Background: Autologous peripheral blood stem cell transplantation (APBSCT) is gradually replacing autologous bone marrow transplantation in many clinical settings. The key question is how to evaluate the quality of grafts. We analyzed the relationship between hematopoietic reconstitution and characteristics of patients and grafts.
Methods and results: Data from 95 APBSCTs were analyzed. Peripheral stem cells were obtained after mobilization using anti-neoplastic chemotherapy followed by Neupogen (G-CSF). After high dose chemotherapy and APBSCT, patients received Leucomax (GM-CSF). Patients were reinfused with a median of 6.1 x 10(6) (range 0.83-29.3) CD34+ cells/kg, and 25.1 x 10(4) (range 1.0-167.0) CFU-GM/kg of body weight. The median time to engraftment was 12 days (both for granulocytes 1 x 10(9)/l and platelets 50 x 10(9)/l). We found a significant correlation between the number of CD34+ cells and CFU-GM reinfused and also between their respective graft sizes and time to leukocytes, platelets, and granulocytes recovery. We did not find a significant correlation between the number of mononuclear cells reinfused and any analyzed parameter (time to engraftment, age, diagnosis, number of previous chemotherapies, type of mobilization or high-dose regimen). However, administration of preparative high-dose chemotherapy consisted of busulphan and cyclophosphamide was associated with the risk of a transient secondary graft failure.
Conclusions: We conclude that the content of progenitors in PBSC grafts and time to booth leukocyte and platelet recovery are best estimated by the number of CD34+ cells (not less than 1 x 10(6)/kg) and CFU-GM (not less than 1 x 10(4)/kg). The number of mononuclear cells in an autologous PBSC graft is not suitable and useful for prediction of engraftment rate. There is probably no additional benefit of reinfusion of more than 8-10 x 10(6) CD34+ cells/kg and/or 50 x 10(4) CFU-GM/kg, because hematopoietic recovery is not more rapid.