Combined effects of genetic polymorphisms in six selected genes on lung cancer susceptibility

Lung Cancer (2007) 57, 135—142

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/lungcan

Combined effects of genetic polymorphisms in six selected genes on lung cancer susceptibility Mihi Yang a,∗, Yunhee Choi b, Bin Hwangbo c, Jin Soo Lee c a

Department of Toxicology, College of Pharmacy, Sookmyung Women’s University, 53-12 Chungpa-Dong, Yongsan-Gu, Seoul 140-742, Republic of Korea b Department of Preventive Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea c Center for Lung Cancer, Research Institute and Hospital, National Cancer Center, Goyang, Republic of Korea Received 10 November 2006; received in revised form 3 March 2007; accepted 5 March 2007

KEYWORDS Lung cancer; Genetic polymorphism; CYP1A1; GSTM1; GSTP1; GSTT1; MPO; NQO1

Summary Various molecular epidemiological studies have been performed to find genetic etiology for lung cancer. Particularly, genetic polymorphisms in NAD(P)H-quinone oxidoreductase (NQO1), cytochrome P450 (CYP)1A1, myeloperoxidase (MPO), glutathione-S-transferase (GST)P1, GSTT1, and GSTM1, and have been suspected to affect lung cancer risk. However, there was no study that examined the combined effects of these genes on lung cancer risk. We studied the combined genetic effects on lung cancer risk in 671 Korean subjects including 318 lung cancer patients and 353 controls. They filled questionnaires, which included lifestyle and childhood- and current environment data. Based on single nucleotide polymorphisms and gene deletions, genetic polymorphisms of the above six genes were determined with PCR-RFLP and TaqMan methods. As results, genetic polymorphisms in the GSTP1, MPO, and CYP1A1 among the genetic factors showed associations with lung cancer risk. The reference, which is supposed to have the lowest risk for cancer, was subjects who were homozygous wild type of the GSTP1 and CYP1A1 and had the MPO- mutant allele. After combination study of the three genepolymorphism, the subjects who were most different with the reference, i.e. had the mutant allele of the GSTP1 and CYP1A1 and homozygous wild type of the MPO, showed approximately 5-fold-higher risk for lung cancer than the reference (95% CI, 2.05—12.05). Therefore, our study suggests that the combination of the GSTP1, MPO, and CYP1A1 variations affects susceptibility to lung cancer. © 2007 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

∗

Corresponding author. Tel.: +82 2 20777179; fax: 82 2 710 9871. E-mail address: [email protected] (M. Yang).

Cigarette smoking is one well-known lung cancer risk. To prevent lung cancer, cessation of smoking has increased over the last decades. Even though it is too soon to see protective effects of cessation of the smoking against lung cancer,

0169-5002/$ — see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.lungcan.2007.03.005

136

M. Yang et al.

incidence of lung cancer has not exactly decreased with smoking cessation as yet [1]. Therefore, various efforts to find etiology of lung cancer besides tobacco smoking have been made. One of these efforts is to study exposure timing of smoking, such as a commencement of smoking at an early age [2], or side stream in childhood [3]. For example, molecular analysis of tissue specimens from resection in lung cancers has shown a relation between increased lung DNA damage and early age at smoking initiation [3,4]. The organs of children are immature. Their differentiation and growth are rapidly progressed, therefore, childhood is considered to be a highly susceptible period for chemical-induced cancer, such as smoking-related lung cancer. Not only the above environmental factors, but also individual variations in genetics can affect susceptibility to lung cancer. Therefore, many molecular or genetic epidemiological studies have been performed to find genetic etiology for lung cancer over the last decades. Among those studies, genetic polymorphisms in metabolic enzymes, which are low penetrative — of which polymorphisms are common, and of which roles of environment are critical — have been the topic of attention [5]. However, most of the molecular or genetic epidemiological studies have a limitation that includes just one or a few of genetic polymorphisms. Thus, combined studies that included various genetic polymorphisms in various genes/enzymes are needed to clarify genetic etiology of lung cancer. In addition, well-organized studies that have correct information of subjects are necessary for reliable molecular epidemiological results. On the other hand, sex difference has shown different susceptibility to lung cancer [6]. Particularly, females show higher risk for adenocarcinoma, which are weakly related to tobacco smoking [3]. However, the reasons for sex differences in lung cancer risk are not yet clear. Therefore, we studied multiple-gene effects of the metabolic enzymes on lung cancer risk, considering sex, lifestyle and childhood- and current environment data.

