Variants of DNA Repair Genes and the Risk of Biliary Tract Cancers and Stones: A Population-Based Study in China

Biliary tract cancers, which encompass tumors of the gallbladder, extrahepatic ducts, and ampulla of Vater, are relatively rare tumors with a high fatality rate. Other than a close link with gallstones, the etiology of biliary tract cancers is poorly understood. We conducted a population-based case-control study in Shanghai, China, to examine whether genetic variants in several DNA repair genes are associated with biliary tract cancers or biliary stones. Genomic DNA from 410 patients with biliary tract cancers (236 gallbladder, 127 bile duct, and 47 ampulla of Vater), 891 patients with biliary stones, and 786 healthy subjects randomly selected from the Shanghai population were genotyped for putative functional single nucleotide polymorphisms in four DNA repair genes (MGMT, RAD23B, CCNH, and XRCC3). Of the five single nucleotide polymorphisms examined, only one (MGMT EX5-25C>T, rs12917) was associated with biliary tract cancer. Independent of gallstones, subjects carrying the CT genotype of the MGMT EX5-25C>T marker had a significantly reduced risk of gallbladder cancer [odds ratio (OR), 0.63; 95% confidence interval (95% CI), 0.41-0.97; P = 0.02] and nonsignificant reduced risks of bile duct (OR, 0.61; 95% CI, 0.35-1.06) and ampulla of Vater (OR, 0.85; 95% CI, 0.39-1.87) cancers. However, this marker was not associated with biliary stones, and the other markers examined were not significantly associated with either biliary tract cancers or stones. Findings from this population-based study in Shanghai suggest that MGMT gene variants may alter susceptibility to biliary tract cancer, particularly gallbladder cancer. Confirmation in future studies, however, is required. (Cancer Epidemiol Biomarkers Prev 2008;17(8):2123–7)

Biliary tract cancers, encompassing cancers of the gallbladder, extrahepatic bile ducts, and ampulla of Vater, are rare malignancies with poor prognosis. Apart from gallstones, the risk factors for biliary tract cancers are not clearly defined. Because each of the three cancer subsites is associated with distinct molecular changes, as well as varying geographic and ethnic patterns, it has been suggested that there is a different etiology for each of the three subsites (1).

Disruptions in genomic stability and integrity due to DNA damage compromise the accuracy of DNA replication, resulting in gene rearrangements, translocations, amplifications, and deletions. These changes can contribute to cancer development (2, 3). DNA damage and mutations are induced by a variety of endogenous processes and exogenous factors, including UV light, cigarette smoke, and dietary elements (4). Under normal circumstances, DNA damage and normal replication errors are corrected by DNA repair mechanisms.

There are several known DNA repair pathways providing distinct but overlapping protection against exposures. The base excision repair pathway is involved in the removal of simple base modifications and oxidative DNA damage, such as single-strand breaks, nonbulky adducts, and alkylation adducts (5). The nucleotide excision repair pathway primarily removes and repairs bulky adducts but has also been reported to play a role in the repair of oxidative DNA damage (6, 7). The homologous recombination pathway repairs double-strand DNA breaks in the S-G₂ phases of the cell cycle (8). O⁶-methylguanine-DNA methyltransferase (MGMT, also named O⁶-alkylguanine-DNA alkyltransferase, AGT) is the principal mechanism for repairing O⁶-alkylguanine adducts (9); it binds to and removes alkyl groups from the O⁶ position of guanine in a single step.

It is unclear which DNA repair pathways or enzymes are most important for protection against biliary tract cancers. Previously, we have reported that single nucleotide polymorphisms (SNP) in certain base excision repair genes, such as XRCC1 and APEX1, are associated with biliary tract cancer (10). To gain further insight into the role of DNA repair in biliary tract cancer etiology, we investigated common putative functional SNPs in four more DNA repair genes, including MGMT for the repair of O⁶-alkylguanine adducts, radiation gene (RAD23B), and cyclin H (CCNH) in the nucleotide excision repair pathway, and the X-ray repair cross-complementary group (XRCC3) for the repair of double-strand DNA breaks, and their associations with biliary tract cancer risk in a population-based case-control study in Shanghai, China.

