Abstract
AIM: To investigate the role of functional genetic polymorphisms of metabolic enzymes of tobacco carcinogens in the development of colorectal adenomas.
METHODS: The study subjects were 455 patients with colorectal adenomas and 1052 controls with no polyps who underwent total colonoscopy in a preretirement health examination at two Self Defense Forces hospitals. The genetic polymorphisms studied were CYP1A1*2A (rs 4646903), CYP1A1*2C (rs 1048943), GSTM1 (null or non-null genotype), GSTT1 (null or non-null genotype) and NQO1 C609T (rs 1800566). Genotypes were determined by the polymerase chain reaction (PCR)-restriction fragment length polymorphism or PCR method using genomic DNA extracted from the buffy coat. Cigarette smoking and other lifestyle factors were ascertained by a self-administered questionnaire. The associations of the polymorphisms with colorectal adenomas were examined by means of OR and 95%CI, which were derived from logistic regression analysis. Statistical adjustment was made for smoking, alcohol use, body mass index and other factors. The gene-gene interaction and effect modification of smoking were evaluated by the likelihood ratio test.
RESULTS: None of the five polymorphisms showed a significant association with colorectal adenomas, nor was the combination of GSTM1 and GSTT1. A borderline significant interaction was observed for the combination of CYP1A1*2C and NQO1 (P = 0.051). The OR associated with CYP1A1*2C was significantly lower than unity among individuals with the NQO1 609CC genotype. The adjusted OR for the combination of the CYP1A1*2C allele and NQO1 609CC genotype was 0.61 (95%CI: 0.42-0.91). Although the interaction was not statistically significant (P = 0.24), the OR for individuals carrying the CYP1A1*2C allele and GSTT1 null genotype decreased significantly compared with those who had neither CYP1A1*2C allele nor GSTT1 null genotype (adjusted OR: 0.69, 95%CI: 0.49-0.97). Smoking did not modify the associations of the individual polymorphisms with colorectal adenomas. There was no measurable effect modification of smoking even regarding the combination of the genetic polymorphisms of the phase I and phase II enzymes.
CONCLUSION: Combination of the CYP1A1*2C and NQO1 609CC genotypes was associated with a decreased risk of colorectal adenomas regardless of smoking status.
Keywords: Colorectal adenoma, Smoking, Polymorphism, CYP1A1, GSTM1, GSTT1, NQO1
Core tip: The study investigated the associations of CYP1A1*2A, CYP1A1*2C, GSTM1, GSTT1 and NQO1 C609T polymorphisms with colorectal adenomas among 455 cases of colorectal adenomas and 1052 controls with no polyps. None of the five polymorphisms showed a significant association with colorectal adenomas, nor was the combination of GSTM1 and GSTT1. A borderline significant interaction was observed for the combination of CYP1A1*2C and NQO1. Combination of the CYP1A1*2C and NQO1 609CC genotypes was associated with a decreased risk of colorectal adenomas regardless of smoking status.
INTRODUCTION
Colorectal cancer is one of the most common cancers, accounting for approximately 10% of incident cancer cases worldwide[1]. Colorectal adenoma is a well-established precursor lesion of colorectal cancer[2,3]. Cigarette smoking has been related to increased risk of colorectal adenomas, whereas the association between smoking and colorectal cancer risk is rather inconsistent and much weaker[4-6]. Despite the consistent association between smoking and colorectal adenomas, biological mechanisms for the association remain unknown. Tobacco smoke contains various types of carcinogens such as polycyclic aromatic hydrocarbons, heterocyclic amines, aromatic amines and N-nitrosamines, which are activated by phase I enzymes and/or detoxified by phase II enzymes, and thus functional genetic polymorphisms of the metabolic enzymes are of interest in colorectal carcinogenesis[7].
CYP1A1 is a phase I enzyme responsible for bioactivation of tobacco carcinogens. Two functional polymorphisms are known in the CYP1A1 gene[8,9]. One is the 3698T>C substitution (CYP1A1*2A, rs 4646903) that creates an MspI restriction site in the 3’-flanking region, and the other is the 2454A>G substitution (CYP1A1*2C, rs 1048943) that results in an amino acid change (Ile462Val) in exon 7. The CYP1A1*2A and CYP1A1*2C alleles are linked to higher inducibility of the enzyme, and have been associated with an increased risk of lung cancer and less consistently of other tobacco-related cancers[8,9]. Several studies reported an increased risk of colorectal cancer associated with CYP1A1*2A[10] and CYP1A1*2C[11], while others showed no association of either CYP1A1*2A or CYP1A1*2C with colorectal cancer[12-15] or adenomas[16]. Isoforms of the glutathione S-transferase (GST) are involved in detoxification of chemical carcinogens and environmental toxic compounds[17,18]. GSTM1 and GSTT1 polymorphisms have been studied most intensively in relation to tobacco-related cancers. The GSTM1 and GSTT1 null genotypes result in a complete loss of enzyme function[17,18]. A meta-analysis suggested an increased risk of colorectal cancer associated with the null genotype of GSTT1, but not of GSTM1[19], while another meta-analysis showed no association of either the GSTM1 or GSTT1 null genotype with colorectal cancer or adenomas[20]. Some recent studies have shown an increased risk of colorectal cancer associated with the GSTM1 and GSTT1 null genotypes in combination[21,22], but others failed to show such an increase in the risk of colorectal cancer[14,15] or adenomas[23].
NAD(P)H-quinone oxidoreductase 1 (NQO1) is involved in detoxification through two electron reductions of quinones to hydroquinones, while NQO1 can also activate procarcinogens in tobacco smoke[24]. The 609C>T polymorphism (rs 1800566) that causes an amino acid substitution (Pro187Ser) results in loss of NQO1 activity[24]. A meta-analysis reported that NQO1 609C>T was associated with a small increase in the risk of colorectal cancer in Caucasians[25], but a recent large Japanese study failed to corroborate such an association[15]. Homozygosity of the NQO1 609T allele was shown to be positively associated with colorectal adenomas[26]. Heavy smokers carrying both the CYP1A1*2C and NQO1 609T variant alleles showed a substantial increase in the risk of adenomas[26].