2. Methods 2.1. Study subjects The study was approved by the Institutional Review Board at the National Cancer Center (NCC), Goyang, Korea. After

Table 1

agreement of informed consent, we could recruit 671 subjects (433 men and 238 women) who visited the NCC Hospital between 2002 and 2003. Cases (215 males and 103 females, total 318 persons) were patients who were histologically diagnosed as primary lung cancer at the NCC. Controls (218 men and 135 women, 353 persons) were visitors at the NCC for any cancer screening and health check up. Based on the screening and health check up, none of the controls had lung cancer or any other cancer. Most of the controls (326 subjects) were not diagnosed into any disease and the other 27 subjects of them were diagnosed into some diseases such as fatty liver, colorectal polyp, pneumonia, hypothyroidism, etc. Our questionnaire included information of occupation, family history of cancer, education, exercise, alcohol drinking, and exposure to smoking and housing environment at childhood, etc. Trained interviewers obtained informative questionnaire from each subject in person.

2.2. Genotype selection We selected polymorphisms of the six genes, which are involved in environmental carcinogen metabolism and have been suggested to affect lung cancer risk in scattered reports (Table 1). Table 1 shows their putative roles for lung cancer and representative polymorphisms that we studied.

2.3. DNA extraction and genotyping Genomic DNA of peripheral blood, which was obtained from each case or control via vein puncture, was isolated with a DNA isolation kit (Gentra, Minneapolis, MN, USA), following the manufacturer’s protocol. We determined each genotype following previously reported methods [7,17—20]. Whole genotyping was performed by investigators who were blinded to the subjects’ information. To perform quality control for genotyping, we included 20 double-blinded DNA samples, which were confirmed to be identical in all genome species. We used the PTC-200 gradient thermocycler (MJ Research, Waltham, MA) and the ABI GeneAmp PCR System 2700 (Applied Biosystems, Foster City, CA) for PCR amplification except TaqMan assay. For the TaqMan assay, we used the ABI PRISM® 7000 Sequence Detection System (Applied Biosystems).

The list of genetic polymorphisms that we studied

Gene name

Polymorphic site

Putative roles for lung cancer

NAD(P)H quinone oxidoreductase: NQO1 Cytochrome P450 1A1: CYP1A1 Myeloperoxidase: MPO

C609T in exon 6: Pro187Ser

Activation of nitrosamines; the Ser allele has a decreased activity [7]. Activation of polycyclic aromatic hydrocarbons (PAHs); The Val allele has a increased activity [8] and is risky [9]. Activation of PAHs: the A allele reduces activity and is protective [10]. Highly expressed in lung [11,12]: the 105Val allele has decreased activity toward 1-chloro-2,4-dinitrobenzene [13] but controversial trends toward PAH diol-epoxides [11,14]. Activation or detoxification of PAHs: the null type is risky [15,16].

A2455G in exon7: Ile462Val G-463 A: promoter region

Glutathione-S-transferase P1: GSTP1

A313G in exon5: Ile105Val

GSTT1 and GST M1

Gene presence or deletion

Six selected genes and lung cancer susceptibility

137

2.4. Statistics In the first phase of this analysis, each covariate was univariately tested for the association with lung cancer: The independent t-test was used for continuous covariates. Chisquare test was used for categorical covariates. The other categorical covariates, such as alcohol drinking, and education level, were tested by Chi-square test or Fisher’s exact test. In the second phase, we estimated adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for the association between lung cancer and genetic polymorphism using

Table 2

logistic regression models. Logistic regression models were constructed separately for sex. The significant confounding factors or independent variables at a level of P < 0.2 from the univariate analyses were selected for the models. Obtained candidates were used for following analyses. In the third phase, the lung cancer cases by each histological subtype were separately compared to the controls using multicategory logit models. Age, sex, smoking status, drinking, education, income, childhood exposure and exercise were adjusted in the models. That analysis was also carried out for each sex because of different pattern of risk factors of lung cancer by sex.

