Colonic mucosal folate concentrations and their association with colorectal cancer

Abstract

Colorectal cancer is the second most common cause of cancer-related death in both men and women. In the United States, there were 134 000 new cases and 59 400 deaths from colorectal cancer in 1996 (1). However, it is estimated that 50‐75% of these deaths are preventable, either through screening and early detection or nutritional methods or chemoprevention. Chemoprevention and nutritional methods involve the use of aspirin and nonsteroidal antiinflammatory agents on the one hand and dietary change on the other. Many studies have suggested that diets high in fat and low in fruit, vegetables, and fiber combined with a lack of physical exercise are associated with an increased risk of colon cancer. Fruit and green, leafy vegetables are the main dietary sources of folate; epidemiologic studies have suggested that a diminished folate status is associated with an increased risk of colon cancer (2, 3). These data are supported by a recent study that used a rodent model of colorectal cancer induced by dimethylhydrazine; the study showed that folate deficiency enhances and dietary folate protects against the development of colonic neoplasia (4). The mechanism by which folate deficiency may induce colon cancer remains obscure. It has been suggested that folate deficiency produces DNA strand breaks in the p53 tumor suppressor gene and site-specific protooncogene DNA hypomethylation (5). The phenotypic effect of folate supplementation on the rectal mucosa of patients with recurrent adenomatous polyps is a reduction in the rate of mucosal cell proliferation, possibly leading to a reduction in the incidence of both polyps and colorectal cancer (6). The epidemiologic studies that have shown an association between colorectal cancer and folate deficiency have relied mostly on either dietary folate intake or blood folate concentrations as measures of folate status. Whether such measurements accurately assess the concentration of folate in the colonic mucosa, which contains the cells of origin in colorectal cancer, is questionable. In a human study in which pains were taken to isolate colonic mucosal cells from contamination with lamina propria cells or surface mucus before the mucosal cell folate content was measured, no evidence of folate deficiency was found in the healthy mucosa adjoining either colonic polyps or cancers, although a slight reduction in folate was found in both polyps and cancer cells (7). Because there was no relation between blood and colonic cell folate concentrations, colonic mucosal cells may not necessarily derive all of their folate content entirely from the blood; at least part may be absorbed from the intestinal lumen. It may therefore be invalid to assume that blood folate concentrations reflect folate concentrations in colonic mucosal cells. Tissue-specific susceptibility to folate deficiency in humans has been reported before. Megaloblastic changes were noted in the mucosal cells of the cervix in women with normal blood folate and cobalamin concentrations who were taking contraceptive pills (2); additionally, buccal mucosal cells in smokers were shown to be depleted of folate when systemic folate concentrations were normal (8). In this issue, Kim et al (9) showed in patients with either hyperplastic or adenomatous polyps that the folate content of sigmoid colon biopsy samples (which included mucosal cells, mucus, and lamina propria cells) was significantly correlated with serum and red blood cell folate concentrations and inversely correlated with plasma homocysteine concentrations. Homocysteine is now recognized to be a particularly sensitive measure of folate deficiency. However, whether tissue folate concentrations accurately reflect the folate content of the colonic mucosal cells from which colorectal cancer is derived is open to question. The differences between these papers can only be resolved by further studies. Nevertheless, as stated by Kim et al (9), there can be no doubt that in the future, colonic mucosal cell folate content will be a much more accurate assessment of folate status than blood folate measurements alone. Kim et al also showed that the folate content of sigmoid mucosal biopsy samples taken ♢ 10 cm away from colonic polyps was significantly lower in patients with adenomatous polyps (which predispose to colon cancer) than in patients with hyperplastic polyps (which are not considered to be premalignant). This is in contrast with the studies by Meenan et al (7) and suggests that there may be a global reduction of folate content in the colonic mucosa of patients with adenomatous polyps. However, there was a marked overlap in folate concentrations in both blood and the colonic mucosa between the 2 groups of patients with hyperplastic and adenomatous polyps. This suggests that the reduced folate concentration in the mucosa of patients with adenomatous polyps may coincide with a metabolic defect in folate metabolism. It seems probable that to induce colon cancer, folate deficiency must be associated with other metabolic defects. Paradoxical