC for either the GA genotype (OR = 0.98; 95%CI 0.77–1.25; P = 0.878) or the AA genotype (OR = 1.31; 95%CI 0.80–2.15; P = 0.282) when compared with the wild-type GG genotype, or in carriers of the A allele (AA and AG genotypes combined) (OR = 1.02; 95%CI 0.81–1.29; P = 0.841) (Table II). Nested analyses following stratification for histological subtype, cumulative tobacco-smoke exposure (median of 46 pack-years), age (median of 64 years) or sex found no significant associations (Table II).

Gene–gene interactions
To analyse the gene–gene interactions, genotypes of two genes were combined and sorted into three categories consisting of one risk allele, two risk alleles and no risk alleles (the reference group). To ensure sufficient power, gene–gene interaction analyses were confined to all NSCLCs and major histological subtypes (AC and SCC), adjusting for age, pack-years and sex (Table III). Our published data for the candidate susceptibility gene CYP1A1 Ile462Val SNP were included in the interaction analyses (16). Due to the protective effect observed for GSTP1 Val/Val genotypes in univariate analyses, GSTP1 Ile/Ile and Ile/Val genotypes were combined as the putative high risk genotype in multivariate analyses. For MPO, GG was classified as the putative high risk genotype. Of 10 possible pair-wise genotype combinations for GSTP1, GSTT1, GSTM1, MPO and CYP1A1, only CYP1A1 and MPO had an interactive effect in NSCLCs overall. This interaction was also identified within the AC subtype where the addition of MPO genotype to the CYP1A1 logistic regression model resulted in a >10% increase in Exp(ß), indicating MPO as a confounder. The CYP1A1 and MPO genotypes were then combined to form a new variable and the logistic regression analyses are displayed in Table III. Using a combined genotype consisting of no putative risk alleles (CYP1A1 Ile/Ile and MPO AA/AG) as a reference, one risk allele (CYP1A1 Ile/Val + Val/Val and MPO AA/AG or CYP1A1 Ile/Ile and MPO GG genotypes) was not associated with an increased risk (OR = 0.93; 95% CI 0.72–1.19; P = 0.724) but both risk alleles (CYP1A1 Ile/Val + Val/Val and MPO GG genotypes) conferred a significantly increased risk of NSCLC (OR = 2.88; 95% CI 1.70–5.00; P < 0.0001). This effect was also observed within the AC subset, where one risk allele conferred no increased risk (OR = 0.80; 95% CI 0.58–1.10; P = 0.170) but the presence of both risk alleles was associated with substantially increased risk (OR = 3.72; 95% CI 2.01–6.88; P < 0.0001). The ORs for the combined CYP1A1 and MPO genotypes in all NSCLCs and for the AC subset were considerably greater than the product of the ORs in univariate analyses (univariate odds ratio in all NSCLCs, 1.85 and 0.98, respectively, and in ACs, 2.38 and 0.88, respectively) suggesting an interactive rather than an additive effect between the two (Figure 1).


View this table:
[in this window]
[in a new window]
  		Table III. Gene–gene interactions of MPO G-463A and CYP1A1 Ile462Val polymorphisms in NSCLCs, ACs and SCCs

 

Figure 1
View larger version (12K):
[in this window]
[in a new window]
  		Fig. 1. Gene–gene interactions of MPO G-463A and CYP1A1 Ile462Val polymorphisms in NSCLCs, ACs and SCCs. Odds ratios and 95% confidence intervals of individual genotype risk effect of each gene and combined genotypes risk effect (for one risk allele and two risk alleles) using logistic regression adjusting for age, sex and tobacco-smoke exposure. The gene–gene effect observed in NSCLCs appears to result from the AC subtype.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
Evidence suggest that the metabolic balance of activating phase I enzymes and detoxifying phase II enzymes influences an individual smoker's risk of developing tobacco-related lung cancer. It follows that genetic SNPs resulting in altered activity of such enzymes may modify cancer susceptibility. Many studies have endeavoured to link SNPs in individual gene to lung cancer risk, and the results have been conflicting. These discrepancies are most probably due to population heterogeneity as ethnicity, histological subtype and level of tobacco-smoke exposure have all been shown to contribute (24), as has publication bias, towards positive studies (25). Several sizable studies have attempted to elucidate the role of gene SNPs in GSTM1, GSTT1, GSTP1, MPO and CYP1A1 (26–30). With a large, well-defined, homogenous population available to us, we took the opportunity to explore the effects of interaction between these gene SNPs on the risk of lung cancer.