Demographic characteristics of the subjects

N

Cases (318)

Controls (353)

p

Age: years Sex: male, %

55.4 ± 9.6 67.6

48.3 ± 9.6 61.8

<0.01 0.11

Self smoking Current smoker Ex-smoker Nonsmoker Recent quittera

43.1 19.2 34.9 2.8

35.1 15.0 45.3 4.5

Alcohol drinking: frequency Never ≤Once/month 2—3 times/month 1—2 times/week 3—4 times/week 5—6 times/weeks Everyday

33.4 11.0 6.6 13.6 13.9 5.1 16.4

25.1 17.1 11.4 24.8 14.5 2.9 4.3

Childhood exposureb Rural Suburb Others

74.3 25.1 0.6

50.7 47.3 2.0

Education (period) <6 years since elementary school 6 years ≤ education period < 9 years 9 years ≤ education period < 12 years 12 years ≤ education period < 16 years ≥16 years

4.7 22.3 16.0 32.7 11.0

0.6 5.4 7.1 29.8 30.1

Exercise: frequency Never <2 times/week 1—2 times/week 3—4 times/week 5—6 times/week Every day

54.1 7.6 9.4 8.2 4.1 16.7

31.2 8.5 17.9 20.9 7.1 14.5

Monthly income after tax: U.S. $ <1000 1000—2000 2000—4000 4000—8000 ≥8000

23.7 29.8 32.9 10.8 2.9

2.3 10.5 43.9 28.8 14.5

a b

Stopped <1 year ago. Residential area at childhood.

0.02

<0.01

0.04

<0.01

<0.01

<0.01

138

M. Yang et al.

Finally, the effects of multi loci-genetic polymorphisms were estimated from a multiple logistic regression model. All statistical analyses were performed using SAS version 8.1 software (SAS Institute, Inc., Cary, North Carolina).

3. Results 3.1. Demographic characteristics of the subjects The mean age of the subjects was 51.6 years (±10.3), and the number of male and female among them was 433 and 238, respectively. Among the cases, adenocarcinoma was the histological main part (54.7% of the cases), followed by squamous cell carcinoma (22.3%), and small cell lung cancer (9.7%). The histological trend of lung cancer in the Korean subjects reflects recent increases in incidence of adenocarcinoma [3]. After univariate analyses in lifestyle and environmental factors, lung cancer patients showed high distribution in old age, male, current smoking, rural living at childhood, low-education, non-exercise, and low-income, compared to controls (Table 2). Due to sex, these distributional associations were somewhat different: These associations were strongly shown in men, while the associations between lung cancer and smoking or alcohol drinking were not significant in women (p = 0.14 and 0.19, respectively).

Table 3

NQO1

MPO

GSTP1

We could determine 96—100% of genotypes of each gene in all subjects, e.g. 643—671 of the 671 subjects. Table 3 shows the distributional differences in the genetic polymorphisms between cases and controls. As results, distribution of genotypes of the NQO1, CYP1A1, GSTP1, MPO, followed the Hardy-Weinberg equilibrium (ps > 0.05). When we compared distributions of each genotype in the Koreans with those of other east-Asians, the distributions in Koreans were almost the same to those in Japanese and Chinese, except those of the MPO [21,22]. Among various genetic polymorphisms, only the allele distribution of the GSTP1 polymorphism was significantly different between cases and controls, i.e. the proportion of GSTP1-mutant, the Val allele was higher in cases than controls.

3.3. Lung cancer-associated genetic factors adjusted for environment We studied effects of the multiple genetic polymorphisms on lung cancer with environmental factors, which were associated with lung cancer by univariate analyses (Table 3). Besides subjects who had any missing datum of questionnaire or genotypes, we could perform a multi-nominal logistic regression in 618 subjects. Caused by sex, distribu-

Different distributions of genetic polymorphisms between lung cancer cases and controlsa

Polymorphic siteb

CYP1A1

3.2. Distribution of the genetic polymorphisms

Case vs. Control

Case Control

Case Control

Case Control

Case Control

Genotype Frequency N (%)

p-value

Pro/Pro

Pro/Ser

Ser/Ser

110(35.03) 120(34.58)

158(50.32) 166(47.84)

46(14.65) 61(17.58)

lle/lle

lle/Val

Val/Val

182(57.96) 220(63.04)

116(36.94) 111(31.81)

16(5.10) 18(5.16)

