Impact of NRAS and/or KRAS mutations on post-transplant outcomes in Acute Myeloid Leukemia: A study on behalf of EBMT-ALWP

Abstract

Abstract Background: NRAS, and/or KRAS mutations are identified in 10–15% of patients diagnosed with acute myeloid leukemia (AML). While their prognostic value in non-transplant settings remains inconsistent and treatment-dependent, their impact in the context of allogeneic hematopoietic cell transplantation (allo-HCT) is underexplored. Given the growing use of targeted and maintenance therapies post allo-HCT, understanding whether NRAS/KRAS mutations influence post-transplant outcomes is crucial for risk stratification and future interventional trial design. Methods: This is a retrospective, registry-based analysis conducted after the approval of the Acute Leukemia Working Party of the European Society for Blood and Marrow Transplantation. Adult AML patients who underwent allo-HCT between 2016 and 2021 with known NRAS and/or KRAS mutation status at diagnosis were included, regardless of the disease status at transplant. The primary objective was to evaluate the impact of NRAS/KRAS mutations on post-transplant outcomes including overall survival (OS), leukemia-free survival (LFS), cumulative incidence of relapse (CIR), and non-relapse mortality (NRM). Results: A total of 770 patients with NRAS and/or KRAS mutant AML were identified, with a median transplant year of 2020 and a median age at allo-HCT of 53 years (range: 18–79); 53% were male. Median time from diagnosis to allo-HCT was 5 months (interquartile range: 4–8) for patients in first complete remission (CR1). Most had de novo AML (77%) and intermediate-risk ELN2022 cytogenetics (64%, with normal karyotype in 43%); 25% had adverse-risk cytogenetics. NPM1 and FLT3-ITD mutations were present in 42% and 30% of patients, respectively; 58% were negative for both. Myeloablative conditioning was used in 56% of cases. Donor types included matched sibling (19%), haploidentical (25%), and matched unrelated (33%). At transplant, 537 patients were in CR1, 93 in second or third CR (CR2+), and 140 had active disease. Compared to CR1 and active disease patients, those in CR2+ were more often female (59% vs. 46% vs. 43%, p=0.03), had de novo AML (91% vs. 80% vs. 57%), and showed higher rates of favorable cytogenetic risk (mainly inversion 16; 26% vs. 10% vs. 5%, p<0.001) and NPM1 mutations (67% vs. 37% vs. 38%, p<0.001). After a median follow-up of 2 years (95% confidence interval [CI]: 1.9–2.1), 2-year OS and LFS for the entire cohort were 65% (95% CI: 61-69) and 54% (95% CI: 50-58), respectively; 2-year CIR was 31% (95% CI: 27-35), and NRM 15% (95% CI: 12-18). In CR1 patients, 2-year OS, LFS and CIR were 72%, 61% and 26%, respectively; for CR2+ patients, 2-year OS was 62%, LFS 57%, and CIR 26%; for active disease, 2-year OS was 41%, LFS 28%, and CIR 51%. According to ELN2022 cytogenetic risk, 2-year LFS among CR1 patients was 77%, 61%, and 55% (p=0.04) for favorable, intermediate, and adverse risk, respectively. Among CR2+ patients, 2-year LFS was 55% for favorable and 63% for intermediate risk, while six of the eight patients with adverse risk died before 2 years and two were alive at 1 year without further follow-up. In patients transplanted with active disease, the 2-year LFS was 100%, 31% and 17% for those with favorable (n=6), intermediate and adverse risk, respectively. Multivariable analysis for CR1 patients showed that LFS was significantly worse in secondary AML (hazard ratio [HR]: 1.57; 95% CI: 1.11–2.22; p=0.01), while mismatched related donors, including haploidentical, were associated with improved LFS (HR: 0.54; 95% CI: 0.34–0.86; p=0.009). Cytomegalovirus positivity in patients was also associated with better LFS (HR: 0.71; 95% CI: 0.5-0.96; p=0.03). Increasing age negatively impacted OS (HR: 1.2 per 10-year increment; 95% CI: 1.0–1.4; p=0.02). Conclusion: This large multicenter study included a substantial number of AML patients harboring NRAS and/or KRAS mutations who underwent allo-HCT. Outcomes in this molecular subgroup were encouraging, with over 60% remaining leukemia-free at 2 years. Notably, patients in CR2+ had predominantly favorable features such as NPM1 mutations and inversion 16, and achieved survival outcomes not different to those in CR1. In addition, haploidentical transplant appeared to confer a better LFS in this cohort, suggesting a potential role for graft-versus-leukemia effects in this setting. These findings add to the growing body of evidence supporting the development of a transplant-specific risk stratification that can predict long-term outcomes.