To clarify the role of CYP1A1, GSTM1, GSTT1 and NQO1 polymorphisms in colorectal carcinogenesis with reference to smoking, we examined the associations of these polymorphisms with colorectal adenomas and the effect of smoking on the associations between the polymorphisms and colorectal adenomas. A particular emphasis was placed on the combination of genetic polymorphisms of phase I and phase II enzymes, because the literature is sparse on the influence of gene-gene interactions on the risk of colorectal cancer and adenomas.
MATERIALS AND METHODS
Subjects
Study subjects were male officials in the Self Defense Forces who received a preretirement health examination at the Self Defense Forces Fukuoka Hospital or Kumamoto Hospital during the period from January 1997 to March 2001. The preretirement health examination is a nationwide program that offers a comprehensive medical examination including colonoscopy to persons retiring from the Self Defense Forces. Details of the preretirement health examination have been described elsewhere[27,28]. The subjects were Japanese in ethnicity. A 7-mL fasting venous blood sample was donated for the purpose of medical research with written informed consent. The study was approved by the Ethics Committee of Kyushu University Faculty of Medical Sciences.
The present study included 455 cases of histologically confirmed colorectal adenoma and 1052 controls with no polyps who underwent total colonoscopy. In a consecutive series of 2454 men, five refused to participate in the study and we excluded 77 who did not undergo colonoscopy. Further exclusions were 242 men with a history of colectomy (n = 17), colorectal polypectomy (n = 212), malignant neoplasm (n = 27) or inflammatory bowel disease (n = 1). For the remaining 2135 men, colonoscopic findings were classified as polyp (n = 938, 43.9%), colorectal cancer (n = 1, 0.0%), non-polyp benign lesion (n = 123, 5.8%) and normal (n = 1073, 50.3%). Of the 938 men with colorectal polyps, 461 were found to have adenoma without in situ or invasive carcinoma. The controls comprised 1067 men who underwent a complete colonoscopy among the 1196 men with normal colonoscopy or non-polyp benign lesions. DNA was not available for 21 men (6 cases and 15 controls), and 455 cases and 1052 controls remained in the analysis.
Lifestyle questionnaire
Smoking habits, alcohol consumption, physical activity and other lifestyle factors were ascertained by a self-administered questionnaire, with a supplemental interview for unanswered questions given prior to colonoscopy. Details of the lifestyle questions have been described elsewhere[27,28]. Lifetime exposure to cigarette smoking was expressed by cigarette-years, which were calculated as the product of total years of smoking and the average number of cigarettes per day. Cigarette smoking was classified into 0, 1-399, 400-799 and ≥ 800 cigarette-years. Daily intake of ethanol was estimated for current alcohol drinkers based on consumption frequencies and amounts of five types of alcoholic beverages on average in the past year. Alcohol use was categorized into never, past and current use with a consumption of < 30, 30-59 or ≥ 60 mL of ethanol per day. Body mass index was categorized into four levels (< 22.5, 22.5-24.9, 25.0-27.4 and ≥ 27.5 kg/m2). The categories for alcohol use and body mass index were arbitrary, but in accordance with those used in the previous studies[27,28]. Leisure-time physical activity was expressed as the sum of products of intensity score (metabolic equivalent) and amount of time for at most three types of regular exercise, and was categorized by quartiles in the control group.
Genotyping
DNA was extracted from the buffy coat by use of a commercial kit (Qiagen GmbH, Hilden, Germany). Genotyping was carried out by the polymerase chain reaction (PCR)-restriction fragment length polymorphism or PCR method, with agarose-gel electrophoresis and visualization by ethidium bromide. The PCR was performed in a mixture of 10 μL containing 1 μL template DNA with a concentration of 50-150 ng/μL. The PCR for CYP1A1*2A polymorphism was done using the primers described by Sivaraman et al[10], and the 340-bp PCR product was digested with MspI, which resulted in fragments of 200 and 140 bp for the CYP1A1*2A allele. The CYP1A1*2C polymorphism was genotyped using the primers previously specified[29], with digestion by restriction enzyme HincII. The 187-bp product was cleaved into three fragments (120, 48 and 19 bp) in the presence of the CYP1A1*2C allele, and otherwise into two fragments (139 and 48 bp). GSTM1 and GSTT1 polymorphisms were determined by the multiplex PCR using the primers for GSTM1, GSTT1 and albumin as described previously[30]. Genotyping for NQO1 609C>T was performed as described previously[31]. The 230-bp PCR product was digested with HinfI, resulting in fragments of 195 and 35 bp for the 609C allele and fragments of 151, 44 and 35 bp for the 609T allele. The assay was repeated at most three times when the PCR was unsuccessful or when the migration pattern on the gel was aberrant.
Statistical analysis
Deviation of genotype frequency from the Hardy-Weinberg equilibrium was tested by χ2 test with one degree of freedom using the Stata version 10 (Stata Corporation, College Station, TX, United States). The associations between the polymorphisms and colorectal adenomas were assessed by means of OR and 95%CI, which were derived from logistic regression analysis. Statistical adjustment was made for age (continuous variable), hospital (dichotomous variable), rank in the Self Defense Forces (low, middle and high), cigarette smoking, alcohol use, body mass index, physical activity and parental colorectal cancer. The gene-gene interaction and effect modification of smoking were evaluated by the likelihood ratio test. In the analysis of the effect modification of smoking, smoking status was categorized into < 400 and ≥ 400 cigarette-years, i.e., < 20 and ≥ 20 pack-years, because an increased risk of adenomas associated with smoking was discernible only in the latter categories (see below). Statistical significance was declared if two-sided P was < 0.05. Statistical analysis was performed with SAS version 9.2 (SAS Institute, Cary, NC, United States).
RESULTS
Selected characteristics of the cases and controls are summarized in Table 1. The age range was 50-57 years for the cases and 47-59 years for the controls. The cases had a greater body mass index and a lower physical activity in leisure time than the controls. Heavy smoking and high alcohol consumption were more frequent in the cases than in the controls.
Table 1.
Selected characteristics of the study subjects
| Characteristics | Cases | Controls |
| (n = 455) | (n = 1052) | |
| Age (yr), mean ± SD | 52.4 ± 0.8 | 52.4 ± 0.9 |
| Body mass index (kg/m2), mean ± SD | 24.1 ± 2.8 | 23.7 ± 2.5 |
| MET-h/wk, median (IQR)1 | 12 (3-24) | 14 (5-24) |
| Smoking (cigarette-yr) | ||
| 0 | 20.90% | 33.70% |
| 1-399 | 14.10% | 18.80% |
| 400-799 | 45.50% | 33.70% |
| ≥ 800 | 19.60% | 13.70% |
| Alcohol use (mL/d) | ||
| Never | 11.20% | 13.80% |
| Past | 2.90% | 3.10% |
| < 30 | 21.10% | 30.70% |
| 30-59 | 34.10% | 28.70% |
| ≥ 60 | 30.80% | 23.70% |
Leisure-time physical activity. IQR: Interquartile range; MET: Metabolic equivalent.
