Chronic myeloid malignancies are categorized to the three main categories myeloproliferative neoplasms (MPNs), myelodysplastic syndromes (MDSs) and MDS/MPN overlap. So far, no specific genetic alteration profiles have been identified in the MDS/MPN overlap category. Recent studies identified mutations in SET-binding protein 1 (SETBP1) as novel marker in myeloid malignancies, especially in atypical chronic myeloid leukemia (aCML) and related diseases. We analyzed SETBP1 in 1 130 patients with MPN and MDS/MPN overlap and found mutation frequencies of 3.8% and 9.4%, respectively. In particular, there was a high frequency of SETBP1 mutation in aCML (19/60; 31.7%) and MDS/MPN unclassifiable (MDS/MPN, U; 20/240; 9.3%). SETBP1 mutated (SETBP1mut) patients showed significantly higher white blood cell counts and lower platelet counts and hemoglobin levels than SETBP1 wild-type patients. Cytomorphologic evaluation revealed a more dysplastic phenotype in SETBP1mut cases as compared with wild-type cases. We confirm a significant association of SETBP1mut with −7 and isochromosome i(17)(q10). Moreover, SETBP1mut were strongly associated with ASXL1 and CBL mutations (P<0.001 for both) and were mutually exclusive of JAK2 and TET2 mutations. In conclusion, SETBP1mut add an important new diagnostic marker for MDS/MPN and in particular for aCML.
According to the new World Health Organization (WHO) classification 2008, chronic myeloid malignancies are separated into three main categories: myeloproliferative neoplasms (MPNs), myelodysplastic syndromes (MDSs) and MDSs/MPNs.1 MDS/MPN cases show overlapping characteristics of both MDS and MPN. This category includes chronic myelomonocytic leukemia (CMML), juvenile myelomonocytic leukemia, atypical BCR–ABL1-negative chronic myeloid leukemia (aCML) and unclassifiable MDS/MPN (MDS/MPN, U). In aCML many clinical features suggest the diagnosis of CML, however, the pathognomonic Philadelphia chromosome and the corresponding BCR–ABL1 fusion transcript are missing.2, 3 In contrast to Philadelphia chromosome/BCR–ABL1-positive CML, the BCR–ABL1-negative MPNs display a very heterogeneous group from genetic aspects. Cytogenetic aberrations are found in around 15% of MPN patients. Most aberrations are unbalanced such as +8, chromosome 7 aberrations, del(20q), i(17q) or -Y. The MDS/MPN overlap category differs from MPN by a higher frequency of cytogenetic aberrations (overall 25–30%) and a higher incidence of chromosome 7 aberrations and isochromosome i(17)(q10).4
Although some molecular mutations are highly associated with MPN cases, there are no specific ones found in MDS/MPN overlap category. The most important role in diagnostics of BCR–ABL1-negative MPNs has the V617F-mutation within exon 14 of the JAK2 kinase gene, occurring in >70% of all MPN cases,5, 6, 7 in contrast to <30% in MDS/MPN patients.4 In 10% of V617F-negative polycythemia vera (PV), JAK2 mutations in exon 12 occur.8 The MPL gene, coding for a thrombopoetin receptor, is mutated in 5–10% of JAK2V617F-negative essential thrombocythemia (ET) or primary myelofibrosis (PMF).9, 10 Mutations in CBL, a gene coding for a E3-ubiquitine-ligase and therefore involved in tyrosine kinase signaling, have been identified with frequencies of 2% and 20% in patients with the MDS/MPN overlap entities aCML and CMML.11 Another gene that is frequently mutated in CMML (40%),12, 13 MDS (10%)12 and MPN (11%)14 is ASXL1, a polycomb group protein involved in epigenetic regulation mechanism. Similarly, TET2 is the most frequently mutated gene in CMML patients (61%).13
