Lung adenocarcinoma has replaced squamous cell lung carcinomaas the most frequent histological subtype in lung cancers. However, genetic factors that affect cancer susceptibility are much less understood in adenocarcinoma than in squamous cell carcinoma. In this study, polymorphisms in five genes involved in the metabolism of carcinogens or in the repair of damaged DNA in lung cells, NQO1-Pro187Ser, GSTT1-positive/null, GSTM1-positive/null, CYP1A1-Ile462Val, and OGG1-Ser326Cys, were examined for association with lung adenocarcinoma risk in a case-control study of 198 patients and 152 control subjects. The NQO1 and GSTT1 polymorphisms were associated with lung adenocarcinoma risk with adjusted odds ratio of 2.15 for the NQO1-Pro/Pro genotype versus the Ser/Ser genotype and adjusted odds ratio of 1.61 for the GSTT1-null genotype versus the positive genotype, respectively. Furthermore, individuals with the combined genotype of NQO1-Pro/Pro and GSTT1-null showed greater risk compared with those of NQO1-Ser/Ser and GSTT1-positive. In contrast, significant association was not observed for the GSTM1, CYP1A1, and OGG1 polymorphisms with lung adenocarcinoma risk, although several studies have shown their implication in the risk for squamous cell lung carcinoma. The result indicates that the NQO1-Pro/Pro and GSTT1-null genotypes are risk factors for lung adenocarcinoma development, and that the genetic factors for susceptibility to adenocarcinoma are different from those to squamous cell carcinoma. The enhanced risk of the NQO1-Pro/Pro genotype combined with the GSTT1-null genotype was more evident in smokers than in nonsmokers. Therefore, carcinogens in tobacco smoke, which are activated by NQO1 and detoxified by GSTT1, could have a role in lung adenocarcinoma development.

Lung cancer is the leading cause of cancer-related deaths in the world. Lung cancer consists of at least three major histological subtypes: adenocarcinoma, squamous cell carcinoma, and small cell carcinoma. It is well known that the development of squamous and small cell carcinomas is strongly associated with smoking, whereas that of adenocarcinoma is less associated compared with those two subtypes, indicating that carcinogenic processes are different among the histological subtypes of lung cancers. In recent years, adenocarcinoma has replaced squamous cell carcinoma as the most frequent histological subtype in lung cancers (1, 2, 3, 4). This is considered as being attributable to the decrease in the number of smokers, because the majority of lung cancers in nonsmokers are adenocarcinomas (1, 2, 3, 4). The changes in tobacco composition associated with the prevalence of filter cigarettes also have been thought to have a large effect on the increase of adenocarcinoma (1, 2, 3, 4). Another possible reason is that smokers of filter cigarettes compensate for the reduced nicotine yields by deep and intense inhalation of smoke, resulting in increased deposition of tobacco carcinogens in the peripheral lung where adenocarcinoma preferentially develops.

Genetic factors responsible for susceptibility to lung cancer have been searched for to establish novel and efficient ways of preventing the disease. The association of cancer risk with polymorphisms in genes that encode enzymes involved in the metabolism of tobacco carcinogens, such as cytochrome P450-1A1 (CYP1A1) and glutathione S-transferase M1 (GSTM1), has been extensively studied, especially in squamous cell lung carcinoma (5, 6, 7, 8, 9, 10, 11, 12, 13). PAHs3 in tobacco smoke have been indicated to largely contribute to the development of squamous cell lung carcinoma (1, 14). CYP1A1 plays a major role in the activation of PAHs, and GSTM1 plays a major role in the detoxification of activated PAH intermediates (7, 8). SNP of CYP1A1 at codon 462 leading to a substitution of valine (Val) for isoleucine (Ile) results in an increase in the enzyme activity, whereas an inherited homozygous deletion of the GSTM1 gene (the GSTM1-null genotype) results in a lack of the enzyme activity (7, 8). The CYP1A1-Val/Val genotype, the GSTM1-null genotype and their combined genotype have been reproducibly associated with an increased risk for squamous cell lung carcinoma (5, 6, 7, 8, 9, 10, 11, 12, 13). In addition, a recent study of ours suggests that a SNP in the OGG1 gene causing an amino acid substitution, Ser326Cys, is also associated with the risk for squamous cell lung carcinoma (15). OGG1 is a DNA glycosyrase for oh8G, which is a mutagenic base formed by oxidative stresses, including the ones caused by tobacco carcinogens. It was shown that the ability of OGG1-Cys326 protein to suppress mutations was significantly lower than that of OGG1-Ser326 protein (16, 17). The Cys/Cys genotype was associated with an increased risk for squamous cell lung carcinoma (15). Therefore, individuals with weak OGG1 activity associated with the Cys/Cys genotype could be susceptible to the development of squamous cell lung carcinoma because of the high rate of accumulation of mutations caused by oh8G. Thus, several genetic factors underlying the risk for the development of squamous cell lung carcinoma have been identified. In contrast, several genetic polymorphisms that were shown to be associated with the risk for squamous cell lung carcinoma have not been clearly associated with the risk for lung adenocarcinoma (5, 6, 7, 8, 9, 10, 11, 12, 13, 15). In addition, carcinogens and other etiological factors strongly associated with the development of lung adenocarcinoma have not been identified. Therefore, genetic factors involved in the risk for lung adenocarcinoma are mostly unknown at present.

Interindividual differences in the activities of NQO1 and GSTT1 have been shown to be involved in lung cancer susceptibility (18, 19, 20, 21, 22, 23, 24, 25). NQO1 metabolically activates several carcinogens such as nitrosamines and heterocyclic amines that are present in tobacco smoke and foods (18). GSTT1 is involved in the detoxification of several carcinogens, such as 1, 3-butadiene and ethylene oxide, that are present in tobacco smoke and ambient air (19). SNP at codon 187 in NQO leads to the substitution of proline (Pro) for serine (Ser), and NQO1-Pro187 protein has a higher enzyme activity than NQO1-Ser187 protein does (18). There is a deletion allele for the GSTT1 gene with null GSTT1 activity (19). Therefore, the homozygotes for the NQO1-Pro187 allele (NQO1-Pro/Pro) and/or those for GSTT1 deletion (GSTT1-null) could be associated with an increased risk for lung cancer. Two case-control studies indicated that the NQO1-Pro/Pro genotype was associated with an increased risk for overall lung cancer, although the association with specific histological subtypes of lung cancer was not examined (20, 21). Subsequently, Lin et al. (22) reported that this genotype was associated with the risk for lung adenocarcinoma in Taiwan. However, in this study, the number of adenocarcinoma subjects examined was small (n = 36); therefore, the contribution of this genotype to lung adenocarcinoma susceptibility should be confirmed by a larger study. The association of the GSTT1-null genotype with the risk for overall lung cancer has also been reported (23, 24). However, in these studies, the association with specific histological subtypes of lung cancer was not examined. Therefore, the involvement of the polymorphisms in NQO1 and GSTT1 in susceptibility to lung adenocarcinoma remains obscure.

In the present study, we performed a case-control study of 198 lung adenocarcinoma patients and 152 control subjects for the distributions of genetic polymorphisms, NQO1-Pro187Ser and GSTT1-positive/null. We also examined the distributions of the CYP1A1-Ile462Val, GSTM1-positive/null, and OGG1-Ser326Cys polymorphisms in these populations. The NQO1-Pro/Pro and GSTT1-null genotypes were significantly associated with the risk for lung adenocarcinoma. Furthermore, the combination of the NQO1-Pro/Pro genotype with the GSTT1-null genotype showed an enhanced effect on the risk. The association of the combined genotype with the risk was more evident in smokers than in nonsmokers. In contrast, significant associations of the CYP1A1, OGG1, and GSTM1 polymorphisms with the risk were not observed. These results indicate that genetic factors affecting the risk for the development of lung cancer are different between adenocarcinoma and squamous cell carcinoma. Investigation of NQO1 and GSTT1, as well as their substrates should have an implication in understanding carcinogenic processes of lung adenocarcinoma and preventing the disease.

Case-Control DNA Set.