The study protocol was approved by the institutional review boards of the National Cancer Institute and the Shanghai Cancer Institute. All participants provided written informed consent. Details of this population-based, case-control study have been reported previously (11-14). Briefly, cancer cases were permanent residents of urban Shanghai, 35 to 74 years of age, and newly diagnosed with biliary tract cancer between 1997 and 2001. Cancer cases were identified through a rapid reporting system established by the Shanghai Cancer Institute and 42 collaborating hospitals, which captured more than 95% of the biliary cancer cases diagnosed in urban Shanghai during the study period. Biliary stone patients without a history of cancer were selected by frequency matching to cancer cases on age (5-year intervals), gender, and hospital. Population controls without a history of cancer were randomly selected from all permanent residents listed in the Shanghai Resident Registry and frequency-matched to cancer cases on age (5-year intervals) and gender. Of the eligible cancer cases and controls, 95% and 82%, respectively, agreed to participate in the study.

For cancer cases, more than 70% of subjects had positive pathology, confirming their diagnosis. For each case with pathology materials, a second independent review and a consensus review were carried out to confirm the diagnosis. For each subject without pathologic confirmation, medical records, surgical reports, and imaging data, including magnetic resonance imaging, endoscopic retrograde cholangiopancreatography, and computed tomography, were reviewed to confirm their diagnosis. Biliary stone cases were confirmed by review of abdominal ultrasound, endoscopic retrograde cholangiopancreatography films, medical records, and surgical records, or by pathologic material for those who underwent a cholecystectomy.

Gallstone status was assessed in nearly all biliary tract cancer cases and population controls. Gallstones were identified among cancer cases by self-reported history, surgical reports, or imaging results from magnetic resonance imaging, endoscopic retrograde cholangiopancreatography, computed tomography, or ultrasound. Among population controls, gallstones were identified by self-reported history or by abdominal ultrasound for those who gave consent for the procedure (85% of all population controls).

In-person interviews were conducted with each participant by trained interviewers using a structured questionnaire to collect information on demographic characteristics, medical histories, and lifestyle factors, including diet and physical activity. Weight and height were measured at interview. Cases were interviewed within 3 weeks of diagnosis. All interviews were taped and verified for accuracy of interview protocol and coding. The response rate for interviews was more than 95% among cases and 82% among controls. To assess the reproducibility of interview responses, we randomly selected 5% of the study subjects for a second interview 3 months after the initial interview. The concordance between the original and follow-up interviews was more than 90%.

Overnight fasting blood samples were collected from the participants who gave consent for blood collection (more than 80% of the total group). For the current study, genomic DNA was available for the genotyping of 410 patients with biliary tract cancers (236 gallbladder, 127 bile duct, and 47 ampulla of Vater), 891 patients with biliary stones (670 gallstones and 221 bile duct stones), and 786 healthy subjects randomly selected from the Shanghai population. Genomic DNA was extracted from buffy coat using the standard phenol-chloroform method. Four candidate genes were selected for their roles in DNA repair and their potential effects on biliary disease pathogenesis. SNPs were selected based on their putative functional significance, having a variant allele frequency of at least 5% among Asians as reported by the National Cancer Institute SNP500Cancer Database¹⁰

(15), and a validated assay at the National Cancer Institute Core Genotyping Facility.¹¹ The SNPs were genotyped at the Core Genotyping Facility using the TaqMan assay (Applied Biosystems)¹⁰: O⁶-methylguanine-DNA-methyltransferase, MGMT EX5-25C>T (rs12917), RAD23B IVS5-15A>G (rs1805335), RAD23B EX7+65C>T (rs1805329), cyclin H, CCNH EX7+49T>C (rs2266690), and the X-ray repair cross-complementary group XRCC3 EX8-53C>T (rs861539). To evaluate the quality of genotyping, DNA aliquots were shipped to the Core Genotyping Facility with 80 internal blind duplicates from four individuals spaced at varying intervals between the study samples. The failure rate for genotyping among all subjects was less than 2%, and the concordance between quality control samples was more than 99%.

