Proliferative liver lesions and sex steroids in rats

Abstract

In standard two year tumorigenicity studies some gestagens, alone or in combination with an estrogen have been shown to produce proliferative liver lesions (PLL) in rats. Estrogens in general have not produced PLLs; for ethinyl estradiol the situation is equivocal. The steroids which increase the incidence of PLLs have not been shown to have the characteristics usually associated with classical hepatocarcinogens such as diethylnitrosamine (e.g. negative mutagenicity and cytoxicity, failure to bind covalently to DNA). Therefore, another mechanism must be assumed. A possible explanation is given by the ability demonstrated for a few steroids (mestranol alone or in combination with norethindrone, ethinylestradiol) to promote preneoplastic islands induced by diethylnitrosamine, resulting in an increase in tumor incidence over rats given only the carcinogen. Spontaneous preneoplastic islands in rats occur in various incidence in different strains and we have recently shown that their reaction to steroids—enhanced proliferation—is the same as the effect on preneoplastic islands produced by classical carcinogens. Sex steroids also enhance the growth or normal hepatocytes. It, therefore, seems reasonable to assume that sex steroids produce PLLs by acting on spontaneous preneoplastic lesions and not by the mechanisms usually associated with classical carcinogens. Extrapolating the incidence of PLLs in rats to a risk in humans may be highly misleading because of different pharmacokinetic patterns. Even in rats treated orally for two years with sex steroids dosed as multiples of the human contraceptive dose, resorption and biovailability may vary by a factor of 10 or more. It is, therefore, inappropriate to state that one sex steroid is more active in the production of PLLs than another, unless the data from pharmacokinetic studies are taken into account. Available evidence shows that sex steroids act as tumor promotors and enhance the growth of normal hepatocytes. It is extremely doubtful whether a sex steroid producing PLLs in the rat after two years of oral treatment can be classified as an initiating tumorigen for the rat. Substantial risk for humans cannot be inferred, because of the high incidence of spontaneous preneoplastic lesions in rats, and because of the very high sensitivity of the rat liver to a variety of chemically unrelated compounds including sex steroids.