Recent hypotheses for the origin of colon cancer.

The origin of many of the cancers seen in Canada, the United States, and other Western countries appears to be associated with diet. International data and migrant studies have shown that cancer of the colon and breast are related to environmental factors (1-4) and international studies suggest that cancers of the colon, breast, pancreas, and prostate are associated with diets high in meat and dietary fat and low in fiber and cereal grains (5-7). However, the specific dietary risk factors respon sible for these cancers are still unclear despite considerable effort in epidemiolÃ3gica! cohort and case-control studies (810). Experimental methods are probably needed to help define the risk factors associated with diet. The colon is perhaps the best site for carrying out such experimental studies. At this site it is possible to examine the chemistry of the environment in the colonie lumen. It is possible to watch the development of the disease directly with the endoscope. It is possible to biopsy the portion of colon at risk and examine it microscopically. All the steps involved in the development of colon cancer can be observed: the effect of diet on the chemistry in the lumen of the colon; the effect of this chemistry on the cellular biology of the epithelial cells of the colon; the effect of cellular changes on the development of the disease. Hypotheses for the origin of colon cancer can be developed and tested with experimental methods that are much more rapid than the long period of time involved between diet exposure and the onset of cancer. Here we will review recent experience in the development and testing of five such hypotheses. They have resulted from studies that were initially directed toward the identification of carcinogenic initiators and promoters in the colonie environ ment (11) and are summarized in Table I. They are being tested for their consistency with laboratory, epidemiolÃ3gica!, and clin ical studies. The present status of these tests is summarized in Table 2.