A study from led by investigators at the Catalan Institute of Oncology (ICO) in Barcelona has revealed that high consumption of beer combined with a single nucleotide polymorphism (SNP) in the alcohol dehydrogenase gene is associated with a nearly nine-fold increased risk of gastric cancer.
Thanks to the lovely folks in the American Association for Cancer Research (AACR) press office who recognize science bloggers as press, I was able to sit in on a press conference this morning at the AACR annual meeting in Orlando where several studies were discussed on genetic and environmental factors in cancer risk.
Lead author of this particular study, cancer epidemiologist Eric Duell, Ph.D., presented a study of Europeans on alcohol consumption and risk of gastric cancer due to SNPs in the alcohol dehydrogenase gene, ADH1. Recall from biochemistry that ADH1 and other ADH forms catalyze the rate-limiting step in the ethyl alcohol oxidation to acetaldehyde, a known carcinogen. Acetaldehyde, in turn, is oxidized to acetate by aldehyde dehydrogenases (ALDHs).
This is an impressive retrospective analysis. The study data was culled the European Prospective Investigation into Cancer and Nutrition (EPIC), a study of 521,000 individuals aged 35 to 50 who completed diet and alcohol use questionnaires at 23 centers across 10 European countries between 1992 and 1998. A subset, or nested-study, called EurGast examined environmental factors and genetic susceptibility to gastric cancer in 364 cases relative to 1272 controls.
When examined as a pool only one SNP was associated with a modest, 30% increased risk for gastric cancer. Combining this SNP with alcohol consumption data revealed that 60 g EtOH/day increased risk by 75%. (Sixty grams of ethanol per day is the amount present in approximately four 12 oz beers at 5% alcohol by volume, four 5 oz glasses of wine at 12% ABV, or four 1oz shots of 100 proof liquor.)
However, the subanalysis of that SNP stratified for alcohol consumption and type of alcohol revealed the big surprise. Consumption as beer, but not wine or liquor, combined with this SNP at both alleles was associated with increased gastric cancer risk of 8.72-fold in this high consumption group (just one allele increased risk by only 33%). This SNP in the ADH1 gene, rs1230025, is an intergenic T→A polymorphism, neither in the promoter or the coding region of the gene.
The influence of this SNP has only been evaluated in one study where it was shown to be associated with a lower breath alcohol concentration – and presumably higher acetaldehyde concentration, although not explicitly measured – at late timepoints when normal volunteers are given a challenge of 0.75 g/kg of ethanol.
But why is beer the only alcoholic beverage with this increased gastric cancer risk in the background of this particular SNP? Alcohol is alcohol, right? I asked Duell about this point because he said that beer has low levels of nitrosamines, the liver and stomach procarcinogens. However, nitrosamines are not activated to their proximal carcinogenic species by ADHs but rather by the cytochrome P450 CYP2E1 (incidentally, the same CYP that oxidizes ethyl alcohol at high concentrations. Duell also noted that the levels of nitrosamines are much lower in beer today than earlier in the lifetime of the participants.
Instead, Duell said he was interested in “what [were] they eating when they were drinking beer.” I find this aspect fascinating but can’t quite figure what other dietary carcinogen people would be eating that was influenced by this particular ADH genotype. I’m also interested to know the carcinogenic potential of other alcohols metabolized by ADH that may be present in beer but not wine or liquor. All sorts of lovely organics are made by yeast depending on the strain used, the specific grains, pH, oxygenation, and other metabolic substrates.
Since subjects evaluated in this work spanned 10 countries, incoming AACR president Judy Garber, MD, asked Duell if the cases could be analyzed by nationality. Unfortunately, Duell said, the study would lose the power necessary to see differences in these smaller subgroups.
So what do these data mean to alcohol consumers? Well, first, drinking ethyl alcohol at 60g or greater per day has many other health risks besides gastric cancer, regardless of which ADH SNPs one might have. (Indeed, there may have been some conference attendees who I saw last night drinking with this degree of enthusiasm.).
However, this particular SNP is really, really bad news. I’m hard-pressed to think of any environmental factor besides smoking that causes a nine-fold increase in risk of any cancer. I’d definitely suggest to the consumer genomics company 23andMe that they add this SNP to their screen. Individuals choosing to drink alcoholic beverages may wish to know if they carry one or both of these ADH1 alleles.
But which alcoholic beverage? Beer is clearly a bad mix if you have both alleles of rs1230025. I’m sure that the European brewing community is none too pleased with Dr. Duell’s group.
However, the team may indeed be the toast of the wine and liquor industries.
Update 4 Apr 2:25 pm: I just had a chance to look more closely at the rs1230025 page at NCBI for the population distribution of this polymorphic region. Turns out that Craig Venter has no worries as his genotype is homozygous for the reference allele T/T. Jim Watson, however, had moderate risk in drinking beer being heterozygous (T/A). However, Celera’s collection of three sets of 30 unrelated individuals (African-American, Asian, and Caucasian) show the demon homozygous A/A at 79%, 96%, and 55% of each respective group. This limited Celera data is suggestive that the high risk genotype for gastric cancer with high beer consumption might be highly prevalent.
Leave a Reply