Recently, DNA methyltransferase 3A (DNMT3A) mutations have been identified in acute myeloid leukemia (AML), the highest frequency being found within cytogenetically normal (CN) AML. In this study, diagnostic samples from 123 adults younger than 60 years with primary CN-AML homogeneously treated in the Acute Leukemia French Association-9801 and -9802 trials were screened for mutations in DNMT3A-conserved domains by direct sequencing. Patients were also assessed for the presence of FLT3 (fms-like tyrosine kinase receptor-3), NPM1 (nucleophosmin), CEBPA, WT1 (Wilms tumor 1), IDH1 (isocitrate dehydrogenase 1) and IDH2 mutations. Thirty-eight mutations were detected in 36 patients (29%): 36 nucleotide substitutions, mostly affecting amino-acid residue R882 and two frameshift deletions. DNMT3A mutations were significantly associated with the French --American --British subtypes M4/M5 and the presence of NPM1 mutations. In the whole cohort, DNMT3A mutated patients had a shorter event-free survival (5-year EFS: 13% vs 32%, P ¼ 0.02) and overall survival (5-year OS: 23% vs 45%, P ¼ 0.02) compared with DNMT3A wild-type patients. In multivariate analysis including age, white blood cell count, NPM1/FLT3-internal tandem duplication/CEBPA risk group and DNMT3A mutational status, the presence of a DNMT3A mutation remained an independent adverse prognostic factor for EFS and OS, suggesting that testing for DNMT3A mutations could help further improve risk stratification in CN-AML.

Acute myeloid leukemia (AML) is a heterogeneous disease. Even within the same NPM1-mutated genetic subgroup, some patients harbor additional mutations in FLT3, IDH1/2, DNMT3A or TET2. Recent studies have shown the prognostic significance of minimal residual disease (MRD) in AML but it remains to be determined which molecular markers are the most suitable for MRD monitoring. Recent advances in next-generation sequencing (NGS) have provided the opportunity to use multiple molecular markers. In this study, we used NGS technology to assess MRD in 31 AML patients enrolled in the ALFA-0701 trial and harboring NPM1 mutations associated to IDH1/2 or DNMT3A mutations. NPM1 mutation-based MRD monitoring was performed by RTqPCR. IDH1/2 and DNMT3A mutations were quantified by NGS using an Ion Torrent Proton instrument with high coverage (2 million reads per sample). The monitoringof IDH1/2 mutations showed that these mutations were reliable MRD markers that allowed the prediction of relapse in the majority of patients. Moreover, IDH1/2 mutation status predicted relapse or disease evolution in 100% of cases if we included the patient who developed myelodysplastic syndrome. In contrast, DNMT3A mutations were not correlated to the disease status, as we found that a preleukemic clone with DNMT3A mutation persisted in 40% of the patients who were in complete remission, reflecting the persistence of clonal hematopoiesis.

No abstract