An important issue in allogeneic peripheral blood progenitor cell transplantation is the optimization of the regimen of mobilization of progenitor cells from normal donors. It has been shown that for G-CSF doses up to 10 microg/kg/day, a dose-response relationship exists for the degree of progenitor cell mobilization. Formal comparisons with doses higher than 10 microg/kg/day, however, have not been reported. The aim of this study was to compare the mobilization and collection results of two different G-CSF (Filgrastim) schedules: 10 microg/kg/12 h (n = 20; group A) vs 10 microg/kg/24 h (n = 20; group B). Apheresis sessions were started on day 5 (after 4 days of G-CSF). Adverse events consisted of bone pain, headache, and fatigue which required treatment with acetaminophen +/- codeine in both donor groups. Discontinuation of G-CSF administration for intolerable side-effects was not necessary in any case. The increase in peripheral leukocyte and lymphocyte counts x 109/l on day 5 was higher in group A (56.2 (37.1-75.2) and 4.4 (2. 1-14.6), respectively) than in group B (27.5 (13.2-53.9) and 2.6 (1. 9-5.1), respectively) (P < 0.0001 and P = 0.008). Platelets x 109/l decreased in group A from 228 (161-286) before G-CSF administration to 207 (155-328) on day 5 (P = 0.03), whereas no change was observed in group B. Following the first apheresis, a significant decrease in platelet count was observed with both G-CSF schedules without any differences between groups. The number x 106/kg of both nucleated and CD34+ cells collected after the first apheresis session was higher in group A (672 (462-992) and 5.9 (3.4-10.4), respectively) than in group B (427 (319-608) and 3.1 (1.1-6.8), respectively) (P = 0.0003 in both cases). The median number of CD3+cells x 106/kg collected after one apheresis session was similar with both G-CSF schedules (212 (91-430) in group A and 170 (110-291) in group B) (P = NS). In conclusion, the schedule of 10 microg/kg/12 h was well tolerated and resulted in the collection of a higher number of progenitor cells than 10 microg/kg/24 h without increasing the T cell content. This approach could avoid the donor having to undergo a second apheresis, and facilitate further graft manipulation.