G/G

G/A

A/A

269(84.59) 283(80.17)

49(15.41) 68(19.26)

0(0.00) 2(0.57)

lle/lle

lle/Val

Val/Val

198(62.46) 248(70.25)

103(32.49) 93(26.35)

16(5.05) 12(3.40)

Allele Frequency

0.58

0.37

0.13

0.09

Pro

Ser

0.60 0.59

0.40 0.41

lle

Val

0.76 0.79

0.24 0.21

G

A

0.92 0.90

0.08 0.10

lle

Val

0.79 0.83

0.21 0.17

Present

Deletion

p-value

0.53

0.27

0.12

0.03c

GSTT1

Case Control

148(46.84) 179(51.88)

168(53.16) 166(48.12)

0.19

GSTM1

Case Control

159(50.16) 167(48.27)

158(46.84) 179(51.73)

0.63

a b c

Two-sided 2 test. Wild/wild = 1; hetero or mutant/mutant = 2; gene present = 1, deletion = 2 for GSTT1 and GSTM1. Significantly different between cases and controls.

Six selected genes and lung cancer susceptibility Table 4

139

Combined effects of genetic polymorphisms on adenocarcinomaa

Genotypeb

Total (N) Adenocarcinoma/control (166/323)

Males (N) Adenocarcinoma/control (83/202)

Females (N) Adenocarcinoma/control (83/121)

NQO1 CYP1A1 MPO GSTP1 GSTT1 GSTM1

1.06 (0.67, 1.68) 1.19 (0.77, 1.84) 0.45 (0.25, 0.82) 1.30 (0.83, 2.04) 1.09 (0.71, 1.67) 1.09 (0.71, 1.68)

1.24 0.99 0.42 0.85 0.97 1.48

0.79 (0.37, 1.69) 1.69 (0.81, 3.52) 0.41 (0.17, 0.99) 2.48 (1.19, 5.16) 1.24 (0.62, 2.49) 0.74 (0.37, 1.46)

(0.67, (0.56, (0.17, (0.46, (0.54, (0.83,

2.30) 1.77) 1.03) 1.59) 1.72) 2.64)

Bold, P < 0.05. a Data show ORs (95% CI) adjusted for age, sex, self-smoking, alcohol drinking, education, income, exercise, and childhood exposure. b Wild/wild = 1; hetero or mutant/mutant = 2; gene present = 1, deletion = 2 for GSTT1 and GSTM1.

tions of lung cancer subtypes and of genotypes of the MPO among the six studied genes were different (p < 0.001 and p = 0.07, respectively). Thus, we studied genetic effects on lung cancer types with sex in a prevalent subtype, adenocarcinoma, and the others (Table 4). As results, we found that presence of the mutant allele of the GSTP1 showed increased risk for adenocarcinoma in females. In a case of the MPO, the presence of the MPO- mutant allele showed significantly protective effects on adenocarcinoma with or without consideration of sex. Considering other histological features, we also studied genetic effects on small cell carcinoma versus non-small cell carcinoma. As results, only the GSTP1 mutant type showed risks in small cell carcinoma cases (p = 0.08, cases, N = 27, controls, N = 336). However, due to small number of the cases, we could not obtain 95% CI in our regression model. In a case of non-small cell carcinoma (N = 275), the GSTP1 mutant type and MPO wild type showed risks (p = 0.08 and 0.005, respectively, adjusted OR [95% CI], 1.44 [0.96—2.17] and 2.13 [1.25—3.57], respectively). Therefore,

Table 5

the GSTP1 mutant type showed somewhat risks in both of small cell and non-small cell carcinoma, while the MPO wild type showed significant risks only in non-small cell carcinoma. In addition, the mutant type of the CYP1A1 showed a significant high risk for the squamous cell carcinoma (OR [95% CI], 1.7 [1.01—2.86]: case versus control, N = 71 versus 349). Focusing on the above three genetic polymorphisms of the GSTP1, MPO, and CYP1A1, which affected adenocarcinoma or the other types, we studied multi-loci genetic effects on lung cancer (Table 5). The reference, which is suspected lowest risk for cancer, was subjects who were homozygous wild type of the GSTP1 and CYP1A1 and had the MPO- mutant allele. Among the present eight genotypes, subjects who had the #8 genotype-which was different in all three genetic poly-morphisms with the references, i.e. the mutant alleles of GSTP1 and CYP1A1 and homozygous wild type of the MPO - showed approximately 5-fold-higher risk for lung cancer than the reference. In addition, we found

