The Clinical Features and Prognostic Impact of DNMT3A Gene Mutation in Japanese Patients with De Novo AML

Several gene mutations were found in acute myeloid leukemia (AML) and were expected to be used as prognostic factors. The nucleophosmin 1 (NPM1) and CCAAT/enhancer-binding protein alpha (CEBPA) gene mutations are associated with a favorable outcome and lack of a transplant benefit, but FMS-like tyrosine kinase 3 (FLT3) mutations leading to an internal tandem duplication (ITD) are associated with adverse outcome. Recently the DNA methyltransferase 3A (DNMT3A) gene mutation was identified by whole-genome sequencing in patients with AML. DNMT3A encodes for the enzyme DNA (cytosine-5)-methyltransferase 3A and belongs to the family of other methyltransferases like DNMT1 and DNMT3B. These enzymes are involved in adding methyl groups to the cytosine residue of CpG dinucleotides and thus play an important role in epigenetic regulation of genes, but the mechanism of DNMT3A mutation -associated leukemogenesis was still unknown. About clinical impact of DNMT3A mutation, several groups reported that DNMT3A mutation have been associated with adverse outcome. We analyzed clinical features and prognostic impact of DNMT3A mutation in patients with Japanese patients with AML.

We retrospectively analyzed 188 cases of de novo AML treated at Nippon Medical School Hospital and its affiliated facilities from 2000 to 2010. We analyzed 188 samples at initial presentation and 70 samples at relapse. Mutation analyses were performed for FLT3ITD by PCR amplification, and NPM1, CEBPA, and DNMT3A mutations by direct sequence. To validate sequencing results, PCR products were inserted into the pCR2.1-TOPO vector using a TOPO TA cloning kit (Invitrogen, Carlsbad, CA). Recombinant plasmids isolated from 8 to 12 white colonies were sequenced.

The median age was 53.0 years (range, 17–87 years) with 62.1% being males. Patients were followed-up for 0.03–126.1 months after initial presentation, with a median of 19.9 months. DNMT3A mutations were detected in 35 cases (18.6%) at initial presentation, and 12 cases (17.1%) at relapse. Most frequently, codon R882 located in exon 23 was mutated, and 32 cases (91.4%) of these mutations were located within methyltrasferase domain.

We evaluated the correlation of clinical and genetic characteristics with DNMT3A mutations. Age (p=0.2884), sex (p=0.6964), and platelet count (p=0.9940) were not significantly different in AML patients with and without DNMT3A mutations. However the frequency of higher white blood cell count (p=0.0001), M5 in FAB classification (p=0.0018), and intermediate risk karyotype (p=0.0032) in patients with DNMT3A mutations were significantly higher than those in patients without them. Also, patients with DNMT3A mutations had a mutation in NPM1 (p<0.0001) and FLT3ITD (p=0.009) more frequently.

We next assessed the prognostic impact of DNMT3A mutations. For total cohort of patients with AML, complete remission rate and rates of relapse free survival (RFS) at 5 years were not statistically different between AML patients with and without DNMT3A mutations. However the overall survival (OS) at 5 years in patients with DNMT3A mutation (11.0%) was significantly lower than those in patients without them (28.9%) (p=0.0005). Among the intermediate risk karyotype or FLT3ITD negative AML patients, RFS at 5 years were not statistically different between AML patients with and without DNMT3A mutations, but OS at 5 years in patients with DNMT3A mutation (intermediate risk karyotype: 12.6%, Flt3ITD negative: 11.6%) was significantly lower than those in patients without them (intermediate risk karyotype: 23.9%, p=0.0231, FLT3ITD negative: 30.9%, p=0.0046).

This study shows that DNMT3A mutation is an important adverse prognostic factor among intermediate risk karyotype or FLT3ITD negative AML patients. Recently TET2 and Isocitrate Dehydrogenase (IDH) 1/2 gene mutations in de novo AML were reported. These genes including DNMT3A play an important role in epigenetic regulation of genes such as methylation etc, and mutations of these genes may be associated with leukemogenesis of AML. Now we are analyzing TET2 and IDH1/2 mutations among same our AML cohort, and we will show the prognostic impact of TET2, IDH1/2 and DNMT3A mutations in patients with de novo AML.

No relevant conflicts of interest to declare.