Response of nitrosomethylurea-induced rat mammary tumor to endocrine therapy and comparison with clinical response.

We have compared the response of the N-methyl-N-nitrosourea-induced rat mammary tumor to various endocrine agents with response in patients with breast cancer. To do this, we have induced tumors in 228 animals (65% of intact rats developed tumors but only 14% of ovariectomized rats developed tumors). In intact rats, 4-hydroxyandrostenedione, tamoxifen, and a combination of tamoxifen, aminoglutethimide (AG), and danazol induced significant tumor regression (P less than 0.001, P = 0.01, P = 0.04, respectively), whereas danazol, AG, and trilostane were ineffective when used singly. Further investigation showed that the inactivity of AG in the rat was due to extensive acetylation of this compound. In order to mimic postmenopausal breast cancer, we ovariectomized rats after N-methyl-N-nitrosourea administration. In ovariectomized animals, AG was again ineffective in inducing tumor regression but 4-hydroxyandrostenedione was highly active when compared to controls (P = 0.002). Comparison with response of this model to endocrine therapy with response in patients indicates that this has good predictive capacity since it shows that agents which have minor activity in human breast cancer such as danazol and trilostane are inactive in the model. The intact rat model does not, however, predict whether a drug will be useful for pre- or postmenopausal patients. We recommend that before committing large numbers of patients to clinical trials on the basis of this model, the metabolism of new compounds should be compared in humans and rats.