NEW ORLEANS-Heavy drinking can lead to cancer of the head, neck, esophagus, and liver, but whether moderate alcohol consumption increases risk at these sites is still unclear, said Matthew P. Longnecker, MD, of the National Institute of Environmental Health Sci-ences, Research Triangle Park, NC.
NEW ORLEANSHeavy drinking can lead to cancer of the head, neck, esophagus, and liver, but whether moderate alcohol consumption increases risk at these sites is still unclear, said Matthew P. Longnecker, MD, of the National Institute of Environmental Health Sci-ences, Research Triangle Park, NC.
Alcohol consumption also appears to have a subtle or weak correlation with large bowel, breast, and pancreatic cancer, he said at the 4th International Symposium on Nutrition in Cancer.
The mechanisms by which consumption of alcoholic beverages causes cancer are not established, he said, although many have been proposed. Recent data from Asia suggest that for a given level of heavy alcohol consumption, persons with the inactive form of aldehyde dehydrogenase are at increased risk of alcohol-induced cancers, compared with persons who have the active enzyme.
Duration and level of exposure may be a factor in the relationship between alcohol and cancer. Dr. Longnecker added that the association of alcohol with certain cancers appears not to be beverage-specific. Specific beverage effects simply do not pan out in the studies, he said. The data do not suggest a particular congener in the alcoholic beverage as being responsible for the carcinogenic effect.
Part of the problem in researching the connection between alcohol and cancer is that people cannot be trusted to tell the truth about their drinking habits. If subjects in epidemiologic studies are understating their alcohol consumption, the dose-response relationship may be overstated, and the true slope of the relationship may be less than that observed in human studies, Dr. Longnecker said.
The increased risk of cancer is probably in the 10% or so of people who are heavy drinkers, defined as having more than two drinks a day. Heavy drinkers often have other reasons to have an increased cancer risk (smoking, poor diet), but these do not account for the effect alcohol has on cancer risk.
Among those cancers known to be influenced by alcohol consumption, the greatest dose-response relationship is seen in cancer of the oropharynx, and cancer of the liver has the lowest.
While smoking is a cofactor, the dose-response relationships hold true for nonsmokers as well. Alcohol and cigarette smoking, when combined, have a greater-than-additive effect on risk of cancers of the mouth, pharynx, larynx, and esophagus.
In cancer of the liver, the dose response is smaller, and it requires about five drinks a day to double the relative risk. A two-step process is most likely involved, whereby alcohol leads to cirrhosis, which greatly increases the risk of cancer. A two-step process probably operates in pancreatic cancer as well, with alcohol being a risk factor for chronic pancreatitis, which increases cancer risk, Dr. Longnecker explained.
A subtle link is also demonstrated between alcohol and large bowel cancer. A metaanalysis concluded that one drink a day raises the risk about 5%, but a high folate level appears to be protective.
A weak association and a dose-response relationship also appear to exist between breast cancer and alcohol. For each drink per day, risk for breast cancer is increased 10%.
For a woman at high risk of breast cancer and low risk of coronary heart disease who has 1 drink per day, I suspect the effect on mortality of decreasing alcohol intake is a wash. It brings you to a lifestyle decision. For those women, the U-shaped relation between alcohol intake and total mortality does not hold. For more typical women, if you go from one drink to none per day, you decrease your risk of breast cancer and increase your risk of coronary heart disease. In that case, decreasing consumption may result in a net increase in mortality. Dr. Longneckers advice: If you choose to drink, dont drink too much.