Abstract
Abstract 2
MDS and other chronic myeloid malignancies such as MDS/MPN are characterized by a frequent progression to secondary AML (sAML), a likely multistep process of acquisition of genetic abnormalities.
Genes involved in congenital genetic cancer susceptibility syndromes are often targets of somatic mutations in various tumors. For instance, germ-line mutations of SETBP1 are associated with Schinzel-Giedion syndrome (SGS), which is characterized by skeletal malformations, mental retardation and frequent neuroepithelial tumors. While SETBP1 overexpression in myeloid malignancies links to poor prognosis, somatic mutations of SETBP1 were not previously identified in leukemias. When we performed whole exome sequencing of 20 cases with myeloid malignancies, in addition to detecting previously described lesions, such as TET2, CBL and ASXL1, we identified a somatic SETBP1 mutation (D868N) in 2 cases with RAEB. Analysis of DNA from CD3+ cells from these patients confirmed its somatic nature. Sanger sequencing was applied to all coding exons in an additional 48 cases, leading to detection of 2 additional somatic mutations (G870S and I871T) in 2 patients with CMML and sAML, respectively. These findings prompted us to further expand our screening cohort: targeted SETBP1 sequencing was performed in a total of 734 patients (283 with MDS, 106 with sAML, 167 with MDS/MPN, 138 MPN and 146 with primary AML): 52 mutations were detected in 52 patients (7.1%); D868N, G870S and I871T alterations were more frequently observed (N=27, N=16 and N=5, respectively), while D868Y, S869N, D880E and D880N were less prevalent.
These mutations, of which 92% (48 out of 52) were identical to those in the SGS germ line, were detected in 15% with CMML (24/156), 15% with sAML (16/106) and 7% CML blast phase (2/28). Clinically, mutant cases were associated with higher age (p=.014), deletion of chromosome 7q (p=.0005) and shorter median survival (28 vs. 13 months, p<.0001). As shown in the analysis of 11 paired samples of progressing MDS patients, all SETBP1 mutations were acquired during leukemic evolution. In addition to mutations, SETBP1 overexpression can be found in 12% and 26% of cases of MDS and sAML, respectively, a finding linking higher activity of SETBP1 to leukemic progression.
To directly test whether SETBP1 mutations represent gain-of-function, we performed retroviral transduction of murine Setbp1 engineered with two of the somatic mutations, D868N and I871T, and evaluated the ability of the mutants to immortalize normal murine myeloid progenitors. With a low viral titer of 1 x105 cfu, both Setbp1 mutants caused efficient immortalization of myeloid progenitors, similar to overexpressed WT Setbp1. In addition, cells immortalized with mutant Setbp1 proliferated faster than cells with WT Setbp1. These data suggest that mutations of SETBP1 in our study represent gain-of-function in leukemias.
The in vitro immortalization effect of overexpressed WT Setbp1 was associated with and dependent on Hoxa9 and Hoxa10 overexpression. We performed quantitative RT-PCR and western blot experiments to evaluate expression of these genes in our mutant cases. Relative HOXA9 and HOXA10 mRNA expression values were higher in all mutant cases (N=7) than median of those in WT cases (N=4). Also, both HOXA9 and HOXA10 proteins were detected in all cases with SETBP1 mutations, suggesting that HOXA9 and HOXA10 induction is consistently associated with SETBP1 mutations similar to observations in forced expression of WT Setbp1. Moreover, in agreement with findings in primary cells showing that SETBP1 mutations or high SETBP1 expression share a common genetic association with RUNX1 mutations, Runx1 expression was reduced after in vitro immortalization of normal bone marrow cells by forced Setbp1 overexpression and two Runx1 promoter sequences were amplified after ChIP performed with antibody specific for exogenous Setbp1 protein. Moreover, Setbp1 shRNA knockdown resulted in enhanced Runx1 transcription consistent with the negative regulation of this gene by Setbp1. These results indicate that SETBP1 is associated with decreased activity of RUNX1 due to hypomorphic mutations or by direct down-modulation WT RUNX1 expression bypassing the need for mutations.
In sum, somatic recurrent SETBP1 mutations are lead to gain of function and are associated with molecular pathogenesis of myeloid leukemic transformation of various primary myeloid subentities.
Makishima:Scott Hamilton CARES Initiative: Research Funding. Maciejewski:NIH: Research Funding; Aplastic Anemia&MDS International Foundation: Research Funding.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal