Need to pursue new leads in the epidemiology of colorectal cancer.
S. Graham,W. Haenszel,F. Bock,J. Lyon
DOI: https://doi.org/10.1093/jnci/63.4.879
1979-10-01
Journal of the National Cancer Institute
Abstract:A close examination of the research evidence suggests that a clear case has not been established for two popular theories: High ingestion of fiber and low ingestion of fats decrease the risk of cancer of the large bowel. The fiber hypothesis rests on the assumption that increased amounts of fiber in the diet lessen the opportunity for carcinogenesis from luminal factors (1). The theory maintains that a high-fiber diet can result in more rapid transit time of material through the gut. This may reduce the opportunity for bacteria in the bowel to produce a carcinogen and it may reduce the length of time a carcinogen could be in contact with the gut wall. In addition, the high-fiber diet increases the total bulk of the content so that any carcinogenic materials passing through do so in a more dilute form (1). Burkitt et al. (2) and Eastwood et al. (3) furnished some evidence for a dilution effect. However, there is disagreement in the literature on the other points. There is no unequivocal evidence that high fiber always shortens transit time. Studies on both sides of the issue may be cited (4). Some authors show that fiber speeds transit time; others show that it has no effect on transit time; still others, like Harvey et al. (5), show that transit is speeded in persons with originally slow transit times, slowed in those with rapid times, and not affected at all in those with moderate transit times. Epidemiologic studies have been few, and most of these studies have been done on small numbers of people. In one study (1), transit times were compared in English and African boarding school boys, and the African boys were found to have shorter transit times. This design, however, compared people of different races in different cultures, which could account for the difference. Moreover, to generalize from healthy boys to adults with colon cancer is difficult. In later studies of occupational groups in England and Africa, findings were similar (2). In a related study (6), Glober and colleagues studied transit time using the Hinton method in which 23 Caucasian and 63 Japanese males swallowed barium-impregnated plastic pellets, and the time until their appearance in the stool was measured. Although Japanese living in Hawaii have rates of colon cancer similar to those of Caucasian Hawaiians, their average transit time was found to be significantly less than that of the Caucasians. A few epidemiologic studies (7-10) indicated that persons with a relatively high intake of vegetables may be at lower risk for bowel cancer. Whether this finding can be attributed to the generally high fiber contents of such diets or whether it should be attributed to other characteristics of the vegetables is not clear. Conceivably, the less frequent use of these foods by patients with colon cancer could be associated with lower levels of vitamins A or C or other constituents of value. The other popular theory suggests that a high-fat diet increases neutral sterols and bile acids in the bowel and modifies bacterial enzymatic activity, which in combination could produce carcinogens. Here the evidence is more impressive (11-13). For example, Reddy et al. (14, 15) studied 8 individuals who first were on a high-meat diet and then on a no-meat diet, 12 so-called ovolacto-vegetarians, and 11 Seventh-Day Adventists (also ovolacto-vegetarians) and compared each group with normal omnivorous people. Aries (16) did a similar study of 15 Vegans, members of 'an English group who eat no animal products. In these inquiries, the concentration of bile acids in feces of individuals who abstain from meat was found to be from 21 to 86% that of the omnivores and the concentration of neutral steroids was from 33 to 82%. Recently, Reddy and Wynder (17) compared the fecal neutral steroid and bile acid concentrations in colon cancer patients and in controls and found significantly higher concentrations in the cancer patients. They also found greater enzymatic activity of the fecal flora in the cancer patients. Interestingly, they showed that animals challenged with dimethylhydrazine and fed either vegetable oil or beef fat exhibited more tumors than did animals on a control diet (18). Apparently, the evidence is accumulating that a high-fat diet might indeed produce