Enzymatic data has shown that the valine isoform of GSTP1 Ile105Val has higher catalytic activity towards bay region diol epoxides such as the diol epoxide of B[a]P, a compound known to induce tumours and present at high levels in cigarette smoke (31). The evidence presented in this study of the association of the valine allele with decreased risk of NSCLC, especially in females, is consistent with GSTP1 modified carcinogenesis of tobacco diol epoxides as a possible mechanism of lung cancer. However, as both associations were only of borderline significance, and even uncorrected for multiple testing, we are cautious in implicating a protective effect of the valine allele in lung cancer susceptibility. Of the 16 previous studies of GSTP1 Ile105Val and lung cancer susceptibility in Caucasians, 2 have associated an increased risk of lung cancer with the valine allele (32,33), and both consist of male populations of fewer than 300 cases. Recent, larger studies were not confirmatory (28–30), with the exception of some subcohorts such as those with low cumulative tobacco-smoke exposure (28,30).

We found no association of GSTT1 deficiency and overall lung cancer risk, which is consistent with the majority of published case–control studies (8,9). Histologic stratification revealed a significant effect (albeit uncorrected for multiple testing) in ACs but not in SCCs. This was also noted in two previous studies (34,35). This effect may not have been observed in other studies due to insufficient sample sizes—only one study had >500 cases and controls. Nonetheless, these data add to the mounting epidemiological and biological data suggesting that AC and SCC, the two most common histological subtypes of NSCLC, have distinct genetic risks, thus deserving a separate analysis (36–38).

Despite much investigation since the first report of an association of GSTM1 deficiency and lung cancer (39), the role of GSTM1 null has not yet been resolved primarily due to inadequate power of many of the studies to date. Two recent meta-analyses suggest that this SNP has a very modest effect, if any, upon lung cancer risk (26,40). Low levels of GSTM1 expression in lung tissue (41) may explain why GSTM1 SNPs do not appear to modify the risk for lung cancer. Here, we aimed to substantiate these meta-analyses in a statistically powered homogenous population and our results confirmed the lack of an effect for GSTM1 deficiency upon overall lung cancer risk. We did find that the null genotype was under-represented (uncorrected) in a subset of NSCLCs with heavier tobacco-smoke exposure. However, as this occurred in an exploratory subset analysis conclusions are limited by the possibility of a false positive result contributed to by chance or by unknown confounding factors.

We found that MPO G-463A conferred no protective effect on NSCLC risk in our population (designed to detect an OR of 0.7 or lower), which is in accord with 4 of the 10 published case–control studies investigating this association in Caucasian populations. Although six studies have reported a statistically significant protective effect of the A/A genotype or the combined A/A + A/G genotype, most of these had small sample sizes—the largest using 307 cases and 307 controls (42). On the other hand the negative studies had a total of 2064 lung cancer cases (>60% of the 3360 cases included by all the10 studies) with the largest reporting on 988 cases and 1128 controls (27).

The advantage of the present study was that the large sample size allowed for statistical analysis of gene–gene interactions between two putative risk SNPs. Such analyses are important as illustrated in the glutathione (GSH) metabolic pathway where multiple enzymes with overlapping functions and shared substrates have been associated with host susceptibility to carcinogens and toxic agents (43). In addition to the four SNPs analysed in this study, we included genotyping results from a previously published gene, CYP1A1. CYP1A1 is the principal enzyme metabolizing PAHs such as B[a]P to highly reactive intermediates (44). A single base substitution at codon 462 (Ile->Val) leads to higher levels of carcinogen activation and was significantly associated with an increased risk of NSCLC in our population (16). A highly statistically significant incremental risk modifying effect for NSCLC was observed with the combination of the risk genotypes of MPO and CYP1A1, and this was pronounced in the AC subtype. The effect was not observed in SCCs, consistent with distinctly different genetic risks between the two most common histological subtypes of NSCLC. The interaction of MPO and CYP1A1 described here, and results from other positive gene–gene interaction studies (45,46), support the notion that genome-based lung cancer risk is likely to be influenced by combinations of single risk genes of modest effect as well as synergistic gene–gene interactions.