The study population consisted of 198 lung adenocarcinoma cases and 152 control subjects, recruited from the National Nishigunma Hospital and the National Cancer Center Hospital from 1999 to 2001 (Table 1). All of the cases and control subjects were Japanese. All of the lung adenocarcinoma cases, from whom informed consent as well as blood samples were obtained, were consecutively included in this study without any particular exclusion criteria. The participation rate was approximately 80%. All of the cases were diagnosed as being lung adenocarcinoma by cytological and/or histological examinations according to the WHO classification (26). The cases consisted of 87 (43.9%) of stage I, 10 (5.1%) of stage II, 44 (22.2%) of stage III, 56 (28.3%) of stage IV, and a case of which the information on the stage was not available. One hundred and twelve (56.6%) were surgical cases, whereas the remaining 86 (43.4%) received other therapies, including chemotherapy and radiation therapy. Fifteen of the cases had histories of other malignant tumors, including breast cancer (5 cases), gastric cancer (3 cases), and thyroid cancer (3 cases). Diagnoses of primary lung cancers and not of metastases of other malignancies were carefully made by pathological and immunohistochemical examinations for these cases. Controls were randomly selected from inpatients and outpatients with no history of cancer at the hospitals during the study periods. The distribution of clinical diagnoses among the controls was as follows: chronic obstructive lung disease, 27 cases; bronchitis or pneumonia, 24 cases; pulmonary tuberculosis, 17 cases; old pulmonary tuberculosis, 8 cases; pulmonary nontuberculous mycobacteriosis, 10 cases; pulmonary abscess, 9 cases; pneumoconiosis, 9 cases; pulmonary mycosis, 8 cases; sarcoidosis, 3 cases; other respiratory diseases, 14 cases; benign tumor, 3 cases; diabetes mellitus, 3 cases; pancreatitis, 2 cases; gastric ulcer, 1 case; uterine myoma, 1 case; hypertension, 1 case; angina pectoris, 1 case; pericarditis, 1 case; rheumatoid arthritis, 1 case; bone fracture, 1 case; and healthy individuals, 8 cases. Ages of the participants were computed from their date of birth. Smoking history was obtained via interview using a questionnaire. Smoking habit was represented by cigarette-years, which was defined as the number of cigarettes smoked daily multiplied by years of smoking, both in current smokers and former smokers. Nonsmokers were defined as those who had never smoked. Smokers were defined as those who had at any time smoked at least one cigarette a day for 1 year or more. Informed consent was taken from all of the participants before blood sampling, and the study was approved by the ethical committees of the Nishigunma Hospital and the National Cancer Center. Ten ml of heparinized whole-blood was obtained from each lung adenocarcinoma patient and from each control subject. Genomic DNA was extracted from the whole-blood by using a QIAamp DNA blood Maxi kit (Qiagen, Tokyo, Japan) according to the supplier’s recommendations.

Genotyping of the CYP1A1, NQO1, and OGG1 Polymorphisms.

The genotypes of the CYP1A1 Ile462Val and NQO1 Pro187Ser polymorphisms were determined by PCR-RFLP methods described by Oyama et al. (27) and Chen et al. (21), respectively. The genotypes of the OGG1 Ser326Cys polymorphism were determined also by a PCR-RFLP method using the fact that the Cys polymorphism at codon 326 of OGG1 creates a Fnu4HI restriction site between codon 326 (TGC) and codon 327 (CGC). The sequences of paired primers used for identification of the CYP1A1 and NQO1 polymorphisms were described (21, 27). The sequences of paired primers for the OGG1 polymorphisms were 5′-ACTGTCACTAGTCTCACCAG-3′ and 5′-GGAAGGTGCTTGGGGAAT-3′. Ten ng of genomic DNA were suspended in a total volume of 20 μl PCR buffer containing 10 mm Tris-HCl (pH 8.3), 50 mm KCl, 1.5 mm MgCl2, 300 nm of each primer, 200 μm of deoxynucleotide triphosphate, and 0.5 units of AmpliTaq Gold DNA polymerase (PE Applied Biosystems, Tokyo, Japan). PCR conditions were as follows: 30 s at 95°C and 30 s at 60°C for 45 cycles, followed by 10 min at 60°C. Three DNA fragments of 665, 868, and 620 bp, that contain a single HincII, HinfI, and Fnu4HI site, respectively, were amplified by PCR from the pcDNA3.1 (−) plasmid DNA (Invitrogen, Tokyo, Japan) to be used as controls to confirm complete digestion with the restriction enzymes. PCR products containing SNP sites for CYP1A1, NQO1, and OGG1 were mixed with these control fragments and were digested with HincII, HinfI, and Fnu4HI (New England Biolabs, Tokyo, Japan).

In the genotyping of CYP1A1, complete digestion of 187-bp PCR products produced 139-bp and 48-bp fragments for the Ile allele, and 120-bp, 48-bp, and 19-bp fragments for the Val allele (21). In the genotyping of NQO1, complete digestion of 230-bp PCR products produced 195-bp and 35-bp fragments for the Pro allele, and 151-bp, 44-bp, and 35-bp fragments for the Ser allele (27). In the OGG1 analysis, complete digestion of 200-bp PCR products produced two 100-bp fragments for the Cys allele, whereas they remained as 200-bp fragments for the Ser allele. These digested products were electrophoresed on a 3% agarose gel and visualized by ethidium bromide staining. Complete digestion of PCR products was confirmed by the fact that control fragments, which were mixed with each PCR product for genotyping, produced 333-bp and 332-bp fragments for HincII, 441-bp and 427-bp fragments for HinfI, and 315-bp and 305-bp fragments for Fnu4HI.

Genotyping of the GSTM1 and GSTT1 Polymorphisms.

The genotypes of the GSTM1 and GSTT1 polymorphisms were determined by a multiplex PCR method described by Arand et al. (28). The GSTM1 or GSTT1 specific primer set and a primer set for albumin were used for the same amplification reaction. Ten ng of genomic DNA were suspended in a total volume of 20 μl PCR buffer containing 10 mm Tris-HCl (pH 8.3), 50 mm KCl, 2.5 mm MgCl2, 3 μg/ml of each GSTM1 primer or 1 μg/ml of each GSTT1 primer, 0.6 μg/ml of each albumin primer, 200 μm of deoxynucleotide triphosphate, and 0.5 units of AmpliTaq Gold DNA polymerase (PE Applied Biosystems). PCR conditions were as follows: 60 s at 95°C, 60 s at 60°C, and 60 s at 72°C for 45 cycles, followed by 10 min at 72°C. The 219-bp GSTM1 fragments and the 459-bp GSTT1 fragments were coamplified with the 350-bp albumin fragments in the same reaction tube. The albumin fragments served as a positive control for the success of the amplification reaction. The absence of either GSTM1 or GSTT1 fragments indicated the corresponding null genotype. The PCR products were electrophoresed on a 3% agarose gel and visualized by ethidium bromide staining.

Statistical Analysis.

The strength of association between lung adenocarcinoma and either the CYP1A1, NQO1, OGG1, GSTM1, or GSTT1 polymorphisms was measured as ORs. ORs adjusted for age, gender, and smoking habit with 95% CIs were calculated using an unconditional logistic regression analysis (29). When dividing the study population into smokers and nonsmokers, ORs adjusted for age and gender with 95% CIs were calculated. Differences in the distributions of gender, age, and smoking habit between cases and controls were tested by the χ2 test. Difference in the genotype distribution of the polymorphism in the control subjects between this study and previous case-control studies for the Japanese population was also tested by the χ2 test. The statistical analyses described above were performed using the StatView version 5.0 software (SAS Institute Inc., Cary, NC). The Markov chain method with the GENEPOP program4 was used to test for deviations of genotype distributions from the HWE. Ps < 0.05 were considered as being statistically significant, with 0.05 ≤ P < 0.1 being marginal.

Genotype Distributions of the NQO1, GSTT1, GSTM1, CYP1A1, and OGG1 Polymorphisms.

