Impact of Multiple Alcohol Dehydrogenase Gene Polymorphisms on Risk of Upper Aerodigestive Tract Cancers in a Japanese Population

Published OnlineFirst October 27, 2009; DOI: 10.1158/1055-9965.EPI-09-0499

3097

Isao Oze,1,7 Keitaro Matsuo,1,8 Takeshi Suzuki,5 Takakazu Kawase,1 Miki Watanabe,1 Akio Hiraki,6 Hidemi Ito,1 Satoyo Hosono,1 Taijiro Ozawa,2 Shunzo Hatooka,3 Yasuhi Yatabe,4 Yasuhisa Hasegawa,2 Masayuki Shinoda,3 Katsuyuki Kiura,7 Kazuo Tajima,1 Mitsune Tanimoto,7 and Hideo Tanaka1,8 1Division of Epidemiology and Prevention, Aichi Cancer Center Research Institute, Nagoya, Japan; Departments of 2Head/Neck Surgery, 3Thoracic Surgery, and 4Pathology and Molecular Diagnostics, Aichi Cancer Center Central Hospital, Nagoya, Japan; 5Department of Medical Oncology and Immunology, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan; 6Health Center, Okayama University, Okayama, Japan; 7Department of Hematology, Oncology and Respiratory Medicine, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University, Okayama, Japan; and 8Department of Epidemiology, Nagoya University Graduate School of Medicine, Nagoya, Japan

Abstract

Alcohol intake is positivelyassociated with the risk of assays. Associations between polymorphisms and UAT upper aerodigestive tract (UAT) cancer. The genes that cancer were assessed byodds ratios and 95% confidence encode alcohol-metabolizing enzymes, primarily alcohol intervals using conditional logistic regression models dehydrogenases (ADH) and aldehyde dehydrogenases that adjusted for age, sex, smoking, drinking, and (ALDH), are polymorphic. In Caucasians, significant ALDH2. Adjusted odds ratios were significant for associations between polymorphisms in ADH1B rs4148887 and rs3805322 in ADH4, rs1229984 in ADH1B, (rs1229984) and ADH1C (rs698 and rs1693482), and UAT rs698 and rs1693482 in ADH1C, and rs284787, rs1154460, cancer have been observed, despite strong linkage dis- and rs3737482 in ADH7. We also observed that ADH7 equilibrium among them. Moreover, UATcancer was sig- rs3737482 and ADH4 rs4148887 had independentlyand nificantlyassociated with rs1573496 in ADH7, and not statisticallysignificant effects on UATcancer. The magni- with rs1984362 in ADH4.However,littleevidenceis tude of effect of these ADH polymorphisms was greater available concerning ADH4 or ADH7 polymorphisms in in subjects who were heavydrinkers, heavysmokers, Asian populations. We conducted a matched case-control and had esophageal cancer. These findings show that studyto clarifythe role of ADH polymorphisms in a Jap- multiple ADH gene polymorphisms were associated anese population. Cases and controls were 585 patients with UAT cancer in this Japanese population. Further with UATcancer and 1,170 noncancer outpatients. Geno- studies in various ethnicities are required. (Cancer Epi- typing for ADHs and ALDH2 wasdoneusingTaqMan demiol Biomarkers Prev 2009;18(11):3097–102)

