A prospective study begun in 1976 to determine the incidence of endometrial cancer in postmenopausal women was undertaken because of the adverse publicity alleging an increased risk of endometrial cancer in women receiving estrogen therapy. A retrospective study for the year 1975 was added. During 8,170 patient-years in the 3-year 1975-1977 period, 14 endometrial malignancies were diagnosed, for an annual incidence of 1.7 per 1,000 women. During the 3,792 patient-years of observation of estrogen-progestogen users, the incidence of endometrial cancer was 0.5:1,000. The incidence of this carcinoma among estrogen users (8 cases during 2,088 patient-years) was 3.8: 1,000.The difference between these two groups was statistically significant (p < 0.01).One adenocarcinoma was detected in a patient who used estrogen vaginal cream for 7 months, for an incidence of 1.7:1,000 during 573 patient-years of observation. In the untreated women, during 1,515 patient-years, there were 3 endometrial cancers, for an incidence of 2.0:1,000. There was no significant difference between the untreated group and the estrogen users, and only a trend (p < 0.21) between the estrogen-progestogen users and the untreated women. Synthetic progestogens were used to treat 199 women with endometrial hyperplasia (a precancerous lesion) for 3-6 months. The hyperplastic endometrium reverted to normal endometrium in 96.5 percent. The authors believe that all postmenopausal women with an intact uterus should be given the Progestogen Challenge Test, and the progestogen continued each month as long as bleeding follows. Such methods should be highly effective in preventing most endometrial cancer in these women.Since the mid 1970's, several reports have implicated estrogen replacement therapy in postmenopausal women as a possible cause of endometrial cancer (1-6). These epidemiologic and retrospective studies indicate that in such estrogen