Distributions of gender, age, and smoking habit among cases with lung adenocarcinoma and controls are shown in Table 1. The difference in the distributions of smoking habit was not significant between the cases and controls as a whole (P = 0.455), and also between the cases and controls of both male and female subjects (P = 0.463 and 0.319, respectively). A large portion of the male subjects were smokers, whereas that of the female subjects were nonsmokers, both in the cases and controls. On the other hand, the difference in the distribution of gender was marginal (P = 0.098), and that of age was statistically significant between the cases and controls (P = 0.006). Thus, the cases and controls were incompletely matched in this study population. The genotypes of the NQO1, GSTT1, GSTM1, CYP1A1, and OGG1 polymorphisms were determined for the controls as shown in Fig. 1, and genotype distributions of these polymorphisms are summarized in Table 2. We first compared the genotype distribution of each polymorphism between this study and that of the previous study for the Japanese population (9, 10, 17, 21, 30). The genotype distributions of the NQO1, GSTT1, GSTM1, CYP1A1, and OGG1 polymorphisms in the control subjects of this study were not significantly different from those of previous case-control studies for the Japanese population (P > 0.05). We additionally tested whether the genotype distributions of these polymorphisms deviated from HWE. Deviations from HWE were not tested for the distributions of the GSTM1 and GSTT1 genotypes, because the genotyping method used in this study did not enable us to discriminate the heterozygote for the positive allele from the homozygote for the positive allele for the GSTM1 and GSTT1 polymorphisms. The genotype distributions of the NQO1, CYP1A1, and OGG1 polymorphisms in the controls were consistent with HWE (P = 0.610, 0.799, and 0.190, respectively). The genotypes of these polymorphisms were determined for the cases, then the relative risks of the possibly high-risk genotypes in these polymorphisms were obtained as ORs adjusted for gender, age, and smoking habit to consider effects of these factors on the risks (Table 2).

The risks of the NQO1 genotype carrying the Pro allele, NQO1-Pro/Ser and -Pro/Pro, were 1.49 and 2.15, respectively. Thus, the risk of the NQO1-Pro/Pro genotype was higher than that of the NQO1-Pro/Ser genotype. Furthermore, the increased risk of the NQO1-Pro/Pro genotype was statistically significant, whereas that of the NQO1-Pro/Ser genotype was not. Therefore, it was indicated that the NQO1-Pro/Pro genotype was associated with lung adenocarcinoma risk.

The risk of the GSTT1-null genotype was 1.61 with statistical significance. Therefore, it was also indicated that the GSTT1-null genotype was associated with lung adenocarcinoma risk.

On the other hand, the risk of the GSTM1-null genotype was 1.49 with marginal significance, indicating that the association of the GSTM1-null genotype with the risk was only marginal.

The risks of the CYP1A1-Ile/Val and -Val/Val genotypes were 1.22 and 1.59, respectively. Thus, the risk of the CYP1A1-Val/Val genotype was higher than that of the CYP1A1-Ile/Val genotype. However, the increased risks of these CYP1A1 genotypes were not statistically significant. Therefore, it was indicated that the association of the CYP1A1 polymorphism with lung adenocarcinoma risk was weaker compared with that of the NQO1, GSTT1, or GSTM1 polymorphism.

The risk of the OGG1-Ser/Cys and -Cys/Cys genotypes were 1.33 and 0.90, respectively. Thus, the risk of the OGG1-Cys/Cys genotype was lower than that of the OGG1-Ser/Cys genotype. Furthermore, the risks of these OGG1 genotypes were not statistically significant. Therefore, we concluded that the OGG1 polymorphism was not associated with lung adenocarcinoma risk.

Genotype Distributions of the NQO1, GSTT1, GSTM1, CYP1A1, and OGG1 Polymorphisms in Smokers and Nonsmokers.

To assess the relationship between smoking and each polymorphism in the contribution to lung adenocarcinoma risk, genotype distributions of these polymorphisms were re-examined after dividing the study population into smokers and nonsmokers. Then, the relative risks of the possibly high-risk genotypes were obtained as ORs adjusted for gender and age (Table 3).

In smokers, the risks of the NQO1-Pro/Ser and -Pro/Pro genotypes were 1.57 and 2.25, respectively. The increased risk of the NQO1-Pro/Pro genotype was statistically marginal and that of the NQO1-Pro/Ser genotype was not statistically significant in smokers. In nonsmokers, the risks of the NQO1-Pro/Ser and -Pro/Pro genotypes were 2.56 and 2.97, respectively, but were not significantly increased. Therefore, the association of the NQO1 polymorphism with lung adenocarcinoma risk appeared to be weak both in smokers and in nonsmokers.

The risk of the GSTT1-null genotype in smokers was 1.66, and was statistically marginal. In nonsmokers, the risk of the GSTT1-null genotype was 1.81, and was higher than that in smokers. However, the risk in nonsmokers was not significantly increased, probably due to the small number of nonsmokers analyzed. Therefore, the association of the GSTT1 polymorphism with lung adenocarcinoma risk appeared to be weak both in smokers and in nonsmokers.

The risk of the GSTM1-null genotype in smokers was 1.05, and was not statistically significant. In contrast, the risk of the GSTM1-null genotype in nonsmokers was 3.32 with statistical significance. Thus, it is likely that the association of the GSTM1-null genotype with lung adenocarcinoma risk is greater in nonsmokers than in smokers. However, the distribution of the GSTM1 genotype in the nonsmoker controls of this study was significantly different from that of a previous case-control study for the Japanese population (P = 0.017; Ref. 9). Therefore, it is also likely that the risk of the GSTM1-null genotype in nonsmokers was enhanced by a biased allele distribution in the nonsmoker controls.

In smokers, the risks of the CYP1A1-Ile/Val and -Val/Val genotypes were 0.96 and 2.01, respectively. Thus, the risk of the CYP1A1-Val/Val genotype was higher than that of the CYP1A1-Ile/Val genotype. However, the risks of these CYP1A1 genotypes were not statistically significant. Therefore, the association of the CYP1A1 polymorphism with lung adenocarcinoma risk appears to be weak in smokers. In nonsmokers, the risks of the CYP1A1-Ile/Val and -Val/Val genotypes were 2.06 and 1.50, respectively. Thus, the risk of the CYP1A1-Val/Val genotype was lower than that of the CYP1A1-Ile/Val genotype. The increased risk of the CYP1A1-Val/Val genotype was not statistically significant, whereas that of the CYP1A1-Ile/Val genotype was marginal. The marginal association of the CYP1A1-Ile/Val genotype in nonsmokers is likely to be because of the few nonsmoker controls or a biased allele distribution of the CYP1A1 polymorphism in the controls, although the genotype distribution did not significantly deviate from HWE (P = 0.178). Therefore, the association of the CYP1A1 polymorphism with the risk was not evident in nonsmokers.

The risks of the OGG1-Ser/Cys and -Cys/Cys genotypes in smokers were 1.11 and 1.15, respectively. Thus, the risk of the OGG1-Ser/Cys genotype was almost the same as that of the OGG1-Cys/Cys genotype. The increased risks of these OGG1 genotypes in smokers were not statistically significant. In nonsmokers, the risks of the OGG1-Ser/Cys and -Cys/Cys genotypes were 1.95 and 0.52, respectively. Thus, the risk of the OGG1-Cys/Cys genotype was lower than that of the OGG1-Ser/Cys genotype in nonsmokers, and the risks of these OGG1 genotypes were not statistically significant. Therefore, it was indicated that the OGG1 polymorphism was not associated with lung adenocarcinoma risk in either smokers or nonsmokers.

Enhanced Effect of the Combined Genotypes for NQO1 and GSTT1 on Lung Adenocarcinoma Risk.

The NQO1-Pro/Pro and GSTT1-null genotypes showed significant association with lung adenocarcinoma risk. Therefore, we additionally examined the relative risk of each combined genotype as ORs adjusted for gender, age, and smoking habit to assess the effect of the combined genotypes for NQO1 and GSTT1 on the risk (Table 4). When using the NQO1-Ser/Ser plus GSTT1-positive genotype as a reference, the risks of all of the other combined genotypes were increased. Especially, the risk of the NQO1-Pro/Pro plus GSTT1-null genotype was highest with OR of 4.61, and was statistically significant. The risk of the NQO1-Pro/Pro plus GSTT1-null genotype versus all of the other combined genotypes was also significantly increased with OR of 2.39. Thus, the combination of the NQO1-Pro/Pro genotype with the GSTT1-null genotype enhanced the effect of each genotype on lung adenocarcinoma risk.

