We established a pilot study to evaluate the safety and efficacy of granulocyte (WBC) transfusions using products prepared by granulocyte colony-stimulating factor (G-CSF) plus dexamethasone stimulation of the donor. Neutropenic patients with life-threatening infections were eligible to receive daily G-CSF/steroid-mobilized WBC components. A different volunteer allogeneic, non-HLA matched donor was used each day. As of August 1998, 14 neutropenic patients (four with aplastic anemia, four following allogeneic marrow/peripheral blood stem cell transplant, two with lymphoma, two with LGL-leukemia, and one each with breast cancer and CML) received a median of 6 (range 2 to 28) daily WBC transfusions. Nine of 14 patients had fungal infections. Specific microbial processes included systemic aspergillus infection in six patients, disseminated fusarium in two, and one patient each with disseminated candida, bacterial pneumonia, vancomycin-resistant enterococcemia, pseudomonas skin ulcer, diverticulitis, and RSV pneumonia. Eleven of 14 patients, including all nine with fungal infections, experienced initial clinical improvement, consisting of devervescence in fever, improvement in easily visible skin lesions, and/or radiographic improvement in pulmonary lesions. Five of these 11 either died of other causes (n=3) or had progression of the infectious process after initial improvement (n=2). Only six of the 14 patients who received WBC transfusions were eventually discharged from the hospital, including three of the nine with fungal infections, and three of five with bacterial processes.The granulocytapheresis products contained 7.2 ' 1.4 x 10 to the 10th power granulocytes in a mean volume of 358 ml. Platelet content of the granulocyte products was 4.8 x 10 to the 11th power, equivalent to a 9-unit platelet transfusion. The mean 1-hour post-transfusion increment in absolute neutrophil count (ANC) was 1.87 + 0.75 in 10 patients without splenomegaly, but decreased to 0.38 and 0.24 x 10 to the 9th power/L in patients with splenomegaly (n=3) and human leukocyte antigens (HLA)-alloimmunization (n=1), respectively. ANCs continued to rise for up to 8 hours after each transfusion and did not decline to pretransfusion levels until 24 to 30 hrs after infusion. The granulocyte transfusions were well tolerated, and coadministration (within 4 hours) of amphotericin did not lead to pulmonary compromise. None of the six surviving patients became HLA-alloimmunized. Donors experienced mild bone pain, headache, and insomnia, but were willing and eager to support such donations again. In summary, transfusions of G-CSF-mobilized granulocyte components were well tolerated and resulted in dramatic and sustained increases in circulating granulocytes. The beneficial effects experienced in our patients in this pilot study suggest that G-CSF mobilized granulocytes deserve evaluation in the form of a large, randomized, prospectively controlled, multicenter trial.
