In an effort to determine whether subcutaneous or continuous intravenous infusion administration of rhGM-CSF results in better hematopoietic progenitor mobilization, the findings of two sequential clinical trials were reviewed. Patients who had received prior chemotherapy for leukemia, lymphoma, multiple myeloma, breast cancer, or other solid tumors and were candidates for high-dose therapy received rhGM-CSF, 250 micrograms/m2/day, either as a continuous intravenous infusion (trial 1) or subcutaneously (trial 2) for stem cell mobilization. At least five apheresis collection procedures were performed to collect a target number of 6.5 x 10(8) mononuclear cells (MNC)/kg. For the 37 patients in trial 1, the collections contained a median of 7.99 x 10(8) MNC/L (range 6.42-21.36) and a median of 5.27 x 10(4) CFU-GM/kg (range 0.28-19.35). In trial 1, 25 patients were autografted with their cells and recovered 0.5 x 10(9) granulocytes/L at a median of 12 days (range 6-16). For the 33 patients in trial 2, the autograft product contained a median of 7.63 x 10(8) MNC/kg (range 6.51-22.66) and 6.31 x 10(4) CFU-GM/kg (range 0.06-60.4). In trial 2, 25 patients were autografted. The median time to reach 0.5 x 10(9) granulocytes/L was 11 days (range 9-26). All patients received rhGM-CSF after peripheral stem cell transplant. No significant differences in the collected products or the time to hematopoietic recovery was found between the two trials (p > 0.05). The mobilization effects of subcutaneous rhGM-CSF in these pretreated patients were similar to those of intravenous rhGM-CSF.