Sequence Variants of NAT1 and NAT2 and Other Xenometabolic Genes and Risk of Lung and Aerodigestive Tract Cancers in Central Europe

Tobacco smoke contains an extensive cocktail of highly carcinogenic chemicals. Individuals with a slower elimination rate of the chemicals in tobacco smoke may have increased exposure to their carcinogenic properties compared with those with a faster rate. Polymorphisms that alter the function of the genes involved in the activation or the detoxification of the chemical carcinogens in tobacco smoke can potentially influence an individual's risk of developing a tobacco-related cancer. To test this hypothesis, we have genotyped polymorphisms in 16 genes involved in metabolism of chemical carcinogens in a Central and Eastern European case-control study comprising 2,250 lung cases, 811 upper aerodigestive cancer (UADT) cases, and 2,704 controls. The N-acetyltransferase (NAT) genes were the most implicated in risk, with the NAT1*10 haplotype showing an inverse association in lung cancer, in both heterozygote carriers [odds ratio (OR), 0.81; 95% confidence interval (95% CI), 0.70-0.93] and homozygote carriers (OR, 0.70; 95% CI, 0.48-1.01), suggesting a genotype dose response (P < 0.001). In UADT cancer, a similar inverse association was noted in NAT1*10 although only in heterozygotes (OR, 0.78; 95%CI, 0.65-0.95). In NAT2, when considering the individuals inferred acetylator phenotypes based on their NAT2 diplotype, “slow” acetylators compared with intermediate or fast acetylators showed no association with risk. None of the other 14 genes provided robust evidence of an association for either lung or UADT cancer. We therefore conclude that, of the genetic variation studied, NAT1 gene was the most likely candidate to influence the risk of developing a tobacco-related cancer. (Cancer Epidemiol Biomarkers Prev 2008;17(1):141–7)

Tobacco smoking is the predominant cause of lung and upper aerodigestive tract (UADT) cancers (1). The mutagenic chemical carcinogens in tobacco smoke are eliminated from the human body via a series of enzymes that metabolize the carcinogenic compounds. Genetic variation in functionally important regions of xenometabolic genes seem to influence the rate of action of these enzymes (2-4). An individual's genetic profile of xenometabolic alleles will have a role in determining the rate at which carcinogens are eliminated and therefore the extent of carcinogen exposure. Individuals who have inherited alleles that result in higher degree of carcinogen exposure are expected to have a higher cancer risk relative to those who inherited lower exposure alleles (2-6).

We hypothesized that polymorphisms in the genes encoding the proteins involved in either the activation (phase I) and/or the detoxification (phase II) of chemical carcinogens in tobacco smoke will influence risk of developing lung or UADT cancers. To test this hypothesis, we have genotyped polymorphisms in 16 xenometabolic genes in a large case-control study of lung cancer patients and UADT cancer cases.

The study was conducted in 15 centers in six countries of Central and Eastern Europe, including Czech Republic (Prague, Olomouc, Brno), Hungary (Borsod, Heves, Szabolcs, Szolnok, Budapest), Poland (Warsaw, Lodz), Romania (Bucharest), Russia (Moscow), and Slovakia (Banska Bystrica, Bratislava, Nitra). Each center followed an identical protocol and was responsible for recruiting a consecutive group of newly diagnosed cases of lung cancer and UADT cancer, as well as a comparable group of population or hospital controls. All subjects were recruited between 1998 and 2003. Lung cancer cases were recruited in all centers, whereas upper aerodigestive cancer cases were recruited only in the centers in Russia, Czech Republic, Romania, Slovakia, and Poland. A total of 2,250 lung cancer cases, 811 upper aerodigestive cancer cases (168 oral cavity, 113 pharynx, 326 larynx, 176 esophagus, and 28 cases of cancers from overlapping oral/pharynx sites), were recruited. All cases were histologically confirmed, and upper aerodigestive cases were restricted to squamous cell carcinoma, the predominant histologic type. Controls in all centers except Warsaw were chosen among subjects admitted as inpatients or outpatients in the same hospital as the cases with conditions unrelated to tobacco, including minor surgical conditions, benign disorders, common infections, eye conditions (except cataract or diabetic retinopathy), and common orthopedic diseases (except osteoporosis). In Warsaw, population controls were selected by random sampling from the Polish Electronic List of Residents. Cases and controls were frequency matched by sex, age (±3 years), center, referral (or of residence) area, and period of recruitment (±6 months). The participation rates for both cases and controls were more than 80% for both cases and controls in all centers. Both cases and controls underwent an identical interview based on the same questionnaire. Written consent for participation was obtained from all study subjects and ethical approval has been obtained for all study centers as well as at IARC, the coordinating center. The characteristics of the study subjects are as displayed in Table 1. Further details on the questionnaire, as well as case and control recruitment, have been reported elsewhere (7).

