BACKGROUND. Myeloid growth factors (G-CSF, GM-CSF) are often administered following autologous stem cell transplantation to reduce the duration of neutropenia and limit infection-related morbidity and mortality.

OBJECTIVES. (1) To determine if prophylactic myeloid growth factors (GF) administered to adult patients undergoing autologous stem cell transplantation for a hematological malignancy improve clinical outcomes post-transplant, and (2) to determine the optimal dose and schedule for the administration of GF post-transplantation.

METHODS. Computerized databases were searched for reports from 1966 to March 2004. Reference lists from published reports were hand searched and published abstracts were also considered. Randomized trials comparing different GF regimens were selected by two independent assessors based on explicit inclusion and exclusion criteria; disagreement was resolved by consensus. Two independent reviewers blinded to authors, institution, journal name, and results used a validated scale to assess study quality. When possible, study results were pooled using a random effects model to obtain a pooled relative risk.

RESULTS. Of 985 citations identified, 27 studies evaluating 2268 patients were included. Studies were grouped and analyzed according to their control and experimental arms: (1) early administration of growth factor (< day +3 post-transplant) compared with placebo or no growth factor (N=18 studies), (2) early administration of growth factor compared with delayed growth factor (> day +3, N=8 studies), and (3) high doses of GF (> 5ug/kg/d) compared to standard/low doses of GF (<5ug/kg/d, N=5 studies). Compared to no GF support, early administration of GF resulted in a statistically significant reduction in the median days to neutrophil engraftment (>0.5x109/L) in 14 of 15 studies (range 2–13 days). The magnitude of this benefit was diminished when only studies of higher methodologic quality and optimal transplant conditions (peripheral blood stem cells) were considered (range 2–3 days). A significant decrease in duration of hospitalization was reported in 8 of 15 studies; however no consistent benefit was observed for median days of i.v. antibiotics (3 of 10 studies), median days of fever (1 of 10 studies), or in the rate of microbiologically documented infections (RR 0.94, 95% CI 0.69–1.29). Compared to delayed administration of GF, early administration of GF was not associated with a significant difference in median duration of neutropenia any of 5 studies. Only 1 study reported a significant difference in the median duration of hospitalization and median days of i.v. antibiotics and it was of lower methodological quality. No difference in the rate of microbiologically documented infections was detected (RR 1.29, 95% CI 0.83–2.03). Compared to standard doses of GF, no study evaluating high dose GF reported a statistically significant improvement in any reported outcome but treatment was more costly owing to an increase in total GF administered.

CONCLUSION. The administration of GF post-autologous stem cell transplant reduces the time to neutrophil engraftment and may reduce length of hospital stay. The effectiveness of delayed administration of standard dose GF (5 ug/kg/d) appears similar to strategies that initiate GF support immediately post-transplant or utilize high doses.

Author notes

Corresponding author