Recent studies identified SET-binding protein 1 (SETBP1) as novel marker in myeloid malignancies.15 SETBP1 localizes on chromosome 18q21.1 and codes for a nuclear localizing protein with mainly unknown function. SETBP1 is a large protein of 1597 amino acids (aa) with an estimated molecular weight of 170 kDa. It was initially found as interaction partner of SET, a small inhibitor of the tumor suppressors protein phosphatase 2 and NM23-HI. Germ-line mutations in SETBP1 are associated with the Schinzel–Giedion syndrome, which is characterized by skeletal malformations, mental retardation and developmental delay.16 However, SETBP1 mutations may also occur as somatic alterations with a role for leukemogenesis. Overexpression of SETBP1 was shown in an acute myeloid leukemia (AML) patient with a t(12;18)(p13;q12) with poor outcome.17 In a t(12;18)(p13;q12)-positive patient, a potential mechanism regulating the progression from PMF to secondary AML with overexpression of SETBP1 was discussed.18 A fusion of SETBP1 to NUP98 corresponding to a t(11;18)(p15;q12) was reported in a pediatric patient with T-lineage acute lymphoblastic leukemia.19 Most recently, mutations in SETBP1 were found to occur in aCML and related diseases, identifying SETBP1 mutation as a possible oncogenic event in patients classified to the MDS/MPN overlap category.15
To further characterize the role of SETBP1 mutations in chronic myeloid neoplasms, we analyzed a large cohort of 1130 MPN and MDS/MPN cases based on the WHO criteria and correlated the findings to clinical data, cytomorphologic subtypes, cytogenetics and the concomitant occurrence of other molecular mutations.
Patients and methods
In total, 1130 cases with BCR–ABL1-negative MPN (n=579) and MPN/MDS overlap (n=551) were analyzed. Thirty-six of these MDS/MPN overlap cases had been included in a previous study.15 The cohort comprised 690 men and 440 women with a median age of 71.1 years (range 20.9–93.3 years; for further details see Table 1). The median follow-up was 12 months in CMML and 17.1 months in aCML patients. Classification of all cases was validated on peripheral blood and/or bone marrow smears according to the WHO.1 Patients were selected for the study as we focused mostly on patients from the MDS/MPN overlap category or on patients with unclassifiable MPN (MPN, U) based on the results of the study from Piazza et al.15 describing higher frequencies of SETBP1 mutations in these categories. Thus, patients with clearly defined MPN such as PV, essential ET or PMF were only of minor interest. We finally subdivided all patients into the specific disease entities as follows: within the MPN category: MPN, U: n=510, PV: n=13, essential ET: n=30, PMF: n=9; within the MDS/MPN category: MDS/MPN, U: n=214, aCML: n=60, CMML: n=294 (CMML-1: n=217, CMML-2: n=77). As we noticed a correlation of SETBP1 mutations with −7 and i(17)(q10) we analyzed in a second step 265 cases of different acute and chronic myeloid neoplasms. Within the cohort of −7 cases, 90 were AML, 6 MDS/MPN overlap, 42 MDS, 4 MPN and 4 CML. Cases with i(17q) were classified as AML (n=40), MDS/MPN overlap (n=15), MDS (n=23), MPN (n=14) and CML (n=27).
In all MPN cases, the t(9;22)/BCR–ABL1 gene fusion had been excluded by reverse transcriptase-PCR and/or interphase fluorescence in situ hybridization. Patients’ samples were sent from different hematologic centers for diagnosis to the MLL Munich Leukemia Laboratory between 2005 and 2012. All patients gave their consent for genetic analyses and the use of laboratory results for research purposes. The study adhered to the tenets of the Declaration of Helsinki and was approved by the laboratorýs institutional review board.
