Alcohol Drinking and One-Carbon Metabolism-Related Gene Polymorphisms on Pancreatic Cancer Risk

A high intake of folate, which is plentiful in vegetables and fruits, has been associated with a reduced risk of several cancers (1). Folate functions within so-called “one-carbon metabolism” to facilitate de novo deoxynucleoside triphosphate synthesis and to provide the methyl groups required for intracellular methylation reactions. Epidemiologic studies have suggested the importance of folate in pancreatic cancer risk (2, 3). Polymorphisms in critical enzymes involved in the one-carbon metabolism pathway, including methylenetetrahydrofolate reductase (MTHFR), methionine synthase (MTR), methionine synthase reductase (MTRR), and thymidylate synthase (TS), play important and interrelated roles in folate metabolism and may thereby influence the risk of pancreatic cancer.

Heavy alcohol consumption is known to be a major cause of chronic pancreatitis, and chronic pancreatitis has been linked to pancreatic cancer (4); however, the association between alcohol consumption and risk of pancreatic cancer has been inconsistent (5, 6). Chronic inflammation in pancreatitis induces DNA damage and mutations and thereby facilitates the development of pancreatic cancer. DNA synthesis for replication and repair is largely dependent on the availability of the one-carbon metabolism pathway. Therefore, the one-carbon metabolism polymorphisms may modify influence of alcohol drinking on pancreatic cancer risk.

Here, we evaluated the effect of alcohol consumption in conjunction with genetic polymorphisms in one-carbon metabolism enzymes on pancreatic cancer risk among Japanese.

The subjects, ages 20 to 79 years, in the present study were enrolled between January 2001 and November 2005 in the framework of Hospital-based Epidemiologic Research Program at Aichi Cancer Center (7, 8). In brief, Hospital-based Epidemiologic Research Program at Aichi Cancer Center -II was launched in 2001, asking all first-visit outpatients in Aichi Cancer Center Hospital to provide 7 mL blood as well as information on lifestyle factors. A total of 35,838 patients visited Aichi Cancer Center Hospital as first-visit outpatients during this period. Among them, 6,300 patients were not enrolled due to miscellaneous reasons; 28,571 (79.7%) completed the questionnaire adequately. Of those who completed an interview, 50.7% donated a blood sample. All participants gave written informed consent and the study was approved by the Ethics Committee of Aichi Cancer Center.

A total of 175 patients who were newly diagnosed as having pancreatic cancer (International Classification of Disease, Tenth Edition code C25) at our hospital were deemed to be potential cases. From these, we excluded 16 patients with a past history of cancer and 2 patients with pancreatic endocrine tumor, leaving 157 cases eligible for analysis. Control subjects were randomly selected from first-visit outpatients. A total of 7,240 individuals who were confirmed not to have cancer according to the cancer registry and medical record was deemed to be potential controls. We excluded 276 patients with a past history of cancer, leaving 6,964 controls eligible for analysis. Eventually, 785 controls were individually matched with case subjects by age (±3 years) and sex at a 1:5 case-control ratio. We assessed the clinical diagnosis among noncancer outpatients in the previous study and confirmed that a large fraction of the patients had no abnormal findings by examination and nonspecific diseases (9). All subjects for the present study were Japanese and most subjects are living in and around Aichi Prefecture, central Japan.

Genotyping for MTHFR C677T (dbSNP ID: rs1801133), MTR A2756G (rs1805087), and MTRR A66G (rs1801394) was based on TaqMan Assays by Applied Biosystems. The TS variable number of tandem repeat polymorphism was defined by PCR using 5′-CGTGGCTCCTGCGTTTCC-3′ and 5′-GAGCCGGCCACAGGCAT-3′ primers.

