Purpose: Many attempts have been made to improve the results of allogeneic bone marrow transplantation (alloBMT) for patients with acute myeloid leukemia (AML) in first complete remission (CR1). Bone marrow cell dose has been reported to be an important factor in alloBMT; however, its true impact on relapse incidence (RI), leukemia-free survival (LFS), and nonrelapse mortality (NRM) in a large cohort of patients is unknown.
Patients and methods: From January 1992 to December 1999, 572 bone marrow transplantation recipients reported to the European Blood and Marrow Transplantation (EBMT) registry underwent allografting from an HLA-identical sibling donor with an unmanipulated bone marrow for AML in CR1.
Results: The median number of nucleated cells (NCs) infused was 2.6 x 10(8)/kg. Estimated 5-year NRM, LFS, and RI for patients receiving more or less than 2.6 x 10(8) NCs/kg were, respectively, 18% +/- 5% v 30% +/- 5% (P =.001), 68% +/- 3% v 46% +/- 3% (P <.00001), and 14% +/- 4% v 24% +/- 5% (P =.004). The association of cell dose with the above outcomes was confirmed in multivariate analyses for NRM (relative risk [RR], 0.53; P =.0007), for LFS (RR, 0.53; P =.00008), and for RI (RR, 0.57; P =.02). The cell dose was also an important factor for neutrophil (RR, 0.76; P =.009) and platelet (RR, 0.77; P =.03) recoveries; however, it did not statistically influence the incidence of acute graft-versus-host disease.
Conclusion: This study shows that marrow cell dose is one of the most important factors influencing relapse, NRM, and LFS after alloBMT for patients with AML in CR1. Therefore, increasing the marrow cell dose should substantially improve the survival of these patients.