We additionally assessed the relationship between smoking and each combined genotype in the contribution to lung adenocarcinoma risk. Distribution of the combined genotypes for NQO1 and GSTT1 was re-examined after dividing the study population into smokers and nonsmokers, and the relative risks for these genotypes were obtained as ORs adjusted for gender and age (Table 5). In smokers and nonsmokers, the risks of the NQO1-Pro/Pro plus GSTT1-null genotype versus the NQO1-Ser/Ser plus GSTT1-positive genotype were increased with ORs of 6.77 and 3.63, respectively. The increased risk in smokers was statistically significant, whereas that in nonsmokers was not. The risks for the NQO1-Pro/Pro plus GSTT1-null genotype versus all of the other combined genotypes in smokers and nonsmokers were also increased with ORs of 2.89 and 1.61, respectively. The increased risk in smokers was also statistically significant, whereas that in nonsmokers was not. Thus, the enhanced effect of the combined genotype, NQO1-Pro/Pro plus GSTT1-null, on lung adenocarcinoma risk was more evident in smokers than in nonsmokers.

In this study, polymorphisms in five genes involved in the metabolism of carcinogens or in the repair of damaged DNA in lung cells, NQO1-Pro187Ser, GSTT1-positive/null, GSTM1-positive/null, CYP1A1-Ile462Val, and OGG1-Ser326Cys, were examined for association with lung adenocarcinoma risk in a case-control study of 198 patients and 152 control subjects. Two of them, NQO1-Pro187Ser and GSTT1-positive/null, were associated with lung adenocarcinoma risk, whereas the other three polymorphisms were not. The NQO1-Pro/Pro genotype was associated with the increased risk, with OR of 2.15 when the NQO1-Ser/Ser genotype was used as a reference. The result of the NQO1 analysis is consistent with that of a previous small study with Taiwanese patients (22), and it further supports the idea that the NQO1-Pro/Pro genotype is a risk factor for the development of lung adenocarcinoma. A similar trend was also observed in the analyses of overall lung cancers of Japanese Hawaiians and Mexican-Americans in the United States (20, 21). The GSTT1-null genotype has been indicated as being a risk factor for the development of overall lung cancer (23, 24). However, its contribution to each histological subtype of lung cancer has not been examined. In the present study, we showed that the GSTT1-null genotype is associated with an increased risk for lung adenocarcinoma with OR of 1.61 when the GSTT1-positive genotype was used as a reference. Furthermore, the association was enhanced with OR of 4.61 when the NQO1-Pro/Pro and GSTT1-null genotypes were combined, and when the NQO1-Ser/Ser genotype combined with the GSTT1-positive genotype was used as a reference. These results strongly indicate that two genetic polymorphisms, NQO1-Pro187Ser and GSTT1-positive/null, are critical genetic factors involved in susceptibility to lung adenocarcinoma.

NQO1 is an oxidoreductase that has dual functions of both activating and detoxifying carcinogens. Therefore, whether the NQO1-Pro/Pro genotype, related to the high enzymatic activity, is at a higher or lower risk depends on the types of cancer. In this study, NQO1-Pro/Pro was associated with an increased risk for lung adenocarcinoma. Therefore, it was suggested that carcinogens activated by NQO1 have a role in lung adenocarcinoma development. Nitrosamines, such as 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), and heterocyclic amines, which are present in tobacco smoke and foods, are candidate carcinogens, because they are activated by NQO1 and are known to induce lung adenocarcinoma in rodents (18, 31, 32, 33). The NQO1 allele associated with high reductase activity has been associated with lung cancer risk in three studies including Mexican-Americans in the United States, Japanese Hawaiians, and Taiwanese, as described above (20, 21, 22). The present study adds to this trend in associating high reductase activity with risk for adenocarcinoma among Japanese. However, contrary to these observations, the high activity allele appears protective in Caucasian populations (34). Studies of ethnic minorities in the United States and several populations in Hawaii indicated that ethnicity rather than histology was a significant factor for the risk of NQO1 genotypes to lung cancer (20, 21). In addition, a recent study of Caucasians indicated that the association of NQO1 genotypes with lung cancer risk is influenced by smoking behavior rather than histology (25). Thus, unmeasured ethnic factors, including genetic and environmental ones, that influence the nature of tobacco carcinogen exposure, may be related to the effect of NQO1 genotypes on lung cancer risk. GSTT1 is known to detoxify several environmental carcinogens, including those in tobacco smoke, but not to detoxify PAHs efficiently (19). In addition, GSTT1 also functions as an activating enzyme for several carcinogens by metabolizing them to mutagenic intermediates (19). The association of the GSTT1-null genotype with the adenocarcinoma risk in this study indicates that carcinogens detoxified by GSTT1, probably those other than PAHs, have a role in lung adenocarcinoma development.

The combination of the NQO1-Pro/Pro genotype with the GSTT1-null genotype showed an enhanced effect on lung adenocarcinoma risk. This result may imply that NQO1 and GSTT1 function in overlapping pathways of metabolism for environmental carcinogens, as in the case of CYP1A1 and GSTM1 (5, 6, 7, 8, 9, 12, 13). Interestingly, the enhanced risk of the NQO1-Pro/Pro genotype combined with the GSTT1-null genotype was more evident in smokers than in nonsmokers. Therefore, it is possible that tobacco carcinogens activated by NQO1 and detoxified by GSTT1 play a major role in the development of lung adenocarcinoma. In tobacco smoke, 3800 compounds have been identified, and >100 of them have been shown to be carcinogenic or mutagenic (33, 35). Therefore, it is possible that some carcinogenic compounds are sequentially metabolized by these enzymes, although, to our knowledge, such compounds have not been identified in tobacco smoke. It was indicated recently by a large case-control study that the risk of NQO1 genotypes for lung cancer is largely influenced by individual smoking behavior (25). Therefore, it would be critical to compare the risks of NQO1 and GSTT1 genotypes among smokers based on smoking behavior to elucidate the significance of the NQO1 and GSTT1 polymorphisms on the adenocarcinoma risk in smokers. However, we did not undertake such analyses because the number of smokers both in the cases and controls in this study was not enough to be analyzed after dividing them according to smoking behavior. Those studies are now in progress in our laboratory by collecting a larger number of cases and controls. The enhanced risk of the NQO1-Pro/Pro genotype with the GSTT1-null genotype was observed in nonsmokers as well as smokers, suggesting the presence of carcinogens other than those in tobacco smoke that are sequentially metabolized by NQO1 and GSTT1. However, their association with lung adenocarcinoma risk was not statistically significant in nonsmokers, probably because the number of nonsmokers examined was smaller than that of smokers in this study population. The contribution of this combined genotype to the risk in nonsmokers should be further investigated in a larger number of subjects.

In contrast to the results of NQO1 and GSTT1, possible risk genotypes CYP1A1-Val/Val, OGG1-Cys/Cys, and GSTM1-null, which have been implicated in the risk for squamous cell lung carcinoma, did not show statistically significant association with lung adenocarcinoma risk. These results are well consistent with previous results that the polymorphisms of CYP1A1 and GSTM1 were not or only marginally associated with lung adenocarcinoma risk, and that the polymorphism of OGG1 was not associated with it (5, 6, 7, 8, 9, 10, 11, 12, 13, 15). These results indicate that the genetic factors responsible for susceptibility to the development of adenocarcinoma are different from those of squamous cell carcinoma. The difference in the effect of these three genetic polymorphisms on the risk among the histological subtypes may be because of the difference in the contribution of carcinogens, which are metabolized by the enzymes encoded by the polymorphic genes, to cancer development among the subtypes. PAHs in tobacco smoke are thought to play a role in the development of squamous cell lung carcinoma (1, 14). CYP1A1 plays a major role in the activation of PAHs, and GSTM1 plays a major role in the detoxification of activated PAH-intermediates (7, 8). Therefore, among individuals exposed to tobacco carcinogens including PAHs, the CYP1A1 and GSTM1 polymorphisms may preferentially modulate the risk for the development of squamous cell lung carcinoma. Several tobacco carcinogens including PAHs have been shown to induce intracellular oh8G (36). The OGG1-Cys/Cys genotype has been shown to be associated with the risk for squamous cell lung carcinoma and another tobacco-related cancer, squamous cell carcinoma of the esophagus (15, 37). In contrast, the association with the risk was not evident in adenocarcinomas of the lung and stomach, which are not strongly related to tobacco smoking (15, 38). Thus, the tobacco carcinogens that lead to the induction of oh8G could contribute to the development of defined types of cancers, and the risk for these cancers could be modulated by the OGG1 polymorphism.