After DNA extraction, genotyping was done by the 5′ nuclease assay [TaqMan, Applied Biosystems; CYP1A1: rs4646903; CYP1A2: rs762551, rs2069514, rs2470890; CYP1B1: rs1800440, rs1056836; CYP2A6: rs1801272; CYP2C9: rs1057910, rs1799853; CYP2E1: rs6413420, rs3813867; CYP3A4: rs2740574; MEH/EPHX1: rs1051740, rs2234922; COMT: rs4680; NQO1: rs1800566, rs4986998; GSTM: rs1799735, rs7483; GSTP1: rs947894; UGT1A7: rs17868323; NAT1: rs4986989 (A40T); NAT2: rs1801280 (I114T), rs1799929 (C481T), rs1799930 (R197Q), rs1208 (L268R), rs1041983 (C282T), rs1799931 (G286E), MGB Eclipse (Nanogen Technologies); CYP1A1: rs1048943; NAT1: rs13249533 (R187Q), rs1057126 (T1088A), rs15561 (C1095A)], or gel discrimination for GSTM1 and GSTT1 alleles, consisting of the deletion of the entire gene. Sequences of primers and probes for the variants were obtained from the SNP500 project¹³

and are available on request. We attempted to choose single nucleotide polymorphisms (SNP) for this study that have a known or inferred effect on the function of the xenometabolic protein and/or that have been implicated in disease based on previous association studies. The order of DNA samples from cases and controls was randomized on PCR plates, and duplicate genotyping was done for a random 10% of the samples for quality control. Genotyping call rates were similar for cases and controls (≥94%). All duplicate quality control genotypes showed >99.5% concordance. The distributions of the SNPs and haplotypes conformed to that expected by Hardy-Weinberg equilibrium in the control population (P > 0.01; Table 2).

All the analyses were conducted with SAS software. We calculated odds ratios (OR) and 95% confidence intervals (95% CI) for SNPs after adjusting for potential confounders, including country of residence, age, sex, smoking pack-years, and years of alcohol for upper aerodigestive cancer, using unconditional logistic regression. Linkage disequilibrium between variants was tested by the measures of D′ and R².

For the determination of the NAT alleles or haplotypes, haplotype dosages were calculated using the tagSNP software (8) to indicate each individual's probability of being heterozygote (Bstat) or homozygote (Cstat) and for a codominant model (Hstat) for each haplotype. The tagSNP dosage variables were then implemented as covariates in the unconditional logistic regression model and risk estimates for haplotypes were calculated using all other haplotypes (i.e., those with zero dosage) as a reference category. Three P values were calculated for heterozygote haplotype carriers, for homozygote haplotype carriers, and for a codominant model. We classified individuals as “fast” or “slow” acetylators using predicted phenotypes. NAT2*4 and NAT2*12 alleles or haplotypes were classified fast acetylators, and NAT2*7B, NAT2*6A NAT2*5C NAT2*5A, and NAT2*5B were classified as slow acetylators, as described elsewhere (9, 10).

Finally, our previous exploratory analysis among lung cancer patients with a young onset of disease served as the basis of hypothesis generation for the current investigation (11). Incorporation of prior probabilities of an association and correction for multiple testing was carried out using a Bayesian false discovery probability (12). Bayesian false discovery probability gives the probability of no association given the data and a specified prior on the presence of an association, and was developed to provide a method for flagging associations as noteworthy.

