IDH2 R172 mutation is associated with improved survival in IDH-mutated AML/MDS patients undergoing allogeneic stem cell transplantation: A geth-TC multicenter study

Abstract

Abstract Introduction: Mutations in isocitrate dehydrogenase (IDH1 and IDH2) occur in approximately 15–20% and 5-15% of AML and MDS patients, respectively. However, their prognostic significance in patients undergoing allogeneic hematopoietic stem cell transplantation (allo-HCT) remains unclear. While prior studies have reported inconsistent results for IDH1 and IDH2R140 mutations, the IDH2R172 variant has been associated with improved survival and is thought to represent a biologically distinct entity. Given the lack of consensus and limited data in the transplant setting, we performed a multicenter analysis to evaluate post-transplant outcomes according to IDH mutational subtype in AML/MDS patients. Methods:We conducted a retrospective, multicenter study across nine EBMT/GETH-CT institutions, Inclusion criteria were: 1)diagnosis of AML or MDS according to WHO 2016 criteria; 2)IDH1 or IDH2 identified via Sanger sequencing or next-generation sequencing (NGS) and 3)recipients of first allo-HCT. All patients signed informed consent. The primary endpoint was 1-year overall survival (OS); secondary endpoints included progression-free survival (PFS), relapse incidence (RI), non-relapse mortality (NRM), and graft-versus-host disease (GvHD) outcomes. Multivariable Cox and Fine-Gray models were used to assess prognostic factors. Results: From 2012 to 2021, a total of 101 AML (n=88, 87%) and MDS(n=13, 13%) patients were included: 46 (46%) carried IDH1R132 and 54 (54%), IDH2 mutations (29 R140, 15 R172, 10 unspecified). Median age was 54.5 years (IQR, 44-63).Most patients (70%) had a normal karyotype, and 76% carried additional mutations (median of 2 per patient). NPM1 was the most frequent co-mutation (30%), enriched in IDH2R140 (45%) and absent in IDH2R172 (p=0.01). IDH2R172 was more often associated with epigenetic alterations such as BCOR.With a median follow-up of 21.2 months (IQR, 18-31), 1-year OS was 79% (95% CI, 71–89). In univariate analysis, patients with IDH2R172 showed superior 1-year OS (93%) compared to those with IDH1 or IDH2R140 (74%) (p=0.05). MRD positivity before allo-HCT correlated with worse OS (HR 3.0; p=0.02). Multivariable analysis (MVA) confirmed the favorable prognostic impact of the IDH2R172 mutation on OS (HR 0.09; 95% CI, 0.01- 0.8) (p=0.03). The 1-year PFS was 75% (95% CI, 67–85), IDH2R172 was associated with a higher 1-year PFS: 93% vs. 74% (p=0.06). MRD positivity before allo-HCT was associated with impaired PFS compared to MRD negative fraction (64% vs. 85%, p=0.01; HR 3.4, 95%CI 1.3-8.9, p=0.01). MVA PFS analysis demonstrated that IDH2R172 conferred a significantly lower risk of progression (HR 0.10; 95% CI, 0.01–0.90; p=0.04). The 1-year RI was 16% (95% CI, 8–24). Notably, patients with IDH2R172 had a 0% repalse incidence vs. 14% in other subtypes (p<0.001). In MVA, IDH2R172 independently predicted absence of relapse (HR 0.0; p<0.001). 1-year NRM was 9.5% (95% CI, 3–15),), and no significant predictors were identified in univariate analysis. The 100-day cumulative incidence (CI) of acute GvHD was 40% (95% CI, 30–49), significantly higher among patients receiving Tacrolimus–SirolimusGVHD prophylaxis compared with PTCy (HR 9.3; p<0.001). The 1-year CI of Chronic GvHD was 27% (95% CI, 18–37). TCD was associated with a lower cGvHD (HR 0.3; 95% CI 0.1-1, p=0.048).Lastly, 1-year GRFS was 51% (95% CI, 37–62), and was significantly improved in IDH2R172 patients (82%) compared to IDH2R140 (46%) and IDH1 (40%) (HR 0.19; p=0.02) (HR 0.19; p=0.02) and among those receiving grafts from sibling donors (HR 0.4; p=0.01). Conclusions: In this large multicenter series of IDH-mutated AML/MDS undergoing allo-HCT, the IDH2R172 mutation was associated with a improved survival, a lower RI, and a higher GRFS. Whether this reflects favorable disease biology or increased sensitivity to graft-versus-leukemia effects following allo-HCT remains unclear. Our findings may have direct clinical implications for risk stratification and transplant decision-making and warrant prospective validation.