Our study used a large homogeneous group of patients with NSCLC to identify combinatorial small effects that may not have been detected by smaller studies. Since the hypothetical basis for investigation of SNPs involved in carcinogen activation or detoxification would not apply to non-smokers they were deliberately excluded from the control group. We considered it important to adjust for carcinogen exposure by the level of exposure (pack-years) as opposed to the use of smoking status (current, former or never). To minimize false positives, we interpreted the statistical tests for the main hypothesis (overall NSCLC, age and smoking history) in the light of the multiple comparisons (Table II), and then explored the possible effects of gender and subtype as hypothesis generating questions. The study design is vulnerable to possible misclassification bias within the control group as this group consists of individuals at risk from smoking who may develop lung cancer in due course. To counter this bias, age and tobacco-smoke exposure were included as confounding factors in the logistic regression analysis. In any case, this source of misclassification bias is likely to underestimate any effect sizes observed.

In conclusion, our study has two important findings for lung cancer molecular epidemiology. First, it reinforces the lack of or minimal risk burden for individual GSTM1 null, GSTT1 null, GSTP1 Ile105Val and MPO G-463A SNPs on NSCLC susceptibility in Caucasians. Second, we report for the first time a significant increase in risk for the interaction between CYP1A1 and MPO—beyond that of each gene individually—thus reinforcing the importance of gene–gene interactions in NSCLC development. Future studies should make choices for gene–gene interaction analysis based on the positive candidate genes on univariate analysis and then combine these with other candidate genes to examine for interaction. Alternatively, SNP chips could be used to analyse many SNPs but there is a risk of type I error. Even a modest increase in risk may be critically important in public health, especially if the SNP (or combination thereof) and the associated cancer are common. For example, effects of polymorphic genes may contribute to risk stratification among smokers, support targeted smoking cessation interventions, intensive surveillance, screening, early detection or other means of reducing the health burden of invasive lung cancer. Future epidemiological studies must focus on appropriate design and power (47) in order to determine the true role of genetic susceptibility in lung cancer.


   Acknowledgments
 
The authors thank the physicians, surgeons and pathologists at The Prince Charles Hospital who endorsed the project; the patients and donors who participated in this study; Linda H. Passmore for subject recruitment; and Ainsley M. Tunnicliffe and Jessie M. Kelly for their help in preparation of DNA. This work was supported by The Prince Charles Hospital Foundation and The Queensland Cancer Fund.