In terms of histological appearance and clinical features, lung adenocarcinoma is known to show a greater diversity than other histological subtypes of lung cancer (26). The fact that no etiological factors have been identified inhibited the elucidation of molecular mechanisms underlying the susceptibility to this disease. In this study, genetic polymorphisms of NQO1 and GSTT1 were identified as candidates for the genetic factors underlying the susceptibility. Additional studies on their associations with the risk of each histological subtype of adenocarcinoma, such as bronchioloalveolar type, would provide us with valuable information for the elucidation of the possible etiological heterogeneity of lung adenocarcinoma. This study included a large number of individuals with respiratory and other diseases associated with smoking in the control subjects; therefore, the majority of the subjects were smokers. In addition, a considerable number of nonsmokers were present in the adenocarcinoma cases. Thus, the smoking habit was almost matched between the cases and controls in this study (Table 1). However, the use of these control subjects could lead to underestimation or overestimation of the risks of NQO1 and GSTT1 genotypes for lung adenocarcinoma if these genotypes were related to the risks for diseases observed in the control subjects. Thus, associations of these polymorphisms should be confirmed in different populations of adenocarcinoma cases as well as controls. Both the NQO1 and GSTT1 genes are expressed in epithelial cells in the peripherally located bronchioles and alveoli where adenocarcinoma preferentially develops (19, 39). Therefore, it is possible that interindividual difference in the activities of NQO1 and GSTT1 enzymes in lung cells because of the polymorphisms leads to the difference in the quantity of some carcinogens and/or DNA adducts in lung cells. Identification of environmental carcinogens, which are activated by NQO1 and detoxified by GSTT1, will be of great help to elucidate the etiology of lung adenocarcinoma and to develop effective methods for its prevention.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Supported in part by Grants-in-Aid from the Ministry of Health, Labor and Welfare for the Second Term Comprehensive 10-Year Strategy for Cancer Control, from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and from the Program for Promotion of Fundamental Studies in Health Sciences of the Organization for Pharmaceutical Safety and Research of Japan. N. S. and N. Y. are recipients of a Research Resident Fellowship from the Foundation for Promotion of Cancer Research.

3

The abbreviations used are: PAH, polycyclic aromatic hydrocarbon; SNP, single nucleotide polymorphism; oh8G, 8-hydroxyguanine; OR, odds ratio; CI, confidence interval; HWE, Hardy-Weinberg equilibrium; NQO1, NAD(P)H:quinone oxidoreductase; GSTT1, glutathione S-transferase T1.

4

Internet address: http://wbiomed.curtin.edu.au/genepop/genepop_op1.html.

Fig. 1.

Genotyping of the CYP1A1, NQO1, OGG1, GSTM1 and GSTT1 polymorphisms. A, CYP1A1. Lane 1, undigested pcDNA 3.1 fragments (665 bp); Lane 2, digested pcDNA 3.1 fragments (333 bp + 332 bp); Lane 3, 139-bp fragments for the CYP1A1-Ile/Ile genotype; Lane 4, fragments of 139 bp, 120 bp, and 48 bp for the CYP1A1-Ile/Val genotype; Lane 5, fragments of 120 bp and 48 bp for the CYP1A1-Val/Val genotype. In Lanes 3–5, digested pcDNA 3.1 fragments were also visualized. B, NQO1. Lane 1, undigested pcDNA 3.1 fragments (868 bp); Lane 2, digested pcDNA 3.1 fragments (441 bp + 427 bp); Lane 3, 195-bp fragments for the NQO1-Pro/Pro genotype; Lane 4, fragments of 195 bp and 151 bp for the NQO1-Pro/Ser genotype; Lane 5, fragments of 151 bp for the NQO1-Ser/Ser genotype. In Lanes 3–5, digested pcDNA 3.1 fragments were also visualized. C, OGG1. Lane 1, undigested pcDNA 3.1 fragments (620 bp); Lane 2, digested pcDNA 3.1 fragments (305 bp + 315 bp); Lane 3, 200-bp fragments for the OGG1-Ser/Ser genotype; Lane 4, fragments of 200 bp and 100 bp for the OGG1-Ser/Cys genotype; Lane 5, fragments of 100 bp for the OGG1-Cys/Cys genotype. In Lanes 3–5, digested pcDNA 3.1 fragments were also visualized. D, GSTM1. Lanes 1, 3 and 4, 350-bp fragments of the albumin gene and 219-bp fragments for the GSTM1-positive genotype, Lanes 2 and 5, 350-bp fragments of the albumin gene and the absence of 219-bp fragments for the GSTM1-null genotype. E, GSTT1. Lanes 1, 4, and 5, 350-bp fragments of the albumin gene and 459-bp fragments for the GSTT1-positive genotype; Lanes 2 and 3, 350-bp fragments of the albumin gene and the absence of 459-bp fragments for the GSTT1-null genotype.

Fig. 1.

Genotyping of the CYP1A1, NQO1, OGG1, GSTM1 and GSTT1 polymorphisms. A, CYP1A1. Lane 1, undigested pcDNA 3.1 fragments (665 bp); Lane 2, digested pcDNA 3.1 fragments (333 bp + 332 bp); Lane 3, 139-bp fragments for the CYP1A1-Ile/Ile genotype; Lane 4, fragments of 139 bp, 120 bp, and 48 bp for the CYP1A1-Ile/Val genotype; Lane 5, fragments of 120 bp and 48 bp for the CYP1A1-Val/Val genotype. In Lanes 3–5, digested pcDNA 3.1 fragments were also visualized. B, NQO1. Lane 1, undigested pcDNA 3.1 fragments (868 bp); Lane 2, digested pcDNA 3.1 fragments (441 bp + 427 bp); Lane 3, 195-bp fragments for the NQO1-Pro/Pro genotype; Lane 4, fragments of 195 bp and 151 bp for the NQO1-Pro/Ser genotype; Lane 5, fragments of 151 bp for the NQO1-Ser/Ser genotype. In Lanes 3–5, digested pcDNA 3.1 fragments were also visualized. C, OGG1. Lane 1, undigested pcDNA 3.1 fragments (620 bp); Lane 2, digested pcDNA 3.1 fragments (305 bp + 315 bp); Lane 3, 200-bp fragments for the OGG1-Ser/Ser genotype; Lane 4, fragments of 200 bp and 100 bp for the OGG1-Ser/Cys genotype; Lane 5, fragments of 100 bp for the OGG1-Cys/Cys genotype. In Lanes 3–5, digested pcDNA 3.1 fragments were also visualized. D, GSTM1. Lanes 1, 3 and 4, 350-bp fragments of the albumin gene and 219-bp fragments for the GSTM1-positive genotype, Lanes 2 and 5, 350-bp fragments of the albumin gene and the absence of 219-bp fragments for the GSTM1-null genotype. E, GSTT1. Lanes 1, 4, and 5, 350-bp fragments of the albumin gene and 459-bp fragments for the GSTT1-positive genotype; Lanes 2 and 3, 350-bp fragments of the albumin gene and the absence of 459-bp fragments for the GSTT1-null genotype.