Among the controls, the frequencies of the CYP1A1*2A, CYP1A1*2C and NQO1 609T alleles were 0.39, 0.23 and 0.38, respectively, and genotype frequencies of the three polymorphisms were all in agreement with the Hardy-Weinberg equilibrium (P = 0.62 for CYP1A1*2A; P = 0.32 for CYP1A1*2C; and P = 0.76 for NQO1 C609T). The CYP1A1*2A and CYP1A1*2C polymorphisms were in complete linkage disequilibrium except for two cases; the deviation of these two was probably due to error in genotyping.
None of the five polymorphisms showed a significant association with colorectal adenomas, nor was the combination of GSTM1 and GSTT1 (Table 2). The gene-gene interaction was examined for the combinations of the CYP1A1 polymorphisms and the GST or NQO1 polymorphism (Table 3). As regards CYP1A1*2A, CYP1A1*2C and NQO1, the homozygous variant genotype was combined with the heterozygous genotype because variant homozygotes were relatively few. A borderline significant interaction was observed for the combination of CYP1A1*2C and NQO1 (P = 0.051). The OR associated with CYP1A1*2C was significantly lower than unity among individuals with the NQO1 609CC genotype. Although the interaction was far from statistical significance, the OR for individuals carrying the CYP1A1*2C allele and GSTT1 null genotype significantly decreased compared with those who had neither the CYP1A1*2C allele nor the GSTT1 null genotype.
Table 2.
Associations between genetic polymorphisms and colorectal adenomas n (%)
| Genotype | Cases | Controls | Crude | Adjusted3 |
| OR (95%CI) | ||||
| CYP1A1*2A | ||||
| 01 | 174 (38.2) | 388 (36.9) | 1.00 (referent) | 1.00 (referent) |
| 1 | 219 (48.1) | 508 (48.3) | 0.96 (0.76-1.22) | 0.97 (0.76-1.24) |
| 2 | 62 (13.6) | 156 (14.8) | 0.89 (0.63-1.25) | 0.86 (0.60-1.23) |
| CYP1A1*2C | ||||
| 01 | 281 (61.8) | 611 (58.1) | 1.00 (referent) | 1.00 (referent) |
| 1 | 152 (33.4) | 389 (37.0) | 0.85 (0.67-1.07) | 0.81 (0.64-1.03) |
| 2 | 22 (4.8) | 52 (4.9) | 0.92 (0.55-1.54) | 0.94 (0.55-1.60) |
| GSTM1 | ||||
| Non-null | 200 (44.0) | 506 (48.1) | 1.00 (referent) | 1.00 (referent) |
| Null | 255 (56.0) | 546 (51.9) | 1.18 (0.95-1.47) | 1.19 (0.94-1.49) |
| GSTT1 | ||||
| Non-null | 258 (56.7) | 552 (52.5) | 1.00 (referent) | 1.00 (referent) |
| Null | 197 (43.3) | 500 (47.5) | 0.84 (0.68-1.05) | 0.87 (0.70-1.10) |
| GSTM1 + GSTT1 | ||||
| 02 | 118 (25.9) | 273 (26.0) | 1.00 (referent) | 1.00 (referent) |
| 1 | 222 (48.8) | 512 (48.7) | 1.00 (0.77-1.31) | 1.04 (0.79-1.37) |
| 2 (both null) | 115 (25.3) | 267 (25.4) | 1.00 (0.73-1.35) | 1.04 (0.76-1.42) |
| NQO1 C609T | ||||
| CC | 161 (35.4) | 412 (39.2) | 1.00 (referent) | 1.00 (referent) |
| CT | 220 (48.4) | 489 (46.5) | 1.15 (0.90-1.47) | 1.18 (0.92-1.52) |
| TT | 74 (16.3) | 151 (14.4) | 1.25 (0.90-1.75) | 1.31 (0.93-1.84) |
Number of variant alleles;
Number of null genotypes;
Adjusted for age, hospital, rank in the Self-Defense Forces, body mass index, cigarette smoking, alcohol use, leisure-time physical activity and parental history of colorectal cancer.
Table 3.
