Conditioning Regimen Intensity and Donor Choice Do Not Impact Outcomes of Patients (pts) with Acute Myeloid Leukemia (AML) after Allogeneic Hematopoietic Cell Transplantation (Allo-HCT) in First Complete Remission (CR1): An EBMT Analysis

Abstract

Allo-HCT is a curative treatment for fit pts with intermediate and adverse risk AML who achieve CR1. The presence of minimal residual disease (MRD) before allo-HCT has been shown to increase risk of relapse and decrease survival. Here, we describe outcomes of pts with AML who received allo-HCT in CR1 and the impact of MRD before allo-HCT on outcomes. We identified 1032 pts with AML who received allo-HCT in CR1 between 2015 and 2020 from the EBMT registry. The median age was 53.6 (range, 18.1-82.5) years and 526 (51%) were male. Most (94%) pts had de novo AML. FLT3-ITD and NPM1 mutations were detected in 46% and 44% of pts, respectively and high-risk cytogenetics were present in 20%. >1 course of induction therapy was required by 18% of pts to achieve CR1 and 39% were MRD positive at time of allo-HCT. The median time from diagnosis to allo-HCT was 4.9 months. Of pts with FTL3-wild type (FLT3-wt) disease, 33% were MRD negative and 22% were MRD positive at time of allo-HCT. In pts with FTL3-ITD mutation, 29% were MRD negative and 17% were MRD positive at time of allo-HCT. Donor type was matched sibling donor (MSD) in 29% of pts (n=297), matched unrelated donor (MUD) in 47% (n=481), mismatched unrelated donor (MMUD 9/10) in 8% (n=88), or haploidentical donor (haplo) in 16% (n=166). HCT-specific comorbidity index (HCT-CI) was >3 in 22% of pts (n=190). Myeloablative conditioning (MAC) was administered to 45% of pts (n=470), while reduced intensity conditioning (RIC) was used in 55% (n=562). The most common conditioning regimen was fludarabine/busulfan in MSD (n=148, 50%), MUD (n=272, 57%) and MMUD (n=38, 43%) recipients, while thiotepa/fludarabine/busulfan was the most used in haplo-HCT (n=99, 60%). Total body irradiation (TBI) was given to only 8% of pts (n=84), mainly in haplo-HCT (n=29). Most (n=1002, 98%) pts received calcineurin inhibitors (CNI), tacrolimus or cyclosporine A, for graft-versus-host disease (GVHD) prophylaxis in addition to mycophenolate mofetil (43%) or methotrexate (38%). Post-transplant cyclophosphamide was used in 18% (n=191) of pts, with 71% (n=136) having received haplo-HCT, and anti-thymocyte globulin was given to 67% (n=689). The median follow-up was 24 months. At day 60, the cumulative incidence of neutrophil engraftment was 99%. The cumulative incidence of grade III-IV acute GVHD (aGVHD) was 9% at day 180. At 2 years, the cumulative incidence of extensive chronic GVHD (cGVHD) was 17%. At 2 years, relapse incidence (RI), NRM, overall survival (OS), leukemia-free survival (LFS), and GVHD-free, relapse-free survival (GRFS) were 26%, 13%, 69%, 61%, and 48%, respectively. The main cause of death was primary disease in 55% of pts (n=157). Other causes of NRM were infections in 16% (n=45), and GVHD in 17% (n=47). In the multivariate analysis, compared to MSD, the use of MUD increased NRM (hazard ratio (HR) 1.86, p=0.02), GRFS (HR 1.28, p=0.039), aGVHD (HR 2.17, p<0.0001), and cGVHD (HR 1.36, p=0.044). The use of MMUD was associated with worse OS (HR 1.6, p=0.042) and higher risk of aGVHD (HR 2.11, p=0.003). Haplo-HCT was associated with higher incidence of aGVHD (HR 1.69, p=0.021) but lower risk of cGHVD (HR 0.63, p=0.033). Older age, lower performance status (PS), requiring >1 induction to achieve CR1, and FLT3-MRD positivity before allo-HCT were associated with worse LFS. Factors associated with lower OS were MMUD 9/10, older age, poor PS, >1 induction to achieve CR1, adverse cytogenetics, and FLT3-MRD positivity before transplant. In vivo TCD was associated with improved GRFS and lower risk of acute and cGVHD without increase of relapse. The use of RIC compared with MAC did not affect relapse, NRM or any other outcome. In conclusion, in pts with AML transplanted in CR1, those requiring >1 induction or with persistent FLT3-MRD positivity at time of allo-HCT had lower OS. Strategies to improve induction therapy, possibly through the addition of targeted agents, aiming to improve CR rates and deepening responses could potentially improve OS after allo-HCT. The use of RIC was associated with similar outcomes compared to MAC. This finding is particularly important for older pts or those with high HCT-CI scores unlikely to tolerate MAC. Survival after MSD allo-HCT is comparable to MUD and haplo, a finding that is in accordance with previous results for haplo and MUD as valid alternative donors in the absence of MSD. The use of MMUD was associated with lower OS compared to MSD and should be used as a last resort donor in this setting.