Effects of multi loci- genetic polymorphisms on lung cancer riska

Genotype#

N (case/control)

Polymorphismb

OR (95% CI)c

GSTP1

MPO

CYP1A1

1 2 3 4

54 (19/35) 24 (12/12) 22 (10/12) 221 (99/122)

0 0 1 0

1 1 1 0

0 1 0 0

Reference 2.12 (0.66, 6.80) 2.00 (0.56, 7.14) 1.91 (0.89, 4.08)

1.92 (0.91, 4.04)d

5 6 7

18 (8/10) 142 (66/76) 105 (54/51)

1 0 1

1 0 0

1 1 0

1.03 (0.30, 3.61) 2.45 (1.11, 5.43) 2.83 (1.24, 6.48)d

2.41 (1.14, 5.08)e

8

77 (46/31)

1

0

1

4.97 (2.05, 12.05)f

f 4.97 (2.05, 12.05)

Bold, significant differences between the reference and #6 (p = 0.028), #7 (p = 0.014), or #8 (p = 0.0004). a Multiple logistic regression. b Wild/wild, coded as ‘0’; wild/mutant or mutant/mutant, coded as ‘1’. c Adjusted for age, sex, self-smoking, alcohol drinking, education, income, exercise, and childhood exposure. d Between the reference and the group that has a different polymorphism with the reference. e Between the reference and the group that has two different polymorphisms with the reference. f Between the reference and the group that has three different polymorphisms with the reference.

140 increase of lung cancer-risk due to increased numbers of different genotypes from the reference by trend analysis (number of the different genotypes, 1, 2, and 3, ORs, 1.92, 2.41, and 4.97, respectively, p = 0.0064).

4. Discussion Adenocarcinoma is the most common type of lung cancer in this Korean population [1]. In this study, 83% of the female cases had adenocarcinoma, while 40% of the male cases had. In addition, only 7.1% of the female subjects were smokers and active tobacco smoking was not risky for female lung cancer as much as it was for male lung cancer. Thus, we focused on genetic variations in metabolic enzymes, which are involved in metabolisms of environmental carcinogens to study the etiology of lung cancer. Considering environmental and lifestyle factors, we compared distributions of genetic polymorphisms in cases with those in controls and found that only the allele distribution of the GSTP1 polymorphism was significantly different between cases and controls, i.e. proportion of the GSTP1 mutant, the Val allele, was higher in cases than controls (Table 3). This result was proved right, particularly in female adenocarcinoma cases (Table 4). There are some reports that support our result, i.e. presence of the GSTP1-Val allele is risky for female lung cancer [23,24]. However, mechanisms how the presence of the GSTP1-Val allele presents risk for lung cancer is not clear, yet. A clear evidence of the GSTP1 involvement for lung cancer is high localization of the GSTP1 in lung, compared to other GST-isozymes, such as GSTM1, GSTA1, etc. [11,12]. In addition, the GSTP1-Val allele was reported to be associated with RARB (retinoic acid report beta)- methylation in Chinese female non-small cell lung cancer [24]. A recent meta-analysis of mixed populations showed that null types of GSTT1 and GSTM1 among GSTs had weakly positive associations with lung cancer [25]. However, we could not find any association between these genotypes and the Korean lung cancer (Tables 3 and 4). Effects of the polymorphism of the GSTM1 on lung cancer have been suspected due to loss of detoxification of carcinogens such as benzo[a]pyrene diolepoxide for 10 years of more [26], however, are still unclear due to some reasons, e.g. low expression of GSTM1 in lung. Myeloperoxidase is a known lysosomal enzyme found in high concentrations in human lung due to recruitment of neutrophils, and activates benzo[a]pyrene as well as aromatic amines in tobacco smoke and generates carcinogen-free radicals. A single base substitution in the promoter region of the myeloperoxidase gene (-463G > A) has been demonstrated to markedly reduce transcription, activity, and DNA adduct levels in bronchoalveolar lavages of smokers [27]. These mechanisms have supported protective effects of the MPO-463 A allele against lung cancer. For example, a meta-analysis of nine studies constituted a homogeneous set, which yielded a significantly reduced summary OR for the A/A genotype (OR, 0.68, 95% CI: 0.5—0.93) [28]. In the present study, we also found that the MPO-463 A allele showed protective effects on lung cancer, particularly adenocarcinoma regardless of sex (Table 4). However, the possible inverse association of the MPO-A

