Human trophoblast xenografts in athymic mice: a model for peripheral aromatization.

A novel procedure was developed for evaluating aromatase inhibitors using human enzyme in a rodent model. Human choriocarcinoma trophoblast (JAr line) cells injected subcutaneously into athymic nude mice develop into tumor xenografts in 7-14 days which represent sites for peripheral aromatization of androgens. The rapid growth of these trophoblast tumors is estrogen independent. The tumors provide a source of nonovarian human tissue which has relatively high levels of enzyme activity (248 +/- 12 pmol estrogen/g/h) for biochemical determination of in vivo aromatase inhibition. These are major advantages for pharmacological evaluations in comparison to the slow tumor growth response of most carcinogen-induced rodent mammary cancers, which are usually devoid of aromatase activity. In addition, the hormonal dependent components of rodent mammary tumors require several weeks to regress as a result of the indirect effects of estrogen deprivation on tumor growth via inhibition of prolactin dependency, a minor component relative to the role estrogen occupies in hormonally-dependent breast cancer in humans. This model of peripheral aromatization was utilized to evaluate in vivo pharmacological parameters of MDL 18,962 (10-(2-propynyl)estr-4-ene-3,17-dione) such as bioavailability of several formulations, time course and dose responses following different routes of drug administration, pharmacokinetics and tissue distribution of [14C]MDL 18,962. Tumor aromatase activities of trophoblast xenografts were significantly (P less than or equal to 0.05) inhibited when MDL 18,962 was administered intravenously, orally, subcutaneously, or via subcutaneous silastic implants. The ED50 of MDL 18,962 for tumor aromatase inhibition at 6 h after a single treatment was 1.4 mg/kg, s.c. and 3.0 mg/kg, orally. MDL 18,962 blocked aromatase activity more effectively in human trophoblast than in mouse ovarian tissue. Human trophoblast aromatase activity was inhibited by 70% following a single oral dose of 100 mg/kg of MDL 18,962, while the host's ovarian aromatase activity exhibited only marginal inhibition. In vitro, the addition of 10 microM MDL 18,962 to trophoblast tumor cytosol or mouse ovarian cytosol resulted in 99.6 and 91.4% inhibition of aromatase activity, respectively. Tissue distribution of [14C]MDL 18,962 was predominantly associated with endocrine tissues with aromatase activity and organ systems involved in steroid metabolism and excretion. These in vivo data show that MDL 18,962 an enzyme-activated aromatase inhibitor, causes prolonged aromatase inhibition in the absence of saturating levels of inhibitor.