Many of the arguments against use of the beagle as a model for studying the carcinogenicity of progestogens in humans center on a consideration of differences between dogs and humans in the pharmacokinetics of the steroids. A careful, direct comparative study of the pharmacokinetics of five steroids in rats, beagles, and humans was reported by Humpel (1989), who confirmed that there were both important similarities and important differences in various pharmacokinetic parameters between dogs and humans. This study also showed, however, that the pharmacokinetics of the five steroids were different in rats and humans. Thus, a consideration of differences between humans and other species in the pharmacokinetics of various steroids would lead one to reject the rat as well as the dog as a model for studying these compounds. For this reason, data from the experiments reported by the U. K.'s Committee on Safety of Medicines are no more reassuring about the effects of steroid hormones and mammary cancer than the beagle experiments are alarming.
An elegant experiment in beagles done by El Etreby and colleagues (1989) gives considerable insight into the mechanism for differences in mammary tumorigenicity of the various progestogens in the beagle. In this important experiment, the investigators explored the relationship between progestogen dose and mammary tumor response using doses of several progestogens that were larger or smaller than the “standard ” multiples of the human contraceptive dose. They were able to show that all of the progestogens tested were mammary tumorigens. The difference in mammary tumorigenicity of various progestogens was a matter of the dose at which the effect occurred and not a matter of some progestogens causing mammary tumors and some not causing mammary tumors. Levonorgestrel, which is very poorly absorbed by the dog, showed high rates of mammary tumorigenicity but only at doses that are many hundreds of times a multiple of the human contraceptive dose.
The most important lessons from animal studies of sex steroids come not from what must be considered fairly crude attempts to study their intrinsic carcinogenicity, but from studies that have gone further to define the mechanisms for steroid-induced mammary car-