Between March 1986 and December 1987, two groups of consecutive patients with advanced breast cancer underwent epidoxorubicin (Epidx) monochemotherapv. Twenty-three patients (group A) received Epidx at a dose of 60 mg/m2and 27 (group B) at a dose of 120 mg/m2(i.v. every 3 weeks). No patient had undergone anthracycline treatment before entering the study. Age ranged from 39 to 70 years (mean 52) in group A and from 35 to 69 (mean 50 in group B). The main sites of involvement were liver (5 patients in group A and 10 in group B), lung (4 and 5 patients. respectively), bone (7 and 8 patients, respectively), and soft tissue (6 and 5 patients, respectively).The number of courses of therapy ranged from 4 to 10 (mean 7.4) in group A and from 3 to 10 (mean6.6)in group B. Tumor response and toxic effects were graded according to World Health Organization criteria. CR + PR were 35% in group A and 67% in group B (chi square = 3.862, p<0.05). Results were analyzed at 130 weeks from the beginning of the therapy. At this time, survival was 9% in group A and 15% in group B, with a median survival time of 61 weeks (range 18–130) and 77 weeks (range 24–130), respectively. No patient in group A showed cardiac toxicity higher than grade 2 during or after the treatment, whereas in group B, 2 patients developed congestive heart failure after a cumulative Epidx dose of 1080 and 1200 mg/m2. Treatment delays, to allow recovery of white blood cells, were infrequent and occurred only in patients previously subjected to chemotherapy. No patient required hospitalization for sepsis, and alopecia was reversible in all patients.Our data demonstrate that there is a relationship between Epidx dose and response rate in advanced breast cancer.