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Nebert,D.W. (1980) Human genetic variation in the enzymes of detoxification. Academic Press, New York.
  2. Nerurkar,P.V., Okinaka,L., Aoki,C., Seifried,A., Lum-Jones,A., Wilkens,L.R. and Le Marchand,L. (2000) CYP1A1, GSTM1, and GSTP1 genetic polymorphisms and urinary 1-hydroxypyrene excretion in non-occupationally exposed individuals. Cancer Epidemiol. Biomarkers Prev., 9, 1119–1122.[Abstract/Free Full Text]
  3. Mannervik,B. and Danielson,U.H. (1988) Glutathione transferases—structure and catalytic activity. CRC Crit. Rev. Biochem., 23, 283–337.[Web of Science][Medline]
  4. Mannervik,B., Alin,P., Guthenberg,C., Jensson,H., Tahir,M.K., Warholm,M. and Jornvall,H. (1985) Identification of three classes of cytosolic glutathione transferase common to several mammalian species: correlation between structural data and enzymatic properties. Proc. Natl Acad. Sci. USA, 82, 7202–7206.[Abstract/Free Full Text]
  5. Meyer,D.J., Coles,B., Pemble,S.E., Gilmore,K.S., Fraser,G.M. and Ketterer,B. (1991) Theta, a new class of glutathione transferases purified from rat and man. Biochem. J., 274, 409–414.
  6. Seidegard,J., Pero,R.W., Markowitz,M.M., Roush,G., Miller,D.G. and Beattie,E.J. (1990) Isoenzyme(s) of glutathione transferase (class Mu) as a marker for the susceptibility to lung cancer: a follow up study. Carcinogenesis, 11, 33–36.[Abstract/Free Full Text]
  7. Bell,D.A., Thompson,C.L., Taylor,J., Miller,C.R., Perera,F., Hsieh,L.L. and Lucier,G.W. (1992) Genetic monitoring of human polymorphic cancer susceptibility genes by polymerase chain reaction: application to glutathione transferase mu. Environ. Health Perspect., 98, 113–117.[Web of Science][Medline]
  8. Vineis,P., d'Errico,A., Malats,N. and Boffetta,P. (1999) Overall evaluation and research perspectives. IARC Sci. Publ., 403–408.
  9. Habdous,M., Siest,G., Herbeth,B., Vincent-Viry,M. and Visvikis,S. (2004) Glutathione S-transferases genetic polymorphisms and human diseases: overview of epidemiological studies. Ann. Biol. Clin. (Paris), 62, 15–24.[Medline]
  10. Sundberg,K., Johansson,A.S., Stenberg,G., Widersten,M., Seidel,A., Mannervik,B. and Jernstrom,B. (1998) Differences in the catalytic efficiencies of allelic variants of glutathione transferase P1-1 towards carcinogenic diol epoxides of polycyclic aromatic hydrocarbons. Carcinogenesis, 19, 433–436.[Abstract/Free Full Text]
  11. Hu,X., Ji,X., Srivastava,S.K., Xia,H., Awasthi,S., Nanduri,B., Awasthi,Y.C., Zimniak,P. and Singh,S.V. (1997) Mechanism of differential catalytic efficiency of two polymorphic forms of human glutathione S-transferase P1-1 in the glutathione conjugation of carcinogenic diol epoxide of chrysene. Arch. Biochem. Biophys., 345, 32–38.[CrossRef][Web of Science][Medline]
  12. Coles,B., Yang,M., Lang,N.P. and Kadlubar,F.F. (2000) Expression of hGSTP1 alleles in human lung and catalytic activity of the native protein variants towards 1-chloro-2,4-dinitrobenzene, 4-vinylpyridine and (+)-anti benzo[a]pyrene-7,8-diol-9,10-oxide. Cancer Lett., 156, 167–175.[CrossRef][Web of Science][Medline]
  13. Pal,A., Desai,D.H., Amin,S., Srivastava,S.K., Hu,X., Herzog,C., Zimniak,P. and Singh,S.V. (2000) Location of the epoxide function determines specificity of the allelic variants of human glutathione transferase Pi toward benzo[c]chrysene diol epoxide isomers. FEBS Lett., 486, 163–166.[CrossRef][Web of Science][Medline]
  14. Petruska,J.M., Mosebrook,D.R., Jakab,G.J. and Trush,M.A. (1992) Myeloperoxidase-enhanced formation of (±)-trans-7,8-dihydroxy-7,8-dihydrobenzo[a]pyrene-DNA adducts in lung tissue in vitro: a role of pulmonary inflammation in the bioactivation of a procarcinogen. Carcinogenesis, 13, 1075–1081.[Abstract/Free Full Text]
  15. Piedrafita,F.J., Molander,R.B., Vansant,G., Orlova,E.A., Pfahl,M. and Reynolds,W.F. (1996) An Alu element in the myeloperoxidase promoter contains a composite SP1-thyroid hormone-retinoic acid response element. J. Biol. Chem., 271, 14412–14420.[Abstract/Free Full Text]