Cytomorphology and cytogenetics
In all cases, bone marrow and peripheral blood smears underwent May Giemsa Gruenwald staining. For cytomorphology, 100 nucleated cells were counted in the peripheral blood, 200 in the bone marrow. Cytochemistry was performed for myeloperoxidase and nonspecific esterase, and iron staining was done for detection of ring sideroblasts in cases with increased erythropoiesis or anemia. Classification of the disease entities and dysplasia was rated according to WHO criteria.1
Chromosome banding analysis was performed in 1116/1130 cases after short-term culture. Karyotypes were analyzed after G-banding and described according to the International System for Human Cytogenetic Nomenclature.20 When needed, cases were investigated additionally by fluorescence in situ hybridization.21
Isolation of mononuclear cells, DNA and mRNA extraction, as well as random primed complementary DNA synthesis was performed as described previously.22 A 451-bp PCR fragment (covering codons for aa 800 to 935), containing the mutational hotspot region of SETBP1,15 was amplified with the Qiagen Taq PCR Master Mix (Qiagen, Hilden, Germany) from either genomic DNA or complementary DNA templates, using the following primers: SETBP1-for: 5′-IndexTermGCCAGCACTGAAACCAATTT-3′, SETBP1-rev: 5′-IndexTermTCCGTTTCCTCTTGTGCTTT-3′. The single amplicon was analyzed by Sanger sequencing in all cases using BigDye Term v1.1 cycle sequencing chemistry (Applied Biosystems, Weiterstadt, Germany). Estimation of the mutational load was based on the electropherograms of the forward and reverse reactions.
In order to obtain any prediction for a damaging character of these specific missense mutations, we used the PolyPhen-2 (http://genetics.bwh.harvard.edu/pph2),23 SIFT (http://sift.jcvi.org)24 and MutationTaster (www.mutationtaster.org)25 online analysis tools.
Additional mutational data obtained by Sanger sequencing or melting curve analyses were available in subcohorts: in the majority of patients analysis for the JAK2V617F (n=1127/1130) was performed by melting curve analysis following previous descriptions,26 analysis for CBL mutations was done in 1108/1130 cases by sequencing exons 8 and 911, 27 and sequencing of ASXL1 exon 12 was performed in 398 cases, subsequently analyzed in aCML, CMML and SETBP1-positive cases.12 The c.1934dupG mutation in ASXL1 was rated as somatic mutation.28 Focusing on JAK2V617 wild-type patients, analyses for JAK2exon12 mutations (n=746) and MPLW515 (n=765) mutations were performed by melting curve analyses.29, 30 TET2 mutation was analyzed by next generation sequencing in 164 CMML cases.11, 13 Mutation of TP53 was also addressed by next generation sequencing selectively in 52 cases with i(17)(q10).31
Dichotomous variables were compared between different groups using the χ2-test and continuous variables by Student’s T-test. Results were considered significant at P<0.05. Adjustment for multiple testing was not done. Statistical analyses were performed using SPSS version 19.0 (IBM Corporation, Armonk, NY, USA); the reported P-values are two-sided. Survival curves were calculated for overall survival (OS) according to Kaplan–Meier and compared using the two-sided log rank test. OS was the time from diagnosis to death or last follow-up.
Characterization of SETBP1 mutations
SETBP1 mutations were detected in 74 of 1130 patients from the total cohort (6.6%). All SETBP1 mutations were missense mutations and occurred with a mutational load of 10–50%, indicating a heterozygous status, only one case showed a homozygous mutation. The mutations clustered in the SKI homologous region of SETBP1 within a stretch of 16 aa (aa 858 to 874). The most prevalent mutation was p.Asp868Asn (n=28), followed by p.Gly870Ser (n=24), p.Ileu871Thr (n=9) and p.Gly870Asp (n=4). All other mutations occurred in single cases only. In two cases, two different mutations were observed, both cases having a p.Asp868Asn mutation, in one case combined with a p.Gly870Ser mutation, in the second case combined with a p.Asp908Asn mutation. The mutation localization, aa exchanges and frequencies are given in detail in Figure 1. Analyzing the potential deleterious character of these missense mutations by SIFT, PolyPhen 2 and MutationTaster for all mutations revealed overall a damaging character.
Frequency of SETBP1 mutations
The frequency of SETBP1 mutations was significantly higher in the MDS/MPN overlap category with 9.4% (52 of 551 patients) as compared with the MPN category with 3.8% (22 of 579 patients; P<0.001). Analyzing MDS/MPN overlap and MPN entities separately by subdividing the cohorts into CMML and MDS/MPN, U cases, as well as into PV, ET, PMF or MPN, U, respectively, revealed most mutated cases within the unclassifiable/unspecific entities with 12.3% for MDS/MPN, U (31/253) and 4.2% for MPN, U (22/526). In contrast, no SETBP1 mutation was found in any of the MPN categories PV (0/13), ET (0/30) and PMF (0/9).
