A randomised phase II/III study was conducted in patients with advanced breast cancer to determine the dose intensity achievable through an acceleration of administration of chemotherapy with epidoxorubicin and cyclophosphamide (EC) alone, as compared with the combination of this regimen with two different schedules of granulocyte-macrophage colony stimulating factor (GM-CSF). 73 patients received EC intravenous (i.v.) (epidoxorubicin 100 mg/m2, cyclophosphamide 600 mg/m2) on day 1 (group A), or the same chemotherapy plus sub-cutaneous (s.c.) GM-CSF (5 micrograms/kg/day) either from days 3 to 12 (group B) or from days -6 to -3 (group C). The primary objective of the study was the investigation of dose intensity delivered in the three treatment arms, whereas the secondary objective was response rate. A significant increase (P = 0.006) in dose intensity of 21% was observed for treatment group B, whereas the increase in dose intensity achieved in group C (7%) was not significant (P = 0.086). Response rates (complete response (CR) + partial response (PR)) of 56% were observed in group A, 65% in group B, and 57% in group C, respectively. This difference in response rates did not reach statistical significance (P = 0.271). We thus conclude that an acceleration of the EC regimen over the standard schedule could be accomplished with postchemotherapeutic GM-CSF support, leading to an increase in dose intensity, whereas pretherapeutic short-term GM-CSF administration did not reach this goal.