tests of the carcinogenicity of sex steroid hormones. It also reviews studies in these three species designed to elucidate the hormonal mechanisms of mammary cancer in general and in relation to exogenous administration of sex steroid hormones.
When the first steroid contraceptives were developed, they were submitted to the same toxicological tests that had been routinely conducted with other drugs. These tests were performed to identify the acute and chronic toxic effects of the drugs, and they were generally carried out in one rodent and one nonrodent species.
Because contraceptive steroids were to be taken by a large number of healthy young women for a considerable length of time, and because many important cancers in women were considered to be hormonally mediated, the carcinogenic risk of contraceptive steroids was of particular concern. This concern led the U.S. Food and Drug Administration (FDA) to require long-term tests of tumorigenicity in at least two species as a precondition for marketing approval. Later, FDA and other drug regulatory agencies began to require additional evaluation of contraceptive steroids in long-term studies in beagle dogs and rhesus monkeys.
Information from the long-term tests conducted to obtain marketing approval constitutes a valuable data resource and is reviewed here in some depth. It is important to recognize, however, that in 1932, Lacassagne (1932) demonstrated that exogenous estrogens caused mammary cancer in mice. By the late 1950s, it was already well established that exogenous estrogens caused mammary cancer in at least some strains of mice and rats.
A 1972 report by the U.K. Committee on Safety of Medicines (CSM, 1972) constitutes the largest data base on tumorigenicity of contraceptive steroids in mice and rats. The studies reported by the CSM involved 7,000 mice and 6,500 rats who were given six different estrogen/progestogen combinations at three doses and were examined for the presence or absence of neoplasia at 18 different sites. The tests of different estrogens and estrogen/progestogen combinations were carried out according to a standard protocol. Almost all experiments in mice used the CF-LP mouse strain; all of the experiments used 120 mice per treatment group. Doses of contraceptive steroids were 2-5, 50, and 200-400 times the human contraceptive dose. The