Analyzing CMML cases separately revealed a mutation frequency of 7.1% (21 of 294; 16 of 217 CMML-1 and 5 of 77 CMML-2 cases). Finally, we categorized the MDS/MPN, U and MPN, U cases in conformable and non-conformable with aCML according to WHO criteria, defined by peripheral leukocytosis ⩾13.0 × 109/l, neutrophil precursors ⩾10% of leukocytes, <20% of blasts, basophils <2%, monocytes <10%, increased and dysplastic granulopoiesis in the bone marrow, and exclusion of BCR–ABL1 and of PDGFRA or PDGFRB rearrangements.2 Sixty cases were classified as being aCML. Mutational analysis of SETBP1 resulted in 19 positive cases (19/60; 31.7%). Thus, SETBP1 mutations significantly associate with aCML, both within the total cohort and within MDS/MPN overlap entities (19/40 vs 50/1045 and 36/483; P<0.001), identifying SETBP1 as a more specific marker for aCML.
Clinical characterization and outcome of SETBP1 mutated cases
In the total cohort, SETBP1 mutations correlated with male gender (male/female ratio 2.7 vs 1.5 in SETBP wt; P=0.035), higher white blood cell (WBC) count (44.6 × 109/l vs 16.0 × 109/l in SETBP1wt; P<0.001), lower platelet counts (130 × 109/l vs 300 × 109/l in SETBP1wt; P<0.001) and lower hemoglobin levels (11.2 vs 11.6 g/dl in SETBP1wt; P<0.001). No overall correlation was observed between SETBP1 mutation and age. However analyzed separately, in MPN patients mutations in SETBP1 correlated with higher age (median 75 years vs 68 years; P=0.025). In MDS/MPN overlap, patients these clinical correlations were restricted to leukocytosis (median: 50.1 × 109/l in SETBP1 mut patients vs 17.9 × 109/l in wild-types; P=0.002; Table 1). Moreover, in CMML cases there were no significant differences in SETBP1mut between CMML-1 and CMML-2 or myeloproliferative (WBC counts ⩾13.0 × 109/l) and dysplastic CMML subtype (WBC counts <13.0 × 109/l).
Calculating the OS for prognostic relevance of SETBP1 in both the aCML (Figure 2a) and CMML (Figure 2b) patients revealed no statistically significant difference between SETBP1mut and SETBP1wt patients, although the outcome was slightly better in SETBP1mut patients. Follow-up data were available in 47 aCML and 180 CMML cases, the median follow-up was 17.1 and 12 months, respectively. The OS of SETBP1mut in aCML was 32.9 months vs 15.6 months in SETBP1wt. In CMML patients, the median OS was not reached for SETBP1mut and 29.6 months for SETBP1wt cases.
Cytomorphological characterization of SETBP1 mutated cases
Bone marrow cytomorphology was available in a total of 36 patients with SETBP1 mutations (23 MDS/MPN; 13 MPN). Overall cellularity was increased in the majority of cases (n=29/36; 80.6%). Considering distinct hematopoietic lineages, granulopoiesis (n=31/36; 86.1%) and megakaryopoiesis (n=21/36; 58.3%) were increased in the majority, whereas erythropoiesis was reduced in most cases (n=28/36; 77.8%). At least 10% of dysplastic cells per hematopoietic lineage were found in megakaryopoiesis in 23 of 25 cases (92.0%), in granulopoiesis in 32 of 35 evaluable cases (91.4%) and in erythropoiesis in 11 of 31 evaluable cases (35.5%). Dysplasia was more pronounced in megakaryopoiesis with a median of 50% of dysplastic cells (range 0–100%) and in granulopoiesis with a median of 40% of cells (range 0–80%), whereas the median proportion of dysplastic cells was only 5% in erythropoiesis (range 0–50%; Table 2).