Close modal
Table 1

Distributions of gender, age, and smoking habit among cases with lung adenocarcinoma and controls

CasesControlsP
No.(%)No.(%)
No. of subjects  198  152   
Gender       
 Male  124 (62.6) 108 (71.1) 0.098 
 Female  74 (37.4) 44 (28.9)  
Age       
 Mean± SD  63± 10  65 ± 13  0.006 
 <50  18 (9.1) 22 (14.5)  
 50–69  124 (62.6) 69 (45.4)  
 ≥70  56 (28.3) 61 (40.1)  
Smoking habit       
 Total       
  Nonsmokers  75 (37.9) 49 (32.2) 0.455 
  Smokers <800 65 (32.8) 47 (30.9)  
  (Cigarette-years) ≥800 55 (27.8) 50 (32.9)  
  Unknown  3 (1.5) 6 (4.0)  
 Male       
  Nonsmokers  15 (12.1) 11 (10.2) 0.463 
  Smokers <800 53 (42.7) 43 (39.8)  
  (Cigarette-years) ≥800 53 (42.7) 50 (46.3)  
  Unknown  3 (2.4) 4 (3.7)  
 Female       
  Nonsmokers  60 (81.1) 38 (86.4) 0.319 
  Smokers <800 12 (16.2) 4 (9.1)  
  (Cigarette-years) ≥800 2 (2.7) 0 (0)  
  Unknown  0 (0) 2 (4.5)  
CasesControlsP
No.(%)No.(%)
No. of subjects  198  152   
Gender       
 Male  124 (62.6) 108 (71.1) 0.098 
 Female  74 (37.4) 44 (28.9)  
Age       
 Mean± SD  63± 10  65 ± 13  0.006 
 <50  18 (9.1) 22 (14.5)  
 50–69  124 (62.6) 69 (45.4)  
 ≥70  56 (28.3) 61 (40.1)  
Smoking habit       
 Total       
  Nonsmokers  75 (37.9) 49 (32.2) 0.455 
  Smokers <800 65 (32.8) 47 (30.9)  
  (Cigarette-years) ≥800 55 (27.8) 50 (32.9)  
  Unknown  3 (1.5) 6 (4.0)  
 Male       
  Nonsmokers  15 (12.1) 11 (10.2) 0.463 
  Smokers <800 53 (42.7) 43 (39.8)  
  (Cigarette-years) ≥800 53 (42.7) 50 (46.3)  
  Unknown  3 (2.4) 4 (3.7)  
 Female       
  Nonsmokers  60 (81.1) 38 (86.4) 0.319 
  Smokers <800 12 (16.2) 4 (9.1)  
  (Cigarette-years) ≥800 2 (2.7) 0 (0)  
  Unknown  0 (0) 2 (4.5)  
Table 2

Distributions of the NQO1, GSTT1, GSTM1, CYP1A1, and OGG1 polymorphisms

GeneGenotypeNo. of cases (%)/controls (%)ORa95% CIP
NQO1 Ser/Ser 22 (11.1)/23 (15.1) 1.00   
 Pro/Ser 93 (47.0)/77 (50.7) 1.49 0.74–2.98 0.263 
 Pro/Pro 83 (41.9)/52 (34.2) 2.15 1.03–4.48 0.042 
GSTT1 Positive 99 (50.0)/93 (61.2) 1.00   
 Null 99 (50.0)/59 (38.8) 1.61 1.04–2.52 0.035 
GSTM1 Positive 105 (53.0)/96 (63.2) 1.00   
 Null 93 (47.0)/56 (36.8) 1.49 0.95–2.33 0.084 
CYP1A1 Ile/Ile 115 (58.1)/99 (65.1) 1.00   
 Ile/Val 70 (35.3)/47 (30.9) 1.22 0.76–1.96 0.419 
 Val/Val 13 (6.6)/6 (4.0) 1.59 0.57–4.48 0.378 
OGG1 Ser/Ser 54 (27.3)/50 (32.9) 1.00   
 Ser/Cys 106 (53.5)/66 (43.4) 1.33 0.80–2.23 0.274 
 Cys/Cys 38 (19.2)/36 (23.7) 0.90 0.48–1.70 0.753 
GeneGenotypeNo. of cases (%)/controls (%)ORa95% CIP
NQO1 Ser/Ser 22 (11.1)/23 (15.1) 1.00   
 Pro/Ser 93 (47.0)/77 (50.7) 1.49 0.74–2.98 0.263 
 Pro/Pro 83 (41.9)/52 (34.2) 2.15 1.03–4.48 0.042 
GSTT1 Positive 99 (50.0)/93 (61.2) 1.00   
 Null 99 (50.0)/59 (38.8) 1.61 1.04–2.52 0.035 
GSTM1 Positive 105 (53.0)/96 (63.2) 1.00   
 Null 93 (47.0)/56 (36.8) 1.49 0.95–2.33 0.084 
CYP1A1 Ile/Ile 115 (58.1)/99 (65.1) 1.00   
 Ile/Val 70 (35.3)/47 (30.9) 1.22 0.76–1.96 0.419 
 Val/Val 13 (6.6)/6 (4.0) 1.59 0.57–4.48 0.378 
OGG1 Ser/Ser 54 (27.3)/50 (32.9) 1.00   
 Ser/Cys 106 (53.5)/66 (43.4) 1.33 0.80–2.23 0.274 
 Cys/Cys 38 (19.2)/36 (23.7) 0.90 0.48–1.70 0.753 
a

Adjusted for gender, age and smoking habit.

Table 3

Distributions of the NQO1, GSTT1, GSTM1, CYP1A1, and OGG1 polymorphisms in smokers and nonsmokers

PopulationGeneGenotypeNo. of cases (%)/controls (%)ORa95% CIP
Smokers NQO1 Ser/Ser 13 (10.8)/16 (16.5) 1.00   
  Pro/Ser 57 (47.5)/50 (51.5) 1.57 0.66–3.73 0.308 
  Pro/Pro 50 (41.7)/31 (32.0) 2.25 0.91–5.56 0.079 
 GSTT1 Positive 60 (50.0)/60 (61.9) 1.00   
  Null 60 (50.0)/37 (38.1) 1.66 0.96–2.89 0.072 
 GSTM1 Positive 69 (57.5)/57 (58.8) 1.00   
  Null 51 (42.5)/40 (41.2) 1.05 0.61–1.83 0.851 
 CYP1A1 Ile/Ile 73 (60.8)/61 (62.9) 1.00   
  Ile/Val 40 (33.3)/33 (34.0) 0.96 0.54–1.73 0.898 
  Val/Val 7 (5.8)/3 (3.1) 2.01 0.49–8.25 0.330 
 OGG1 Ser/Ser 36 (30.0)/31 (35.1) 1.00   
  Ser/Cys 58 (48.3)/45 (43.3) 1.11 0.60–2.08 0.698 
  Cys/Cys 26 (21.7)/21 (21.6) 1.15 0.53–2.50 0.724 
Nonsmokers NQO1 Ser/Ser 9 (12.0)/7 (14.3) 1.00   
  Pro/Ser 34 (45.3)/25 (51.0) 2.56 0.67–9.75 0.169 
  Pro/Pro 32 (42.7)/17 (34.7) 2.97 0.75–11.78 0.121 
 GSTT1 Positive 36 (48.0)/29 (59.2) 1.00   
  Null 39 (52.0)/20 (40.8) 1.81 0.80–4.11 0.156 
 GSTM1 Positive 35 (46.7)/36 (73.5) 1.00   
  Null 40 (53.3)/13 (26.5) 3.32 1.41–7.84 0.006 
 CYP1A1 Ile/Ile 40 (53.3)/34 (69.4) 1.00   
  Ile/Val 29 (38.7)/12 (24.5) 2.06 0.85–4.99 0.108 
  Val/Val 6 (8.0)/3 (6.1) 1.50 0.30–7.48 0.622 
 OGG1 Ser/Ser 18 (24.0)/16 (32.6) 1.00   
  Ser/Cys 47 (62.7)/19 (38.8) 1.95 0.74–5.17 0.180 
  Cys/Cys 10 (13.3)/14 (28.6) 0.52 0.15–1.78 0.300 
PopulationGeneGenotypeNo. of cases (%)/controls (%)ORa95% CIP
Smokers NQO1 Ser/Ser 13 (10.8)/16 (16.5) 1.00   
  Pro/Ser 57 (47.5)/50 (51.5) 1.57 0.66–3.73 0.308 
  Pro/Pro 50 (41.7)/31 (32.0) 2.25 0.91–5.56 0.079 
 GSTT1 Positive 60 (50.0)/60 (61.9) 1.00   
  Null 60 (50.0)/37 (38.1) 1.66 0.96–2.89 0.072 
 GSTM1 Positive 69 (57.5)/57 (58.8) 1.00   
  Null 51 (42.5)/40 (41.2) 1.05 0.61–1.83 0.851 
 CYP1A1 Ile/Ile 73 (60.8)/61 (62.9) 1.00   
  Ile/Val 40 (33.3)/33 (34.0) 0.96 0.54–1.73 0.898 
  Val/Val 7 (5.8)/3 (3.1) 2.01 0.49–8.25 0.330 
 OGG1 Ser/Ser 36 (30.0)/31 (35.1) 1.00   
  Ser/Cys 58 (48.3)/45 (43.3) 1.11 0.60–2.08 0.698 
  Cys/Cys 26 (21.7)/21 (21.6) 1.15 0.53–2.50 0.724 
Nonsmokers NQO1 Ser/Ser 9 (12.0)/7 (14.3) 1.00   
  Pro/Ser 34 (45.3)/25 (51.0) 2.56 0.67–9.75 0.169 
  Pro/Pro 32 (42.7)/17 (34.7) 2.97 0.75–11.78 0.121 
 GSTT1 Positive 36 (48.0)/29 (59.2) 1.00   
  Null 39 (52.0)/20 (40.8) 1.81 0.80–4.11 0.156 
 GSTM1 Positive 35 (46.7)/36 (73.5) 1.00   
  Null 40 (53.3)/13 (26.5) 3.32 1.41–7.84 0.006 
 CYP1A1 Ile/Ile 40 (53.3)/34 (69.4) 1.00   
  Ile/Val 29 (38.7)/12 (24.5) 2.06 0.85–4.99 0.108 
  Val/Val 6 (8.0)/3 (6.1) 1.50 0.30–7.48 0.622 
 OGG1 Ser/Ser 18 (24.0)/16 (32.6) 1.00   
  Ser/Cys 47 (62.7)/19 (38.8) 1.95 0.74–5.17 0.180 
  Cys/Cys 10 (13.3)/14 (28.6) 0.52 0.15–1.78 0.300 
a

