The initial evidence that folate intake at levels achievable through the use of supplements and/or fortification might cause a paradoxical acceleration of carcinogenesis in individuals who harbor neoplastic lesions dates back to the 1940s when two groups of investigators gave large doses of folic acid to patients with acute leukemia and observed what Sidney Farber politely termed ‘‘the acceleration phenomenon’’, whereby the rate of expansion of the leukemic clone increased tremendously.3,4
This is almost certainly due to the fact that neoplastic cells have a high rate of proliferation and therefore require large quantities of folate to maintain thymidine synthesis at a pace that is necessary to meet the heightened needs for DNA synthesis.29
Controlled studies in a variety of animal models of colon cancer have provided similar findings: in settings where there is a particularly strong underlying predisposition to colon cancer or in a setting where neoplastic tumors are already established, supplemental folic acid is protective only before neoplastic foci appear in the intestine. Once such foci are established, the more folic acid that is given, the faster microscopic foci and macroscopic tumors arise.30–34
The results of a recent randomized trial of folic acid in the prevention of colorectal adenomas are consistent with such as effect as well.35
This study, which enrolled subjects with a history of prior adenomas, did not observe a modulation of recurrence rates after the first follow-up colonoscopy at 3 years. However, at the time of the second follow-up, 6–8 years after initial randomization, a significant 2.3-fold increase in the multiplicity of recurrent adenomas occurred among those receiving folic acid supplements (as well as a marginally significant 1.7-fold increase in so-called ‘high risk’ adenomas). These results are consistent with the concept that the abundant intake of folic acid may accelerate tumorigenesis among those who harbor existing foci of neoplasia. Recent data indicating that the ‘miss rate’ for advanced adenomas is ~10%36
and that ‘flat adenomas’, which are difficult to detect by conventional colonoscopy, are common and much more likely to contain foci of carcinoma than polypoid adenomas37
underscore the plausibility that trials such as this contain sizeable numbers of individuals with high risk neoplasms in spite of undergoing a baseline ‘cleansing’ colonoscopy. To date, three other folic acid supplementation trials that have used adenoma recurrence as an endpoint have been reported38–40
and none have observed a similar increase in tumorigenesis, but importantly, none of these other three have had the extended follow-up over 6–8 years that was provided for in the Cole et al
trial. Moreover, in an ecological study recently reported by myself and colleagues, we observed an increase of ~4 cases/100,000 individuals in the nationwide rates of colorectal cancer diagnoses in conjunction with the institution of mandatory folic acid fortification of cereal grains in both the U.S.A. and Canada in the 1990s.41
Studies of an ecological nature have inherent weaknesses; moreover, the question arises as to whether colorectal cancer rates could feasibly increase within the first year after the institution of fortification, as was observed to occur. Nevertheless, data from the abovementioned studies collectively raise a concern: 1) that is consistent with the known biochemical functions of folate in DNA synthesis, 2) which can be readily replicated in animal models of colorectal carcinogenesis, and 3) which is a phenomenon that has been observed to occur in small, inadvertent studies of human subjects with hematologic malignancies.
The possibility of a paradoxical promoting effect of excessive quantities of folate may be operable in human breast cancer as well: in a recent prospective cohort study among ~25,000 postmenopausal women the risk of developing breast cancer was significantly increased among those taking folic acid supplements;42
a group among whom consumption was ≥853 mcg of folate per day. Like colorectal cancer, the promoting effect of folic acid has also been observed to occur in an animal model of breast carcinogenesis.43
The concept has therefore evolved that higher intakes of folate are cancer-protective in nearly all circumstances except among those individuals who have existing neoplastic lesions and who are consuming exceptionately large quantities of the vitamin, in which case the folate presents additional risk for tumorigenesis. Cancers that slowly evolve over years through an indolent dysplastic phase, such as the colorectal adenoma or a dysplastic nodule of the prostate, are exceedingly common in the general population of healthy, elder adults and would be particularly prone to such an effect.
The fact that folic acid, which is not a naturally-occurring form of the vitamin, is used by food and pharmaceutical industries for fortification and supplementation is potentially of importance. Folic acid is converted to a natural biological form of the vitamin as it passes through the intestinal wall, with enzymatic reduction and methylation resulting in the circulating form of the vitamin, 5-methyltetrahydrofolate. Nevertheless, it has been known for some years that oral doses of folic acid in physiologic quantities can saturate this conversion mechanism, resulting in detectable levels of circulating folic acid, and there has been some concern that this oxidized, nonsubstituted form of folate might feasibly be detrimental because it is not a naturally occurring coenzymatic form of the vitamin. Several studies have shown that as little as 200 mcg of oral folic acid may produce transiently detectable levels of folic acid in the bloodstream,44–46
and more recently, it has been shown that the daily ingestion of 400 mcg produces a sustained appearance of folic acid in the blood.47
In the U.S., 40% of adults 60 years and older are thought to regularly consume a supplement containing folic acid, most of which contain 400 mcg per pill.48
When this folic acid intake is added to that which appears in voluntarily fortified foods such as ready-to-eat breakfast cereals plus the additional quantity that is now consumed in the form of mandatory cereal grain fortification, it should come as no surprise that a large percentage of the population has detectable quantities of unmetabolised folic acid chronically circulating in the bloodstream. Consistent with these projections are recent observations from the Framingham study that demonstrate that those who are regular users of vitamin supplements have a mean concentration of unmetabolised folic acid in fasting plasma that is ~40% higher than that of non-users; moreover, in this era of mandatory fortification, 81% of vitamin users have detectable levels of unmetabolised folic acid in their plasma.49
Of additional concern are recent observations of Troen et al,50
who observed that increasing concentrations of plasma folic acid among elder women who took folic acid-containing supplements were inversely associated with decreases in the cytotoxicity of circulating natural killer cells. Natural killer cells are a population of lymphocytes thought to play a role in the destruction of arising clones of neoplastic cells. Thus, the coenzymatic form of folate used for fortification is perhaps relevant in determining whether a detrimental effect is produced from excessive intake.