Objective: Because a categorical refusal of estrogen replacement therapy (ERT) from postmenopausal patients with a history of breast cancer is not based on any research evidence and may be more harmful than beneficial, we evaluated the safety and efficacy of ERT in these women. Methods: We recruited 131 patients who had been treated for breast cancer for a mean of 4.2 years (range 1 month to 20 years) before. Eighty-eight decided to use ERT, whereas 43 refused or had no need for ERT. At recruitment, the patients were carefully examined for breast and gynaecologic findings. Non-hysterectomized patients wishing to receive ERT ( n=54) then started using estradiol as oral tablets (2 mg/day) ( n=44) or as transdermal gel (1.5 mg/day) ( n=10) in combination with 10-day courses of oral medroxyprogesterone acetate at 4-week intervals, whereas hysterectomized patients ( n=34) used only estradiol, orally (2 mg/day) ( n=31) or transdermally (1.5 mg/day) ( n=3). The patients using ERT were carefully examined 6 and 12 months later, and then annually at a specific outpatient department, and the mean follow-up time is now 2.5 years (range from 1 month to 5.2 years, 216 woman-years). The 43 patients not wishing to receive ERT were followed annually at the oncologic department for a mean of 2.6 years (range from 1 month to 4.7 years), and served as a control group. Results: ERT significantly reduced climacteric symptoms, and the Kupperman score fell by 63%, from 26.9±8.6 to 9.9±6.7 (mean±SD). In non-hysterectomized women, medroxyprogesterone acetate triggered withdrawal bleeding in all except seven women. Seven patients (13%) experienced spotting during ERT. In 27 women, endometrial thickness exceeded 10 mm, and two of the total of 54 patients (3.7%) had simple hyperplasia. This vanished spontaneously in 3–6 months. Ten patients terminated the use of ERT within the first 12 to 39 months due to the lack of severe vasomotor symptoms ( n=4) or due to the recurrence of breast cancer or to cancer of the contralateral breast ( n=6). Eighty-one of the 88 patients (92%) using ERT showed no evidence of recurrence, whereas five patients (5.7%) had recurrence in 12–36 months and two patients (2.3%) developed a cancer of the contralateral breast in 14–24 months; another one of those wanted to continue with ERT. Thus the combined risk of recurrence or a new cancer of the contralateral breast in ERT users was 7/216 woman-years (3% per year). In the control group, 38 of 43 patients (88.4%) showed no evidence of recurrence or contralateral cancer, whereas four patients had recurrence and one developed a contralateral breast cancer (5/112 woman-years, 4% per year). Conclusions: Symptomatic climacteric patients with a history of breast cancer benefited from ERT without increasing their risk of recurrence, but the short follow-up and the small number of patients limit any definitive recommendations.