Validation of the ELN 2017 Classification for AML with Intermediate Risk Cytogenetics with or without NPM1 -Mutations and High or Low Ratio FLT3-ITD s

Introduction

Nucleophosmin 1-mutations (NPM1) and internal tandem duplications in the FMS-like tyrosine kinase 3 receptor gene (FLT3-ITDs) are the most frequently detected molecular abnormalities in patients with AML and are used for risk assessment and stratification of post-remission therapy. In the 2017 ELN recommendations (Döhner et al, Blood 2017) FLT3-ITD mutations are grouped according to the ratio of the mutant (mt) to wild type (wt) alleles into low ratio (LR) and high allelic ratio (HR) mutations using a cut-off of 0.5. AML with intermediate risk (IR) cytogenetic abnormalities and NPM1 -mutant FLT3- wt AML is grouped together with NPM1 -mutant and LR FLT3-ITD as favorable risk AML, only NPM1 -wt HR FLT3-ITD AML is considered adverse risk. Since validation studies have not been published so far for the 2017 ELN recommendations, we analyzed survival of patients with IR cytogenetic abnormalities who were treated in randomized controlled trials on intensive induction chemotherapy of the Study Alliance Leukemia (SAL) in the past two decades.

Methods

We performed a retrospective study on all patients who had received intensive induction chemotherapy for newly diagnosed AML within randomized SAL trials with available data on ELN 2017 classification. All molecular studies were performed centrally in one laboratory. Cox regression analyses were fitted with baseline risk factors including genetic markers for the full analysis set. Interaction tests were used to determine the impact of gene-gene interactions. Extended Cox regression models with a time-dependent covariate modeling the time of allogeneic hematopoietic stem cell transplantation (alloHCT) were used to study the role of alloHCT for patients who had reached first complete remission (CR1).

Results

In total, data on 1610 IR-patients was evaluable. The median age at first diagnosis of AML was 55 years (range, 18 to 87 years), both sexes were equally represented and 72% of patients had normal karyotype AML. Eighty-three percent of patients had de novo AML, 12% secondary AML and 3% therapy-related AML. The median follow-up was 81 months (interquartile range, 63 to 103 months). Numbers of patients in the respective genetic subgroups defined by the 2017 ELN classifications and 5-year overall survival (5-yr-OS) probabilities from diagnosis were as follows (see also Figure 1): 418 patients with FLT3-ITD wt NPM1 mt AML with 50% 5-yr-OS and 119 patients with LR FLT3-ITD with NPM1 mt AML with 41% 5-yr-OS were classed together as favorable AML, 741 patients with FLT3-ITD wt NPM1 wt AML with 27% 5-yr-OS, 170 patients with HR FLT3-ITD with NPM1 mt AML with 31% 5-yr-OS, and 70 patients with LR FLT3-ITD with NPM1 wt AML also with 31% 5-yr-OS were classed as intermediate risk AML and finally 92 patients with HR FLT3-ITD with NPM1 wt AML with 20% 5-yr-OS were classed as adverse risk AML. In any combination NPM1 mutations had a protective effect. Interaction tests indicated that the combination of an NPM1 -mutation and LR FLT3-ITD increased the risk of mortality compared to the single mutations (Wald-test for interaction, p=0.08).

When alloHCT in first complete remission was analyzed in an extended Cox regression analysis as time-dependent covariate, the impact differed among patients assigned to the three risk groups defined by the 2017 ELN recommendations: Results indicated that alloHCT had no impact on overall survival in patients with favorable AML HR 1.0 (95%-CI, 0.8 - 1.2), but a beneficial impact in patients with intermediate risk AML HR 0.8 (95%-CI 0.7 - 0.9) or high risk AML HR 0.7 (95%-CI 0.5 - 1.0).

Conclusions

The risk classification of the 2017 ELN recommendations according to NPM1 and FLT3-ITD genotypes was validated in this large retrospective analysis including data from 1610 patients with newly diagnosed AML with intermediate risk cytogenetic abnormalities or normal karyotype. The time-dependent analysis of alloHCT indicated that patients with favorable AML should not receive alloHCT as post-remission therapy in CR1, whereas patients with intermediate risk or high risk AML would benefit from alloHCT in CR1.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal