How many food additives are rodent carcinogens?

Our plain-English summary of How many food additives are rodent carcinogens?, by F.M. Johnson (Environmental and Molecular Mutagenesis, 2002).

The study

• Author: F.M. Johnson
• Journal: Environmental and Molecular Mutagenesis
• Published: 2002;39(1):69-80
• Cited by: 11 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

One generally assumes that chemical agents added to foods are reasonably free of risks to human health, and practically everyone consumes some additives in his or her food daily throughout life. In the United States, the 1958 Food Additives Amendment to the Federal Food, Drug and Cosmetic Act of 1938 requires food manufacturers to demonstrate the safety of food additives to the Food and Drug Administration (FDA). The Amendment contains a provision that prohibits approval of an additive if it is found to cause cancer in humans or animals. In the present study, data from the National Toxicology Program rodent bioassay (NTPRB) were used to identify a sample of approximately 50 rodent-tested additives and other chemicals added to food that had been evaluated independently of the FDA/food industry. Surprisingly, the sample shows more than 40% of these food chemicals to be carcinogenic in one or more rodent groups. If this percentage is extrapolated to all substances added to food in the United States, it would imply that more than 1000 of such substances are potential rodent carcinogens. The NTP and FDA test guidelines use similar, though not necessarily identical, rodent test procedures, including near lifetime exposures to the maximum tolerated dose. The FDA specifies that test chemicals should be administered by the oral route. However, the oral route includes three methods of delivering chemicals, that is, mixed in the food or water or delivered by stomach tube (gavage). The NTP data show only 1 of 18 food chemicals mixed in the food are rodent carcinogens, but 16 of 23 gavage-administered food chemicals are carcinogenic to rodents. The distribution suggests that among orally delivered chemicals, those administered in the feed will more likely prove to be noncarcinogens than chemicals given by gavage. The rodent data also reveal that effects may vary according to dose and genotype, as well as by route of administration, to further complicate extrapolation to humans. Human experience with known carcinogens such as tobacco, asbestos, and benzidine convinces us that environmental carcinogens constitute a real threat to human health, although predicting human carcinogens from rodent tests involves a number of uncertainties. These uncertainties do not mean that we should simply ignore the presence of carcinogens. Rather, in the interests of public safety, a serious effort should be made to resolve the questions surrounding the presence of chemicals identified as rodent carcinogens in our food. Environ. Mol. Mutagen. 39:69-80, 2002 Published 2002 Wiley-Liss, Inc.

In plain English

This 'Perspectives' article matched an FDA list of substances added to food against National Toxicology Program rodent cancer tests, yielding 58 test entries (benzyl acetate was tested twice). Of these, 27 were positive in at least one of four groups (male and female rats and mice) and 28 in none. Excluding banned or no-longer-used chemicals, 20 of 46 (43%) were rodent carcinogens. Of 18 chemicals mixed into feed, 1 was a rodent carcinogen, versus 16 of 23 given by stomach tube (gavage). The author said this suggests route may matter and called for testing the same chemicals by different routes. Applying the 43.5% rate, the author estimated 1078 carcinogenic food substances; the article says this is for the full list, but the figure matches its 2480 currently used entries. The author cautioned that the sample was small and possibly biased and that rodent results may not apply to humans. The author concluded that the Delaney clause may not protect against carcinogens as much as assumed. Asbestos is mentioned only as an example of a known human carcinogen.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Analysis of existing data: an FDA list of substances added to food was combined with the National Toxicology Program list of rodent-tested chemicals, and the rodent cancer test results were tabulated
• Who was studied: 58 test entries for chemicals on the FDA's "Everything" Added to Food in the United States (EAFUS) list that had been tested in National Toxicology Program rodent bioassays (male and female rats and mice). Benzyl acetate appears twice because it was tested twice.
• Exposure: Food additives and other chemicals added to food, given to rodents in feed, in drinking water, by stomach tube (gavage), or by other routes such as inhalation or skin application
• Main results:
  • Of the 58 entries, 27 were rodent carcinogens (a positive response in one to four sex–species groups), 28 were noncarcinogens, and 3 were incompletely evaluated
  • After excluding chemicals banned or no longer reported as used in food, 20 of 46 (43%) were rodent carcinogens
  • By route: chemicals given in feed included 17 noncarcinogens and 1 carcinogen; chemicals given by gavage included 7 noncarcinogens and 16 carcinogens
  • Chemicals given by drinking water, inhalation, skin painting or injection included 4 noncarcinogens and 10 carcinogens
  • The author reported that noncarcinogens were significantly more common in feed studies than in gavage and other non-feed studies combined (Pearson chi-square; the comparison symbol before the P value of 0.001 is lost in the text). The author said this suggests non-feed routes may increase the likelihood of a positive result.
  • The author noted that few studies have given the same compounds by different routes, and called for systematic comparisons
  • Benzyl acetate given by gavage was reported carcinogenic in male and female mice but negative in rats; given in feed it was negative in all four groups
  • Among presently used EAFUS chemicals, 2 of 42 (5%) were positive in the Salmonella mutation test
  • The author extrapolated that if 43.5% applied to the EAFUS list, 1078 food substances would be carcinogens. The article calls this the full list (3125 agents), but 1078 corresponds to 43.5% of the 2480 currently used entries.
  • On a separate FDA list of indirect additives (substances used in food-contact materials), 48 of 88 adequately tested agents (55%) were rodent carcinogens
  • Generally recognized as safe (GRAS) substances had the lowest proportion of carcinogens: 4 of 14, compared with 13 of 23 direct or secondary direct additives and 6 of 11 indirect-only additives
• The authors' conclusion: The author concluded that public attention should be given to the possibility that the Delaney clause may not provide the high level of protection from carcinogens commonly assumed. The author also stated that confounding factors prevent a firm conclusion.
• Limitations the authors state:
  • The sample was small and possibly not representative of all chemicals added to food
  • Chemicals chosen for NTP testing could be biased toward those suspected of causing cancer
  • The sample was small and possibly biased, and rodent test results may not be relevant to human exposures
  • Testing at the maximum tolerated dose, especially by gavage, may far exceed intended food-additive levels, so a positive result may not always be meaningful
  • Negative results from feed studies could be falsely reassuring, because animals may eat less of feed containing the chemical

Topics

Medical terms in this study, explained in our Lexicon: Carcinogenic.

Related studies

• Mineralogical features associated with cytotoxic and proliferative effects of fibrous talc and asbestos on rodent tracheal epithelial and pleural mesothelial cells — A.G. Wylie, H.C. Skinner, et al. (1997)
• Talc-treated rice and Japanese stomach cancer — R.R. Merliss (1971)
• Effect of exposure route on potency of carcinogens — W.E. Pepelko (1991)
• Review of occupational lung carcinogens — K. Steenland, D. Loomis, et al. (1996)
• Human risk assessment of carcinogens from an administrative perspective — I. Tesseraux, A. Lommel, et al. (1990)
• Inhalation carcinogenesis from various forms of asbestos — A.L. Reeves, H.E. Puro, et al. (1974)

Browse more research in Environmental and Molecular Mutagenesis, or return to medical research.