The Optimal Busulfan Dose of Conditioning Regimens for Allogeneic Hematopoietic Cell Transplantation in Younger Patients with Acute Myeloid Leukemia

Background: In allogeneic hematopoietic cell transplantation (HCT) for younger patients with acute myeloid leukemia (AML), myeloablative conditioning (MAC) regimens have been considered as standard conditioning. MAC can reduce the risk of relapse, but increase non-relapse mortality, including organ toxicity and severe graft versus host disease. Therefore, the optimal busulfan dose in conditioning regimens has not been established in younger patients with AML. Thus, we aimed to investigate post-HCT outcomes according to the busulfan dose of conditioning regimens in younger patients with AML.

Methods: We retrospectively analyzed long-term outcomes according to the busulfan dose (2 days vs. 3 days vs. 4 days) in 686 adult patients (age <55 years) with AML who had undergone HCT between 2000 and 2020 at the Asan Medical Center, Seoul, Korea. The clinical data was collected from the Asan Medical Center HCT registry. Patients with AML who had undergone a second or third HCT were excluded from this study.

Results: The median age was 41 (range 15-54). The 5-year overall survival (OS) and event-free survival (EFS) rate was 52.2% and 46.4%, respectively. The 3 days of busulfan (Bu3) (n=88) showed significantly higher OS than 4 days of busulfan (Bu4) (n=229) and 2 days of busulfan (Bu2) (n=369) (2-year OS; 76.1% vs. 65.1% and 56.3%, respectively; P<0.001). The EFS rate of Bu3 and Bu4 was significantly higher than that of Bu2 (2-year EFS; 60.2% and 59.8% vs. 46.9%, respectively; P<0.001). Among the three groups, there was no significant difference in GRFS (P=0.952). The cumulative incidence of relapse was significantly lower in Bu4 (26.2% at 2 years) than in Bu2 (40.4%) and Bu3 (36.4%) (P<0.001). However, the NRM was significantly higher in Bu4 (13.9% at 2 years) and Bu2 (12.7%) than in Bu3 (3.4%) (P=0.005). In HCT at complete remission (CR) status, Bu3 showed significantly better OS than Bu4 and Bu2 (P=0.039), whereas there was no significant difference in EFS according to the busulfan dose (P=0.121). In patients with refractory disease, however, the higher dose of busulfan (Bu3 or Bu4) did not show survival advantages in terms of OS (P=0.153) and EFS (P=0.829) over Bu2. Regarding donor types, Bu3 showed significantly better OS than Bu2 in HCT from matched sibling donors (MSD) (P=0.030) and haploidentical family donors (HFD) (P<0.001). However, in HCT from unrelated donors (URD), the Bu2 group showed the best GRFS, and OS in Bu2 or Bu3 was significantly longer than in Bu4 (P=0.028).

Conclusion: For younger patients with AML, Bu3 would be the optimal conditioning intensity in HCT from MSD or HFD at CR status. In URD, Bu2 or Bu3 would be appropriate. In refractory disease, increasing busulfan dose intensity could not improve post-HCT outcomes.