We have confirmed the high incidence of DNMT3A mutations in patients with AML with IR cytogenetics. Patients with DNMT3A mutations relapse more often and have inferior OS when only patients achieving CR are analyzed. 'Double-mutated' patients have a very poor prognosis.

Mutational analysis of C-KIT, fms-like tyrosine kinase 3 (FLT3), and JAK2 genes was performed in 60 patients with core binding factor acute myeloid leukemia (CBF-AML). Patients reaching molecular remission had lower incidence of relapse and better overall survival (OS) than those not achieving molecular remission (p = 0.008 and 0.044, respectively). The overall incidence of C-KIT mutations was 33.3%, FLT3/internal tandem duplication (ITD) 6.6%, FLT3(D835) 10.0% and JAK2(V617F) mutations 3.3%. C-KIT mutations did not predict for clinical/molecular relapse (p = 0.33). OS of patients with C-KIT mutations was identical to patients without them when all patients with CBF-AML were analyzed together (p = 0.58). When AML1/ETO-positive patients were evaluated separately, OS in C-KIT-mutated patients was slightly inferior to unmutated ones (p = 0.14). Patients with CBF-AML with a mutated C-KIT gene were also more prone to extramedullary disease (p = 0.08). Of six patients harboring various FLT3(D835) mutations, four (66.7%) relapsed, whereas among 43 cases without these mutations, 16 relapses (37%) were observed (p = 0.08). Our results on minimal residual disease, C-KIT, and FLT3/ITDs are in line with previous studies. Surprisingly, a possible role for FLT3(D835) mutations was noted in addition. These results need validation in even larger patient cohorts than ours. For routine clinical practice, it may be meaningful to screen for C-KIT mutations in AML1/ETO-positive patients, as well as for FLT3(D835) mutations in CBF-AML.