  16. Larsen,J.E., Colosimo,M.L., Yang,I.A., Bowman,R.V., Zimmerman,P.V. and Fong,K.M. (2005) Risk of non-small cell lung cancer and the cytochrome P4501A1 Ile462Val polymorphism. Cancer Causes Control, 16, 579–585.[CrossRef][Web of Science][Medline]
  17. Mountain,C.F. (1986) A new international staging system for lung cancer. Chest, 89, 225S–233S.[Free Full Text]
  18. Miller,S.A., Dykes,D.D. and Polesky,H.F. (1988) A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res., 16, 1215.[Free Full Text]
  19. Levi,S., Urbano-Ispizua,A., Gill,R., Thomas,D.M., Gilbertson,J., Foster,C. and Marshall,C.J. (1991) Multiple K-ras codon 12 mutations in cholangiocarcinomas demonstrated with a sensitive polymerase chain reaction technique. Cancer Res., 51, 3497–3502.[Abstract/Free Full Text]
  20. Fong,K.M., Kida,Y., Zimmerman,P.V., Ikenaga,M. and Smith,P.J. (1995) Loss of heterozygosity frequently affects chromosome 17q in non-small cell lung cancer. Cancer Res., 55, 4268–4272.[Abstract/Free Full Text]
  21. Harries,L.W., Stubbins,M.J., Forman,D., Howard,G.C. and Wolf,C.R. (1997) Identification of genetic polymorphisms at the glutathione S-transferase Pi locus and association with susceptibility to bladder, testicular and prostate cancer. Carcinogenesis, 18, 641–644.[Abstract/Free Full Text]
  22. London,S.J., Lehman,T.A. and Taylor,J.A. (1997) Myeloperoxidase genetic polymorphism and lung cancer risk. Cancer Res., 57, 5001–5003.[Abstract/Free Full Text]
  23. Shi,M.M., Myrand,S.P., Bleavins,M.R. and de la Iglesia,F.A. (1999) High-throughput genotyping method for glutathione S-transferase T1 and M1 gene deletions using TaqMan probes. Res. Commun. Mol. Pathol. Pharmacol., 103, 3–15.[Web of Science][Medline]
  24. Le Marchand,L., Sivaraman,L., Pierce,L., Seifried,A., Lum,A., Wilkens,L.R. and Lau,A.F. (1998) Associations of CYP1A1, GSTM1, and CYP2E1 polymorphisms with lung cancer suggest cell type specificities to tobacco carcinogens. Cancer Res., 58, 4858–4863.[Abstract/Free Full Text]
  25. Little,J., Bradley,L., Bray,M.S. et al. (2002) Reporting, appraising, and integrating data on genotype prevalence and gene–disease associations. Am. J. Epidemiol., 156, 300–310.[Abstract/Free Full Text]
  26. Houlston,R.S. (1999) Glutathione S-transferase M1 status and lung cancer risk: a meta-analysis. Cancer Epidemiol. Biomarkers Prev., 8, 675–682.[Abstract/Free Full Text]
  27. Xu,L.L., Liu,G., Miller,D.P., Zhou,W., Lynch,T.J., Wain,J.C., Su,L. and Christiani,D.C. (2002) Counterpoint: the myeloperoxidase –463G->a polymorphism does not decrease lung cancer susceptibility in Caucasians. Cancer Epidemiol. Biomarkers Prev., 11, 1555–1559.[Abstract/Free Full Text]
  28. Miller,D.P., Neuberg,D., de Vivo,I., Wain,J.C., Lynch,T.J., Su,L. and Christiani,D.C. (2003) Smoking and the risk of lung cancer: susceptibility with GSTP1 polymorphisms. Epidemiology, 14, 545–551.[CrossRef][Web of Science][Medline]
  29. Wang,Y., Spitz,M.R., Schabath,M.B., Ali-Osman,F., Mata,H. and Wu,X. (2003) Association between glutathione S-transferase p1 polymorphisms and lung cancer risk in Caucasians: a case–control study. Lung Cancer, 40, 25–32.[CrossRef][Web of Science][Medline]
  30. Schneider,J., Bernges,U., Philipp,M. and Woitowitz,H.J. (2004) GSTM1, GSTT1, and GSTP1 polymorphism and lung cancer risk in relation to tobacco smoking. Cancer Lett., 208, 65–74.[CrossRef][Web of Science][Medline]
  31. Hecht,S.S. (1999) Tobacco smoke carcinogens and lung cancer. J. Natl Cancer Inst., 91, 1194–1210.[Abstract/Free Full Text]
  32. Ryberg,D., Skaug,V., Hewer,A., Phillips,D.H., Harries,L.W., Wolf,C.R., Ogreid,D., Ulvik,A., Vu,P. and Haugen,A. (1997) Genotypes of glutathione transferase M1 and P1 and their significance for lung DNA adduct levels and cancer risk. Carcinogenesis, 18, 1285–1289.[Abstract/Free Full Text]