Daily alcohol consumption in grams was determined by summing the pure alcohol amount in the average daily consumption of Japanese sake (rice wine), shochu (distilled spirit), beer, wine, and whiskey, with one cup of Japanese sake (180 mL) considered equivalent to 23 g ethanol, one drink of shochu (180 mL) to 46 g, one large bottle of beer (720 mL) to 23 g, one glass of wine (80 mL) to 9.2 g, and one shot of whiskey (28.5 mL) to 11.5 g. Heavy drinkers were defined as those currently drinking alcoholic beverages ≥5 days/wk in a daily amount of ≥46 g (two Japanese drinks), whereas moderate drinkers were defined as those currently consuming less frequently than 5 days/wk, in lower amounts, or both. Former drinkers or smokers were defined as those who quit drinking or smoking at least 1 year before the survey, respectively. Dietary intake of folate was computed based on the food frequency questionnaire consisted of 47 single food items (10). The deattenuated correlation coefficients for energy-adjusted intakes of folate using 3-day weighed dietary records were 0.36 [95% confidence interval (95% CI), 0.12-0.58] in men and 0.38 (95% CI, 0.25-0.62) in women (11).

To assess the strength of the associations between alcohol consumption, polymorphisms of folate metabolism enzyme, and pancreatic cancer risk, odds ratios (OR) with 95% CI were estimated using unconditional logistic models adjusted for potential confounders. Potential confounders considered in the multivariate analyses were age, sex, drinking habit (never, former, moderate, or heavy drinkers), smoking habit (never, former, or current smokers of <40 or ≥40 pack-years), current body mass index (<18.5, 18.5-24.9, or ≥25.0), total nonalcohol energy intake (as a continuous variable), dietary folate intake (μg/d, tertiles), history of diabetes mellitus (yes or no), and referral pattern to our hospital (patient discretion, family or friend recommendation, referral from another clinic, secondary screening after primary screening, or others). Accordance with the Hardy-Weinberg equilibrium was checked for controls using the χ² test and used to assess any discrepancies between genotype and allele frequencies. To exclude the subjects who stop drinking due to pancreatic cancer, analysis without former drinkers was conducted for association with alcohol drinking. Interactions were assessed by the logistic model, which included interaction terms between alcohol consumption and genes with scores of genotype and drinking habit. P < 0.05 was considered statistically significant. Analyses of the risk estimate were performed using STATA version 10 (Stata).

Table 1 shows the distribution of cases and controls by background characteristics. The proportion of current smokers of ≥40 pack-years was significantly higher in cases than controls (P = 0.006). A significantly high frequency of a history of diabetes mellitus was seen in cases (P < 0.001).

Genotype frequencies for all polymorphisms were in accordance with the Hardy-Weinberg law in controls (Table 2). None of the polymorphisms showed any significant effect on pancreatic cancer risk by genotype.

Heavy alcohol drinking was marginally associated with an increased risk of pancreatic cancer in overall analysis (OR, 1.90; 95% CI, 1.00-3.62; P_trend = 0.036; Table 3). To assess the effect of alcohol consumption and the one-carbon metabolism-related gene polymorphisms in pancreatic cancer risk, furthermore, we conducted the stratified analysis by the genotypes. Among subjects with the MTHFR 677 CC genotype, adjusted OR (95% CI) of pancreatic cancer was 4.50 (1.44-14.05) for heavy drinkers relative to never drinker (P_trend = 0.008). In contrast, the trend was not significant among those with MTHFR 677 CT or TT genotype. Heavy drinkers with MTR AA genotype or MTRR 66 G allele had higher risk of pancreatic cancer relative to never drinkers with these genotypes, whereas no association was observed in other genotypes. We examined the association between alcohol consumption and pancreatic cancer risk by folate intake; no clear interaction was found (data not shown).

We found that the risk of pancreatic cancer was increased with alcohol consumption in subjects with the MTHFR 677 CC genotype, MTR 2756 AA genotype, and MTRR 66 G allele. These finding suggest that the association between alcohol drinking and pancreatic cancer risk may be modified by these polymorphism.

Previous two studies in the United States and China showed an increased risk with the MTHFR 677 TT genotype (12, 13), whereas no association was found in a second U.S. study (14). The China study also reported an association between TS and pancreatic cancer risk (12).