Coincidence of SETBP1 with cytogenetic alterations
In the total cohort, aberrant karyotypes were found in 227 of 1116 (20.3%) cases. We performed correlation of SETBP1 mutations with the following cytogenetic subgroups: normal karyotype, -Y, +8, −7, del(20q) and i(17)(q10) as sole alterations (Table 1 and Figures 3a and b). In the total cohort, SETBP1mut correlated significantly with −7 or i(17)(q10) as sole aberrations. Half of the patients with −7 were SETBP1 mutated (7/14, 50%), whereas only 6.0% were mutated in cases with any other karyotype (66/1102; P<0.001). This was also true in cases with i(17)(q10), where 5 of 10 cases (50%) were SETBP1 mutated in contrast to 6.2% (68/1106) in cases with any other karyotype (P<0.001). In contrast, SETBP1mut never occurred in patients with +8 (0 of 49 cases), whereas 6.8% (73/1067) were SETBP1 mutated in all other karyotypes (Table 1 and Figures 3a and b). These correlations were also reflected in the MDS/MPN category, and at a lower significance level in the MPN category (Table 1).
Analysis of SETBP1 in −7 and i(17)(q10) in other entities
This high coincidence of SETBP1 mutations with the cytogenetic groups i(17)(q10) and −7 prompted us to enlarge the total cohort by including different myeloid neoplasms with these two specific aberrations. Therefore, we analyzed SETBP1 mutations in another completely independent cohort (n=265) of 119 patients with a myeloid malignancy with i(17)(q10) and 146 patients with monosomy 7. We analyzed cases with i(17)(q10) or −7 as sole aberrations as well as in combination with other cytogenetic abnormalities. Mutations in SETBP1 were observed in de novo MDS, therapy-related-MDS, MDS/MPN, U, CMML, MPN, de novo AML, secondary-AML, therapy-related-AML, but never in CML (see Table 3). Further categorizing the cohorts in cases with sole cytogenetic aberration, with one additional aberration and with more than one additional aberration resulted in a significant difference in the i(17)(q10) cohort. In 54.3% (19/35) of patients with a sole i(17)(q10) a SETBP1 mutation was detectable, in 38.2% (13/34) within the i(17)(q10) with one additional aberration and in 10% (5/50) with more than one additional aberration (P<0.001). Similar differences were seen in the −7 cohort, but did not reach significance (P=0.08).
Analyzing TP53 mutation in cases with a i(17)(q10) resulted in mutual exclusiveness of SETBP1 and TP53 mutations (24/43 SETBP1mut in TP53wt vs 0/9 SETBP1mut in TP53mut; P=0.002).
Coincidence of SETBP1 with other mutations
The total cohort consisted of 779 JAK2wt (779/1127; 69.1%) and 348 JAK2V617Fmut (30.9%) cases. Only 3 of 746 analyzed patients showed a JAK2exon12 mutation (0.4%) and 38 of 765 (5.0%) an MPLW515 mutation. Mutations in CBL occurred in 91 of 1108 analyzed cases (8.2%) and in ASXL1 in 208 of 398 analyzed cases (52.3%).
