The present phase III trial was carried out to verify whether a kinetic recruitment induced by low doses of diethylstilbestrol (DES) could increase the killing efficacy of chemotherapy in patients with locally advanced breast cancer. One-hundred and seventeen untreated patients with locally advanced breast cancer (stage IIIA/IIIB) were randomized to receive 3 courses of primary chemotherapy consisting of cyclophosphamide (600 mg/m2 i.v.), doxorubicin (50 mg/m2 i.v.) and fluorouracil (600 mg/m2 i.v.) (CAF) on day 1, or DES-CAF (DES, 1 mg orally days 1-3, CAF on day 4). The courses were repeated every 3 weeks. The patients who achieved an objective response were submitted to mastectomy followed by 3 courses of CAF alternated with 3 courses of CMF (cyclophosphamide, 600 mg/m2 i.v.; methotrexate, 40 mg/m2 i.v.; fluorouracil, 600 mg/m2 i.v.), with or without DES. The two treatment arms were well balanced in terms of clinical and pathologic features. There was no significant difference in response rates to induction chemotherapy between the two treatment arms (objective response rate, 63.3% for CAF and 56.1% for DES-CAF). Median overall survival was 49 and 47 months and median progression-free survival was 24 and 21 months for CAF and DES-CAF patients, respectively. Toxicity was not significantly different in the two groups, with the exception of leukopenia: DES chemotherapy was significantly more myelotoxic than the standard treatment, which resulted in a significant reduction in the actual dose intensity. In spite of the attractive experimental evidence, we conclude that so far there is no clinical advantage in the combination of estrogen and chemotherapy. Further research is needed to investigate different schedules of chemotherapy and hormones, or to test the possibility of combining various mitogens.