Impact of Granulocyte Colony-Stimulating Factor during Induction Therapy in Children with Acute Myelogenous Leukemia: Results from the Prospective and Randomized Trial AML-BFM 98.

Background: Treatment for acute myelogenous leukemia (AML) is one of the most intensive treatment modalities in pediatric oncology. Infection-related mortality is particularly high among these patients during chemotherapy-induced neutropenia. Granulocyte colony-stimulating factor (G-CSF) expands the circulating pool of neutrophils by stimulating proliferation and maturation of myeloid progenitor cells. Although G-CSF is widely used for the prevention of infectious complications in cancer patients, no larger randomized study has evaluated the effect of G-CSF in pediatric cancer patients undergoing treatment for AML.

Patients and Methods: The impact of G-CSF on hematopoetic recovery, infectious complications and five-year event-free survival (5-EFS) was prospectively evaluated in the multicenter clinical trial AML-BFM 98. Patients (inclusion criteria: primary diagnosis of AML, age 0–18 years) were randomized to receive or not to receive G-CSF (5 μg/kg/day) after induction 1 (cytarabine, idarubicin, etoposide; AIE) and induction 2 (high-dose cytarabine and mitoxantrone; HAM). Patients who had more than 5% blasts in the bone marrow on day 15 were excluded from randomization. Patients with Down Syndrome were included in the clinical trial, but received only 60% of idarubicin in induction 1 and no second induction (HAM).

Results: Between July 1998 and June 2003, 546 patients entered the clinical trial AML-BFM 98 (Down Syndome: n=45). Three-hundred-and-twenty-seven patients were eligible for randomization. Hundred-fifty eight of these patients were randomized to receive G-CSF, and 169 patients not to receive G-CSF. Both groups did not differ in their clinical characteristics. The G-CSF group had a significantly shorter duration of neutropenia (ANC<500/μl) both after induction 1 (16.3 vs 20.0 days, P=.019) and after induction 2 (12.4 vs 16.3 days, P=.0024). The time of thrombocytopenia (<20,000/μl) was similar in both groups (9.3, 9.4, 5.5, and 6.3 days, respectively). No impact of G-CSF was found on the incidence of grade II infections (fever without an identified pathogen) and grade III/IV infections (fever due to an identified pathogen/sepsis syndrome). Five-year event-free survival (5-EFS) did not significantly differ between the groups with and without G-CSF (65.6% + 3.9% vs 65.7% + 3.8%, P-logrank=.86). No significant difference was found when comparing the results of children randomized for G-CSF (intended-to treat) and of children who actually received the hematopoietic growth factor. Children with Down Syndrome showed also similar 5-EFS with and without G-CSF (P-logrank=.72).

Conclusion: G-CSF significantly shortens the duration of neutropenia in children undergoing induction therapy for AML, but does not influence the incidence of infectious complications nor outcome (5-EFS probability). Whether there is a subgroup of children with AML who benefit from the administration of the hematopoetic growth factor has to be evaluated.