Introduction

Alcohol consumption is one of the most important risk (ALDH), mainly ALDH2(3). Because the genes that en- factors for upper aerodigestive tract (UAT) cancer (1). Ac- code these representative ethanol-metabolizing enzymes etaldehyde, an oxidative product of ethanol, is suspected contain polymorphisms that modulate individual differ- to be a major carcinogen behind this association (2). In ences in ethanol- and acetaldehyde-oxidizing capacity general, ethanol is oxidized to acetaldehyde by alcohol (4), they have been hypothesized to explain individual dehydrogenase enzymes (ADH), which is then further differences in UAT cancer susceptibility. Identification of oxidized to acetate by aldehyde-dehydrogenase enzymes differences among individuals in susceptibility to UAT cancer will help our understanding of the mechanisms of UAT carcinogenesis and assist in the development of tailored UAT cancer prevention strategies. Received 5/26/09; revised 9/2/09; accepted 9/3/09; published OnlineFirst 10/27/09. There are seven ADH genes, namely ADH5, ADH4, Grant support: Grants-in-Aid for Scientific Research from the Ministry of Education, ADH6, ADH1A, ADH1B, ADH1C, and ADH7 (4). The as- Science, Sports, Culture and Technology of Japan, for Cancer Research from the Ministry of Health, Labour and Welfare of Japan, and for the Third Term sociation between polymorphisms in some of these ADH Comprehensive 10-year Strategy for Cancer Control from the Ministry of Health, genes and UAT cancers has been investigated, although Labour and Welfare of Japan. This study was also supported in part by the Foundation ADH1B for Promotion of Cancer Research in Japan and by a research grant from the Uehara most studies have focused on rs1229984 or Memorial Foundation. These grantors were not involved in the study design, subject ADH1C (rs698 and rs1693482; refs. 5-13). Results sug- enrollment, study analysis or interpretation, or submission of the manuscript for this study. gested that individuals carrying alleles conferring slow Note: Supplementary data for this article are available at Cancer Epidemiology, oxidizing capacity had higher risk of UAT cancers. The Biomarkers & Prevention Online (http://cebp.aacrjournals.org/). largest, most comprehensive epidemiologic study recently Requests for reprints: Keitaro Matsuo, Division of Epidemiology and Prevention, conducted in Europe showed that the effect of ADH1C Aichi Cancer Center Research Institute, 1-1 Kanokoden, Chikusa-ku, Nagoya 464-8681, ADH1B Japan. Phone: 81-52-762-6111; Fax: 81-52-763-5233. E-mail: [email protected] rs1693482and rs698 was independent of that of Copyright © 2009 American Association for Cancer Research. rs1229984, despite strong linkage disequilibrium (LD) doi:10.1158/1055-9965.EPI-09-0499 among them (14). This study also showed that UATcancer

