Isocitrate Dehydrogenase (IDH) 1 and 2 Mutation Is an Independent Predictor of Better Outcome in Patients with Acute Myeloid Leukemia Undergoing Allogeneic Hematopoietic Stem Cell Transplantation: A Study of the ALWP of EBMT

Isocitrate dehydrogenase (IDH) 1 and 2 mutations occur in 20% of acute myeloid leukemia (AML). Patients with AML carrying IDH1-2 mutations have a similar prognosis compared to patients without these mutations (DiNardo et al, AM J H, 2015). However, the impact of IDH1-2 mutations on patients with AML undergoing allogeneic hematopoietic stem cell transplantation (allo-HCT) is not well known. In this study, we investigate the prognostic impact of IDH1-2 mutational status on AML patients undergoing alloHCT in complete remission (CR).

In this retrospective registry-based analysis, we identified 710 consecutive adult patients (46.2% female; median age: 58.5 years [range, 18-78]) with AML undergoing allo-HCT in CR between 2015 and 2019 at 85 EBMT participating centers. Cord blood, ex-vivo graft manipulated transplants, and patients with favorable cytogenetics were excluded. Median follow-up was 15 months [95% CI 13.4-16.6]. Patients were categorized based on IDH1-2 mutational status, with 300 (42%) mutated and 410 (58%) wild type. Six hundred and fifty-two (92%) and 58 (8%) patients had de novo and secondary AML, respectively, and 141 (20%) patients had poor-risk cytogenetics. IDH1-2 mutation was positively correlated with NPM1 mutation (40% in IDH1-2 mutated vs 27% in wild type, p=0.0001) and more frequently encountered in middle-aged patients (p=0.01). No correlation was noted between IDH1-2 and FLT3 mutation or other patient characteristics. Minimal residual disease (MRD) data was available for 344 patients, 53% of which were MRD-negative at transplant in both groups. Six hundred and twenty-three (88%) and 87 (12%) patients were in first and second CR at time of transplant, respectively. Patients received grafts from a matched sibling (24%), unrelated (62%), or haploidentical (14%) donor, and myeloablative conditioning (MAC) was used in 42%. Ninety-three percent of the patients received peripheral blood as the stem cell source.

At day 180, the cumulative incidence of grade II-IV acute graft-versus-host disease (aGVHD) was significantly lower in IDH1-2 mutated compared to wild-type patients (22% vs 33%, p=0.002). No differences in chronic GVHD rates were noted between the 2 groups (39% vs 40%, p=0.87). The 2-year cumulative relapse incidence (RI) was significantly lower and the GVHD-free, relapse-free survival (GRFS) was also improved in IDH1-2 mutated compared to wild-type patients (14.4% vs 27%, p=0.001 and 47% vs 39%, p=0.006, respectively). This led to an improved leukemia-free survival (LFS) in IDH1-2 patients (69% vs 59%, p=0.01), however, it did not translate into an overall survival (OS) difference. No significant difference was noted in non-relapse mortality (NRM) between the 2 groups (17% vs 14.2%, p=0.26).

These findings were confirmed in multivariate analysis. In fact, IDH1-2 mutation was associated with significant improvement in RI (hazard ratio [HR]=0.4 [95%CI 0.25-0.64], p=0.0001), LFS (HR=0.7 [95%CI 0.51-0.95], p=0.022), aGVHD II-IV (HR=0.63 [95%CI 0.45-0.87], p=0.005) and GRFS (HR=0.69 [95%CI 0.54-0.89], p=0.004). Conversely, the presence of adverse cytogenetics and undergoing allo-SCT in second CR increased the RI (HR=2.29 [95%CI 1.46-3.61], p=0.0003 and HR=2.84 [95%CI 1.64-4.91], p=0.0002, respectively) and were associated with a shorter LFS (HR=1.67 [95%CI 1.18-2.36], p=0.004 and HR=1.61 [95%CI 1.06-2.44], p=0.025) while reduced intensity (RIC) conditioning had a worse impact on OS compared to MAC (HR=1.56 [95%CI 1.07-2.29], p=0.022). Additionally, in the subgroup of patients with available MRD data, MRD positivity at transplant significantly increased RI (HR=2.15 [95%CI 1.09-4.23], p=0.027) with no impact on survival.

In conclusion, our data suggest that the presence of IDH1-2 mutations acts as an independent prognostic factor and is associated with improved outcome in patients with AML in CR undergoing allo-HCT. Indeed, patients with IDH1-2 mutations had significantly lower rates of RI and aGVHD, which translated into improved LFS and GRFS. Nevertheless, patients with MRD positivity at time of transplant had significantly increased RI. Further studies investigating allo-HCT outcomes in IDH1-2 mutated patients with AML in the era of IDH inhibitors (both in the pre- and post-transplant setting) would help to further define the impact of these mutations in this setting and thus optimize an individualized treatment approach.