Bone marrow or peripheral blood for reduced-intensity conditioning unrelated donor transplantation

There have been no randomized trials that have compared peripheral blood (PB) with bone marrow (BM) grafts in the setting of reduced-intensity conditioning (RIC) transplantations for hematologic malignancy. Because immune modulation plays a significant role in sustaining clinical remission after RIC, we hypothesize that higher graft-versus-host disease (GVHD) associated with PB transplantation may offer a survival advantage.

The primary outcome evaluated was overall survival. Cox regression models were built to study outcomes after transplantation of PB (n = 887) relative to BM (n = 219) for patients with acute myeloid leukemia, myelodysplastic syndrome, or non-Hodgkin lymphoma, the three most common indications for unrelated RIC transplantation. Transplantations were performed in the United States between 2000 and 2008. Conditioning regimens consisted of an alkylating agent and fludarabine, and GVHD prophylaxis involved a calcineurin inhibitor (CNI) with either methotrexate (MTX) or mycophenolate mofetil (MMF).

After adjusting for age, performance score, donor-recipient HLA-match, disease, and disease status at transplantation (factors associated with overall survival), there were no significant differences in 5-year rates of survival after transplantation of PB compared with BM: 34% versus 38% with CNI-MTX and 27% versus 20% with CNI-MMF GVHD prophylaxis.

Survival after transplantation of PB and BM are comparable in the setting of nonirradiation RIC regimens for hematologic malignancy. The effect of GVHD prophylaxis on survival merits further evaluation.