Cancer Epidemiol Biomarkers Prev 2009;18(11). November 2009

3098 ADH Polymorphisms and Upper Aerodigestive Tract Cancer

was significantly associated with rs1573496 in ADH7 in to date about functional significance of polymorphisms Caucasians, and not with rs1984362in ADH4 (14). in ADH4/ADH7 genes. TagSNPs were selected as those However, little evidence is available concerning ADH4 satisfying a minor allele frequencies of over 20% and an or ADH7 polymorphisms in Asian populations. ALDH2 is R2 of >0.8 in JPT using the tagSNP Picker function at the monomorphic in Caucasians, whereas ALDH2 polymor- International HapMap Web site. phism (Glu504Lys, rs671) is prevalent in East Asians Assessment of Alcohol Intake and Smoking Expo- (15, 16). Those with ALDH2 Lys allele (null type) have sure. Lifetime alcohol consumption of various common higher risk for UAT cancer than those with ALDH2 Glu beverages (Japanese sake, beer, shochu, whiskey, and allele due to the catalytic inactivity for acetaldehyde elim- wine) was determined in terms of the average number ination (17, 18). Although polymorphisms between ADHs of drinks per day, which was then converted into a Japa- and ALDH2 were not in LD, ADHs and ALDH2 might af- nese sake (rice wine) equivalent. We asked about the fect each other through their function for acetaldehyde amount consumed in terms of one “go” (180 mL) of Jap- creation and elimination. Therefore, the association be- anese sake equivalent, which contains 23 g ethanol, one tween ADHs and UAT cancer in Asians might be different large bottle (720 mL) of beer, two shots (57 mL) of whis- from that in Caucasians. key, or two and a half glasses of wine (200 mL). One drink We conducted a case-control study to clarify differences of “shochu” (distilled spirit), which contains 25% ethanol, in the magnitude of effect of ADH4, ADH7, ADH1B, and was rated as 108 mL. In this analysis, we defined one unit ADH1C polymorphisms on UAT cancer in a Japanese of drink as a half go. Total alcohol consumption (grams population. per consumption) of Japanese sake, beer, shochu, whiskey, and wine was calculated for current and former reg- Materials and Methods ular drinkers, who were then categorized into the four levels of never drinker, moderate drinker, high-moderate Subjects. The subjects were 585 patients with no prior drinker, and heavy drinker. Heavy drinking was defined history of cancer who were histologically diagnosed with as consumption on 5 d or more per week and four units UAT cancer (oral cavity and pharynx cancer in 217, larynx (46 g ethanol) or more on each occasion; high-moderate cancer in 103, and esophageal cancer in 265) between Jan- drinking as consumption on 5 d or more per week and uary 2001 and December 2005 at Aichi Cancer Center fewer than four units (46 g ethanol) on each occasion; Hospital (ACCH). All of the subjects were recruited in moderate drinking as consumption on 4 d or fewer per the framework of the Hospital-based Epidemiologic Re- week; and never drinking as never having drunk alcoho- search Program at Aichi Cancer Center, as described else- lic beverages. Information on smoking status was ob- where (19, 20). UAT cancer was defined according to the tained in the three categories of nonsmoker, former following codes of the International Classification of Dis- smoker, and current smoker, with former smokers defined eases and Related Health Problems (ICD10): oral cavity as those who had quit smoking at least 1 y before the sur- and oropharynx (C00.3-C00.9, C01, C02.0-C02.4, C03, vey. Cumulative smoking dose was evaluated as pack- C04, C05.0-C05.2, C06, C09, C10), hypopharynx (C12, years (PY), the product of the number of packs consumed C13), oral cavity-oropharynxhypopharynx not otherwise per day and the number of years of smoking. specified (C02.8, C02.9, C05.8, C05.9, C14), larynx (C32), Statistical Analysis. Associations between polymorph- and esophagus (C15). Malignant neoplasms of the sali- isms and UAT cancer were assessed by odds ratio (OR) vary glands (C07, C08), nasopharynx (C11), nasal (C30), and 95% confidence interval (CI) using conditional logis- and paranasal (C31) were excluded as they have quite tic regression models. P values for heterogeneity were as- distinct etiologies. The controls were 1,170 first-visit out- sessed by adding interaction terms between each ADH patients at ACCH during the same period who were con- locus and confounders in the models. ORs and 95% CIs firmed to have no cancer and no history of neoplasia. were estimated for per-allele, dominant, and recessive Noncancer status was confirmed by medical examinations models. Genotypes were included as scores. Potential including radiographic examinations. Those who sus- confounders considered in the multivariate analyses were pected of having UAT cancer were examined by physical age as a continuous variable, sex (male, female), smoking or endoscopic inspection. Radiographic examinations (PY < 5; 5 < PY < 20; 20 < PY < 40; 40 < PY), alcohol con- were carried out for subjects suspected of having cancer sumption (never, moderate, high-moderate, heavy), and after inspection. Controls were selected randomly and fre- ALDH2 genotype (Glu/Glu, Glu/Lys, Lys/Lys or Glu/ quency matched with group of age category (<40, 40-49, Glu, Glu/Lys and Lys/Lys). 50-59, 60-69, >70) and sex (male, female) at a case-control We applied a retrospective profile-likelihood method ratio of 1:2. (21, 22) in logistic regression to evaluate haplotype effects Genotyping of ADH4, ADH7, ADH1B, ADH1C, and by using “haplologit” command (23). We constructed ALDH2. For each subject, DNA was extracted from the haplotypes by using unphased information of loci that buffy coat fraction with a DNA Blood mini kit (Qiagen) were in LD (ADH4 rs4148887, ADH4 rs3805322, ADH1B or BioRobot EZ1 and EZ1 DNA Blood 350mL kit (Qia- rs1229984, ADH1C rs698, ADH1C rs1693482, and ADH7 gen). Genotyping for ADH4 rs4148887, ADH4 rs3805322, rs284787) on chromosome 4. As ADH7 rs3737482was ADH7 rs284787, ADH7 rs1154460, ADH7 rs3737482, ADH7 not in LD with other loci, we put this locus as a covariate rs1573496, ADH1B rs1229984, ADH1C rs1693482, ADH1C in the model with other potential confounders (smoking, rs698, and ALDH2 rs671 was based on TaqMan Assays alcohol consumption, and indicator variables for ALDH2 (Applied Biosystems). We used tagSNPs for ADH4 and genotypes) considered in conditional logistic regression ADH7 on the basis that minor allele frequencies for signif- for individual loci analysis. We estimated ORs in log ad- icant loci identified by the European study (14) are mark- ditive model for haplotypes of their frequencies over edly low in Japanese and limited evidence was available 0.001 (Supplementary Table S1) compared with the most

