Long Term Immunological and Clinical Effects after a Randomized, Double-Blind Trial of Filgrastim Versus Placebo in Allogeneic Blood Transplantation.

Steven Pavletic, Stefano Tarantolo, James C. Lynch, Sunil Abhyankar, Philip J. Bierman, Julie M. Vose, Robert B. Geller, Joseph McGuirk, James Foran, Robert Gregory Bociek, Abdul Hadi, Suzanne D. Day, James O. Armitage, Anne Kessinger, Michael R. Bi

Author address: 

University of Nebraska Medical Center, Omaha, NB, USA; Saint Lukes Hospital of Kansas City, Kansas City, MO, USA; Arizona Cancer Center, Scottsdale, AZ, USA; National Cancer Institute, Bethesda, MD, USA


Enthusiasm regarding early benefits of posttransplant granulocyte colony-stimulating factor (G-CSF) administration is tempered by concerns regarding delays in functional immunological recovery in patients (pts) receiving T-cell depleted HLA-mismatched related donor transplants (Volpi, Blood 2001; 97:2514). To address the clinical importance of these reports in T-cell replete HLA-matched transplants we analyzed the long-term results of 50 pts enrolled in the randomized trial of filgrastim 10 mg/kg (G-CSF) vs. placebo(Bishop, Blood 2000; 96:80). All patients received related donor HLA-matched T cell replete G-CSF mobilized blood transplants for a hematological malignancy and were treated using an identical transplant protocol. Neutrophil engraftment was superior in the G-CSF arm and there was no difference in 100 day survival or acute GVHD. This report describes the long term outcome and immune function of pts who were enrolled in this trial. 28 pts survived one-year after transplant and 15 were eligible for the comparison of immune function by randomization assignment. At 12 months posttransplant immune parameters (data presented in median values) in G-CSF vs. placebo groups showed the following: a) Lymphocyte subsets (%): CD3 55 vs. 73 (P=0.15), CD4 25 vs. 27 (P=0.78), CD8 23 vs. 45 (P=0.32), CD19 22 vs 18 (P=1.0), CD56 8 vs 6 (P=0.69); b) Immunoglobulins (mg/dL): IgG1 649 vs. 585 (P=0.48), IgG2 266 vs. 129 (P=0.22), IgG3 64 vs 64 (P=0.93), IgG4 36 vs. 23 (P=0.11), IgA 72 vs. 81 (P=0.79), IgM 171 vs 93 (P=0.25); c) Tests of cytotoxicity (effector:target ratios 50:1): NK activity against K562 4% vs. 3.5% (P=0.76), LAK activity against Raji 22% vs.15.6% (P=0.26); and d) Mitogen-induced proliferation (cpm): PHA (2.5 g/ml) 22,116 vs. 18,148 (P=0.63), LPS (200 g/ml) 4,573 vs. 5,436 (P=0.73). 14/26 patients are still alive in the G-CSF arm and 9/24 in the placebo arm. Median follow-up of surviving patients is 43 months. Four year overall and event-free survival in the G-CSF and the placebo group were 53% (95% CI 33-72%) vs. 35% (95% CI 15-55%) (log rank P = 0.40), and 44% (95% CI 24-64%) vs. 36% (95% CI 17-56%) (log rank P = 0.65). Bacterial or fungal infections were the cause of 6/12 deaths in the G-CSF arm (all bacterial) and of 4/15 deaths in the placebo arm (two deaths from aspergillus) (P= 0.26). Two pts died of progressive malignancy and one of a second malignancy (colon cancer) all in the placebo arm. Two patients relapsed in the G-CSF arm and three in the placebo arm. Four year cumulative incidences of relapse were 8% (95% CI 0-35%) vs. 13% (95% CI 0-38%) in G-CSF vs. placebo arms, respectively (log rank P = 0.52). Chronic GVHD developed in 14/19 100 day survivors (11 extensive stage) after G-CSF, and in 17/20 (14 extensive stage) in the placebo arm. While pts receiving G-CSF had a lower incidence of cGVHD this difference was not statistically significant. The four year cumulative incidence of cGVHD was 56% (95% CI 24-88%) after G-CSF and 71% (95% CI 48-94%) after placebo (log rank P = 0.41). In summary, this analysis did not identify any late immunological or clinical detrimental effects of the posttransplant G-CSF administration.

abstract No: 


Full conference title: 

43rd American Society of Hematology (ASH) Annual Meeting
    • ASH 43rd (2001)