M. Yang et al. allele with lung cancer risk has remained controversial [29]. Therefore, effects of the MPO polymorphism on lung cancer should be further studied in order to find definitive conclusion. History of studies concerning the CYP1A1 genetic effects on lung cancer is one of the oldest ones in molecular epidemiology, as similar to the case of the above GSTM1 [30]. The mechanism of high risk in the CYP1A1 Val allele presence was explained with its high enzyme activity compared to the CYP1A1 Ile allele [8]. Presence of the CYP1A1 mutant, the Val allele, showed high risk in the squamous cell carcinoma rather than adenocarcinoma in the present study. Our result follows a hypothesis that the CYP1A1 polymorphism shows its effects on strongly tobacco-related cancer, because the CYP1A1 enzyme is involved in bioactivation of tobacco carcinogens and the squamous cell carcinoma showed stronger association with tobacco-smoking than adenocarinoma: Le Marchand and colleagues also suggested that the CYP1A1 Ile462Val polymorphism might confer an increased risk of lung cancer-squamous cell carcinoma rather than adenocarcinoma and especially in those who have never smoked and in women in a pooled analysis with 11 case-controls [9]. Recently, the CYP1A1 genetic polymorphism reported to influence adenocarinoma with interaction of the MPO-463 polymorphism [31]. In a case of NQO1, the C609T substitution encodes an enzyme with reduced quinone reductase activity in vitro and thus was hypothesized to be protective to cancer [7]. The Pro/Ser and Ser/Ser genotypes combined of NQO1 showed significant association with decreased risk of lung cancer in a Japanese meta-analysis study (random effects OR [95%CI] = 0.70 [0.56—0.88]: cases versus controls, N = 499 versus 959) [32]. However, Chao and colleagues recently reported no clear association between the NQO1 Pro187Ser polymorphism and lung cancer risk in the three ethnic groups examined: OR (95%CI), White, ‘C/T + T/T’ versus C/C = 1.04 (0.96—1.13), Asian, 0.99 (0.72—1.34), and Blacks, 0.95 (0.66—1.36) in a meta-analysis [33]. In our study which was performed in an ethnically identical population and adjusted for life and environmental factors, such as smoking status, the ‘C/T + T/T’ type showed some what protective effects in female adenocarcinoma, even though it was not significant (Table 4). As there were inconsistent results, effects of the NQO1 polymorphism on lung cancer should be further studied with careful consideration of environmental factors. We tried to find the genetic etiology of lung cancer with integrated approaches, i.e. multiple gene effects with adjustment for life and environmental effects. As results, we found combination effects of the three genetic polymorphisms of the GSTP1, MPO, and CYP1A1 (Table 5). It follows the hypothesis that multiple genes interact to confer genomic-based susceptibility to lung cancer. However, a recent Australian study showed an opposite result to our result concerning the GSTP1 (Table 4), i.e. it showed that the GSTP1 105Val genotype was protective to female NSCLC, and that there was interaction only between the MPO and CYP1A1 risk genotypes, ‘the MPO GG and the CYP1A1 Ile/Val or Val/Val’, to increase the overall risk of NSCLC [31]. Therefore, the combination effects of multiple genes may be different in various ethnical populations, as ethnical difference results in different effects of each gene.

Six selected genes and lung cancer susceptibility In conclusion, we suggest that the combination of the GSTP1, MPO, and CYP1A1 variations affects susceptibility to lung cancer. Particularly, presence of the GSTP1 Val allele showed higher risk for adenocarcinoma in Korean female nonsmokers. In the near future, the mechanisms should be clarified how the three genetic polymorphisms are involved in lung cancer initiation.

141

[14]

[15]

Conflict of interest statement None declared.

[16]

Acknowledgements This work was supported by the grant of the NCC (Goyang, South Korea). We appreciate the help of Ms. Yeong-In Kim, Hyo Won Park, Choon Seon Park, and Hee Sun Kim at the NCC.