High consumption of dietary folate has been associated with a lower risk of pancreatic cancer (2, 3). Folate status is determined by both folate intake and folate metabolism, which is influenced by folate metabolism enzyme polymorphisms. On the other hand, alcohol consumption may impair folate metabolism (15). These observations suggest that folate intake, the polymorphisms, and alcohol consumption may share some common pathways contributing to pancreatic carcinogenesis.

In this study, stratification of analyses by the respective genotypes revealed a significant effect of alcohol consumption on pancreatic cancer risk among subjects with MTHFR 677 CC genotype. Alcohol produces inflammation in pancreatic tissues via various mechanisms, such as the generation of reactive oxygen species and alterations in immune responses (16), which in turn lead to pancreatitis (17). Chronic pancreatitis increases genomic damage and cellular proliferation, leading to the malignant transformation of pancreatic cells (18) and ultimately affecting the development of pancreatic cancer. Considering folate metabolism pathway, low activity of MTHFR, MTHFR 677 CT or TT genotypes, leads to accumulate of 5,10-methylentetrahydrofolate, which is required for conversion of uridylate to thymidylate, whereas it is thought that individuals harboring the MTHFR 677 CC genotype have less DNA synthesis and repair capacity. These finding suggest that the individuals with the CC genotype would be more susceptible to DNA damage by large amounts of alcohol than those with the CT or TT genotypes. Furthermore, alcohol ingestion reduces the intestinal absorption of folate (15), which may accelerate the reduction of DNA repair capacity.

Our results showed that the polymorphism of MTR and MTRR also may modify pancreatic cancer risk by alcohol drinking. Aberrant DNA methylation by MTR and MTRR polymorphisms and large amount of alcohol drinking results in altered expression of critical proto-oncogenes and tumor suppressor genes. MTR and MTRR catalyze the remethylation of homocysteine to methionine; thus, their polymorphism may influence the pancreatic cancer risk, although the functional effect of these gene polymorphisms has been well unknown. Frequency of variant allele in MTR and MTRR was small; thus, the interpretation should be caution.

Frequency of heavy alcohol drinkers in this study is approximately in agreement with that of prospective study in Japan (19). According to WHO report, the percentage of heavy drinkers is relatively high in men compared with other countries (e.g., 6.4% for men in the United States; ref. 20). Therefore, it is important to classify the risky genetic background for heavy drinkers in pancreatic cancer development.

Several studies have shown protective effect of folate on cancers, whereas, in some animal studies and a recent supplement study, adverse effect of folate was observed (21, 22). Future research must clarify the dual effect of folate on pancreatic cancer.

Potential limitations of the present study should be considered. First, our study is hospital-based case-control design; thus, selection bias may be concerned. We used noncancer patients at our hospital for this purpose on the basis that our subjects arose within this population, thereby warranting internal validity. We have confirmed previously the similarity of this population to the general population in terms of a range of exposures of interest, in this case alcohol consumption, thereby warranting external validity (23). Equivalence in the genotype distribution for the MTHFR C677T polymorphism between our controls and the general population has also been reported (24). Referral patterns to our hospital were different between cases and controls; thus, relative geographic distributions of residences between cases and controls may be different. However, this finding would not lead to bias, because frequency of MTHFR C677T genotype is not different in Japanese (24, 25). Second, this study may suffer from recall bias. Some pancreatic patients might be severe condition at the time of interview. Third, this study focused on four candidate polymorphisms rather than a comprehensive evaluation of the genetic variants. Fourth, we could not obtain the information about the history of pancreatitis, which is risk factor of pancreatic cancer. Lastly, the results in this study may be a false-positive due to small size, particularly in subgroup analysis; thus, large-scale study would assist our results.

In conclusion, our case-control study suggested that effect of alcohol drinking on pancreatic cancer risk may be modified by the folate metabolism enzyme polymorphism in the Japanese population.

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 all the doctors, nurses, technical staff, and hospital management staff of Aichi Cancer Center Hospital for the daily administration of the Hospital-based Epidemiologic Research Program at Aichi Cancer Center study and the staff of the Department of Breast Oncology, Aichi Cancer Center Hospital, for support and helpful discussion.