Cancer Epidemiol Biomarkers Prev 2009;18(11). November 2009

Cancer Epidemiology, Biomarkers & Prevention 3099

Table 1. Characteristics of the cases and controls 0.77 1.30 1.17 1.31 1.33 1.32 1.41 Characteristic Case (%) Control (%) 1.50 OR* model (0.61-0.97) (1.04-1.63) (0.94-1.46) (0.96-1.83) (0.95-1.81) (1.06-1.64) (1.12-1.76) (1.21-1.87) (95% CI) n = 585 n = 1170 Recessive Sex Male 487 (83.2) 974 (83.2) Female 98 (16.8) 196 (16.8) Age at interview (y) <40 20 (3.4) 42 (3.6) 0.73 1.62 1.29 7.39 7.98 2.20 1.96 1.52 40-49 46 (7.9) 101 (8.6) OR* model (95% CI) (0.55-0.96) (0.96-1.73) (1.18-2.24) (0.98-65.08) (0.89-61.09) (1.46-3.32) (1.15-2.01) 50-59 186 (31.8) 355 (30.3) (1.34-2.87) 60-69 217 (37.1) 460 (39.3) >70 116 (19.8) 212 (18.1) Smoking status Nonsmoker 96 (16.4) 399 (34.1) Former smoker* 161 (27.5) 381 (32.6)

Current smoker 326 (55.7) 389 (33.2) OR* (0.69-0.94) (1.04-1.42) (1.04-1.43) (1.01-1.85) (1.02-1.87) (1.15-1.62) (1.23-1.72) (1.13-1.54) Cumulative smoking PY < 5 103 (17.6) 448 (38.3) (95% CI) 5 ≤ PY < 20 67 (11.5) 164 (14.0) 20 ≤ PY < 40 161 (27.5) 258 (22.1) ≤ † 40 PY 249 (42.6) 288 (24.6)

Alcohol drinking Never 94 (16.1) 361 (30.9) minor Moderate 89 (15.2) 332 (28.4) Minor/ High-moderate 134 (22.9) 287 (24.5) Heavy 253 (43.2) 170 (14.5) Cancer site Oral/pharynx 217 (37.1) Larynx 103 (17.6) minor Esophagus 265 (45.3) Major/

*Former smoker was defined as those who quit smoking at least 1 y. † Moderate drinker was defined as consumption <4 d/wk. High-moderate drinker was defined as consumption of <46 g ethanol >5 d/wk. Heavy

drinker was defined as consumption of >46 g ethanol >5 d/wk. major Major/

frequent haplotype (CGAACT) as well as an OR for Genotyped counts Allelic Dominant ADH7

rs3737482. Discrepancies between expected and minor observed genotype and allele frequencies in the controls Minor/ were assessed by accordance with the Hardy-Weinberg 2 equilibrium using the χ test. Case Control minor Statistical analyses were done using STATA version 10 Major/ (Stata Corporation). A P value of <0.05 was considered statistically significant. LD estimates were calculated using Haploview. major

Results allele

Table 1 shows the characteristics of cases and controls. Al- Minor cohol consumption was more prevalent in the cases than frequency in the matched controls. In addition, prevalence of current Case Control Major/ smoking and cumulative exposure to smoking was also . more prevalent among cases than controls. ADH4, Table 2shows the genotype distributions for ALDH2 genotype frequency and ORs allele ADH1B, ADH1C, and ADH7 and their ORs and 95% CIs minor for UAT cancer. Genotype distributions of each locus among controls were accordant with the Hardy-Weinberg ADH7 equilibrium. We genotyped ADH7 rs1573496 in 412sub- jects in this study, and confirmed that all carried homozygous C-alleles. This locus was therefore excluded from , and further analysis. Adjusted per allele ORs were significant for rs4148887 and rs3805322 in ADH4, rs1229984 in ADH1B, rs698 and rs1693482in ADH1C, and rs284787, rs1154460, and rs3737482in ADH7. ADH4 and ADH1B polymorphisms consistently showed statistically significant associations in both dominant and recessive models, and we did not observe marked changes in the ORs. Al- though ADH7 polymorphisms showed similar OR trends