  33. Stucker,I., Hirvonen,A., De Waziers,I., Cabelguenne,A., Mitrunen,K., Cenee,S., Koum-Besson,E., Hemon,D., Beaune,P. and Loriot,M.A. (2002) Genetic polymorphisms of glutathione S-transferases as modulators of lung cancer susceptibility. Carcinogenesis, 23, 1475–1481.[Abstract/Free Full Text]
  34. Sunaga,N., Kohno,T., Yanagitani,N., Sugimura,H., Kunitoh,H., Tamura,T., Takei,Y., Tsuchiya,S., Saito,R. and Yokota,J. (2002) Contribution of the NQO1 and GSTT1 polymorphisms to lung adenocarcinoma susceptibility. Cancer Epidemiol. Biomarkers Prev., 11, 730–738.[Abstract/Free Full Text]
  35. Sorensen,M., Autrup,H., Tjonneland,A., Overvad,K. and Raaschou-Nielsen,O. (2004) Glutathione S-transferase T1 null-genotype is associated with an increased risk of lung cancer. Int. J. Cancer., 110, 219–224.[CrossRef][Web of Science][Medline]
  36. Devesa,S.S., Shaw,G.L. and Blot,W.J. (1991) Changing patterns of lung cancer incidence by histological type. Cancer Epidemiol. Biomarkers Prev., 1, 29–34.[Abstract]
  37. Yokoyama,S., Yamakawa,K., Tsuchiya,E., Murata,M., Sakiyama,S. and Nakamura,Y. (1992) Deletion mapping on the short arm of chromosome 3 in squamous cell carcinoma and adenocarcinoma of the lung. Cancer Res., 52, 873–877.[Abstract/Free Full Text]
  38. Sato,S., Nakamura,Y. and Tsuchiya,E. (1994) Difference of allelotype between squamous cell carcinoma and adenocarcinoma of the lung. Cancer Res., 54, 5652–5655.[Abstract/Free Full Text]
  39. Seidegard,J., Pero,R.W., Miller,D.G. and Beattie,E.J. (1986) A glutathione transferase in human leukocytes as a marker for the susceptibility to lung cancer. Carcinogenesis, 7, 751–753.[Abstract/Free Full Text]
  40. Benhamou,S., Lee,W.J., Alexandrie,A.K. et al. (2002) Meta- and pooled analyses of the effects of glutathione S-transferase M1 polymorphisms and smoking on lung cancer risk. Carcinogenesis, 23, 1343–1350.[Abstract/Free Full Text]
  41. Anttila,S., Hirvonen,A., Vainio,H., Husgafvel-Pursiainen,K., Hayes,J.D. and Ketterer,B. (1993) Immunohistochemical localization of glutathione S-transferases in human lung. Cancer Res., 53, 5643–5648.[Abstract/Free Full Text]
  42. Wu,X., Schabath,M.B. and Spitz,M.R. (2003) Myeloperoxidase promoter region polymorphism and lung cancer risk. Methods Mol. Med., 75, 121–133.[Medline]
  43. Sellers,T.A. and Yang,P. (2002) Familial and genetic influences on risk of lung cancer. In King,R.A., Rotter,J.I. and Motulsky,A.G. (eds) The Genetic Basis of Common Diseases. Oxford University Press, NY, pp. 700–712.
  44. Kawajiri,K., Nakachi,K., Imai,K., Watanabe,J. and Hayashi,S. (1993) The CYP1A1 gene and cancer susceptibility. Crit. Rev. Oncol. Hematol., 14, 77–87.[Web of Science][Medline]
  45. Jourenkova-Mironova,N., Wikman,H., Bouchardy,C., Voho,A., Dayer,P., Benhamou,S. and Hirvonen,A. (1998) Role of glutathione S-transferase GSTM1, GSTM3, GSTP1 and GSTT1 genotypes in modulating susceptibility to smoking-related lung cancer. Pharmacogenetics, 8, 495–502.[Web of Science][Medline]
  46. Kelsey,K.T., Spitz,M.R., Zuo,Z.F. and Wiencke,J.K. (1997) Polymorphisms in the glutathione S-transferase class mu and theta genes interact and increase susceptibility to lung cancer in minority populations (Texas, United States). Cancer Causes Control, 8, 554–559.[CrossRef][Web of Science][Medline]
  47. Davey Smith,G. and Ebrahim,S. (2005) What can mendelian randomisation tell us about modifiable behavioural and environmental exposures? BMJ, 330, 1076–1079.[Free Full Text]
Received June 8, 2005; revised August 17, 2005; accepted September 11, 2005.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Am J EpidemiolHome page
M. L. Cote, W. Chen, D. W. Smith, S. Benhamou, C. Bouchardy, D. Butkiewicz, K. M. Fong, M. Gene, A. Hirvonen, C. Kiyohara, et al.
Meta- and Pooled Analysis of GSTP1 Polymorphism and Lung Cancer: A HuGE-GSEC Review
Am. J. Epidemiol., April 1, 2009; 169(7): 802 - 814.
[Abstract] [Full Text] [PDF]

