Single Cord Blood Transplantation with Fludarabine, Busulfan and Melphalan Is Crucial Therapeutic Strategy for AML Not in Remission: A Single-Center Retrospective Study of 286 Patients Including Prognostic Factor Analysis

Background The results of allogeneic hematopoietic stem cell transplantation (allo-HSCT) for non-remission acute myeloid leukemia (AML) remain poor, and the development of new treatment strategies is urgently needed. Cord blood (CB) is an alternative graft, but several recent studies have shown high graft-versus-leukemia (GVL) effects of CB , which may be an important advantage in the setting of non-remission transplantation. We previously demonstrated the efficacy and safety of a novel conditioning regimen using fludarabine 180 mg/m2, intravenous busulfan 12.8 mg/kg, and melphalan 80 mg/m2 (Flu/Bu/Mel) for single cord blood transplantation (sCBT). However, we did not validate how disease or patient background might affect the outcome of this method. This study retrospectively aimed to investigate the usefulness of Flu/Bu/Mel regimen for non-remission AML patients stratified based on patient background, disease status and genetic abnormalities. Methods Patients who received first allo-HSCT using single CB at our institution for non-remission AML between January 2010 to March 2024 were enrolled. The probability of overall survival (OS) and progression-free survival (PFS) were estimated using the Kaplan-Meier method, and between-group differences were analyzed using the log-rank test. Multivariate analyses were performed using the Cox proportional hazards model, and hazard ratios (HRs) together with their corresponding 95% confidence intervals (CIs) were calculated. Relapse, nonrelapse mortality (NRM), and neutrophil engraftment were evaluated by accommodating the competing risks, and the Fine-Gray model was used for univariate and multivariate analyses. Results A total of 286 patients were enrolled. The median age of the patients was 59 years (range, 49 to 66 years), with 77 patients having HCT-CI scores over 3 (26.9%). 102 patients (35.9%), 165 patients (58.1%) and 17 patients (6.0%) harbored revised Medical Research Council (rMRC) adverse, intermediate and favorable risks, respectively. The entire OS and PFS at 2 years were 46.5% and 42.9%, respectively. Cox proportional hazards model showed adverse risk group (AR) and intermediate group (IR) relative to favorable risk (FR) stratified by rMRC (hazard ratio (HR) 3.12 (1,22-7.97), p = 0.017, (HR) 2.86 (1,13-7.20), p = 0.026), 60 years and older ((HR) 1.45 (1.05-2.01), p = 0.025), pre-transplantation PS equal 2 or 1 relative to PS equal 0 ((HR) 3.62 (1.70-7.73), p <0.001) to be significantly associated with lower OS. Two-year NRM and relapse incidence (RI) were 35.4% and 21.3%, respectively. Fine and Gray method revealed a negative effect of 60 years and older ((HR) 1.64 (1.07- 2.51), p = 0.024) and PS equal 2 or 1 relative to 0 ((HR) 4.60 (1.42-14.95), p = 0.011, (HR) 2.93 (1.02-8.43), p = 0.046, respectively) for NRM. Regarding RI, AR relative to FR ((HR) 9.61 (1.14-81.01), p = 0.037) and bone marrow blast ≥ 30% before transplantation ((HR) 1.90 (1.11-3.25), p = 0.02) had a negative impact. Subgroup analysis showed Core binding factor AML (CBF-AML) was an excellent 2-year OS compared with non-CBF-AML (80.8% and 40.5%, respectively, p = 0.009). The complex cytogenetics group had dismal outcomes with 2-year OS compared to the others (30.2% and 53.1%, respectively, p < 0.001). There were no significant differences in OS with the presence of FLT3-ITD mutation (58.4% and 50.9%, mutation or wild type, respectively), Killer immunoglobulin-like receptor ligand mismatch (KIR-MM) (53.7% and 45.4%, presence or not, respectively). Discussion and Conclusions sCBT with the Flu/Bu/Mel regimen has demonstrated excellent outcomes for non-remission AML. In particular, it has shown very promising outcomes for non-remission CBF-AML and has the potential to overcome the poor prognosis of patients with FLT3 mutations. AR or IR stratified rMRC and complex cytogenetics cohorts still have low survival rates. In these subgroups, innovations aimed at further improving conditioning regimens or enhancing GVL effects and/or establishing pre- and post-transplant disease control methods using novel agents would be desirable. The results of this study identified contributing factors that can be expected to achieve long-term remission with CBT using the Flu/Bu/Mel regimen, as well as risk factors that may lead to NRM or relapse, and we hope that these findings will lead to new developments in the future.