Main effects in terms of genotype dosage for lung cancer and UADT cancer for the SNPS in the metabolizing enzymes are displayed in Table 2. No associations were observed for any of the genotypes with the exception of the N-acetyltransferase (NAT) genes. In NAT1, heterozygotes for NAT1*10 haplotype were inversely associated with both lung and UADT cancers with ORs of 0.81 (95% CI, 0.71-0.93) and 0.78 (95% CI, 0.65-0.95), respectively. There was also evidence consistent with a dose response in lung cancer for the NAT1*10 haplotype (P < 0.001) but not for UADT (P = 0.40). The NAT1*11 haplotype was also associated with both cancers, with heterozygotes having an OR of 1.30 (95% CI, 1.03-1.64) for lung cancer and an OR of 1.47 (95% CI, 1.08-1.99) for UADT cancer. The NAT1*4 haplotype also showed a borderline inverse association with risk in UADT (OR, 0.69; 95% CI, 0.49-0.98; P = 0.04).

We also investigated the effects of the NAT1*10 haplotype among histologic subtypes, country of origin, and smoking status (never, ever, current; Fig. 1). The effects did not seem clearly confined to a single histologic subtype or smoking intensity subgroup, although the power to detect heterogeneity was minimal.

In the NAT2 gene, we noted an association between the NAT2*5B haplotype and lung cancer risk, conferring an OR of 0.85 (95% CI, 0.74-0.97; P = 0.02) for heterozygotes. NAT2*5B and NAT1*10 are in linkage disequilibrium, but the association with NAT2*5B was unchanged when adjusting for the NAT1*10. When considering predicted acetylator phenotypes, we observed no association between type of acetylator and lung or UADT cancer risk, with slow acetylators having an OR of lung cancer of 0.94 (95% CI, 0.84-1.06) and an OR of 0.97 for UADT (95% CI, 0.82-1.14) when compared with fast and intermediate acetylators.

As observed by others (13), there is linkage disequilibrium between the NAT1 and NAT2 genes. We therefore considered the six NAT1/NAT2 locus-wide haplotypes that had a carrier frequency of >4% haplotypes in the context of lung and UADT cancer risk (Table 3). One locus-wide haplotype with the combination of NAT1*10 and NAT2*5B haplotypes (frequency 0.07) was more strongly associated with lung cancer risk than each of the NAT1*10 and NAT2*5B individually, with heterozygotes having an OR of 0.63 (95% CI, 0.48-0.82) with evidence in favor of a genotype dosage effect (P = 0.0003). By contrast, the haplotype containing the putative “fast” NAT acetylator haplotypes, NAT1*10 and NAT2*4, did not present with stronger evidence for association with lung cancer or UADT cancer than that observed with the NAT1*10 alone.

Regarding the 14 other metabolism genes, we observed a protective effect against UADT cancer for individuals who were homozygous for the CYP1B1 N453S polymorphism (OR, 0.53; 95% CI, 0.29-0.97) and an increased risk of UADT cancer for individuals who were homozygous for the MEH H139R polymorphism (OR, 1.60; 95% CI, 1.06-2.41), although the P value for trend for neither was significant. We did not observe any main effects for polymorphisms in the GSTs, CYPs, MEH, COMT, NQO1, or UGT for lung cancer (Table 2). There was no evidence for heterogeneity among the different lung cancer subtypes or UADT subtypes (data not shown).

Using Bayesian false discovery probability to determine the noteworthiness of the associations that achieved a P < 0.05, the association between the NAT1*10 haplotype and NAT1*10/NAT2*5B and lung cancer remained noteworthy with a prior of 1% or greater (Table 4).

Overall, we found little evidence for association between lung and UADT cancer and the genetic variation in the xenometabolic genes tested by this study. The exception was in the NAT genes, where we provide further evidence for a role of the NAT genes in lung and UADT cancers. The NAT1 and NAT2 enzymes have similar overlapping functions in the metabolism of aromatic and heterocyclic amines. The NATs are involved in detoxification of certain carcinogenic amines found in tobacco smoke, and the activation to reactive carcinogenic electrophilic intermediates of others, making them relevant to tobacco-related cancers (13, 14). The expression profiles of NAT1 and NAT2 are complex (14, 15), but NAT2 function is believed to be limited to xenobiotic metabolism in specific tissues and NAT1 is proposed to act in a broader range of cell types and functions (14, 16). The functional consequence of NAT2 genetic variation is well described, enabling individuals to be classified as NAT2 fast, intermediate, or slow acetylators (3, 6). Individuals with a NAT2 slow acetylator phenotype are thought to have a decreased NAT2 metabolic capacity and therefore have a higher risk of cancer. Consistent with this hypothesis, NAT2 slow acetylator haplotypes have been associated with an increased risk of bladder cancer, an effect that seems modulated by smoking (5). By contrast, and consistent with our results here, a recent meta-analysis found no association with the NAT2 slow acetylators when comparing them with the fast and intermediate acetylators and lung cancer risk (17).

