Molecular Characterization Of De Novo and Therapy-Related MDS/AML With Der(1;7)(q10;p10)

The majority of patients with myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) has acquired cytogenetic and molecular abnormalities of diagnostic and prognostic importance. Lately, focus has been on genes involved in epigenetic regulation, such as IDH, TET2, ASXL1 and DNMT3A, in which mutations have been shown to affect prognosis.

The subgroup of MDS/AML with chromosome 7 abnormalities, are associated with somatic mutations of the RUNX1 gene, and so far monosomy 7 and der(1;7)(q10;p10) have been regarded as similar cytogenetic entities because they both result in loss of 7q. However, we have recently shown that mutations of the IDH gene are significantly associated with der(1;7)(q10;p10), but are inversely correlated with other chromosome 7 abnormalities in therapy-related MDS (t-MDS) and AML (t-AML) (Westman et.al. Leukemia 2013; 27(4):957-9).

The aim of the study was to further molecularly characterize a larger cohort of patients with der(1;7)(q10;p10) and to compare the frequency of IDH and other mutations to MDS/AML cases with monosomy 7 as the sole abnormality.

Genomic DNA from 19 de novo and therapy-related MDS/AML cases with der(1;7)(q10;p10) was analyzed for mutations of FLT3, NPM1, IDH1/2, RUNX1 and DNMT3A genes by Sanger sequencing. For comparison 22 cases with monosomy 7 were investigated for mutations of the same genes. Additional investigations of possible mutations of ASXL1 and TET2 are ongoing and will be presented. Statistical evaluations were performed using Fisherxs exact test (two-tailed) or Wilcoxonxs two-sample test.

There was no difference between patients with der(1;7) and monosomy 7 in clinical characteristics such as sex, age, presentation as MDS or AML, or de novo or therapy-related disease (Table 1).

In total, 14 of 19 patients with der(1;7)(q10;p10) had mutations of IDH, RUNX1 or DNMT3A. Seven patients had a mutation in only one of these 5 genes, while the remaining 7 patients had mutations in 2 of the genes (Table 1). Nine patients had RUNX1 mutations (47%), 7 patients had IDH mutations (37%), and 3 patients had DNMT3A mutations (16%).

As for patients with monsomy 7, nine of 22 patients had mutations of IDH, RUNX1 or DNMT3A but only one of the patients had mutations in more than one gene (Table 1). Five patients had RUNX1 mutations (23%), 3 patients had DNMT3A mutations (14%), and one patient had a mutation of IDH1 (5%). No mutations were detected in NPM1 or FLT3 in any of the patients.

When comparing molecular characteristics, mutations of RUNX1, IDH, and DNMT3A were significantly more common in patients with der(1;7) compared to patients with monosomy 7 (p=0.03). IDH mutations were significantly associated with der(1;7) (p=0.02), whereas there was no difference in the distribution of RUNX1 and DNMT3A mutations between patients with der(1;7) and patients with monosomy 7 (p= 0.1 and 0.7, respectively). There was no difference in mutation frequency between patients with de novo and therapy-related MDS/AML (p=0.3).

In MDS/AML with chromosome 7 abnormalities, IDH mutations are significantly associated with der(1;7) compared to cases with monosomy 7, whereas mutations of RUNX1 and DNMT3A are equally distributed between the two cytogenetic subgroups. This difference in mutation status of IDH supports that der(1;7) and monosomy 7 should not be regarded as similar entities and suggests that der(1;7) has a specific biological effect in leukemogenesis different from that of other chromosome 7 defects.

Our findings are in line with a recent multicenter study showing different clinical outcomes for patients with der(1;7) compared to patients with -7/7q- (Ganster et.al., Prognostic impact of der(1;7) in MDS is different from del(7q), EHA 2013). More studies are needed to determine if der(1;7) and monosomy 7 show other molecular differences than IDH1/2 mutation status.

No relevant conflicts of interest to declare.