In recent years, evidence has accumulated of an increased incidence of hepatocellular carcinoma and other cancers of the alimentary tract in the alcoholic (1-3). The thrust of this paper will be to review only briefly the well-known epidemiologic evidence and to discuss in more depth possible mechanisms whereby alcohol abuse may promote the development of cancer.Clinically, an association between heavy drinking and certain types of cancer has been observed for many years. In 1964, the World Health Organization surveyed the research on alcoholism and cancer and concluded that an association existed between excessive drinking of alcoholic beverages and cancer of the mouth, the larynx and the esophagus (4). In a series of studies (5-7), heavy drinkers were found to have roughly a 10-fold increased risk of developing cancer of the mouth and larynx. Subjects who drink excessively often are also heavy smokers and, in this regard, it appears that alcohol plays a more important role than smoking with respect to cancer of the esophagus (8), whereas smoking seems to be more strongly associated with cancer of the mouth and pharynx (9). Additional sites of cancer associated with alcohol abuse include the pancreas (10, ll), the cardia of the stomach (12, 13), the colon (14-18) and the liver (2, 13, 19-22). Over the last 20 years, a number of animal studies have been conducted in which ethanol, either applied topically or administered as part of the diet, was shown to have a syncarcinogenic or cocarcinogenic effect. Examples of these experiments include: the enhanced carcinogenesis of 7,12-dimethylbenzanthracene applied to the cheek epithelium of young hamsters (23, 24) or the skin of mice (25), the induction of esophageal tumors in rats by diethylnitrosamine (DEN) (26), the induction of liver tumors by vinyl chloride (27), the induction of nitrosopyrrolidine-initiated nasopharyngeal cancers in pair-fed hamsters (28) and dimethylhydrazine-initiated rectal cancer in rats (29). However, it also should be noted that the simultaneous addition of DEN and ethanol in drinking water (25%) did not enhance DEN induction of hepatocarcinogenesis (30) nor did consumption of diets containing ethanol (35% total calories) change the induction of the number of liver tumors by dimethylnitrosamine (DMN) (31) or modify the development of preneoplastic hepatic foci by aflatoxin B, in rats (32). Similar results were obtained by Schwarz et al. Address reprint requests to: Charles S. Lieber, M.D., Alcohol Research and Treatment Center, Veterans Administration Medical Cent,er, 130 West Kingsbridge Road, Bronx, New York 10468.(33) using DEN or N-nitrosomorpholin as carcinogens and 10% ethanol in drinking water. In any event, while there is no evidence from animal feeding studies that ethanol per se is carcinogenic (28, 341, it nevertheless may increase the susceptibility of various tissues to chemical carcinogens by a variety of mechanisms such as activation of chemical carcinogens, altering the metabolism and/or distribution of carcinogens, interferen...