Genotype frequencies for each polymorphism were examined among population controls for deviation from fitness for Hardy-Weinberg equilibrium, using the χ² test. Unconditional logistic regression was used to calculate odds ratios (OR) and 95% confidence intervals (95% CI) to assess the relationship of each polymorphism with biliary tract cancer and stone risk at each anatomic subsite, adjusting for age and gender. Although gallstones may be an intermediate step in the causal pathway between genetics and biliary tract cancer, genetic variants may affect cancer risk independently of stones, especially for bile duct and ampulla of Vater cancers, which have a much lower prevalence of stones than gallbladder cancer. Thus, additional logistic regression models were run with further adjustment for biliary stone status to evaluate potential confounding by gallstones. Estimates were calculated for the heterozygous, homozygous variants, and their combined genotypes, with the homozygous genotype of the more frequent allele as the reference category for each marker. Bile duct and ampulla of Vater cancer cases were compared with all population controls. Gallbladder cancer cases were compared with controls without a history of cholecystectomy, and gallstone and bile duct stone cases were compared with controls without biliary stones.

Because preserved foods contain nitrosamines whose DNA-damaging effects are mitigated by DNA repair mechanisms, we examined the joint effects of the DNA repair variants and the consumption of preserved foods on biliary tract cancer and stone risk. Preserved food consumption (g/d) was dichotomized into <14.4 and ≥14.4, based on the median consumption among the control subjects. We also examined whether the effect of genetic variants on biliary tract cancer or stone risk was modified by other potential risks or protective factors, such as regular cigarette smoking (no, yes), body mass index (<23, ≥23 kg/m²), intake of red meat (<30.1 g/d, ≥30.1 g/d), allium vegetables (<9.7 g/d, ≥9.7 g/d), all vegetables (<720.7 g/d, ≥720.7 g/d), all fruits (<653.6 g/d, ≥653.6 g/d), and gallstones for cancer risk only (no, yes). To maximize statistical power in exploring potential interactions, we assumed the dominant model, with heterozygous and homozygous variant genotypes combined; selected characteristics were categorized dichotomously based on the median among the controls. Multiple interactions were assessed, using a genotype exposure cross-product term in the logistic regression models. In addition, we calculated the false-positive reporting probability to assess the likelihood of having false-positive findings (16). All analyses were done using SAS version 8.0. All reported P values are two sided.

Selected characteristics of the study subjects are shown in Table 1. Compared with controls, more gallbladder cancer cases (72.6%, P = 0.001) and gallstone cases (66.7%, P = 0.001) were female. Gallbladder cancer, gallstone, and bile duct stone cases had higher body mass index than controls. All three types of biliary tract cancer had significantly higher proportion of gallstones (59.6-84.8%) compared with controls (24.7%, P < 0.0001).

Among population controls, the genotype distribution for each of the five SNPs did not deviate from HWE. The risk estimates for biliary tract cancers and biliary stones associated with these DNA repair SNPs are shown in Table 2. Of the five SNPs tested, only the MGMT EX5-25C>T marker (rs12917) was associated with biliary tract cancers. Specifically, carriers of the CT genotype had a significantly reduced risk of gallbladder cancer (OR, 0.63; 95% CI, 0.41-0.97) compared with those with the CC genotype. This finding for the MGMT EX5-25C>T marker and gallbladder cancer was noteworthy at the 0.5 false-positive reporting probability level, assuming a prior probability of association of 0.1. This marker was also associated with a lower risk of bile duct cancer (OR, 0.61; 95% CI, 0.35-1.06) and ampulla of Vater cancer (OR, 0.85; 95% CI, 0.39-1.87), but the associations were not statistically significant. The associations between the MGMT EX5-25C>T marker and biliary tract cancers changed a little after further adjustment for biliary stones. MGMT EX5-25C>T was not associated with biliary stones nor were any of the other markers examined.