Associations between combinations of genetic polymorphisms and colorectal adenomas
| Genotype 1 | Genotype 2 | n1 | OR (95%CI)2 | Interaction (P) |
| CYP1A1*2A | GSTM1 | 0.94 | ||
| 03 | Non-null | 75/182 | 1.00 (referent) | |
| 0 | Null | 99/206 | 1.20 (0.83-1.74) | |
| ≥ 1 | Non-null | 125/324 | 0.96 (0.67-1.35) | |
| ≥ 1 | Null | 156/340 | 1.12 (0.80-1.58) | |
| CYP1A1*2A | GSTT1 | 0.08 | ||
| 03 | Non-null | 92/210 | 1.00 (referent) | |
| 0 | Null | 82/178 | 1.14 (0.79-1.65) | |
| ≥ 1 | Non-null | 166/342 | 1.15 (0.83-1.58) | |
| ≥ 1 | Null | 115/322 | 0.85 (0.61-1.19) | |
| CYP1A1*2A | GSTM1 + GSTT1 | 0.44 | ||
| 03 | 04 | 41/102 | 1.00 (referent) | |
| 0 | 1 | 85/188 | 1.19 (0.75-1.88) | |
| 0 | 2 (both null) | 48/98 | 1.35 (0.81-2.27) | |
| ≥ 1 | 0 | 77/171 | 1.17 (0.74-1.86) | |
| ≥ 1 | 1 | 137/324 | 1.12 (0.73-1.72) | |
| ≥ 1 | 2 (both null) | 67/169 | 1.03 (0.64-1.66) | |
| CYP1A1*2A | NQO1 C609T | 0.09 | ||
| 03 | CC | 63/133 | 1.00 (referent) | |
| 0 | CT + TT | 111/255 | 0.92 (0.63-1.36) | |
| ≥ 1 | CC | 98/279 | 0.73 (0.49-1.07) | |
| ≥ 1 | CT + TT | 183/385 | 1.02 (0.71-1.47) | |
| CYP1A1*2C | GSTM1 | 0.77 | ||
| 03 | Non-null | 127/294 | 1.00 (referent) | |
| 0 | Null | 154/317 | 1.16 (0.86-1.55) | |
| ≥ 1 | Non-null | 73/212 | 0.79 (0.56-1.12) | |
| ≥ 1 | Null | 101/229 | 0.98 (0.71-1.36) | |
| CYP1A1*2C | GSTT1 | 0.24 | ||
| 03 | Non-null | 154/321 | 1.00 (referent) | |
| 0 | Null | 127/290 | 0.97 (0.73-1.31) | |
| ≥ 1 | Non-null | 104/231 | 0.93 (0.68-1.27) | |
| ≥ 1 | Null | 70/210 | 0.69 (0.49-0.97) | |
| CYP1A1*2C | GSTM1 + GSTT1 | 0.37 | ||
| 03 | 4 | 74/156 | 1.00 (referent) | |
| 0 | 1 | 133/303 | 0.96 (0.67-1.37) | |
| 0 | 2 (both null) | 74/152 | 1.13 (0.75-1.69) | |
| ≥ 1 | 0 | 44/117 | 0.79 (0.50-1.25) | |
| ≥ 1 | 1 | 89/209 | 0.92 (0.63-1.35) | |
| ≥ 1 | 2 (both null) | 41/115 | 0.72 (0.45-1.15) | |
| CYP1A1*2C | NQO1 C609T | 0.05 | ||
| 03 | CC | 102/220 | 1.00 (referent) | |
| 0 | CT + TT | 179/391 | 0.99 (0.73-1.34) | |
| ≥ 1 | CC | 59/192 | 0.61 (0.42-0.91) | |
| ≥ 1 | CT + TT | 115/249 | 0.98 (0.71-1.37) |
1Number of cases/controls;
Adjusted for age, hospital, rank in the Self-Defense Forces, body mass index, cigarette smoking, alcohol use, leisure-time physical activity and parental history of colorectal cancer;
Number of variant alleles;
Number of null genotypes.
Smoking was positively associated with colorectal adenomas; the multivariate-adjusted ORs for the smoking categories of 0, 1-399, 400-799 and ≥ 800 cigarette-years were 1.00 (referent), 1.18 (95%CI: 0.82-1.71), 2.11 (95%CI: 1.58-2.82) and 2.11 (95%CI: 1.47-3.03), respectively. However, smoking did not modify the associations of the CYP1A1, GSTM1, GSTT1 and NQO1 polymorphisms with colorectal adenomas (Table 4). The ORs associated with heavy smoking were consistently increased, regardless of genotype of the polymorphism. There was no measurable effect modification of smoking even regarding the combination of the genetic polymorphisms of the phase I and phase II enzymes (Table 5). The OR was lowest for the combination of the CYP1A1*2C and NQO1 609CC genotypes among the four composite genotypes in each stratum of smoking. The OR varied according to the combination of CYP1A1*2C and the composite genotypes of GSTM1 and GSTT1 within each stratum of smoking. The OR for the combination of the CYP1A1*2C allele and the non-null genotypes of both GSTM1 and GSTT1 was significantly lower than unity in the category of < 20 pack-years, and the OR increased significantly among heavy smokers without the CYP1A1*2C allele who had null genotypes for both GSTM1 and GSTT1.
Table 4.
Associations between genetic polymorphisms and colorectal adenomas with stratification by smoking category
| Genotype |
< 20 pack-years |
≥ 20 pack-years |
Interaction (P) | ||
| n1 | OR (95%CI)2 | n1 | OR (95%CI)2 | ||
| CYP1A1*2A | |||||
| 03 | 59/207 | 1.00 (referent) | 115/181 | 2.22 (1.52-3.24) | 0.45 |
| ≥ 1 | 100/346 | 1.05 (0.73-1.53) | 181/318 | 1.95 (1.37-2.76) | |
| CYP1A1*2C | |||||
| 03 | 106/329 | 1.00 (referent) | 175/282 | 1.90 (1.41-2.56) | 0.59 |
| ≥ 1 | 53/224 | 0.76 (0.52-1.11) | 121/217 | 1.65 (1.20-2.27) | |
| GSTM1 | |||||
| Non-null | 69/264 | 1.00 (referent) | 131/242 | 1.99 (1.41-2.82) | 0.96 |
| Null | 90/289 | 1.19 (0.83-1.71) | 165/257 | 2.35 (1.68-3.29) | |
| GSTT1 | |||||
| Non-null | 89/268 | 1.00 (referent) | 169/284 | 1.74 (1.27-2.38) | 0.26 |
| Null | 70/285 | 0.74 (0.52-1.06) | 127/215 | 1.69 (1.21-2.36) | |
| GSTM1 + GSTT1 | |||||
| 04 | 39/134 | 1.00 (referent) | 79/139 | 1.87 (1.18-2.96) | 0.63 |
| 1 | 80/264 | 1.03 (0.67-1.61) | 142/248 | 1.91 (1.26-2.91) | |
| 2 (both null) | 40/155 | 0.89 (0.54-1.48) | 75/112 | 2.14 (1.34-3.43) | |
| NQO1 C609T | |||||
| CC | 52/221 | 1.00 (referent) | 109/191 | 2.23 (1.51-3.29) | 0.46 |
| CT + TT | 107/332 | 1.35 (0.93-1.97) | 187/308 | 2.51 (1.75-3.60) | |
1Number of cases/controls;
Adjusted for age, hospital, rank in the Self-Defense Forces, body mass index, cigarette smoking, alcohol use, leisure-time physical activity and parental history of colorectal cancer;
Number of variant alleles;
Number of null genotypes.
Table 5.