in the dominant and recessive models, the ORs for ADH4, ADH1B, ADH1C rs1154460 in the dominant model and rs284787 in the recessive model were not statistically significant. In con- trast, rs698 and rs1693482in ADH1C showed nonsignif- ADH7 Chr 4 100568489 rs3737482 T/C 0.41 0.48 202 283 100 320 566 284 0.81 ADH7ADH7 Chr 4 Chr 4 100552579 rs284787 100560666 rs1154460 T/C G/A 0.37 0.42 0.29 0.36 259 194 224 288 102 103 600 476 464 534 106 160 1.22 1.22 ADH1C Chr 4 100482988 rs1693482 C/T 0.09 0.05 494 82 9 1053 116 1 1.37 ADH1C Chr 4 100479812rs698 A/G 0.09 0.05 493 82 10 1053 116 1 1.38 ADH1B Chr 4 100458342 rs1229984 A/G 0.31 0.22 292 222 71 709 408 53 1.37 ADH4ADH4 Chr 4 Chr 4 100274885 rs4148887 100276021 rs3805322 C/T G/A 0.34 0.25 0.46 0.36 263 185 247 265 75 135 664 484 438 523 68 163 1.45 1.32 icant associations in the dominant and recessive models. Table 2. Gene Chromosome Location SNP Major/ *Adjusted for age, sex, smoking status, drinking status, and

Cancer Epidemiol Biomarkers Prev 2009;18(11). November 2009

3100 ADH Polymorphisms and Upper Aerodigestive Tract Cancer

Figure 1. Allelic ORs and their 95% CIs for UAT cancer by rs4148887 (ADH4) genotype and rs3737482 (ADH7). ORs were calculated using a conditional logistic regression model. Heavy drinking was defined as drinking ≥46 gethanol on ≥5 d/wk; high moderate drinkingas drinking<46 gethanol on ≥5 d/wk; moderate drinkingas drinking ≤4 d/wk; and never drinkingas never havingconsumed alcohol. Smokingstatus was evaluated as PYs.

These loci showed the suggestive dominant impact of were those for smoking status, except for rs698 and minor alleles. rs1693482in ADH1C.AllADH polymorphisms geno- We further evaluated these loci according to back- typed in this study were significantly associated with ground factors (Supplementary Figs. S1-6; Fig. 1). The ORs for esophageal cancer, whereas no significant asso- magnitudeofeffectoftheADH polymorphisms was ciations were observed between any ADH polymorphism more evident in subjects who were heavy drinkers and and larynx cancer. The magnitudes of effect for the ADH smokers with more PYs. P values for heterogeneity were polymorphisms in subjects heterozygous for ALDH2 were statistically significant for drinking status in each locus, as larger than those for subjects with homozygous ALDH2 Glu alleles. Figure 2shows the LD of the ADH4, ADH1B, ADH1C, and ADH7 polymorphisms. Higher LD was found among ADH4, ADH1B,andADH1C polymorphisms and rs284787 in ADH7 polymorphisms. Although rs1154460

Table 3. ORs of polymorphism effect of rs3737482 and haplotype effects

OR* 95% CI P

ADH7† rs37374821.241.06-1.45 0.006 Haplotype† CGAACT 1.00 Reference Haplotype† CAAACC 1.15 0.76-1.73 0.500 Haplotype† CAAACT 0.89 0.44-1.80 0.748 Haplotype† CAGACC 1.31 0.97-1.77 0.081 Haplotype† CAGACT 1.19 0.39-3.70 0.758 Haplotype† CAGGTC 1.48 0.36-6.15 0.588 Haplotype† CGAACC 1.13 0.77-1.69 0.536 Haplotype† CGAGTC 0.93 0.29-2.97 0.897 Haplotype† CGGACC 3.020.81-11.29 0.100 Haplotype† TAAACC 2.09 0.92-4.75 0.080 Haplotype† TAAACT 1.25 0.96-1.62 0.096 Haplotype† TAGACC 1.68 1.34-2.12 <0.001 Haplotype† TAGACT 1.90 0.98-3.68 0.058 Haplotype† TAGGTC 1.921.36-2.73 <0.001 Haplotype TAGGTT 3.05 1.52-6.15 0.002

NOTE: Haplotype frequencies of >0.001 were included in this model. rs3737482 Figure 2. LD within seven ADH gene polymorphisms. In- *Model was included for polymorphism, haplotype, age, sex, alcohol consumption, cumulative smoking, and ALDH2 genotype. † creasingshades of red indicate a highdegreeof correlation. Haplotype was constructed by following loci; rs4148887, rs3805322, Numbers in the panel indicate pair-wise D' values. rs1229984, rs698, rs1693482, and rs284787.