We examined the effect of the MGMT EX5-25C>T variant by subjects' smoking behavior, body mass index, gallstone status, and consumption of preserved food, red meat, vegetables, and fruits. No significant interaction by these lifestyle factors was found (P_interaction range = 0.1-0.9). In addition, we genotyped four additional SNPs, LIG3 EX18-75G>A (rs3136025), MGMT EX7+13A>G (rs2308321), MGMT EX7+119A>G (rs2308327), and RAD23B IVS5-57T>A (rs1805331), which we found to be nonpolymorphic or with a minor allele frequency less than 1% in this Asian population.

In this population-based study, we found that the MGMT EX5-25C>T marker in the DNA repair pathway was associated with a reduced risk of biliary tract cancer. No significant associations with biliary tract cancers were observed for the other four SNPs examined, although small-to-modest effects of the markers cannot be ruled out because of limited sample sizes. Theses results, although in need of confirmation, suggest that variants in MGMT may play a role in biliary tract cancer etiology, particularly for gallbladder cancer.

The MGMT EX5-25C>T marker has been associated with reduced risks of oral (17) and colon cancers (18), but increased risks of prostate (19), lung (20), and bladder cancers (21), and no association with gastric (22) and rectal cancers (18) or colorectal polyps (23). The reasons for the differential effect of this marker in various cancers are unclear, but they may be related in part to substrate (exposure) specificity for the different organ sites and differences in the frequencies of the variant allele in the study populations.

Our observation that the MGMT EX5-25C>T marker is associated with a reduced gallbladder cancer risk is biologically plausible. The protein encoded by the MGMT gene, O⁶-methylguanine-DNA methyltransferase (also named O⁶-alkylguanine-DNA alkyltransferase), repairs DNA damage such as O⁶-alkylguanine DNA adducts (24, 9) caused by alkylating agents, including N-nitroso compounds. These DNA adducts, if left unrepaired, cause initiation of mutations and cellular cytotoxic actions (25). In vitro and in vivo data showed that MGMT plays a critical role in protecting cells from the mutagenic or carcinogenic action of alkylating agents (24, 9). It is important to note that the MGMT EX5-25C>T variant may affect MGMT function and gallbladder cancer risk indirectly, as it is in linkage disequilibrium with 12 other polymorphisms on the MGMT gene (26).

The strengths and limitations of the study should be noted. Selection bias is minimal due to the high case ascertainment (more than 95%) and high response rate (more than 85%). Misclassification of cancer and stone cases was minimal due to the detailed review of pathology and clinical data. Misclassification of genotypes was likewise minimal, judging from the high concordance and reproducibility of genotyping results among quality control samples. Despite being the largest biliary tract cancer study to date, the study had limited statistical power to evaluate the main effects, especially for ampulla of Vater cancer, due to the rarity of some high-risk alleles and the limited number of cases by anatomic subsite. Because of this, we cannot rule out the possibility of small-to-modest effects of these markers. Because gene coverage was limited, the effect of potentially important markers in the DNA repair pathway may have been missed. Also, the observed effect of the MGMT EX5-25C>T marker on biliary tract cancer may represent the effect of other functional SNPs that are in linkage disequilibrium with the MGMT EX5-25C>T locus. Finally, because the population in Shanghai is ethnically homogeneous, our results may not be generalizable to other populations.

In summary, this population-based study in Shanghai showed that the MGMT EX5-25C>T marker was associated with decreased risks of biliary tract cancers, particularly gallbladder cancer. Future studies with larger sample size and broader coverage of the MGMT gene and other genes in the DNA repair pathway are needed to confirm these results and to identify causal gene variants involved in biliary tract cancer etiology.

No potential conflicts of interest were disclosed.

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

We thank the collaborating surgeons and pathologists in Shanghai for assistance in patient recruitment and pathology review; Chia-Rong Cheng, Lu Sun, and Kai Wu of the Shanghai Cancer Institute for coordinating data and specimen collection; and Shelley Niwa of Westat for support with study and data management.