In the total cohort, SETBP1mut correlated significantly with the occurrence of CBLmut. A total of 16 of 91 CBLmut cases showed an additional SETBP1 mutation, whereas only 57 of 1014 CBLwt cases were SETBP1mut positive (17.6% vs 5.6%; P<0.001). Furthermore, SETBP1mut occurred significantly more frequently together with mutations in ASXL1 than in ASXL1wt cases. Nearly one-third of ASXL1mut cases also carried a mutation in SETBP1 (63/208; 30.3% vs 11/190; 5.8% in ASXL1wt; P<0.001). In contrast, SETBP1mut were nearly mutually exclusive of JAK2V617F. Only 3 of 348 cases with JAK2V617F carried a SETBP1 mutation as compared with 70 of 779 JAK2wt cases (0.9% vs 9.0%; P<0.001). No case had both an MPLW515 or JAK2exon12 mutation and a SETBP1 mutation (Figure 3c). All these correlations were also observed in the MPN and MDS/MPN categories when both entities were separately analyzed (data are given in Table 1). We furthermore analyzed these correlations in aCML cases (Figure 3d). Surprisingly, 39/60 (65%) patients carried an ASXL1 mutation and 19/39 an additional SETBP1 mutation, whereas only 4/21 were SETBP1mut within the ASXL1wt group (P=NS). No other correlation reached significance. However, in CMML cases SETBP1mut co-occurred more frequently with TET2wt than TET2mut (8/64 in TET2wt vs 3/100 in TET2mut; P=0.025). Analyzing the co-occurrence of the three genes ASXL1, SETBP1 and CBL in aCML patients revealed that 16 cases showed no mutation in any of these genes, 22 carried an ASXL1 mutation, whereas only 1 and 4 cases showed a CBL or SETBP1 mutation. In 11 patients, ASXL1mut was combined with SETBP1mut, in 2 cases with CBLmut. Only four patients carried a mutation in all three genes. Taking additionally the mutational loads into account supports also ASXL1mut as the earliest event, having higher mutational loads than the other two gene mutations. This may indicate that SETBP1mut is rather a later event in disease progression than an initial mutation.
According to the WHO classification,1 MPNs are first separated into t(9;22)/BCR–ABL1-positive CML and BCR–ABL1-negative MPNs. As a further step, based on laboratory parameters and cyto- and histomorphology focusing on cellularity of distinct hematopoietic lineages or bone marrow fibrosis, some more entities can be clearly discriminated such as PV, essential ET and PMF. Presence of a JAK2V617F or JAK2exon12 mutation may confirm a suspected PV even in case of isolated polyglobulia,32 whereas JAK2V617F-negative ET or PMF may be observed with MPLW515 mutations.9, 10 Despite these approaches,1 a large proportion of cases with so-called MPN, U or MDS/MPN, U remains with highly heterogeneous morphologic phenotypes and clinical profiles. A more detailed subclassification of these cases could not be realized by the presently established diagnostic armamentarium including cyto- and histomorphology, cytogenetics and the Janus kinase (JAK)-signal transducers and activators of transcription (STAT)-activating mutations. In very recent years, some novel molecular markers were detected to be also relevant in the MPNs such as ASXL1, IDH1, IDH2 and TET2 mutations,33 but all these showed no specificity for a distinct entity.
Recently, Piazza et al.15 described SETBP1 mutations to occur at a high frequency of 24.3% (n=17/70) in aCML patients. In this study, we investigated 1130 cases with different chronic myeloid neoplasms and were able to further confirm the previous observation. We detected SETBP1 mutations in 32% (n=19/60) of patients fulfilling the criteria of aCML.2 Furthermore, our study underlines that SETBP1 mutations show a preference for the MDS/MPN category with a frequency of 9.4% as compared with only 3.8% in the MPN category (P<0.001). Within the MDS/MPN category, SETBP1 mutations were detected more frequently in MDS/MPN, U than in CMML (12.3% vs 4.2%). These results are corresponding to the data from Piazza et al.15 publishing incidences of 13% and 4% for MDS/MPN, U and CMML.
MPNs are outlined by myeloproliferation, whereas the MDS/MPN overlap category shows a combination of myeloproliferation and dysplastic hematopoiesis. Both categories (MPNs and MDS/MPN) show overlapping cytogenetic and molecular features, albeit the frequency of the cytogenetic aberrations and of some distinct abnormalities, for example, isochromosome i(17)(q10) and +21, is higher in MDS/MPN, whereas the JAK2V617F is more frequent in the MPNs.4 In PV, a high proliferation level goes in line with an extraordinary high frequency of mutations within the JAK2 gene (that is, the V617F and exon12 mutations).32 In refractory anemia with ring sideroblasts and thrombocytosis, the JAK2V617 and SF3B1 mutations show frequent co-occurrence34 suggesting that the JAK2V617 represents ‘thrombocytosis’ whereas SF3B1 mutations are linked to the occurrence of ring sideroblasts. Accordingly, in the overlap MDS/MPN category, JAK2 mutations may be linked to myeloproliferation, whereas SETBP1 mutations are associated to dysplasia.