References [1] Cancer Registration & Biostatistics Branch NCC. Cancer statistics in Korea, 2000. http://wwwnccrekr/files/statistics/ Cancer%20Statistics%20in%20KOREAdoc. [2] International Agency for Research on Cancer. IARC monographs on the evaluation of the carcinogenic risk of chemicals to humans: Tobacco Smoking Lyon. IARC; 1986. [3] Bilello KS, Murin S, Matthay RA. Epidemiology, etiology, and prevention of lung cancer. Clin Chest Med 2002;23:1—25. [4] Wiencke JK, Thurston SW, Kelsey KT, Varkonyi A, Wain JC, Mark EJ, et al. Early age at smoking initiation and tobacco carcinogen DNA damage in the lung. J Natl Cancer Inst 1999;91:614—9. [5] Vines P, Malats N, Lang M, d’Errico A, Caporaso N, Cuzick J, Boffetta P, editors. Metabolic polymorphisms and susceptibility to cancer. Lyon: IARC; 1999. [6] Stabile LP, Siegfried JM. Sex and sex differences in lung cancer. J Gend Specif Med 2003;6:37—48. [7] Chen H, Lum A, Seifried A, Wilkens LR, Le Marchand L. Association of the NAD(P) H:quinone oxidoreductase 609C—>T polymorphism with a decreased lung cancer risk. Cancer Res 1999;59:3045—8. [8] Cosma G, Crofts F, Taioli E, Toniolo P, Garte S. Relationship between genotype and function of the human CYP1A1 gene. J Toxicol Environ Health 1993;40:309—16. [9] Le Marchand L, Guo C, Benhamou S, Bouchardy C, Cascorbi I, Clapper ML, et al. Pooled analysis of the CYP1A1 exon 7 polymorphism and lung cancer (United States). Cancer Causes Control 2003;14:339—46. [10] Dally H, Gassner K, Jager B, Schmezer P, Spiegelhalder B, Edler L, et al. Myeloperoxidase (MPO) genotype and lung cancer histologic types: the MPO -463 A allele is associated with reduced risk for small cell lung cancer in smokers. Int J Cancer 2002;102:530—5. [11] Coles B, Yang M, Lang NP, Kadlubar FF. Expression of hGSTP1 alleles in human lung and catalytic activity of the native protein variants towards 1-chloro-2,4-dinitrobenzene, 4vinylpyridine and (+)-anti benzo[a]pyrene-7,8-diol-9,10-oxide. Cancer Lett 2000;156:167—75. [12] Yang M, Coles BF, Delongchamp R, Lang NP, Kadlubar FF. Effects of the ADH3, CYP2E1, and GSTP1 genetic polymorphisms on their expressions in Caucasian lung tissue. Lung Cancer 2002;38:15—21. [13] Watson MA, Stewart RK, Smith GB, Massey TE, Bell DA. Human glutathione-S-transferase P1 polymorphisms: relationship to