Cancer Epidemiol Biomarkers Prev 2009;18(11). November 2009

Cancer Epidemiology, Biomarkers & Prevention 3101

and rs3737482in ADH7 had higher LD, these two single be another possible reason. To date, few studies have in- nucleotide polymorphisms (SNP) had lower LD with the vestigated these differences in multiple ADH loci, but a other six SNPs in ADH4, ADH1B, ADH1C, and ADH7. broader understanding in other populations would help We further investigated haplotype effects to estimate in- elucidate mechanisms in the carcinogenesis of UATcancer. dependence effect of rs3737482in ADH7. Supplementary LD within the ADH gene polymorphisms was ob- Table S1 shows the estimated haplotypes and their fre- served. Despite this strong LD, rs1229984 in ADH1B quencies. Considering haplotypes, rs3737482in ADH7 and rs1693482in ADH1C showed a significant indepen- had a significant and independent effect for UAT cancer dent association with UAT cancer (14). In this study, we (Table 3). We also examined whether each locus of showed an independence of ADH7 rs3737482as well as ADH4 had an independent effect. Because polymorphisms ADH4 rs4148887. This supports substantial contribution in ADH4 were in LD with other neighboring loci, we eval- of ADH4 and ADH7 in addition to other ADHs in UAT uated the ORs in subjects with ADH1B homozygous major cancer risk. Further studies in various populations are alleles. We found a significant effect of rs4148887 in ADH4 warranted. (OR, 1.50; 95% CI, 1.12-2.01; P = 0.007), whereas we ob- Our study had several methodologic strengths. First, it served marginally significant association with rs3805322 was conducted in a single region in central Japan. Second, (OR, 1.28; 95% CI, 0.99-1.66; P = 0.057). potential confounding by age and sex was adjusted for by matching of these factors. Lastly, given that our allele frequencies were comparable with those previously reported Discussion in public databases such as HapMap JPT (16), bias in the distribution of selected polymorphisms was negligible. In this study, we found that multiple ADH gene poly- Several potential limitations of our study also warrant morphisms showed significant associations with UAT mention. One methodologic issue was the selection of the cancers. Moreover, rs3737482in ADH7 and rs4148887 in hospital-based noncancer patients as controls. However, ADH4 had a significant and independence effect on because cases and controls were selected from the same UAT cancers in this Japanese population. hospital and almost all patients lived in the Tokai area of We showed that ADH7 polymorphisms were signifi- central Japan, the internal validity of this case-control cantly associated with UAT cancer in a Japanese popula- study is likely to be acceptable. External validity has been tion. Of interest, these loci were different from that confirmed in our previous study (26). In addition, to dilute reported by Hashibe et al. (14) in Caucasians, rs1573496. any bias that might have resulted from the inclusion of a We observed rs1573496 was monomorphic in our popula- specific diagnostic group that is related to the exposure, tion. This might suggest the possibility that the risk of we did not set eligibility criteria for control diseases. Sec- UAT cancer is in fact affected by an unknown truly func- ond, the figures for self-reported life-style factors consid- tional polymophism in LD with both rs1573496 and those ered as potential confounders may be inaccurate. If we found within ADH7. Moreover, we observed a signif- present, however, any such misclassification would likely icant association with ADH4 rs4148887, whereas the be nondifferential, and would likely underestimate the study in Caucasian did not found any significant locus causal association. Lastly, the moderate number of cases in ADH4 (14). Different patterns of recombination across indicates the need for replication of our findings in a larger ethnicities might lead to the identification of different loci. study in a population with the same ethnicity. The relatively small size of our study warrants the need In conclusion, we showed that multiple ADH gene for evaluation in other Asian populations, which in turn polymorphisms are significantly associated with UAT more generally affirms the importance of cross-ethnic cancer in this Japanese population. Given the effect of collaborations (24). LD within the ADH gene cluster, the independence of In this study, the impact of ADH polymorphism was these SNPs was found in the Japanese population. How- stronger in heavy drinkers, heavy smokers, and subjects ever, evaluation is required in further large-scale studies with the ALDH2 Glu/Lys genotype. These factors have in other ethnicities. common characteristics in terms of acetaldehyde exposure. High alcohol intake results in high acetaldehyde Disclosure of Potential Conflicts of Interest exposure; owing to their low catalytic activity in the No potential conflicts of interest were disclosed. elimination of acetaldehyde, subjects with the ALDH2 Glu/Lys genotype are thus exposed to high concentra- tions of acetaldehyde following alcohol consumption. Acknowledgments Smoking increases acetaldehyde exposure via its effect The costs of publication of this article were defrayed in part by the of increasing salivary acetaldehyde levels after alcohol payment of page charges. This article must therefore be hereby advertisement intake (25). Given these findings, our observation of marked in accordance with 18 U.S.C. Section 1734 the marked impact of multiple ADH polymorphisms in solely to indicate this fact. We thank the doctors, nurses, technical staff, and hospital ad- these subpopulations may suggest the central role of ministration staff at ACCH for the daily administration of the acetaldehyde. Hospital-based Epidemiologic Research Program at Aichi Cancer A second interesting finding of this study was the dif- Center study, the staff of the Department of Thoracic Surgery ference in the magnitude of effect of ADH polymorphisms and Head and Neck Surgery, ACCH, for their support, and Dr. across cancer sites. In particular, the associations were Hashibe at IARC for her helpful comments and discussion. stronger for esophageal than for other cancers. Hashibe et al. (14) reported a similar phenomenon for ADH1B References ADH7 rs1229984 and rs1573496. A possible reason for 1. World Cancer Research Fund/American Institute for Cancer Re- thedifferencemightbethesamplesizeforeachcancer search. Food, Nutrition, Physical Activity, and the Prevention of Can- site. Or the variability in various types of exposure might cer: a Global Perspective. Washington (DC): AICR; 2007.

