BACKGROUND.
Women treated with tamoxifen for breast cancer are at increased risk of endometrial cancer. This study examines the experience at Roswell Park Cancer Institute (RPCI) with women diagnosed with both endometrial carcinoma (EC) and breast carcinoma (BC) to determine the risk and stage of endometrial carcinoma among women treated with tamoxifen. METHODS. The tumor registry was searched for women with diagnoses of both BC and EC between 1980 and 1993. Systemic therapy was classified for all analytic cases of breast carcinoma (women who received primary BC treatment at RPCI). Medical records of all women with both BC and'EC were reviewed, including all analytic and nonanalytic cases. RESULTS. There were 1947 analytic and 1534 nonanalytic BC cases and 877 analytic and 239 nonanalytic EC cases. Thirty-six women in the nonanalytic breast cancer group also had endometrial carcinoma. Fifteen had endometrial carcinoma before breast carcinoma, and 20 of 21 women with breast cancer first had no record of tamoxifen use. Thirty-seven women in the analytic breast carcinoma group had endometrial carcinoma. Endometrial carcinoma preceded breast carcinoma in 29 women. Breast carcinoma preceded endometrial carcinoma in eight women, and two of these developed endometrial carcinoma during or after tamoxifen therapy. Therefore, a total of three women developed endometrial cancer during or after tamoxifen therapy (two analytic and one nonanalytic). The EC was classified as International Federation of Gynecology and Obstetrics (FIGO) Stage L4 (1 patient) and IB (2 patients) with one patient each with histologic Grade I, 11, and 111 after 1, 2, and 5 years of tamoxifen therapy, respectively. No patients had recurrence or died from endometrial carcinoma. The risk of endometrial carcinoma with tamoxifen was determined from the number of women in the breast cancer analytic group receiving tamoxifen. Hormonal therapy was coded as part of systemic treatment in 652 of 1947 analytic patients (33%; 510 as adjuvant therapy and 142 for metastatic cancer). Of these patients, 172 of 652 women (26%) had undergone hysterectomy prior to breast cancer diagnosis, and another 71 women (11%) received nontamoxifen hormone therapy (e.g., prednisone). Tamoxifen therapy was documented in 402 women in the analytic group. (The median age of these women at BC diagnosis was 63 years). Therefore, the maximum estimate of endometrial carcinoma risk is 2 of 402 cases (0.5%). CONCLUSIONS. The risk of endometrial carcinoma with tamoxifen use is low. The value of routine invasive screening for endometrial carcinoma for women receiving tamoxifen should be determined by prospective study.