By evaluating bone marrow cytomorphology in SETBP1 mutated cases (including MPN, MDS/MPN, U, or aCML) we found a characteristic phenotype with an increased and dysplastic granulopoiesis and megakaryopoiesis. These results suggest that SETBP1 mutations have a causative role for the phenomenon of dysplasia in granulopoiesis and megakaryopoiesis and are strongly linked to the MDS/MPN category and in particular to aCML.
With regards to CMML, SETBP1 mutations may define a new molecular subgroup, as they were identified in our study in 7.1% of cases. Damm et al.35 reported a similar frequency in patients with CMML of 6.2% and described an adverse prognostic impact of SETBP1 mutations in CMML patients with regards to overall and AML-free survival.
Interestingly, we found a close association of SETBP1 mutations to monosomy 7 and isochromosome 17q (P<0.001 for both), both prognostically unfavorable cytogenetic aberrations. Correlation between SETBP1 overexpression and monosomy 7 has previously been observed in AML patients.17 By our subsequent investigation of an independent cohort of 265 cases with different acute and CMLs selected for these specific cytogenetic alterations, we were able to show that particularly in patients with a sole i(17)(q10) SETBP1 mutations were overrepresented with 54.3% whereas the frequency was lower in cases with additional cytogenetic aberrations. Cases with i(17)(q10) and SETBP1 showed mutual exclusiveness of TP53 mutations (P=0.002). Occurrence of i(17)(q10) as sole cytogenetic abnormality was only rarely reported in myeloid neoplasms defining a distinct clinicopathologic entity with MDS/MPN features and a high risk for leukemic progression. Sequencing of the uninvolved TP53 allele revealed no TP53 mutation in i(17)(q10) cases.36 This suggests a potential influence of any other oncogene or tumor-suppressor gene on pathogenesis of myeloid neoplasms associated with sole i(17)(q10).
In addition, we observed a strong correlation of SETBP1 mutations to mutations in CBL and ASXL1 (P<0.001 for each). The CBL gene, coding for an E3-ubiquitin-ligase, is known to be frequently mutated in patients with CMML and aCML.11, 27, 37, 38 The protein stabilization of SETBP1 is thought to be altered by the mutational abrogation of the ubiquitination site that marks proteins to be degraded.15 CBL mutations may function in a synergistic manner, as most of them are missense substitutions abrogating CBL ubiquitin ligase activity.39 Mutations in ASXL1 have been shown in CMML, MDS and MPN cases,12, 13, 14 but so far not in aCML patients. The study of Piazza et al.15 showed an association of ASXL1 mutated with SETBP1 mutated cases in aCML patients, what was confirmed in this study. In CMML, most gene mutations showed no mutual exclusiveness and were indicating an inferior prognosis compared with patients with a single gene mutation,40 what was also true for ASXL1 mutated cases.41 Damm et al.35 described a frequent co-occurrence of SETBP1mut with ASXL1mut in CMML cases, and furthermore a poor prognosis of SETBP1 mutated patients. Possibly, also ASXL1mut and SETBP1mut function in a synergistic manner. It has been shown that truncated forms of ASXL1 have decreased potential of recruiting PRC2 and therefore result in diminished H3K27me and activation of genes,42 what is also suggested for HOX genes,43 which are also regulated by overexpression of SETBP1.44 In contrast, SETBP1 mutations were nearly exclusive of the JAK2V617F mutation, and were not found together with JAK2exon12 and MPLW515 mutations. Furthermore in CMML patients, SETBP1mut occurred more frequently with TET2wt than TET2mut, what was also described by Damm et al.35 Therefore, distinct cooperation patterns exist, as SETBP1 mutations cooperate with i(17q), −7, CBL and ASXL1 mutations, but do not interact with JAK2-STAT-activating mutations and TET2 mutations in our cohort. This fits well to the observation of Piazza et al.15 that SETBP1 mutations do not occur in ‘classical’ PV, PMF or ET.