Cancer Epidemiol Biomarkers Prev 2009;18(11). November 2009

3102 ADH Polymorphisms and Upper Aerodigestive Tract Cancer

2. Seitz HK, Stickel F. Molecular mechanisms of alcohol-mediated carci- are associated with upper aerodigestive cancers. Nat Genet 2008;40: nogenesis. Nat Rev Cancer 2007;7:599–612. 707–9. 3. Li Y, Zhang D, Jin W, et al. Mitochondrial aldehyde dehydroge- 15. Hamajima N, Takezaki T, Tajima K. Allele frequencies of 25 poly- nase-2(ALDH2)Glu504Lys polymorphism contributes to the vari- morphisms pertaining to cancer risk for Japanese, Koreans and Chi- ation in efficacy of sublingual nitroglycerin. J Clin Invest 2006;116: nese. Asian Pac J Cancer Prev 2002;3:197–206. 506–11. 16. The International HapMap Consortium. A haplotype map of the hu- 4. Osier MV, Pakstis AJ, Soodyall H, et al. A global perspective on genetic man genome. Nature 2005;437:1299–320. variation at the ADH genes reveals unusual patterns of linkage dis- 17. Matsuo K, Hamajima N, Shinoda M, et al. Gene-environment interac- equilibrium and diversity. Am J Hum Genet 2002;71:84–99. tion between an aldehyde dehydrogenase-2(ALDH2)polymorphism 5. Hashibe M, Boffetta P, Zaridze D, et al. Evidence for an important role and alcohol consumption for the risk of esophageal cancer. Carcino- of alcohol- and aldehyde-metabolizing genes in cancers of the upper genesis 2001;22:913–6. aerodigestive tract. Cancer Epidemiol Biomarkers Prev 2006;15: 18. Yokoyama A, Muramatsu T, Ohmori T, Higuchi S, Hayashida M, Ishii 696–703. H. Esophageal cancer and aldehyde dehydrogenase-2genotypes in 6. Matsuo K, Hiraki A, Hirose K, et al. Impact of the alcohol-dehydrogenase Japanese males. Cancer Epidemiol Biomarkers Prev 1996;5:99–102. (ADH) 1C and ADH1B polymorphisms on drinking behavior in non- 19. Hamajima N, Matsuo K, Saito T, et al. Gene-environment interactions alcoholic Japanese. Hum Mutat 2007;28:506–10. and polymorphism studies of cancer risk in the Hospital-based Epide- 7. Yokoyama A, Kato H, Yokoyama T, et al. Genetic polymorphisms of miologic Research Program at Aichi Cancer Center II (HERPACC-II). alcohol and aldehyde dehydrogenases and glutathione S-transferase Asian Pac J Cancer Prev 2001;2:99–107. M1 and drinking, smoking, and diet in Japanese men with esophageal 20. Tajima K, Hirose K, Inoue M, Takezaki T, Hamajima N, Kuroishi T. squamous cell carcinoma. Carcinogenesis 2002;23:1851–9. A model of practical cancer prevention for out-patients visiting a 8. Asakage T, Yokoyama A, Haneda T, et al. Genetic polymorphisms of hospital: the Hospital-based Epidemiologic Research Program at Aichi alcohol and aldehyde dehydrogenases, and drinking, smoking and diet Cancer Center (HERPACC). Asian Pac J Cancer Prev 2000;1:35–47. in Japanese men with oral and pharyngeal squamous cell carcinoma. 21. Lin DY, Zeng D. Likelihood-based inference on haplotype effects in Carcinogenesis 2007;28:865–74. genetic association studies. J Am Stat Assoc 2006;101:89–104. 