A Phase 3, Randomized, Placebo-Controlled Trial of Filgrastim in Patients (Pts) with Hematologic Malignancies (HM) Treated with a Myeloablative Conditioning Regimen and Allogenic Bone Marrow Transplantation (allo BMT).

Background. Recombinant methionyl granulocyte colony-stimulating factor (methuG-CSF; Filgrastim) is commonly used to reduce the duration of neutropenia that follows allo BMT performed after high-dose chemo-/radiotherapy (HDT) in pts with HM. Recently, a retrospective chart review suggested that leukemic pts treated with G-CSF after allo BMT had faster neutrophil recovery, but slower platelet engraftment, increased risk of graft vs host disease (GvHD), and reduced survival relative to pts who did not receive G-CSF (Ringdén et al, 2004). The present double-blind, randomized, placebo-controlled, multicenter phase 3 study was designed to evaluate the effects of Filgrastim treatment on neutrophil recovery. Pts were to be followed for 2 years.

Methods. Eligible pts were aged 12–55 and scheduled for allo BMT from a sibling donor preceded by high-dose chemotherapy +/− total body irradiation for treatment of acute lymphoblastic leukemia (ALL) or lymphoblastic lymphoma (LBL), high-risk non-Hodgkin’s lymphoma (NHL), acute myeloid leukemia (AML), or chronic myeloid leukemia (CML). Pts were randomized to receive either Filgrastim (5μg/kg/day) or placebo from transplant until recovery of an absolute neutrophil count (ANC) of ≥0.5x10⁹/L or for a maximum of 42 days. Randomization was stratified by age (<18 and ≥18 years) and remission status (1^st complete remission/subsequent remissions) within each center. The primary endpoint was the time to ANC of ≥0.5x10⁹/L. Assuming a standard deviation of 3.8 days, the study was powered to have a probability of at least 80% of detecting a difference of at least 3.6 days between the groups for the endpoint of time to ANC ≥0.5x10⁹/L using a two-tailed significance of 5%.

Results. The first pt was enrolled in 1993; however, the planned sample size of 100 pts (50 per arm) was not reached due to slow enrollment and the study was terminated in 1996. A total of 66 pts were enrolled; of these, 51 were analysed (25 Filgrastim, 26 placebo). Filgrastim-treated pts achieved an ANC of ≥0.5x10⁹/L at a median (quartiles) time of 15 (13, 16) days compared with 19 (17, 22) days for placebo; ANC ≥1.0x10⁹/L was reached by 16 (15, 17) and 22 (21, 28) days, respectively. Median (quartiles) time to discharge from hospital was 30 (26, 43) days for Filgrastim vs 36 (28, 42) days for placebo. The number of pts with platelet count ≥25x10⁹/L within 56 days was 16 for Filgrastim vs 23 for placebo, and 12 vs 18, respectively, for platelet count ≥50x10⁹/L within 56 days. The incidence, maximum severity and number of days to onset of acute GvHD were similar across the groups. No difference in adverse events was reported; 4 Filgrastim pts died (0 of disease progression) and 8 placebo pts died (3 of disease progression).

Conclusion. The results of this phase 3, double-blind, placebo-controlled, randomized trial in pts with HM receiving allo BMT may suggest that Filgrastim treatment enhances neutrophil engraftment with no negative impact on platelet recovery or GvHD.