Background and objectives: There is only limited experience with conditioning regimens based on busulfan for patients with acute lymphoblastic leukemia (ALL). Therefore, the aim of this study was to compare the event-free survival (EFS), transplant-related mortality (TRM) and the probability of relapse (PR) of patients undergoing hematopoietic cell transplantation (HCT) for ALL conditioned with or without total body irradiation (TBI).
Design and methods: The study sample consisted of 156 patients conditioned with regimens based on TBI (n=114) or on high doses of oral busulfan (BU) (n=42). Most of the BU group received phenytoin as prophylaxis for seizures. The median follow-up was 6 years.
Results: EFS at 6 years was 43% (95% CI 35%-51%) versus 22% (95% CI 10%-34%) in the TBI and BU subsets respectively (p=0.01). TRM at 18 months was 22% and 17% in the BU and TBI groups (p=0.24), respectively. At 3 years actuarial PR was 71% in the BU group and 47% in the TBI group (p=0.01). In the multivariable analysis, a worse EFS was associated with BU, relative risk (RR) 1.7; advanced disease versus 1st and 2nd complete remission (CR) at HCT, RR 2.5; absence of chronic graft-versus-host disease, RR 1.8; development of veno-occlusive disease RR 2.2 and shorter CR duration before transplant. INTERPRETATION AND CONCLUSIONS. TBI was associated with a lower relapse rate and better EFS, even in patients in 1(st )and 2(nd) CR, than schemes based on high doses of busulfan. This suggests that conditioning regimens based on TBI should remain the standard method of preparative regimen for patients with ALL.