Unlike NAT2, the metabolic phenotype of common NAT1 haplotypes is less clear. NAT1*10 has been described a fast acetylator, although experimental results have been conflicting (3, 18). A fast acetylator phenotype, with greater metabolic capacity, leading to carriers having a lower exposure to the carcinogenic aromatic and heterocyclic amines found in tobacco smoke, would be consistent with our observations of an inverse association with lung and UADT cancer risk.

Previous studies on the NAT1*10 haplotype and lung cancer risk have been limited by study size and have produced heterogenous results. Bouchardy et al. (19) studied 150 lung cancer patients and 172 controls from France and similarly found an inverse association with the NAT1*10 haplotype in both heterozygotes (OR, 0.6; 95% CI, 0.4-1.1) and homozygotes (OR, 0.2; 95% CI, 0.03-0.74). A nonsignificant inverse association between the NAT1*10 haplotype was noted in 257 lung cases matched to controls in the Physicians Health cohort study (OR, 0.87; 95% CI, 0.5-1.5; ref. 19). Wikman et al. (9) studied 392 lung cancer cases and 351 controls and found no clear effect when comparing slow carriers with fast (NAT1*10) carriers (OR, 0.84; 95% CI, 0.34-2.1) or with intermediate carriers (OR, 1.22; 95% CI, 0.87-1.7), but noted an increase in risk when considering adenocarcinoma only. One small study of 45 cases and 45 controls observed an increase in risk with the NAT1*10 haplotype (OR, 3.7; 95% CI, 1.2-16.0; ref. 21). Our previous study reported an inverse association with the NAT1*10 haplotype in a small subset of the young lung cancer cases included in the current study (11). It is worth noting that when excluding young onset cancers, the current study independently replicates this finding (P for trend = 0.005; data not shown).

As a consequence of linkage disequilibrium between the two NAT1 and NAT2 genes, similar to other reports (13), the NAT1*10 (fast) haplotype was found in three common locus-wide haplotypes in combination with the NAT2*4 (fast), NAT2*5B (slow), and NAT2*6 (fast) haplotypes in our population. We postulated that individuals that have a “fast-fast” acetylator locus-wide haplotype, NAT1*10/NAT2*4, might be at further decreased risk than just that observed with the NAT1*10 alone. However, the inverse association noted with NAT1*10 did not seem to be stronger when considering the NAT1*10/NAT2*4 fast/fast haplotype. One NAT locus-wide haplotype, NAT1*10/NAT2*5B, showed stronger effect than that observed with either haplotype individually. In terms of phenotype, it is unclear how a “fast/slow” NAT1/NAT2 acetylator could account for this association. One explanation is this may be simply exaggerated by chance. An alternate possibility is that unknown genetic variation at the NAT locus, partially tagged by the NAT1*10 and NAT2*5B but more efficiently tagged by the NAT1*10/NAT2*5B locus-wide haplotype, is influencing lung cancer risk. This exploratory result clearly requires independent confirmation.

In conclusion, of the genetic variation examined by this study, haplotypes in the NAT genes and in particular NAT1*10, seem to be the most relevant to lung and UADT cancer. The association between NAT1*10 and lung cancer remains noteworthy for priors of 1% or greater (Table 4). The fact that our observations seem consistent with the biological action of this gene and several smaller studies have noted similar results would support a prior of 1% or more. Further study of the NAT1*10 haplotype in other large studies or pooling results from multiple studies, and extensive investigation of the genetic variation in the regions responsible for alternatively spliced forms of the NAT genes, would seem warranted.

We thank all participants of the Central European study and staff in the collaborating centers.