Somatic SETBP1 mutations in myeloid malignancies

Abstract

Here we report whole-exome sequencing of individuals with various myeloid malignancies and identify recurrent somatic mutations in SETBP1, consistent with a recent report on atypical chronic myeloid leukemia (aCML). Closely positioned somatic SETBP1 mutations encoding changes in Asp868, Ser869, Gly870, Ile871 and Asp880, which match germline mutations in Schinzel-Giedion syndrome (SGS), were detected in 17% of secondary acute myeloid leukemias (sAML) and 15% of chronic myelomonocytic leukemia (CMML) cases. These results from deep sequencing demonstrate a higher mutational detection rate than reported with conventional sequencing methodology. Mutant cases were associated with advanced age and monosomy 7/deletion 7q (-7/del(7q)) constituting poor prognostic factors. Analysis of serially collected samples indicated that SETBP1 mutations were acquired during leukemic evolution. Transduction with mutant Setbp1 led to the immortalization of mouse myeloid progenitors that showed enhanced proliferative capacity compared to cells transduced with wild-type Setbp1. Somatic mutations of SETBP1 seem to cause gain of function, are associated with myeloid leukemic transformation and convey poor prognosis in myelodysplastic syndromes (MDS) and CMML.

Figure 1. Somatic SETBP1 mutations as detected by next-generation whole exome sequencing and Sanger sequencing

(a) Distribution of SETBP1 mutations detected in 52 out of 727 myeloid neoplasms (top middle), all of which were located within the SKI homologous domain. Representative mutations confirmed by Sanger sequencing are shown in the bottom and right panels. In the left panels are a somatic SETBP1 mutation (encoding a p.Asp868Asn alteration) detected by whole exome sequencing of paired tumor (bone marrow) and normal (CD3+ T-cells) DNA from a case with RAEB (whole exome #4) (left panels), where red and blue bars indicate positive and negative strands, respectively. The mutated nucleotides (c.2602G>A) are shown in green. A small amount of tumor cell contamination caused occasional mutant reads in the CD3+ sample, where the presence of multiple SNVs of similar frequencies precluded the possibility of somatic mosaicism. (b) Upper panel shows allele frequencies in paired bone marrow (blue) and CD3+ T-cells (yellow) samples in 2 RAEB and 1 sAML cases and paired RCMD/sAML and RCMD/CMML samples from the same patients (upper panel) as measured by deep sequencing. Depth of independent reads is shown in the bottom panel. In paired tumor/normal samples, small tumor contaminations were detected in CD3+ cells. Paired RCMD/sAML and RCMD/CMML samples demonstrate very small numbers of mutant reads in the initial MDS presentation (RCMD), indicating the presence of a minor SETBP1 mutated clone, which evolved later into more aggressive disease.

Figure 1. Somatic SETBP1 mutations as detected by next-generation whole exome sequencing and Sanger sequencing

(a) Distribution of SETBP1 mutations detected in 52 out of 727 myeloid neoplasms (top middle), all of which were located within the SKI homologous domain. Representative mutations confirmed by Sanger sequencing are shown in the bottom and right panels. In the left panels are a somatic SETBP1 mutation (encoding a p.Asp868Asn alteration) detected by whole exome sequencing of paired tumor (bone marrow) and normal (CD3+ T-cells) DNA from a case with RAEB (whole exome #4) (left panels), where red and blue bars indicate positive and negative strands, respectively. The mutated nucleotides (c.2602G>A) are shown in green. A small amount of tumor cell contamination caused occasional mutant reads in the CD3+ sample, where the presence of multiple SNVs of similar frequencies precluded the possibility of somatic mosaicism. (b) Upper panel shows allele frequencies in paired bone marrow (blue) and CD3+ T-cells (yellow) samples in 2 RAEB and 1 sAML cases and paired RCMD/sAML and RCMD/CMML samples from the same patients (upper panel) as measured by deep sequencing. Depth of independent reads is shown in the bottom panel. In paired tumor/normal samples, small tumor contaminations were detected in CD3+ cells. Paired RCMD/sAML and RCMD/CMML samples demonstrate very small numbers of mutant reads in the initial MDS presentation (RCMD), indicating the presence of a minor SETBP1 mutated clone, which evolved later into more aggressive disease.

Figure 2. The relationship of SETBP1 mutations with other common mutations

Clonological profiles of gene mutations in two representative cases with MDS transformed to RAEB (a) and CMML (b). Initially, hypocellular MDS (RCMD) was diagnosed based on hypocellular bone marrow with normal karyotype in both cases. (c) Coexisting mutations in the SETBP1 mutated cohort are shown in a matrix, in which 36 out of 52 cases (69%) were positive for other somatic concomitant mutations tested by Sanger sequencing. Sequenced genes are listed in Supplementary Table 8. (d) Circos plots illustrating coexisting mutations of the selected 12 genes in the whole cohort. No mutually exclusive manner was observed.

Figure 2. The relationship of SETBP1 mutations with other common mutations

Clonological profiles of gene mutations in two representative cases with MDS transformed to RAEB (a) and CMML (b). Initially, hypocellular MDS (RCMD) was diagnosed based on hypocellular bone marrow with normal karyotype in both cases. (c) Coexisting mutations in the SETBP1 mutated cohort are shown in a matrix, in which 36 out of 52 cases (69%) were positive for other somatic concomitant mutations tested by Sanger sequencing. Sequenced genes are listed in Supplementary Table 8. (d) Circos plots illustrating coexisting mutations of the selected 12 genes in the whole cohort. No mutually exclusive manner was observed.

Figure 3. Impact of SETBP1 mutations on clinical outcomes

In the whole cohort, patients with SETBP1 mutations (MT) had worse overall survival, compared with wild type (WT). SETBP1 mutations were poor prognostic factors for patients with normal karyotype. SETBP1 mutations were poor prognostic factors for patients regardless of age (>60 and ≤60 years). In the CMML cohort, patients with double mutations of SETBP1 and CBL represented worse prognosis compared to those with both WT genes. Dots indicate events and censors.

Figure 4. Immortalization of murine myeloid progenitors by SETBP1 mutations

(a) Efficiencies of empty pMYs retroviral vector (Control) and pMYs constructs expressing wild-type Setbp1 (WT) and Setbp1 mutants (p.Asp868Asn and p.Ile871Thr) in immortalizing mouse myeloid progenitor cells in three independent experiments. (b) Wright-Giemsa staining of cells immortalized by retroviral transduction of indicated mutant and WT Setbp1. Scale bar: 50µm. (c) Mean and s.d. of colony-forming potentials of murine myeloid progenitors immortalized by WT or mutant Setbp1 on methylcellulose media in the presence of SCF (100 ng/ml) and IL-3 (20 ng/ml). Combined results from three independent myeloid progenitor populations immortalized by each retroviral construct are shown. *, P<0.05. **, P<0.005. (d) Expansion of myeloid progenitors immortalized by WT or mutant Setbp1 in liquid media with SCF and IL-3 over a 96-hour period. Results from three independent populations immortalized by each retroviral construct are presented. Cell numbers were counted by trypan blue staining. Error bars represent s.d.