Prospective randomised double-blind trial of the in vivo use of recombinant human erythropoietin in bone marrow transplantation from HLA-identical sibling donors. The Australian Bone Marrow Transplant Study Group.

Ninety one patients between the ages of 17 and 58 years undergoing histocompatible allogeneic transplants from sibling donors were entered into a double-blind randomised trial to evaluate the effect of human erythropoietin (rhu EPO) at a dose of 300 units per kg/day given thrice weekly by intravenous injection on erythropoiesis and on erythrocyte and platelet transfusion requirements. Dose was ceased when the haemoglobin exceeded 12 g/dL and recommenced if the haemoglobin fell below 12 g/dL, at 150 units/kg/day. If the haemoglobin exceeded 12 g/dL on a further occasion, the dose of rhu EPO was not given. Patients received two units of erythrocytes when the haemoglobin dropped below 8.5 g/dL and received platelet transfusions when the count dropped below 20×10 9 /L. Univariate analysis revealed a significantly higher reticulocyte count, haemoglobin concentration and bone marrow erythropoiesis after day 14 in the group receiving rhu EPO but this was not reflected in decreased erythrocyte transfusion (7±5 in controls versus 6±5 in rhu EPO group) or in platelet transfusions (11±7 in controls versus 11±9 in rhu EPO group). Hospitalisation in each group was the same (29±8 in the control group and 28±8 in the rhu EPO group). However, in the multivariate analysis, the administration of rhu EPO was associated with an 18% reduction in erythrocyte transfusion requirement when other variables were taken into account. No side-effects due to rhu EPO were detected in this study. Despite the less than optimal secretion of erythropoietin noted in patients after allogeneic bone marrow transplantation, the administration of intermittent, low doses of rhu EPO in such patients did not have a significant effect on the transfusion of erythrocytes or platelets and did not result in shorter hospitalisation