Low Risk of Relapse and Favorable Survival after Allogeneic Transplantation with Fludarabine and Melphalan Conditioning for Measurable Residual Acute Myeloid Leukemia

INTRODUCTION Pre-transplant detectable measurable residual disease (MRD) is associated with increased risk of relapse and mortality after allogeneic hematopoietic cell transplantation (alloHCT) in patients with acute myeloid leukemia (AML) in morphological complete remission (CR). In post hoc analysis of randomized phase 3 trial, myeloablative (MAC) but not the reduced-intensity conditioning (RIC) is shown to mitigate the negative impact of MRD positivity (MRDpos) on alloHCT outcomes of AML (Hourigan, JCO 2019). However, only minority of patients in the RIC cohort of that study received fludarabine and melphalan (FluMel), which in several registry and single institutional reports showed outcomes comparable to MAC. We studied the effect of FluMel RIC in comparison to fludarabine and busulfan (FluBu) MAC on alloHCT outcomes of MRDpos AML in CR. METHODS We retrospectively studied patients with pre-HCT MRDpos AML in CR who received their first alloHCT at our institution from 2015-2022. FluMel consisted of Mel total dose of 100-140 mg/m2 and FluBu with PK-targeted IV daily average Bu AUC of 5300, both in combination with Flu total dose of 150-160 mg/m2. MRD status prior to alloHCT was assessed by either multiparametric flow cytometry, molecular testing (next generation sequencing or polymerase chain reaction) or by cytogenetics. Primary objective was to compare the overall survival (OS) between FluMel and FluBu. Secondary objectives included leukemia-free survival (LFS), relapse and non-relapsed mortality (NRM). RESULTS A total of 191 eligible patients were identified: FluMel was used in 120 (63%) patients while 71 (37%) patients received FluBu. Median age was 61 years (range, 21-77) at HCT and 43.5% of patients had HCT comorbidity index (CI) of 3+. At diagnosis, 63.9% of patients had adverse risk AML by 2022 European LeukemiaNet stratification and 28.7% of all patients had secondary AML. Graft versus host disease (GVHD) prophylaxis consisted of posttransplant cyclophosphamide (PTCy)-based regimen in 44.5% of patients, tacrolimus/sirolimus in 38.2% and tacrolimus/methotrexate in 16.8% of patients. Peripheral blood was the stem cell source in 98% of patients, while most used donor type was matched unrelated (MUD, 59.7%), followed by matched sibling (17.3%), haploidentical (15.7%) and mismatched unrelated (7.3%) donor. FluMel and FluBu groups were comparable in baseline characteristics, except for FluMel group having higher proportion of patients with age ³60 years (76.7% vs 5.6%, p<0.001), Karnofsky score <90 (28.3% vs 12.7%, p=0.02), ELN adverse risk AML (71.1% vs 50.7%, p=0.01) and MUD (66.7% vs 47.9%, p<0.001). Comparing FluMel vs FluBu groups, the 2-year probability of OS was 64% vs 68% (p=0.20), LFS was 58% vs 70% (p=0.09), relapse was 20% vs 22% (p=0.77) and NRM was 22.0% vs 8.7% (p=0.01). In multivariable analysis after adjusting for patient age, HCT-CI and ELN risk, there were no statistically significant differences between the FluMel and FluBu groups in OS (HR=0.7, p=0.41), LFS (HR=0.9, p=0.75), relapse (HR=0.7, p=0.31) or NRM (HR=1.5, p=0.68) CONCLUSIONS RIC FluMel provides clinical outcomes similar to MAC FluBU in patients receiving alloHCT for AML in MRDpos CR. These data suggest that low risk of relapse and favorable survival can be extended to older and less fit patients if FluMel RIC is used for alloHCT in high-risk MRDpos AML.