Associations between combinations of genetic polymorphisms and colorectal adenomas with stratification by smoking category
| Genotype 1 | Genotype 2 |
< 20 pack-years |
≥ 20 pack-years |
Interaction (P) | ||
| n1 | OR (95%CI)2 | n1 | OR (95%CI)2 | |||
| CYP1A1*2A | GSTM1 | 0.44 | ||||
| 03 | Non-null | 25/87 | 1.00 (referent) | 50/95 | 1.78 (1.00-3.14) | |
| 0 | Null | 34/120 | 0.94 (0.52-1.69) | 65/86 | 2.53 (1.45-4.43) | |
| ≥ 1 | Non-null | 44/177 | 0.86 (0.49-1.50) | 81/147 | 1.82 (1.07-3.10) | |
| ≥ 1 | Null | 56/169 | 1.19 (0.69-2.05) | 100/171 | 1.92 (1.14-3.22) | |
| CYP1A1*2A | GSTT1 | 0.56 | ||||
| 03 | Non-null | 29/98 | 1.00 (referent) | 63/112 | 1.94 (1.15-3.28) | |
| 0 | Null | 30/109 | 0.95 (0.53-1.71) | 52/69 | 2.53 (1.44-4.44) | |
| ≥ 1 | Non-null | 60/170 | 1.26 (0.75-2.11) | 106/172 | 2.07 (1.27-3.39) | |
| ≥ 1 | Null | 40/176 | 0.81 (0.47-1.40) | 75/146 | 1.70 (1.02-2.82) | |
| CYP1A1*2A | GSTM1 + GSTT1 | 0.40 | ||||
| 03 | 04 | 14/42 | 1.00 (referent) | 27/60 | 1.34 (0.62-2.89) | |
| 0 | 1 | 26/101 | 0.76 (0.36-1.62) | 59/87 | 2.02 (1.00-4.07) | |
| 0 | 2 (both null) | 19/64 | 0.87 (0.39-1.94) | 29/34 | 2.46 (1.11-5.46) | |
| ≥ 1 | 0 | 25/92 | 0.83 (0.39-1.78) | 52/79 | 1.89 (0.93-3.86) | |
| ≥ 1 | 1 | 54/163 | 1.01 (0.51-2.01) | 83/161 | 1.52 (0.78-2.98) | |
| ≥ 1 | 2 (both null) | 21/91 | 0.73 (0.33-1.59) | 46/78 | 1.66 (0.81-3.40) | |
| CYP1A1*2A | NQO1 C609T | 0.76 | ||||
| 03 | CC | 20/73 | 1.00 (referent) | 43/60 | 2.55 (1.34-4.86) | |
| 0 | CT + TT | 39/134 | 1.04 (0.56-1.94) | 72/121 | 2.15 (1.20-3.86) | |
| ≥ 1 | CC | 32/148 | 0.81 (0.43-1.54) | 66/131 | 1.69 (0.94-3.04) | |
| ≥ 1 | CT + TT | 68/198 | 1.29 (0.72-2.29) | 115/187 | 2.23 (1.28-3.89) | |
| CYP1A1*2C | GSTM1 | 0.78 | ||||
| 02 | Non-null | 47/150 | 1.00 (referent) | 80/144 | 1.74 (1.13-2.69) | |
| 0 | Null | 59/179 | 1.04 (0.67-1.63) | 95/138 | 2.15 (1.40-3.31) | |
| ≥ 1 | Non-null | 22/114 | 0.63 (0.36-1.11) | 51/98 | 1.59 (0.98-2.57) | |
| ≥ 1 | Null | 31/110 | 0.94 (0.55-1.58) | 70/119 | 1.77 (1.13-2.78) | |
| CYP1A1*2C | GSTT1 | 0.11 | ||||
| 02 | Non-null | 60/151 | 1.00 (referent) | 94/170 | 1.39 (0.93-2.07) | |
| 0 | Null | 46/178 | 0.66 (0.42-1.03) | 81/112 | 1.80 (1.18-2.75) | |
| ≥ 1 | Non-null | 29/117 | 0.65 (0.39-1.09) | 75/114 | 1.62 (1.06-2.48) | |
| ≥ 1 | Null | 24/107 | 0.59 (0.34-1.01) | 46/103 | 1.06 (0.66-1.69) | |
| CYP1A1*2C | GSTM1 + GSTT1 | 0.22 | ||||
| 02 | 04 | 29/71 | 1.00 (referent) | 45/85 | 1.26 (0.71-2.23) | |
| 0 | 1 | 49/159 | 0.73 (0.42-1.26) | 84/144 | 1.41 (0.84-2.37) | |
| 0 | 2 (both null) | 28/99 | 0.69 (0.38-1.28) | 46/53 | 2.03 (1.12-3.69) | |
| ≥ 1 | 0 | 10/63 | 0.39 (0.17-0.87) | 34/54 | 1.46 (0.79-2.72) | |
| ≥ 1 | 1 | 31/105 | 0.75 (0.41-1.36) | 58/104 | 1.31 (0.76-2.26) | |
| ≥ 1 | 2 (both null) | 12/56 | 0.53 (0.25-1.15) | 29/59 | 1.09 (0.58-2.05) | |
| CYP1A1*2C | NQO1 C609T | 0.34 | ||||
| 02 | CC | 34/127 | 1.00 (referent) | 68/93 | 2.64 (1.60-4.36) | |
| 0 | CT + TT | 72/202 | 1.33 (0.83-2.12) | 107/189 | 2.10 (1.33-3.31) | |
| ≥ 1 | CC | 18/94 | 0.75 (0.39-1.41) | 41/98 | 1.41 (0.83-2.41) | |
| ≥ 1 | CT + TT | 35/130 | 1.03 (0.60-1.77) | 80/119 | 2.47 (1.53-4.00) | |
1Number of cases/controls;
Adjusted for age, hospital, rank in the Self-Defense Forces, body mass index, cigarette smoking, alcohol use, leisure-time physical activity and parental history of colorectal cancer;
Number of variant alleles;
Number of null genotypes.
DISCUSSION
According to recent meta-analyses[32,33], the CYP1A1*2C polymorphism, but not CYP1A1*2A polymorphism, was significantly associated with a modest increase in the risk of colorectal cancer. Neither CYP1A1*2A nor CYP1A1*2C polymorphism was related to colorectal adenomas individually in the present study. The findings are consistent with the previous observation regarding colorectal adenomas[16,26]. Further investigation is needed to clarify whether the associations with the CYP1A1*2C polymorphism is differential for colorectal cancer and adenomas.
