Oral Cavity and Pharynx Gina L. Day, Ph.D.*	Nearly 30,000 new cases of oral cancer--cancers of the lip, tongue, mouth, and pharynx--were estimated for the United States in 1994, and about 7,900 people died of the disease (Boring et al., 1994). These cancers account for almost 4 percent of all malignancies. Except for salivary gland tumors, which are rare, almost all oral cancers are squamous cell carcinomas and share a common etiology. Overall five-year relative survival for oral cancer has remained stable at 40 to 50 percent for several decades. Prognosis varies considerably according to the site of the tumor, stage of disease at treatment, gender, and the age of the patient. In Americans, oral cancer is two to three times more common among males than females. Over 90 percent of cases occur in persons over age 45, with an average age of 64 for whites and 57 for blacks. Like most epithelial tumors, risk of oral cancer increases with age. This cancer occurs more frequently in blacks than whites. During 1987-91, the average annual age-adjusted incidence rate for oral cancer in the United States was 15.8 cases per 100,000 persons per year among white men, 6.2 among white women, 23.7 among black men, and 6.5 among black women (Ries et al., 1994). The most commonly involved sites are the tongue, floor of mouth, gum and other parts of the mouth, lip, tonsil (oropharynx), and hypopharynx. Among white males, the gums and the floor of the mouth are the most frequently affected sites. Among black males, the most common site is the pharynx, or throat. In this country, differences in alcohol and tobacco use account for the bulk of the racial differences in oral cancer (Day et al., 1993). The highest oral cancer rates in the world (world standardized rates > 40/100,000) are reported in parts of France, India, and Southeast Asia (Parkin et al., 1992). Tobacco and alcohol account for approximately three-fourths of all oral cancers in the U.S. (Blot et al., 1988) and are the primary causes of these cancers in most Western countries (IARC 1986, 1988). In some parts of the country, smokeless tobacco use contributes to the high rates of gum and buccal cancers (Surgeon General, 1986). Recent epidemiologic evidence indicates that smoking and drinking are independent risk factors for oral cancer that produce a synergistic effect when combined. In the largest population-based case-control study of oral cancer yet conducted (Blot et al., 1988), strong positive trends in risk were observed according to amount and duration of each type of tobacco and for amount of alcohol consumption. Relative to nonsmokers, heavy cigarette smokers (40+/day for 20+ years) experienced a four-fold risk (men) and ten-fold risk (women) after adjusting for alcohol intake. After controlling for smoking, moderate drinkers (15-29 alcoholic drinks/week) had a three-fold risk of oral cancer and heavy drinkers (> 30 drinks/week) experienced an eight- to nine-fold risk. Combined heavy smoking and drinking resulted in a greater than 35-fold excess risk. The decline in risk of oral cancer following cessation of smoking is particularly rapid, providing further evidence of the role of smoking. In the study by Blot and coworkers, the risk of quitters was less than half that of continuing smokers five years following cessation of smoking, and little or no elevation in risk was found among those who had quit smoking for ten or more years. The chewing of quids containing betel leaves, tobacco, and lime and the smoking of bidi (a tobacco preparation rolled in betel leaf) contribute to the majority of cases in parts of India and Southeast Asia (Mahboubi and Sayed, 1982; Jayant and Deo, 1986). In the southern United States, use of snuff is the primary cause of cancers arising in the cheek and gum (Winn et al., 1981). Among users of snuff, cancerous lesions typically arise at the site where smokeless tobacco, or quid, is held in contact with the buccal mucosa or gingiva. Although not as prevalent as cigarette smoking, habitual use of pipes, cigars, and smokeless tobacco is associated with relative risks for cancers of the mouth as great as that for cigarette smoking. Among other potential etiologic factors, recent epidemiologic studies have indicated that diet may play an important role in the origins of these cancers. Findings have pointed to the protective effects of a diet consistently high in fresh fruits; vegetables; vitamins A, C, and E; and carotenoids, even with adjustment for alcohol intake and smoking (McLaughlin et al., 1988; Franco et al., 1989; Gridley et al., 1990; Franceschi et al., 1991; La Vecchia et al., 1991). A reduced risk of oral cancer associated with vitamin E supplementation has been shown in one study (Gridley et al., 1992). Poor dentition and oral hygiene, trauma due to ill-fitting dentures or jagged teeth, and use of mouthwashes high in alcohol content may enhance oral cancer risk (Winn et al., 1991). Iron deficiency anemia, a relatively common disorder, may produce atrophic oral changes (as seen in patients with Plummer-Vinson syndrome) that may predispose to malignant transformation. Certain generalized skin diseases, such as syphilis and erosive lichen planus, have been implicated as a contributory factor in the development of cancers of the tongue (Mahboubi and Sayed, 1982). DNA viruses, including human papillomaviruses and herpes simplex viruses, have been suggested as possible risk factors, but their role remains to be clarified (Maden, 1992). Associations with other factors, such as urban residence, lower socioeconomic status, and social instability have been found (Greenberg et al., 1991); however, these factors are highly correlated with patterns of exposure to the major risk factors, tobacco and alcohol. Although some early studies have suggested that certain occupational groups, including textiles, metal, and steel workers, as well as persons exposed to asbestos and polyvinyl chloride, have a higher than expected rate of oral cancer (Mahboubi and Sayed, 1982), available epidemiologic evidence indicates that the occupational component does not play a large role in oral cancer etiology. Outdoor work (or sunlight exposure) has been linked to lip cancer, and an increased risk of salivary gland tumors has been associated with ionizing radiation (Mahboubi and Sayed, 1982). Oral leukoplakia is the most common precancerous lesion, with a small percent of such lesions ultimately becoming malignant. Oral cancer is also characterized by an exceptionally high incidence of subsequent tumors in the same site or adjacent sites in the upper digestive and respiratory tracts. Major determinants of new primaries following oral cancer are high levels of tobacco smoking and alcohol drinking (Day et al., 1994). Based on available information, oral cancer is highly preventable. Quitting smoking and limiting alcohol intake would greatly limit deaths from this disease.
continue

* From the Epidemiology and Extramural Programs Branch, Division of Cancer Etiology, National Cancer Institute, Bethesda, Maryland