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


VOL.: 23 (1980) (p. 39)

5. Summary of Data Reported and Evaluation

5.1 Experimental data

Arsanilic acid, arsenic trioxide, sodium arsenite, potassium arsenite (Fowler's solution) and dimethylarsinic acid were tested by the oral route in mice. Lead arsenate, calcium arsenate, arsenic trioxide, sodium arsenate and sodium arsenite were tested by the oral route in rats. Sodium arsenate and arsenite were tested orally in dogs.

Potassium arsenite, arsenic trioxide and sodium arsenate were tested by skin application in mice. Sodium arsenite was tested by inhalation in mice; and arsenic trioxide, a calcium arsenate-copper mixture and copper ore or flue dust containing arsenic were tested by intratracheal administration in rats. Sodium arsenate was tested by intravenous administration in mice. Dimethylarsinic acid was tested by subcutaneous injection in mice; and calcium arsenate was tested by subcutaneous injection in rats. Metallic arsenic was tested by intramedullary injection in rats and rabbits.

In addition, sodium arsenate was tested by subcutaneous injection in mice in an experimental model which included exposures extending from the prenatal to the postnatal period.

Of all these studies, only one involving the subcutaneous administration to mice of sodium arsenate throughout pregnancy and one involving the intratracheal administration of a calcium arsenate-copper mixture to rats provided some evidence of a carcinogenic effect. However, all of the studies, both positive and negative, suffer from some inadequacies.

There is evidence that arsenite and arsenate cross the placenta in mammals. Sodium arsenate and arsenite have embryolethal effects and a teratogenic potential in several mammalian species. A variety of malformations can be induced. When given orally, high doses of arsenate are required to induce a small percentage of abnormalities.

The evidence that arsenic compounds cause mutations and allied effects in bacteria is inconclusive. However, arsenic compounds induce chromosomal aberrations and morphological transformation in mammalian cells.

5.2 Human data

A large number of cases of skin cancer have been reported among people exposed to inorganic arsenic through drugs, drinking-water or pesticides. The clinical presentation and sites of these tumours are different from those of cancers caused by other known skin carcinogens, suggesting that they are causally associated with exposure to arsenic. In one epidemiological study, skin cancer was positively correlated with high arsenic levels in the drinking-water; a second study showed no such correlation, however, the water arsenic levels were substantially lower than those in the first study.

Three cohort studies of workers manufacturing arsenical pesticides showed an excess mortality from respiratory cancer. A further cohort study of workers exposed to lead arsenate during spraying showed no excess mortality from any cancer; however, these people may have been exposed to lower levels than were manufacturing workers.

Case-control and cohort studies in copper smelters demonstrated a significantly increased mortality from respiratory cancer among the workers; however, smelter workers are exposed not only to arsenic compounds but also to other factors in the working environment, some of which may be carcinogenic. An attempt was made to control for exposure to sulphur dioxide, copper, lead, nickel, selenium, antimony and bismuth in one case-control study, and the excess lung cancer remained. Smoking habits were examined in two of the studies and could not account for the excess.

The descriptive epidemiological studies on the mortality of people living in the neighbourhood of copper, lead and/or zinc smelters suggest increased mortality from respiratory cancer. One indicated excess mortality from lung cancer for both men and women, which was not associated with socioeconomic or geographical factors and could not be explained by occupational exposure alone. In the other, the excess mortality from respiratory cancer (which was present only for men) became insignificant when the deaths of workers in the smelter were excluded. These data are inadequate to evaluate the risk of nonoccupational exposure to low levels of airborne arsenic.

Four cases of haemangiosarcoma and one case of carcinoma of the liver have been reported in individuals exposed to medicinal arsenical preparations. One additional case of haemangiosarcoma of the liver was reported in association with general environmental exposure, and two further cases in workers exposed to arsenical pesticides; four cases of liver sarcoma and three of liver carcinoma were associated with vineyard exposure. An excess of lymphomas has been reported in workers in arsenic pesticide manufacture; and excesses of leukaemia, myeloma and colon and liver cancer have been found in smelter workers. An excess of oral cancers has been reported in a population exposed during the spinning of wool which may have been contaminated with arsenical sheep-dip.

Arsenite crosses the placenta. Smelter workers exposed during pregnancy to arsenic compounds (and possibly to other toxic substances) had an excess of infants with low birth weights, an increased frequency of abortions and an increased occurrence of multiple malformations.

An increased incidence of chromosomal aberrations was observed in patients treated with arsenical compounds and in workers exposed occupationally to arsenic compounds in a smelter environment.

5.3 Evaluation

There is inadequate evidence for the carcinogenicity of arsenic compounds in animals. There is sufficient evidence that inorganic arsenic compounds are skin and lung carcinogens in humans. The data suggesting an increased risk for cancer at other sites are inadequate for evaluation.

For definition of the italicized terms, see Preamble Evaluation.

Previous evaluations: Vol. 1 (1972); Vol. 2 (1973)

Subsequent evaluation: Suppl. 7 (1987)

Last updated: 7 April 1998

    See Also:
       Toxicological Abbreviations