The Impact of DNMT3A/FLT3-ITD/NPM1 on Patients with Acute Myeloid Leukemia after Allogeneic Hematopoietic Stem Cell Transplantation

To the Editor,

Recently, Ardestani et al. [1] published their excellent findings in this journal. They found that DNMT3A mutations alone do not affect the clinical outcomes of acute myeloid leukemia (AML) patients undergoing allogeneic hematopoietic stem cell transplantation (allo-HSCT), but when accompanied by FLT3-ITD mutations, the overall survival (OS) was significantly reduced and the relapse rate increased. NPM1 mutations had no impact on either relapse-free survival or OS, but there was a significant difference between AML patients with and without NPM1 mutations for relapse [1].

Integrative genomic analysis of de novo AML identified a subset of AML patients in which DNMT3A, FLT3, and NPM1 mutations coexisted at a higher frequency than would be expected from chance occurrence [2]. Our unpublished data also showed that a close association could be observed among DNMT3A, FLT3, and NPM1 mutations in patients with AML by factor analysis (p<0.05) based on 357 de novo AML patients analyzed by next-generation sequencing. A previous study demonstrated that younger (<60 years) patients with DNMT3A/FLT3/NPM1 mutations had significantly shorter event-free survival (p=0.047) and a tendency towards shorter OS (p=0.095) compared to those in other mutation groups [3]. The adverse impact of DNMT3A mutations is more pronounced than that of FLT3-ITD among patients with NPM1 mutations [3]. Accordingly, how did DNMT3A/FLT3-ITD/NPM1 triple mutations influence the prognoses of AML patients who underwent allo-HSCT in this study? What about the impact of DNMT3A or FLT3-ITD on NPM1-mutated AML patients? Recent studies reported that variant allele frequencies of the NPM1 and FLT3-ITD genes were closely related to long-term outcomes in patients with AML [4,5]. I wonder if there is information available on variant allele frequency in this subset of patients in order to re-analyze the impact of NPM1 and FLT3-ITD on the prognoses of patients following allo-HSCT.

We appreciate Dr. Long Su for his interest in our study and his useful comments about our article published in Turkish Journal Hematology, entitled “The impact of DNMT3A/FLT3-ITD/NPM1 on patients with acute myeloid leukemia after allogeneic hematopoietic stem cell transplantation” [1]. Our study demonstrated that DNMT3A R882 mutations are not related to inferior survival in AML patients after allogeneic HSCT. Regarding his comments, we analyzed impact of DNMT3A and FLT3-ITD on NPM1 mutated AML patients. Our finding indicated that considering NPM1 and DNMT3A mutations together, no significant difference in OS and DFS was revealed (Figure 1).

Figure 1

Survival curves of acute myeloid leukemia patients according to mutational status of DNMT3A-/NPM1: relapse free survival (left) and overall survival (right) (A=DNMT3A-/NPM1-, B=DNMT3A+/NPM1+, C=DNMT3A+/NPM1-, D=DNMT3A-/NPM1-).

FLT3-ITD mutation had a negative impact on both RFS and OS of patients with mutated NPM1. DFS and OS were significantly more favorable in patients with FLT3-ITD-/NPM1+ compared to patients with other mutation subgroups (Figure 2).

Figure 2

Survival curves of acute myeloid leukemia patients according to mutational status of FLT3-ITD/NPM1: relapse free survival (left) and overall survival (right) (A=FLT3-ITD-/NPM1-, B=FLT3-ITD+/NPM1+, C=FLT3-ITD+/NPM1-, D=FLT3-ITD-/NPM1+).

Also, patients with DNMT3A/FLT3-ITD/NPM1 triple mutations demonstrated a significantly worse DFS (p=0.009) and OS (p=0.028) (Figure 3) compared to all other patients (Figure 3).

Figure 3

Survival curves of acute myeloid leukemia patients according to DNMT3A/FLT3-ITD/NPM1 triple mutations: relapse free survival (left) and overall survival (right) (A=DNMT3A/FLT3-ITD/NPM1, B=others).

Unfortunately, in this study, information about the allele frequency of mutations was not available for most of patients, but in an ongoing prospective study, we are studying the effect of the allelic frequency of FLT3-ITD and NPM mutations on the clinical outcome of AML patients after HSCT.

Although there is more agreement on the poor prognostic role of DNMT3A mutations in AML patients [2], few studies have been conducted on the effect of these mutations on the clinical outcome of patients after allogeneic transplantation and there are disagreements over the results of these limited studies [3,4].

These differences can be due to coexistence of variable cytogenetic or molecular genetic aberrations in leukemia cells that could be effective in the clinical outcome of the patients. In the present study, only three genetic aberration were tested and additional genetic abnormality might impact the outcome of the patients.

Taken together, our findings indicated that FLT3-ITD may be a more powerful adverse prognostic biomarker than NPM1 and DNMT3A in AML patients after HSCT. Although the comments have allowed us to add useful information to our previous paper, our conclusion remains the same as in our published work.

Best Regards,