The null effects of the GSTM1 and GSTT1 polymorphisms in the present study are in agreement with previous observations regarding colorectal adenomas[23,34]. Some of the previous studies showed an increased risk of colorectal cancer among individuals with the GSTM1 null genotype[11,21,22], the GSTT1 null genotype[21], or both the GSTM1 and GSTT1 null genotypes[21,22]. However, these findings were not replicated in other studies on colorectal cancer[14,15,35]. Previous studies found no effect of smoking on the association with GSTM1 and GSTT1, either singly or in combination, in relation to colorectal adenomas[23] and cancer[15,22,33]. The GSTM1 and GSTT1 non-null genotypes as determined by gel electrophoresis include heterozygous genotypes (i.e., one active and one inactive allele). One study differentiated the heterozygous genotype from the homozygous non-null genotype for GSTM1 and GSTT1 by TaqMan assay[36]. Both heterozygous and homozygous null genotypes of GSTM1 were associated with a decreased risk of colorectal adenomas irrespective of smoking status, while adenoma risk was increased in association with both heterozygous and homozygous null genotypes of GSTT1 among ever-smokers, but not among never-smokers[36]. However, it is unclear whether heterozygosity in either GSTM1 or GSTT1 affects enzyme activity[18].
Few studies have addressed the combined effect of the CYP1A1 polymorphisms and the GSTM1 and/or GSTT1 null genotype in relation to colorectal adenomas and cancer. The combination of CYP1A1*2A and GSTM1 null genotype was shown to be unrelated to colorectal adenomas[16]. There was no interaction between the two CYP1A1 polymorphisms and either GSTM1 or GSTT1 null genotype on the risk of colorectal cancer[14]. A Japanese study showed a decreased risk of colorectal cancer for the combination of CYP1A*2C and GSTT1 non-null genotype[15]. The present study indicated a decreased risk of colorectal adenomas for the combination of CYP1A1*2C and GSTT1 null genotype, although the interaction was not significant. Inconsistent findings regarding the combination of CYP1A1*2C and GSTT1 are probably ascribed to random fluctuation, although further studies are needed.
The risk of adenomas was lowest for the combination of CYP1A1*2C and the GSTM1 and GSTT1 non-null genotypes among never-smokers or light smokers, and an increased risk associated with heavy smoking was most evident among men without CYP1A1*2C who had the GSTM1 and GSTT1 null genotypes. Caution is required in interpreting the findings because of the small number of each combination of the three polymorphisms when stratified by smoking. Nonetheless, further investigation is warranted because the association of the three-polymorphism combination with colorectal cancer or adenomas has not been investigated previously.
The present study did not corroborate an increased risk of colorectal adenomas associated with the combination of CYP1A1*2C and NQO1 609T alleles reported among non-Hispanic whites[26], but showed a decreased risk among men harboring the CYP1A1*2C allele and the NQO1 609CC genotype. In that study[26], however, there were only a few subjects with both the CYP1A1*2C and NQO1 609T alleles (12 cases and 26 controls), accounting for only 4.0% of the cases and 1.8% of the controls, because these variant alleles are much less frequent in Caucasians than in Asians[10,31]. In a study of Caucasians[26], the OR for the combination of CYP1A1*2C allele and NQO1 609CC genotype was 0.6 (95%CI: 0.3-1.2), as compared with the same referent combination as used in the present study. This finding is thus compatible with the present observation. It should be noted that the borderline significant interaction between CYP1A1*2C and NQO1 C609T resulted from the eight statistical tests (Table 3). The probability of detecting at least one statistically significant result is 0.34, even when none of the eight interactions are present. The combined effect of these two polymorphisms requires careful interpretation, but requires further studies for mechanistic plausibility. Exposure to activated carcinogens may be lowered in individuals with both the CYP1A1*2C allele and the NQO1 609CC genotype for faster activation and detoxification.
The advantages of our study were that colonoscopy was done non-selectively in a defined population and that the absence of polyp lesions was confirmed in the control subjects by complete colonoscopy. Ethnic homogeneity was another advantage. There were several limitations. Statistical adjustment was not made for dietary factors because such data were not available. Only men were included, and the findings may not be applicable to women. Smoking is much less prevalent in women than in men in Japan[15]. The study subjects were not representative of Japanese men in the general population, but selection was unlikely to have occurred with regard to the genetic polymorphisms under study. The allele and genotype frequencies in the present study population were almost the same as those observed among Japanese individuals elsewhere. Among community controls (n = 778) in a Japanese case-control study of colorectal cancer[15], the frequencies of the CYP1A1*2A, CYP1A1*2C and NQO1 609T alleles were 37%, 23% and 38%, respectively, and the GSTM1 and GSTT1 null genotypes accounted for 54% and 44%, respectively. In a random sample of approximately 300 Japanese adults[37], the frequencies of the CYP1A1*2C and NQO1 609T alleles were 21% and 38%, respectively. Finally, the study size was not sufficiently large to detect a moderately increased risk for the variant homozygote. With two-sided α = 0.05, the power of detecting a 1.5-fold increase in the risk for the variant homozygote in the additive model was 0.66 for CYP1A1*2A, 0.39 for CYP1A1*2C and 0.65 for NQO1 C609T.
In conclusion, the combination of CYP1A1*2C and NQO1 609CC genotype was associated with a decreased risk of colorectal adenomas, regardless of smoking status, in Japanese men. Future studies are needed to clarify the biological mechanisms involved.
COMMENTS
Background
Cigarette smoking has consistently been related to an increased risk of colorectal adenomas, and possibly of colorectal cancer, but the biological mechanisms remain unknown.
Research frontiers
Tobacco smoke contains various types of carcinogens, which are activated by phase I enzymes and/or detoxified by phase II enzymes. It is a matter of interest whether or not functional genetic polymorphisms of the metabolic enzymes are related to colorectal adenomas and cancer.
Innovations and breakthroughs
Few studies have examined the association of genetic polymorphisms of phase I and phase II enzymes in combination with colorectal adenomas or cancer. Adenoma risk differed by the combination of genetic polymorphisms of CYP1A1 (phase I enzyme) and NAD(P)H-quinone oxidoreductase 1 (NQO1) (phase II enzyme), and the association was not modified by smoking.
Applications
The findings confer clues to understanding the biological mechanisms of the association between smoking and colorectal adenomas and cancer.
Terminology
CYP1A1 is responsible for bioactivation of tobacco carcinogens. CYP1A1*2A and CYP1A1*2C polymorphisms are putatively functional, and have been related to increased risk of tobacco-related cancers; Glutathione S-transferases are involved in detoxification of chemical carcinogens, and individuals with the GSTM1 and/or GSTT1 null genotype may be susceptible to increased risk of cancer; NQO1 also acts as a phase II enzyme, and the 609C>T polymorphism results in loss of NQO1 activity.
Peer review
This was a good study. The study subjects were all male officials in the SelfDefense Forces. Besides smoking, they may have risk factors, such as alcohol drinking and dietary habits, similar to those in the general population.
Footnotes
Supported by The Scientific Support Programs for Cancer Research, Grant-in-Aid for Scientific Research on Innovative Areas, No. 221S0001; the Ministry of Education, Culture, Sports, Science and Technology, Japan
P- Reviewers Fei J, Makambi KH, Singh JP, Wang N, Zhang YJ, Zhou TB S- Editor Gou SX L- Editor Kerr C E- Editor Li JY
References
- 1.Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin. 2005;55:74–108. doi: 10.3322/canjclin.55.2.74. [DOI] [PubMed] [Google Scholar]
- 2.Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell. 1990;61:759–767. doi: 10.1016/0092-8674(90)90186-i. [DOI] [PubMed] [Google Scholar]
- 3.Winawer SJ, Zauber AG, Stewart E, O’Brien MJ. The natural history of colorectal cancer. Opportunities for intervention. Cancer. 1991;67:1143–1149. doi: 10.1002/1097-0142(19910215)67:4+<1143::aid-cncr2820671507>3.0.co;2-d. [DOI] [PubMed] [Google Scholar]
- 4.Giovannucci E. An updated review of the epidemiological evidence that cigarette smoking increases risk of colorectal cancer. Cancer Epidemiol Biomarkers Prev. 2001;10:725–731. [PubMed] [Google Scholar]
- 5.Botteri E, Iodice S, Raimondi S, Maisonneuve P, Lowenfels AB. Cigarette smoking and adenomatous polyps: a meta-analysis. Gastroenterology. 2008;134:388–395. doi: 10.1053/j.gastro.2007.11.007. [DOI] [PubMed] [Google Scholar]
- 6.Botteri E, Iodice S, Bagnardi V, Raimondi S, Lowenfels AB, Maisonneuve P. Smoking and colorectal cancer: a meta-analysis. JAMA. 2008;300:2765–2778. doi: 10.1001/jama.2008.839. [DOI] [PubMed] [Google Scholar]
- 7.Hecht SS. Tobacco carcinogens, their biomarkers and tobacco-induced cancer. Nat Rev Cancer. 2003;3:733–744. doi: 10.1038/nrc1190. [DOI] [PubMed] [Google Scholar]
- 8.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 Biomarkers Prev. 2000;9:3–28. [PubMed] [Google Scholar]
- 9.Agundez JA. Cytochrome P450 gene polymorphism and cancer. Curr Drug Metab. 2004;5:211–224. doi: 10.2174/1389200043335621. [DOI] [PubMed] [Google Scholar]
- 10.Sivaraman L, Leatham MP, Yee J, Wilkens LR, Lau AF, Le Marchand L. CYP1A1 genetic polymorphisms and in situ colorectal cancer. Cancer Res. 1994;54:3692–3695. [PubMed] [Google Scholar]
- 11.Sachse C, Smith G, Wilkie MJ, Barrett JH, Waxman R, Sullivan F, Forman D, Bishop DT, Wolf CR. A pharmacogenetic study to investigate the role of dietary carcinogens in the etiology of colorectal cancer. Carcinogenesis. 2002;23:1839–1849. doi: 10.1093/carcin/23.11.1839. [DOI] [PubMed] [Google Scholar]
- 12.Ishibe N, Stampfer M, Hunter DJ, Hennekens C, Kelsey KT. A prospective study of cytochrome P450 1A1 polymorphisms and colorectal cancer risk in men. Cancer Epidemiol Biomarkers Prev. 2000;9:855–856. [PubMed] [Google Scholar]
- 13.Slattery ML, Samowtiz W, Ma K, Murtaugh M, Sweeney C, Levin TR, Neuhausen S. CYP1A1, cigarette smoking, and colon and rectal cancer. Am J Epidemiol. 2004;160:842–852. doi: 10.1093/aje/kwh298. [DOI] [PubMed] [Google Scholar]
- 14.Little J, Sharp L, Masson LF, Brockton NT, Cotton SC, Haites NE, Cassidy J. Colorectal cancer and genetic polymorphisms of CYP1A1, GSTM1 and GSTT1: a case-control study in the Grampian region of Scotland. Int J Cancer. 2006;119:2155–2164. doi: 10.1002/ijc.22093. [DOI] [PubMed] [Google Scholar]
- 15.Nisa H, Kono S, Yin G, Toyomura K, Nagano J, Mibu R, Tanaka M, Kakeji Y, Maehara Y, Okamura T, et al. Cigarette smoking, genetic polymorphisms and colorectal cancer risk: the Fukuoka Colorectal Cancer Study. BMC Cancer. 2010;10:274. doi: 10.1186/1471-2407-10-274. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Inoue H, Kiyohara C, Marugame T, Shinomiya S, Tsuji E, Handa K, Hayabuchi H, Onuma K, Hamada H, Koga H, et al. Cigarette smoking, CYP1A1 MspI and GSTM1 genotypes, and colorectal adenomas. Cancer Res. 2000;60:3749–3752. [PubMed] [Google Scholar]
- 17.Rebbeck TR. Molecular epidemiology of the human glutathione S-transferase genotypes GSTM1 and GSTT1 in cancer susceptibility. Cancer Epidemiol Biomarkers Prev. 1997;6:733–743. [PubMed] [Google Scholar]
- 18.Cotton SC, Sharp L, Little J, Brockton N. Glutathione S-transferase polymorphisms and colorectal cancer: a HuGE review. Am J Epidemiol. 2000;151:7–32. doi: 10.1093/oxfordjournals.aje.a010124. [DOI] [PubMed] [Google Scholar]
