A randomized, placebo-controlled trial of recombinant human granulocyte colony-stimulating factor administration in newborn infants with presumed sepsis: significant induction of peripheral and bone marrow neutrophilia

Host defenses in the human neonate are limited by immaturity in phagocytic immunity. Such limitations seem to predispose infected newborns to neutropenia from an exhaustion of the neutrophil reserve. Among the critical defects thus far identified in neonatal phagocytic immunity is a specific reduction in the capacity of mononuclear cells to express granulocyte colony-stimulating factor (G-CSF) after stimulation. However, the safety, pharmacokinetics, and biological efficacy of administration of recombinant human (rh)G-CSF to infected human newborns to compensate for this deficiency is unknown. Forty-two newborn infants (26 to 40 weeks of age) with presumed bacterial sepsis within the first 3 days of life were randomized to receive either placebo or varying doses of rhG-CSF (1.0, 5.0 or 10.0 micrograms/kg every 24 hours [36 patients] or 5.0 or 10.0 micrograms/kg every 12 hours [6 patients]) on days 1, 2, and 3. Complete blood counts with differential and platelet counts were obtained at hours 0, 2, 6, 24, 48, 72, and 96. Circulating G-CSF concentrations were determined at hours 0, 2, 6, 12, 14, 16, 18, 24, and 36. Tibial bone marrow aspirates were obtained after 72 hours for quantification of the bone marrow neutrophil storage pool (NSP), neutrophil proliferative pool, granulocyte progenitors, and pluripotent progenitors. Functional activation of neutrophils (C3bi expression) was determined 24 hours after rhG-CSF or placebo administration. Intravenous rhG-CSF was not associated with any recognized acute toxicity. RhG-CSF induced a significant increase in the blood neutrophil concentration 24 hours after the 5 and 10 micrograms/kg doses every 12 and 24 hours and it was sustained as long as 96 hours. A dose-dependent increase in the NSP was seen following rhG-CSF. Neutrophil C3bi expression was significantly increased at 24 hours after 10 micrograms/kg every 24-hour dose of rhG-CSF. The half-life of rhG-CSF was 4.4 +/- 0.4 hours. The rhG-CSF was well tolerated at all gestational ages treated. The rhG-CSF induced a significant increase in the peripheral blood and bone marrow absolute neutrophil concentration and in C3bi expression. Future clinical trials aimed at improving the outcome of overwhelming bacterial sepsis and neutropenia in newborn infants might include the use of rhG-CSF.