9. Peters ES, McClean MD, Liu M, Eisen EA, Mueller N, Kelsey KT. The 22. Spinka C, Carroll RJ, Chatterjee N. Analysis of case-control studies of ADH1C polymorphism modifies the risk of squamous cell carcinoma genetic and environmental factors with missing genetic information of the head and neck associated with alcohol and tobacco use. Cancer and haplotype-phase ambiguity. Genet Epidemiol 2005;29:108–27. Epidemiol Biomarkers Prev 2005;14:476–82. 23. Marchenko Yulia V, Carroll Raymond J, Lin Danyu Y, Amos 10. Homann N, Stickel F, Konig IR, et al. Alcohol dehydrogenase 1C*1 Christopher I, Gutierrez Roberto G. Semiparametric analysis of allele is a genetic marker for alcohol-associated cancer in heavy case-control genetic data in the presence of environmental factors. drinkers. Int J Cancer 2006;118:1998–2002. Stata J 2008;8:305–33. 11. Solomon PR, Selvam GS, Shanmugam G. Polymorphism in ADH and 24. Hashibe M, Brennan P, Benhamou S, et al. Alcohol drinking in never MTHFR genes in oral squamous cell carcinoma of Indians. Oral Dis users of tobacco, cigarette smoking in never drinkers, and the risk of 2008;14:633–9. head and neck cancer: pooled analysis in the International Head and 12. Brennan P, Lewis S, Hashibe M, et al. Pooled analysis of alcohol de- Neck Cancer Epidemiology Consortium. J Natl Cancer Inst 2007;99: hydrogenase genotypes and head and neck cancer: a HuGE review. 777–89. Am J Epidemiol 2004;159:1–16. 25. Homann N, Tillonen J, Meurman JH, et al. Increased salivary acetal- 13. Yang CX, Matsuo K, Ito H, et al. Esophageal cancer risk by ALDH2 dehyde levels in heavy drinkers and smokers: a microbiological ap- and ADH2polymorphisms and alcohol consumption: exploration of proach to oral cavity cancer. Carcinogenesis 2000;21:663–8. gene-environment and gene-gene interactions. Asian Pac J Cancer 26. Inoue M, Tajima K, Hirose K, et al. Epidemiological features of first- Prev 2005;6:256–62. visit outpatients in Japan: comparison with general population and 14. HashibeM,McKayJD,CuradoMP,etal.MultipleADHgenes variation by sex, age, and season. J Clin Epidemiol 1997;50:69–77.

Cancer Epidemiol Biomarkers Prev 2009;18(11). November 2009

Impact of Multiple Alcohol Dehydrogenase Gene Polymorphisms on Risk of Upper Aerodigestive Tract Cancers in a Japanese Population

Isao Oze, Keitaro Matsuo, Takeshi Suzuki, et al.

Cancer Epidemiol Biomarkers Prev 2009;18:3097-3102. Published OnlineFirst October 27, 2009.

Updated version Access the most recent version of this article at: doi:10.1158/1055-9965.EPI-09-0499

Supplementary Access the most recent supplemental material at: Material http://cebp.aacrjournals.org/content/suppl/2009/10/27/1055-9965.EPI-09-0499.DC1

Cited articles This article cites 25 articles, 3 of which you can access for free at: http://cebp.aacrjournals.org/content/18/11/3097.full#ref-list-1

Citing articles This article has been cited by 2 HighWire-hosted articles. Access the articles at: http://cebp.aacrjournals.org/content/18/11/3097.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cebp.aacrjournals.org/content/18/11/3097. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.