Adjusted for gender and age.

Table 4

Enhanced risk of the NQO1-Pro/Pro genotype by combination with the GSTT1-null genotype

Combined genotypeNo. of cases (%)/controls (%)ORa95% CIP
NQO1-Ser/Ser+ GSTT1-positive 11 (5.6)/15 (9.9) 1.00   
NQO1-Pro/Ser+ GSTT1-positive 46 (23.2)/43 (28.3) 1.65 0.66–4.16 0.286 
NQO1-Pro/Pro+ GSTT1-positive 42 (21.2)/35 (23.0) 1.65 0.64–4.29 0.303 
NQO1-Ser/Ser+ GSTT1-null 11 (5.6)/8 (5.3) 1.54 0.44–5.47 0.502 
NQO1-Pro/Ser+ GSTT1-null 47 (23.7)/34 (22.4) 2.21 0.84–5.77 0.107 
NQO1-Pro/Pro+ GSTT1-nullb 41 (20.7)/17 (11.2) 4.61 1.59–13.34 0.005 
Combined genotypeNo. of cases (%)/controls (%)ORa95% CIP
NQO1-Ser/Ser+ GSTT1-positive 11 (5.6)/15 (9.9) 1.00   
NQO1-Pro/Ser+ GSTT1-positive 46 (23.2)/43 (28.3) 1.65 0.66–4.16 0.286 
NQO1-Pro/Pro+ GSTT1-positive 42 (21.2)/35 (23.0) 1.65 0.64–4.29 0.303 
NQO1-Ser/Ser+ GSTT1-null 11 (5.6)/8 (5.3) 1.54 0.44–5.47 0.502 
NQO1-Pro/Ser+ GSTT1-null 47 (23.7)/34 (22.4) 2.21 0.84–5.77 0.107 
NQO1-Pro/Pro+ GSTT1-nullb 41 (20.7)/17 (11.2) 4.61 1.59–13.34 0.005 
a

Adjusted for gender, age and smoking habit.

b

NQO1-Pro/Pro + GSTT1-null versus all the other combined genotypes: adjusted OR, 2.39; 95% CI, 1.25–4.57; P = 0.009.

Table 5

Risk of the NQO1-Pro/Pro plus GSTT1-null genotype in smokers and nonsmokers

Genotypes comparedPopulationNo. of cases/controlsORa95% CIP
NQO1-Pro/Pro+ GSTT1-null vs. Smoker 27/9 vs. 7/11 6.77 1.65–27.73 0.008 
 NQO1-Ser/Ser+ GSTT1-positive Nonsmoker 14/6 vs. 4/4 3.63 0.57–23.24 0.174 
NQO1-Pro/Pro+ GSTT1-null vs. Smoker 27/9 vs. 93/88 2.89 1.28–6.52 0.011 
 all the other combined genotypes Nonsmoker 14/6 vs. 61/43 1.61 0.52–5.04 0.411 
Genotypes comparedPopulationNo. of cases/controlsORa95% CIP
NQO1-Pro/Pro+ GSTT1-null vs. Smoker 27/9 vs. 7/11 6.77 1.65–27.73 0.008 
 NQO1-Ser/Ser+ GSTT1-positive Nonsmoker 14/6 vs. 4/4 3.63 0.57–23.24 0.174 
NQO1-Pro/Pro+ GSTT1-null vs. Smoker 27/9 vs. 93/88 2.89 1.28–6.52 0.011 
 all the other combined genotypes Nonsmoker 14/6 vs. 61/43 1.61 0.52–5.04 0.411 
a

Adjusted for gender and age.

We thank Ayaka Otsuka for technical assistance. We also thank Dr. Masatomo Mori of the Gunma University School of Medicine, Gunma, Japan, for encouragement throughout this study.