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

lung tissue enzyme activity and population frequency distribution. Carcinogenesis 1998;19:275—80. Sundberg K, Johansson AS, Stenberg G, Widersten M, Seidel A, Mannervik B, et al. Differences in the catalytic efficiencies of allelic variants of glutathione transferase P1-1 towards carcinogenic diol epoxides of polycyclic aromatic hydrocarbons. Carcinogenesis 1998;19:433—6. Perera FP, Mooney LA, Stampfer M, Phillips DH, Bell DA, Rundle A, et al. Associations between carcinogen-DNA damage, glutathione-S-transferase genotypes, and risk of lung cancer in the prospective Physicians’ Health Cohort Study. Carcinogenesis 2002;23:1641—6. Sorensen M, Autrup H, Tjonneland A, Overvad K, RaaschouNielsen O. Glutathione-S-transferase T1 null-genotype is associated with an increased risk of lung cancer. Int J Cancer 2004;110:219—24. Oyama T, Mitsudomi T, Kawamoto T, Ogami A, Osaki T, Kodama Y, et al. Detection of CYP1A1 gene polymorphism using designed RFLP and distributions of CYP1A1 genotypes in Japanese. Int Arch Occup Environ Health 1995;67:253—6. Cancer Genome Anatomy Project SNP500Cancer Database. http://snp500cancer.nci.nih.gov/taqman assays.cfm?snp id= MPO-02. Gao CM, Takezaki T, Wu JZ, Li ZY, Liu YT, Li SP, et al. Glutathione-S-transferases M1 (GSTM1) and GSTT1 genotype, smoking, consumption of alcohol and tea and risk of esophageal and stomach cancers: a case-control study of a high-incidence area in Jiangsu Province. China Cancer Lett 2002;188:95—102. Kelada SN, Stapleton PL, Farin FM, Bammler TK, Eaton DL, Smith-Weller T, et al. Glutathione-S-transferase M1, T1, and P1 polymorphisms and Parkinson’s disease. Neuroscience Lett 2003;337:5—8. Hamajima N, Saito T, Matsuo K, Suzuki T, Nakamura T, Matsuura A, et al. Genotype frequencies of 50 polymorphisms for 241 Japanese non-cancer patients. J Epidemiol 2002;12: 229—36. Hamajima N, Takezaki T, Tajima K. Allele frequencies of 25 polymorphisms pertaining to cancer risk for Japanese, Koreans and Chinese. Asian Pac J Cancer Prev 2002;3:197—206. Miller DP, Neuberg D, de Vivo I, Wain JC, Lynch TJ, Su L, et al. Smoking and the risk of lung cancer: susceptibility with GSTP1 polymorphisms. Epidemiology 2003;14:545—51. Chan EC, Lam SY, Fu KH, Kwong YL. Polymorphisms of the GSTM1, GSTP1, MPO, XRCC1, and NQO1 genes in Chinese patients with non-small cell lung cancers: relationship with aberrant promoter methylation of the CDKN2A and RARB genes. Cancer Genet Cytogenet 2005;162:10—20. Ye Z, Song H, Higgins JP, Pharoah P, Danesh J. Five glutathioneS-transferase gene variants in 23,452 cases of lung cancer and 30,397 controls: meta-analysis of 130 studies. PLoS Med 2006;3:e91. Nazar-Stewart V, Motulsky AG, Eaton DL, White E, Hornung SK, Leng ZT, et al. The glutathione-S-transferase mu polymorphism as a marker for susceptibility to lung carcinoma. Cancer Res 1993;53(10 Sup):2313—8. Van Schooten FJ, Boots AW, Knaapen AM, Godschalk RW, Maas LM, Borm PJ, et al. Myeloperoxidase (MPO) −463G->A reduces MPO activity and DNA adduct levels in bronchoalveolar lavages of smokers. Cancer Epidemiol Biomarkers Prev 2004;13:828—33. Feyler A, Voho A, Bouchardy C, Kuokkanen K, Dayer P, Hirvonen A, et al. Point: myeloperoxidase −463G—> a polymorphism and lung cancer risk. Cancer Epidemiol Biomarkers Prev 2002;11:1550—4. Xu LL, Liu G, Miller DP, Zhou W, Lynch TJ, Wain JC, et al. Counterpoint: the myeloperoxidase −463G—> a polymorphism does not decrease lung cancer susceptibility in Caucasians. Cancer Epidemiol Biomarkers Prev 2002;11:1555—9.

142 [30] Kawajiri K, Nakachi K, Imai K, Yoshii A, Shinoda N, Watanabe J. Identification of genetically high risk individuals to lung cancer by DNA polymorphisms of the cytochrome P450IA1 gene. FEBS Lett 1990;263:131—3. [31] Larsen JE, Colosimo ML, Yang IA, Bowman R, Zimmerman PV, Fong KM. CYP1A1 Ile462Val and MPO G-463A interact to increase risk of adenocarcinoma but not squamous cell carcinoma of the lung. Carcinogenesis 2006;27:525—32.

M. Yang et al. [32] Kiyohara C, Yoshimasu K, Takayama K, Nakanishi Y. NQO1, MPO, and the risk of lung cancer: a HuGE review. Genet Med 2005;7:463—78. [33] Chao C, Zhang ZF, Berthiller J, Boffetta P, Hashibe M. NAD(P)H:quinone oxidoreductase 1 (NQO1) Pro187Ser polymorphism and the risk of lung, bladder, and colorectal cancers: a meta-analysis. Cancer Epidemiol Biomarkers Prev 2006;15:979—87.