Home page
 ThoraxHome page
I A Yang, K M Fong, P V Zimmerman, S T Holgate, and J W Holloway
Genetic susceptibility to the respiratory effects of air pollution
Thorax, June 1, 2008; 63(6): 555 - 563.
[Abstract] [Full Text] [PDF]

Home page
 CarcinogenesisHome page
T. Kohno, H. Kunitoh, K. Suzuki, S. Yamamoto, A. Kuchiba, Y. Matsuno, N. Yanagitani, and J. Yokota
Association of KRAS polymorphisms with risk for lung adenocarcinoma accompanied by atypical adenomatous hyperplasias
Carcinogenesis, May 1, 2008; 29(5): 957 - 963.
[Abstract] [Full Text] [PDF]

Home page
 CarcinogenesisHome page
P. Vineis, S. Anttila, S. Benhamou, M. Spinola, A. Hirvonen, C. Kiyohara, S. J. Garte, R. Puntoni, A. Rannug, R. C. Strange, et al.
Evidence of gene gene interactions in lung carcinogenesis in a large pooled analysis
Carcinogenesis, September 1, 2007; 28(9): 1902 - 1905.
[Abstract] [Full Text] [PDF]

Home page
 CarcinogenesisHome page
K.-A. Yoon, S. Park, B. Hwangbo, H. D. Shin, H. S. Cheong, H.-R. Shin, and J. S. Lee
Genetic polymorphisms in the Rb-binding zinc finger gene RIZ and the risk of lung cancer
Carcinogenesis, September 1, 2007; 28(9): 1971 - 1977.
[Abstract] [Full Text] [PDF]

Home page
 Ann OncolHome page
P Vineis, F Veglia, S Garte, C Malaveille, G Matullo, A Dunning, M Peluso, L Airoldi, K Overvad, O Raaschou-Nielsen, et al.
Genetic susceptibility according to three metabolic pathways in cancers of the lung and bladder and in myeloid leukemias in nonsmokers
Ann. Onc., July 1, 2007; 18(7): 1230 - 1242.
[Abstract] [Full Text] [PDF]

Home page
 Journal of Pharmacy PracticeHome page
Amal Al Omari and D. J. Murry
Pharmacogenetics of the Cytochrome P450 Enzyme System: Review of Current Knowledge and Clinical Significance
Journal of Pharmacy Practice, June 1, 2007; 20(3): 206 - 218.
[Abstract] [PDF]

Home page
 CarcinogenesisHome page
J. Yang, C. B. Ambrosone, C.-C. Hong, J. Ahn, C. Rodriguez, M. J. Thun, and E. E. Calle
Relationships between polymorphisms in NOS3 and MPO genes, cigarette smoking and risk of post-menopausal breast cancer
Carcinogenesis, June 1, 2007; 28(6): 1247 - 1253.
[Abstract] [Full Text] [PDF]

Home page
 CarcinogenesisHome page
A. G. Schwartz, G. M. Prysak, C. H. Bock, and M. L. Cote
The molecular epidemiology of lung cancer
Carcinogenesis, March 1, 2007; 28(3): 507 - 518.
[Abstract] [Full Text] [PDF]

Home page
 CarcinogenesisHome page
K.-A. Yoon, B. Hwangbo, I.-J. Kim, S. Park, H. S. Kim, H. J. Kee, J. E. Lee, Y. K. Jang, J.-G. Park, and J. S. Lee
Novel polymorphisms in the SUV39H2 histone methyltransferase and the risk of lung cancer
Carcinogenesis, November 1, 2006; 27(11): 2217 - 2222.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
27/3/525    most recent
bgi227v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (13)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Larsen, J. E.
Right arrow Articles by Fong, K. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Larsen, J. E.
Right arrow Articles by Fong, K. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?