1
Wynder E. L., Hoffmann D. Smoking and lung cancer: scientific challenges and opportunities.
Cancer Res.
,
54
: 
5284
-5295,  
1994
.
2
Gazdar A. F., Minna J. D. Cigarettes, sex, and lung adenocarcinoma.
J. Natl. Cancer Inst.
,
89
: 
1563
-1565,  
1997
.
3
Thun M. J., Lally C. A., Flannery J. T., Calle E. E., Flanders W. D., Heath C. W., Jr. Cigarette smoking and changes in the histopathology of lung cancer.
J. Natl. Cancer Inst.
,
89
: 
1580
-1586,  
1997
.
4
Charloux A., Quoix E., Wolkove N., Small D., Pauli G., Kreisman H. The increasing incidence of lung adenocarcinoma: reality or artifact? A review of the epidemiology of lung adenocarcinoma.
Int. J. Epidemiol.
,
26
: 
14
-23,  
1997
.
5
Taningher M., Malacarne D., Izzotti A., Ugolini D., Parodi S. Drug metabolism polymorphisms as modulators of cancer susceptibility.
Mutat. Res.
,
436
: 
227
-261,  
1999
.
6
Fryer A. A., Jones P. W. Chapter 22. Interactions between detoxifying enzyme polymorphisms and susceptibility to cancer.
IARC Sci. Publ.
,
148
: 
303
-322,  
1999
.
7
Bartsch H., Nair U., Risch A., Rojas M., Wikman H., Alexandrov K. Genetic polymorphism of CYP genes, alone or in combination, as a risk modifier of tobacco-related cancers.
Cancer Epidemiol. Biomark. Prev.
,
9
: 
3
-28,  
2000
.
8
Rebbeck T. R. Molecular epidemiology of the human glutathione S-transferase genotypes GSTM1 and GSTT1 in cancer susceptibility.
Cancer Epidemiol. Biomark. Prev.
,
6
: 
733
-743,  
1997
.
9
Hayashi S., Watanabe J., Kawajiri K. High susceptibility to lung cancer analyzed in terms of combined genotypes of P450IA1 and Mu-class glutathione S-transferase genes.
Jpn. J. Cancer Res.
,
83
: 
866
-870,  
1992
.
10
Kihara M., Kihara M., Noda K. Lung cancer risk of GSTM1 null genotype is dependent on the extent of tobacco smoke exposure.
Carcinogenesis (Lond.)
,
15
: 
415
-418,  
1994
.
11
Sugimura H., Wakai K., Genka K., Nagura K., Igarashi H., Nagayama K., Ohkawa A., Baba S., Morris B. J., Tsugane S., Ohno Y., Gao C., Li Z., Takezaki T., Tajima K., Iwamasa T. Association of Ile462Val (Exon 7) polymorphism of cytochrome P450 IA1 with lung cancer in the Asian population: further evidence from a case-control study in Okinawa.
Cancer Epidemiol. Biomark. Prev.
,
7
: 
413
-417,  
1998
.
12
Le Marchand L., Sivaraman L., Pierce L., Seifried A., Lum A., Wilkens L. R., Lau A. F. Associations of CYP1A1, GSTM1, and CYP2E1 polymorphisms with lung cancer suggest cell type specificities to tobacco carcinogens.
Cancer Res.
,
58
: 
4858
-4863,  
1998
.
13
Kihara M., Kihara M., Noda K. Risk of smoking for squamous and small cell carcinomas of the lung modulated by combinations of CYP1A1 and GSTM1 gene polymorphisms in a Japanese population.
Carcinogenesis (Lond.)
,
16
: 
2331
-2336,  
1995
.
14
Deutsch-Wenzel R. P., Brune H., Grimmer G., Dettbarn G., Misfeld J. Experimental studies in rat lungs on the carcinogenicity and dose-response relationships of eight frequently occurring environmental polycyclic aromatic hydrocarbons.
J. Natl. Cancer Inst.
,
71
: 
539
-544,  
1983
.
15
Sugimura H., Kohno T., Wakai K., Nagura K., Genka K., Igarashi H., Morris B. J., Baba S., Ohno Y., Gao C., Li Z., Wang J., Takezaki T., Tajima K., Varga T., Sawaguchi T., Lum J. K., Martinson J. J., Tsugane S., Iwamasa T., Shinmura K., Yokota J. hOGG1 Ser326Cys polymorphism and lung cancer susceptibility.
Cancer Epidemiol. Biomark. Prev.
,
8
: 
669
-674,  
1999
.
16
Shinmura K., Yokota J. The OGG1 gene encodes a repair enzyme for oxidatively damaged DNA and is involved in human carcinogenesis.
Antiox. Redox Signal
,
3
: 
597
-609,  
2001
.
17
Kohno T., Shinmura K., Tosaka M., Tani M., Kim S. R., Sugimura H., Nohmi T., Kasai H., Yokota J. Genetic polymorphisms and alternative splicing of the hOGG1 gene, that is involved in the repair of 8-hydroxyguanine in damaged DNA.
Oncogene
,
16
: 
3219
-3225,  
1998
.
18
Ross D., Kepa J. K., Winski S. L., Beall H. D., Anwar A., Siegel D. NAD(P)H:quinone oxidoreductase 1 (NQO1): chemoprotection, bioactivation, gene regulation and genetic polymorphisms.
Chem. Biol. Interact.
,
129
: 
77
-97,  
2000
.
19
Landi S. Mammalian class theta GST and differential susceptibility to carcinogens: a review.
Mutat. Res.
,
463
: 
247
-283,  
2000
.
20
Wiencke J. K., Spitz M. R., McMillan A., Kelsey K. T. Lung cancer in Mexican-Americans and African-Americans is associated with the wild-type genotype of the NAD(P)H: quinone oxidoreductase polymorphism.
Cancer Epidemiol. Biomark. Prev.
,
6
: 
87
-92,  
1997
.
21
Chen H., Lum A., Seifried A., Wilkens L. R., Le Marchand L. Association of the NAD(P)H: quinone oxidoreductase 609C->T polymorphism with a decreased lung cancer risk.
Cancer Res.
,
59
: 
3045
-3048,  
1999
.
22
Lin P., Wang H. J., Lee H., Lee H. S., Wang S. L., Hsueh Y. M., Tsai K. J., Chen C. Y. NAD(P)H: quinone oxidoreductase polymorphism and lung cancer in Taiwan.
J. Toxicol. Environ. Health A
,
58
: 
187
-197,  
1999
.
23
Spitz M. R., Duphorne C. M., Detry M. A., Pillow P. C., Amos C. I., Lei L., de Andrade M., Gu X., Hong W. K., Wu X. Dietary intake of isothiocyanates: evidence of a joint effect with glutathione S-transferase polymorphisms in lung cancer risk.
Cancer Epidemiol. Biomark. Prev.
,
9
: 
1017
-1020,  
2000
.
24
Hou S. M., Falt S., Nyberg F. Glutathione S-transferase T1-null genotype interacts synergistically with heavy smoking on lung cancer risk.
Environ. Mol. Mutagen.
,
38
: 
83
-86,  
2001
.
25
Xu L. L., Wain J. C., Miller D. P., Thurston S. W., Su L., Lynch T. J., Christiani D. C. The NAD(P)H:quinone oxidoreductase 1 gene polymorphism and lung cancer: differential susceptibility based on smoking behavior.
Cancer Epidemiol. Biomark. Prev.
,
10
: 
303
-309,  
2001
.
26
Travis W. D., Colby T. V., Corrin B., Shimosato Y., Brambilla E. Histological typing of lung and pleural tumors.
WHO International Histological Classification of Tumors
, Springer Berlin  
1999
.
27
Oyama T., Mitsudomi T., Kawamoto T., Ogami A., Osaki T., Kodama Y., Yasumoto K. Detection of CYP1A1 gene polymorphism using designed RFLP and distributions of CYP1A1 genotypes in Japanese.
Int. Arch. Occup. Environ. Health
,
67
: 
253
-256,  
1995
.
28
Arand M., Muhlbauer R., Hengstler J., Jager E., Fuchs J., Winkler L., Oesch F. A. multiplex polymerase chain reaction protocol for the simultaneous analysis of the glutathione S-transferase GSTM1 and GSTT1 polymorphisms.
Anal. Biochem.
,
236
: 
184
-186,  
1996
.
29
Breslow N. E., Day N. E. Statistical methods in cancer research. Volume I. The analysis of case-control studies.
IARC Sci. Publ.
,
32
: 
5
-338,  
1980
.
30
Katoh T., Nagata N., Kuroda Y., Itoh H., Kawahara A., Kuroki N., Ookuma R., Bell D. A. Glutathione S-transferase M1 (GSTM1) and T1 (GSTT1) genetic polymorphism and susceptibility to gastric and colorectal adenocarcinoma.
Carcinogenesis (Lond.)
,
17
: 
1855
-1859,  
1996
.
31
Ohgaki H., Hasegawa H., Kato T., Suenaga M., Sato S., Takayama S., Sugimura T. Carcinogenicities in mice and rats of IQ, MeIQ, and MeIQx.
Princess Takamatsu Symp.
,
16
: 
97
-105,  
1985
.
32
Hoffmann D., Rivenson A., Hecht S. S. The biological significance of tobacco-specific N-nitrosamines: smoking and adenocarcinoma of the lung.
Crit. Rev. Toxicol.
,
26
: 
199
-211,  
1996
.
33
Hecht S. S. Tobacco smoke carcinogens and lung cancer.
J. Natl. Cancer Inst.
,
91
: 
1194
-1210,  
1999
.
34
Rosvold E. A., McGlynn K. A., Lustbader E. D., Buetow K. H. Identification of an NAD(P)H: quinone oxidoreductase polymorphism and its association with lung cancer and smoking.
Pharmacogenetics
,
5
: 
199
-206,  
1995
.
35
IARC. .
Monographs on the Evaluation of the Carcinogenic Risks of Chemicals to Humans. Vol. 38. Tobacco Smoking
, IARC Lyon, France  
1985
.
36
Kim K. B., Lee B. M. Oxidative stress to DNA, protein, and antioxidant enzymes (superoxide dismutase and catalase) in rats treated with benzo(a)pyrene.
Cancer Lett.
,
113
: 
205
-212,  
1997
.
37
Xing D. Y., Tan W., Song N., Lin D. X. Ser326Cys polymorphism in hOGG1 gene and risk of esophageal cancer in a Chinese population.
Int. J. Cancer
,
95
: 
140
-143,  
2001
.
38
Hanaoka T., Sugimurabf H., Nagura K., Ihara M., Li X., Hamada G. S., Nishimoto I., Kowalski L. P., Yokota J., Tsugane S. hOGG1 exon7 polymorphism and gastric cancer in case-control studies of Japanese Brazilians and non-Japanese Brazilians.
Cancer Lett.
,
170
: 
53
-61,  
2001
.
39
Siegel D., Franklin W. A., Ross D. Immunohistochemical detection of NAD(P)H: quinone oxidoreductase in human lung and lung tumors.
Clin. Cancer Res.
,
4
: 
2065
-2070,  
1998
.