For patients with disseminated endometrial cancer the prognosis is poor. Radiotherapy, chemotherapy or high-dose progestins have been of limited value in the clinic, with low response rates and a usually short duration. Because of the role of estrogen in the etiology of this disease, a rationale exists for therapies using estrogen antagonists. In order to test this strategy, we used the EnDA endometrial carcinoma of the rat recently described by us. The nonsteroidal antiestrogen ZK 119.010 inhibited the primary-tumor growth of the s.c. implanted EnDA endometrial carcinoma by 50%, being superior to high-dose progestin and tamoxifen (TAM). Moreover, in intact as well as in castrated estrogen (E2)-substituted rats, ZK 119.010 substantially reduced metastatic-tumor growth in the lymph nodes and lungs. With TAM, however, the number of lung metastases in intact and in castrated E2-substituted rats either rose or remained stable and the weight of lymph nodes in intact rats increased. After TAM treatment, almost no low-salt-extractable (cytosolic) estrogen receptor (ER) was measurable in the tumor, whereas ZK 119.010 did not alter ER concentrations. The stimulation of metastatic tumor growth, as well as the loss of cytosolic ER under TAM therapy, may reflect the well-known agonist activity of this compound in uterine tissues. ZK 119.010, however, not only lacks this agonist activity, but it exerts a strong antagonistic one. In conclusion, pure antiestrogens may help to improve treatment of endometrial cancer.