Epidemiologic investigations have provided definitive evidence that
the drinking of alcoholic beverages can induce cancer in humans (IARC,
1988).
Risks of all cancers combined rise as the level of intake
increases; the largest study shows a detectable increase in risk
following consumption of three alcoholic drinks per day which rises to a
60 percent excess with six or more drinks per day (Boffetta and
Garfinkel, 1990).
Risks due to alcohol vary considerably by type of cancer. The strongest
associations are with oral, pharyngeal, esophageal, and laryngeal
cancer. Cigarette smoking, which is also a major determinant of each of
these cancers, tends to combine synergistically with alcohol in
enhancing risk.
In the mid-1980s, a large U.S. case-control study of oral and pharyngeal
cancer provided the most detailed measurements available of the effects
of drinking (and its interaction with smoking) on these cancers (Blot et
al. 1988). Within each category of smoking, risks of oral and pharyngeal
cancer tended to increase as alcohol intake increased. People who
consumed an average of more than four drinks per day incurred a
nine-fold increase in risk of oral and pharyngeal cancer, while there
was about a four-fold increase in risk associated with smoking two or
more packs of cigarettes per day. Heavy drinkers who also were heavy
smokers experienced a greater than 36-fold excess compared to abstainers
from both products, suggesting that much of the effect of alcohol on
these cancers is enhanced by the effect of tobacco, and that reduction
in either one of the products will substantially reduce risk. Indeed, it
is estimated that 75 percent of all oral and pharyngeal cancers in the
United States are due to drinking and smoking. But, even among
nonsmokers, risks of oral and pharyngeal cancers rise with increasing
intake of alcohol. While heavy consumption of all types of alcoholic
beverages seems to increase risk of these cancers, there is some
suggestion that the association is more pronounced for hard liquor and
beer than for wine.
Esophageal cancer is also induced by the combined effects of alcohol
and tobacco, although alcohol alone can increase risk of this cancer in
nonsmokers as well (IARC, 1988). In some areas of the world, particular
alcoholic beverages have been associated with exceptionally high rates
of esophageal cancer. For example, in the Calvados region of northern
France, very high rates of esophageal cancer have been linked to
consumption of local apple brandies (Tuyns et al., 1979). Home-brewed
rum has been associated with high rates of esophageal (and oral) cancers
in Puerto Rico (Martinez, 1969), and cachaca (another distilled spirit
of sugar cane) has been implicated in the elevated risk of esophageal
cancer in Brazil (Victora, 1987). In coastal areas of South Carolina,
which has led the nation in death rates from esophageal cancer for
several decades, nearly 90 percent of black male patients with this
cancer reported they were regular consumers of moonshine whiskeys (Brown
et al., 1988).
Drinkers run a higher relative risk for cancers of the extrinsic
larynx, which comes in contact with alcohol during drinking, than the
intrinsic larynx (Elwood et al., 1984), suggesting that alcohol may act
through topical exposure to increase risk of upper aerodigestive tract
cancers. A recent report has also associated long-term use of
mouthwashes high in alcohol content with increased risk of oral cancer,
also suggesting an effect of topical rather than systemic exposure since
mouthwashes are seldom swallowed (Winn et al., 1991).
The observations of increased risks of oral, pharyngeal, esophageal,
and laryngeal cancer associated with nearly all types of alcoholic
beverages implicate a dominant effect for common ingredients,
particularly ethanol (Blot, 1992). Although ethanol itself and alcoholic
beverages have generally not induced cancer in experimental animals, the
epidemiologic evidence is sufficient to establish carcinogenicity (IARC,
1988). The variation in risk by type of beverage, however, suggests that
substances other than the alcohol may in some instances contribute to
the risk.
Alcohol is also a recognized cause of liver cancer.
Deaths from this
cancer are increased about 50 percent among alcoholics and other heavy
drinkers (IARC, 1988). By altering liver function and the liver's
ability to metabolize some substances into compounds that may be
carcinogenic or to deactivate certain existing carcinogens, alcohol's
effects on the liver may influence not only liver cancer but cancers at
other sites as well.
More than 50 epidemiologic
studies in the past decade have found small to modest increases in risks
of breast cancer associated with drinking alcoholic beverages (Schatzkin
et al., 1994). Two large cohort studies involving nearly 600,000 women
detected a 20 to 30 percent excess of breast cancer associated with
consumption of about one drink per day, with 60 to 70 percent excesses
among heavy drinkers (Colditz
et al., 1990; Garfinkel
et al., 1988). Further research is needed to clarify whether the
association between alcohol and breast cancer is causal in nature.
Several epidemiologic studies suggest that alcohol may induce
colorectal cancers, particularly rectal cancer, but the findings are not
totally consistent. The overall evidence suggests that alcohol is not
causally related to cancers of the stomach, pancreas, lung, bladder, and
other tumors not mentioned earlier (IARC, 1988).
Moderation of intake is the key to prevention of alcohol-induced
cancer. Because many of the cancers induced by alcohol result from heavy
consumption, reducing levels of consumption will considerably lessen the
risk. Because of the interaction of alcohol and tobacco, smoking
cessation will also reduce the effect of alcohol on oral, pharyngeal,
esophageal, and laryngeal cancers.
* From the Biostatistics Branch, Division of Cancer
Etiology, National Cancer Institute, Bethesda, Maryland
More Alcohol and Cancer Info
