The purpose of this study was to determine the safety and efficacy of filgrastim as an adjunct to induction and consolidation chemotherapy in poor risk patients with myelodysplastic syndrome (MDS). Filgrastim was given both during and after chemotherapy with the objective to accelerate hematopoietic repopulation and enhance the efficacy of chemotherapy. In a prospective randomized multicentre phase II trial, a total of 64 patients with poor risk primary MDS were randomized to receive either granulocyte colony-stimulating factor (G-CSF, filgrastim, AMGEN, Breda, The Netherlands) 5 microg/kg/day subcutaneously or no G-CSF in addition to daunomycin (30 mg/m2/days 1, 2 and 3 intravenous bolus) and cytarabine (200 mg/m2 days 1-7, continuous infusion). The overall complete response rate was 63%: 73% for patients receiving filgrastim as compared to 52% in the standard arm (P = 0.08). Overall survival at 2 years was estimated at 29% for patients assigned to the filgrastim arm and 16% for control patients (P = 0.22). The median time for recovery of granulocytes towards 1.0 x 10(9)/l post-chemotherapy was 23 days in the filgrastim-treated patients vs 35 days in the standard arm (P = 0.015). There were no differences in time of platelet recovery, length of hospital stay, duration of antibiotic use or infectious complications between the two treatment groups. However the earlier recovery of neutrophils in the filgrastim group was associated with a reduced interval of 9 days between the induction and consolidation cycle. In patients with poor risk MDS the use of filgrastim during and after induction therapy results in a significantly reduced neutrophil recovery time. Further study may be warranted to see if the apparent trend of the improved response to chemotherapy in combination with filgrastim can be confirmed in greater number of patients and to assess the effect of the addition of filgrastim on survival.