- 19.Chen K, Jiang QT, He HQ. Relationship between metabolic enzyme polymorphism and colorectal cancer. World J Gastroenterol. 2005;11:331–335. doi: 10.3748/wjg.v11.i3.331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Raimondi S, Botteri E, Iodice S, Lowenfels AB, Maisonneuve P. Gene-smoking interaction on colorectal adenoma and cancer risk: review and meta-analysis. Mutat Res. 2009;670:6–14. doi: 10.1016/j.mrfmmm.2009.06.013. [DOI] [PubMed] [Google Scholar]
- 21.Martínez C, Martín F, Fernández JM, García-Martín E, Sastre J, Díaz-Rubio M, Agúndez JA, Ladero JM. Glutathione S-transferases mu 1, theta 1, pi 1, alpha 1 and mu 3 genetic polymorphisms and the risk of colorectal and gastric cancers in humans. Pharmacogenomics. 2006;7:711–718. doi: 10.2217/14622416.7.5.711. [DOI] [PubMed] [Google Scholar]
- 22.Wang J, Jiang J, Zhao Y, Gajalakshmi V, Kuriki K, Suzuki S, Nagaya T, Nakamura S, Akasaka S, Ishikawa H, et al. Genetic polymorphisms of glutathione S-transferase genes and susceptibility to colorectal cancer: a case-control study in an Indian population. Cancer Epidemiol. 2011;35:66–72. doi: 10.1016/j.canep.2010.07.003. [DOI] [PubMed] [Google Scholar]
- 23.Inoue H, Kiyohara C, Shinomiya S, Marugame T, Tsuji E, Handa K, Hayabuchi H, Eguchi H, Fukushima Y, Kono S. Glutathione S-transferase polymorphisms and risk of colorectal adenomas. Cancer Lett. 2001;163:201–206. doi: 10.1016/s0304-3835(00)00692-3. [DOI] [PubMed] [Google Scholar]
- 24.Nebert DW, Roe AL, Vandale SE, Bingham E, Oakley GG. NAD(P)H: quinone oxidoreductase (NQO1) polymorphism, exposure to benzene, and predisposition to disease: a HuGE review. Genet Med. 2002;4:62–70. doi: 10.1097/00125817-200203000-00003. [DOI] [PubMed] [Google Scholar]
- 25.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;15:979–987. doi: 10.1158/1055-9965.EPI-05-0899. [DOI] [PubMed] [Google Scholar]
- 26.Hou L, Chatterjee N, Huang WY, Baccarelli A, Yadavalli S, Yeager M, Bresalier RS, Chanock SJ, Caporaso NE, Ji BT, et al. CYP1A1 Val462 and NQO1 Ser187 polymorphisms, cigarette use, and risk for colorectal adenoma. Carcinogenesis. 2005;26:1122–1128. doi: 10.1093/carcin/bgi054. [DOI] [PubMed] [Google Scholar]
- 27.Toyomura K, Yamaguchi K, Kawamoto H, Tabata S, Shimizu E, Mineshita M, Ogawa S, Lee KY, Kono S. Relation of cigarette smoking and alcohol use to colorectal adenomas by subsite: the self-defense forces health study. Cancer Sci. 2004;95:72–76. doi: 10.1111/j.1349-7006.2004.tb03173.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Morita M, Tabata S, Tajima O, Yin G, Abe H, Kono S. Genetic polymorphisms of CYP2E1 and risk of colorectal adenomas in the Self Defense Forces Health Study. Cancer Epidemiol Biomarkers Prev. 2008;17:1800–1807. doi: 10.1158/1055-9965.EPI-08-0314. [DOI] [PubMed] [Google Scholar]
- 29.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. 1995;67:253–256. doi: 10.1007/BF00409407. [DOI] [PubMed] [Google Scholar]
- 30.Arand M, Mühlbauer R, Hengstler J, Jäger 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. 1996;236:184–186. doi: 10.1006/abio.1996.0153. [DOI] [PubMed] [Google Scholar]
- 31.Kelsey KT, Ross D, Traver RD, Christiani DC, Zuo ZF, Spitz MR, Wang M, Xu X, Lee BK, Schwartz BS, et al. Ethnic variation in the prevalence of a common NAD(P)H quinone oxidoreductase polymorphism and its implications for anti-cancer chemotherapy. Br J Cancer. 1997;76:852–854. doi: 10.1038/bjc.1997.474. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Jin JQ, Hu YY, Niu YM, Yang GL, Wu YY, Leng WD, Xia LY. CYP1A1 Ile462Val polymorphism contributes to colorectal cancer risk: a meta-analysis. World J Gastroenterol. 2011;17:260–266. doi: 10.3748/wjg.v17.i2.260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Theodoratou E, Montazeri Z, Hawken S, Allum GC, Gong J, Tait V, Kirac I, Tazari M, Farrington SM, Demarsh A, et al. Systematic meta-analyses and field synopsis of genetic association studies in colorectal cancer. J Natl Cancer Inst. 2012;104:1433–1457. doi: 10.1093/jnci/djs369. [DOI] [PubMed] [Google Scholar]
- 34.Tiemersma EW, Bunschoten A, Kok FJ, Glatt H, de Boer SY, Kampman E. Effect of SULT1A1 and NAT2 genetic polymorphism on the association between cigarette smoking and colorectal adenomas. Int J Cancer. 2004;108:97–103. doi: 10.1002/ijc.11533. [DOI] [PubMed] [Google Scholar]
- 35.Gertig DM, Stampfer M, Haiman C, Hennekens CH, Kelsey K, Hunter DJ. Glutathione S-transferase GSTM1 and GSTT1 polymorphisms and colorectal cancer risk: a prospective study. Cancer Epidemiol Biomarkers Prev. 1998;7:1001–1005. [PubMed] [Google Scholar]
- 36.Moore LE, Huang WY, Chatterjee N, Gunter M, Chanock S, Yeager M, Welch B, Pinsky P, Weissfeld J, Hayes RB. GSTM1, GSTT1, and GSTP1 polymorphisms and risk of advanced colorectal adenoma. Cancer Epidemiol Biomarkers Prev. 2005;14:1823–1827. doi: 10.1158/1055-9965.EPI-05-0037. [DOI] [PubMed] [Google Scholar]
- 37.Yoshimura K, Hanaoka T, Ohnami S, Ohnami S, Kohno T, Liu Y, Yoshida T, Sakamoto H, Tsugane S. Allele frequencies of single nucleotide polymorphisms (SNPs) in 40 candidate genes for gene-environment studies on cancer: data from population-based Japanese random samples. J Hum Genet. 2003;48:654–658. doi: 10.1007/s10038-003-0096-1. [DOI] [PubMed] [Google Scholar]