Alcohol and cancer.

The cancers consistently associated with ingestion of alcohol, the head and neck cancers, are also associated with tobacco use and arise from epithelia that are in direct contact with both agents. Tobacco smoking-related cancers at sites not directly in contact with alcoholic beverages, that is, lung, bladder, and perhaps pancreas, do not consistently show a relationship to alcohol consumption, although lung and pancreatic tumors are associated in some studies. Liver cancer was thought to be strongly related to alcohol consumption on epidemiological grounds and because of its relationship to cirrhosis. As knowledge of the viral etiology of some cirrhoses has evolved and as methods to detect viruses have developed, the significant association between hepatitis B virus and hepatocellular carcinoma has become clear. Alcohol and hepatitis B virus may interact in the etiology of the disease and have important separate roles as well. There are epidemiologic and experimental data showing that malnutrition (resulting from poor food choice), economic deprivation, or alcoholism contributes to the risk for head, neck, and liver cancers. Colon cancers occur about equally in men and women, are found in well-nourished populations, and are not associated with tobacco smoking. Rectal cancers show a preponderance of cases in men but are frequently found in women as well and are not thought to be associated with smoking or malnutrition. The association between colorectal cancers and alcohol consumption, when it is found, apparently occurs at even relatively low alcohol intakes and is often stronger for consumption of beer than of other beverages. Nutritional and metabolic mechanisms proposed for the influence of alcohol on carcinogenesis are supported by studies in human subjects and laboratory animals. Animal models are needed in which effects of ethanol on carcinogenesis can be consistently demonstrated and which can then be used to examine mechanisms.