International Agency for Research on Cancer (IARC) - Summaries & Evaluations

(Group 3)

For definition of Groups, see Preamble Evaluation.

Supplement 7: (1987) (p. 208)

Sodium fluoride
CAS No.: 7681-49-4

Fluosilicic acid
CAS No.: 16961-83-4
Chem. Abstr. Name: Silicate (2-), hexafluoro-, dihydrogen

Sodium silicofluoride
CAS No.: 16893-85-9
Chem. Abstr. Name: Silicate (2-), hexafluoro-, disodium

CAS No.: 14542-23-5
Chem. Abstr. Name: Fluorite [CaF2]

Stannous fluoride
CAS No.: 7783-47-3
Chem. Abstr. Name: Tin fluoride [SnF2]

Sodium monofluorophosphate
CAS No.: 10163-15-2
Chem. Abstr. Name: Phosphorofluoridic acid, disodium salt

A. Evidence for carcinogenicity to humans (inadequate)

Only studies on water fluoridation and cancer were reviewed. Comparisons have been made of mortality from cancers at all sites and from particular types of cancer between areas with high concentrations of inorganic fluoride in drinking-water (either occurring naturally or as a consequence of fluoridation) and areas with low concentrations, or before and after fluoridation; the areas or groups of areas most frequently studied are Australia, Canada, China, England and Wales, New Zealand, Norway and the USA [ref: 1-6]. When possible, confounding of fluoride concentration with relevant variables such as age, sex, race and ethnic composition of the populations was taken into account. Fluoridation of drinking-water was introduced in the USA in 1950 [ref: 1], and thus the studies in the USA encompass periods of observation of 20 years or more. Studies of areas with different levels of naturally fluoridation cover longer periods of exposure [ref: 1,6]. The studies have shown no consistent tendency for people living in areas with high concentrations of fluoride in the water to have higher cancer rates than those living in areas with low concentrations or for cancer mortality rates to increase following fluoridation.

In several studies, trends in cancer incidence or mortality in naturally or artificially fluoridated areas and in areas with low natural fluoride content and no artificially fluoridated water were evaluated according to individual cancer sites or groups of sites [ref: 1,3,4,6]. Since a large number of comparisons were made, some would be expected by chance alone to show differences. However, no consistent difference has been seen, and there have been as many significant negative associations between fluoridated water supplies and cancer incidence or mortality as there have been positive associations.

Many studies, therefore, cover the range of doses of fluorides in drinking-water to which humans are exposed, and these are mutually consistent in not showing a positive association between exposure to fluoride and overall cancer rates or rates of different cancers. The Working Group noted that the studies involved were of the ecological or correlation type. The Group was therefore unable to classify the evidence for inorganic fluorides used in drinking-water as 'suggesting lack of carcinogenicity'.

B. Evidence for carcinogenicity to animals (inadequate)

Sodium fluoride was tested in three experiments in three different strains of mice by oral administration. The available data are insufficient to allow an evaluation to be made [ref: 1].

C. Other relevant data

Epidemiological studies have shown no association between the presence of fluorides in drinking-water and the incidence of Down's syndrome [ref: 7].

Sodium fluoride did not induce DNA strand breaks in testicular cells of rats treated in vivo and did not cause chromosomal aberrations in bone-marrow or testicular cells or sister chromatid exchanges in bone-marrow cells of mice treated in vivo. It was reported to induce unscheduled DNA synthesis in cultured human cells, and conflicting results were obtained on the induction of chromosomal aberrations; it did not induce sister chromatid exchanges. It induced transformation of Syrian hamster embryo cells in vitro. At high doses and low cell survival, sodium fluoride induced dose-related increases in mutations in cultured mouse lymphoma cells. It did not induce aneuploidy in Drosophila. It induced chromosomal aberrations in plants. It did not induce gene conversion in yeast and was not mutagenic to bacteria [ref: 7].

Stannous fluoride, sodium monofluorophosphate and sodium silicofluoride did not induce sex-linked recessive lethal mutation in Drosophila, and sodium monoflourophosphate did not induce dominant lethal mutations in Drosophila [ref: 7].

Overall evaluation

Fluorides (inorganic, used in drinking-water) are not classifiable as to their carcinogenicity to humans.

For definition of the italicized terms, see Preamble Evaluation.

Also see previous evaluation: Vol. 27 (1982)


1. IARC Monographs, 27, 237-303, 1982

2. Wang, B. (1981) The relationship between malignancy and fluorine content in the drinking water in Xilinkuolemeng, Nei Menggu (Chin.). Zhonghua Zhongliu Zazhi, 3, 19-21

3. Neuberger, J.S. (1982) Fluoridation and cancer: an epidemiologic appraisal. J. Kansas med. Soc., 83, 134-139

4. Chilvers, C. (1982) Cancer mortality by site and fluoridation of water supplies. J. Epidemiol. Commun. Health, 36, 237-242

5. Chilvers, C. (1983) Cancer mortality and fluoridation of water supplies in 35 US cities. Int. J. Epidemiol., 12, 397-404

6. Chilvers, C. & Conway, D. (1985) Cancer mortality in England in relation to levels of naturally occurring fluoride in water supplies. J. Epidemiol. Commun. Health, 39, 44-47

7. IARC Monographs, Suppl. 6, 312-315, 1987

Synonyms for Sodium fluoride

Synonyms for Fluosilicic acid

Synonyms for Sodium silicofluoride

Synonyms for Fluorspar

Synonyms for Stannous fluoride

Synonyms for Sodium monofluorophosphate

Last updated: 9 March 1998

    See Also:
       Toxicological Abbreviations
       Fluorides (EHC 227, 2002)