The effect of SETBP1 mutations on leukemogenesis has been explored by different study groups: overexpression of SETBP1 was suggested to protect SET from protease cleavage. According to the results of Cristobal et al.17, the increasing amount of full length SET protein resulted in inhibition of protein phosphatase 2 by formation of a protein complex consisting of SETBP1, SET and protein phosphatase 2, finally leading to cell proliferation and expansion of leukemic cells. In vitro experiments demonstrated in immortalized BM70 cells a dramatic increase in mRNA levels of HOXA9 and HOXA10 by SETBP1 overexpression. This suggests that SETBP1 targets directly the expression of HOXA9 and HOXA10 by binding to the promoters of these genes.44 Piazza et al.15 demonstrated that expression of mutant SETBP1Gly870Ser in the TF1 cell line resulted in higher SETBP1 protein levels, SET protein stabilization, protein phosphatase 2 inhibition and higher proliferation rates, indicating that SETBP1 mutation reflects a similar mechanism like SETBP1 overexpression.
Although nearly all different SETBP1 missense mutations detected in this study were known from investigation of patients with Schinzel–Giedion syndrome or were recently reported in myeloid disorders,15, 16, 45 we found several new mutations in the MDS/MPN and MPN cohorts, as well as in the cytogenetically defined i(17)(q10)/ −7 cohort: the p.Ile822Thr, p.Asp868Gly, p.Ser869Arg, p.Gly870Val, p.Thr873Lys/Arg, p.Asp874Asn/His as well as p.Asp908Asn mutations, which all localize to the SKI homologous region. All mutations were predicted to have a damaging character on protein level. As the vast majority of the mutations were heterozygous and as all were missense, the SETBP1 protein probably retains structural integrity and a possibly modified function in case of mutation.
SETBP1 mutations were associated with more adverse clinical profiles in our study, that is, they correlated with higher WBC counts, lower hemoglobin levels and lower platelet counts in the total cohort. In the MPN category, SETBP1 mutations were significantly associated with higher age. Piazza et al.15 described significantly worse survival for SETBP1 mutated aCML cases as compared with wild-type patients. This was also reported by Damm et al.35 for CMML patients. This is also supported by the concept that SETBP1mut is rather a later event than an initial one and therefore might possibly influence the clinical course, as also suggested by Damm et al.35 However, this poor outcome was not reflected in the present OS analyses, where OS was not significantly different between SETBP1wt and SETBP1mut patients. This may be due to the fact that the median follow-up was relative short with 12 months for CMML and 17.1 months for aCML patients.
In conclusion, SETBP1 mutations represent an important novel molecular marker, which is highly associated with aCML, the MDS/MPN overlap category and a dysplastic phenotype. SETBP1 mutations co-occurred frequently with CBL and ASXL1 mutations and concomitant monosomy 7 and i(17)(q10). Considering that the mutations were associated with an adverse prognosis, SETBP1 mutation analysis should not only be further investigated for diagnostic purposes but also for clinical decision making in patients with chronic myeloid neoplasms.
Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues 4th edn. International Agency for Research on Cancer (IARC): Lyon, 2008.
Vardiman JW, Bennett JM, Bain B, Brunning RD, Thiele J . Atypical chronic myeloid leukaemia, BCR-ABL1 negative. In: Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, et al. (eds) WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues 4th edn. International Agency for Research on Cancer (IARC): Lyon, p 80–81 2008.
Fend F, Horn T, Koch I, Vela T, Orazi A . Atypical chronic myeloid leukemia as defined in the WHO classification is a JAK2 V617F negative neoplasm. Leuk Res 2008; 32: 1931–1935.
Bacher U, Schnittger S, Kern W, Weiss T, Haferlach T, Haferlach C . Distribution of cytogenetic abnormalities in myelodysplastic syndromes, Philadelphia negative myeloproliferative neoplasms, and the overlap MDS/MPN category. Ann Hematol 2009; 88: 1207–1213.
Jones AV, Kreil S, Zoi K, Waghorn K, Curtis C, Zhang L et al. Widespread occurrence of the JAK2 V617F mutation in chronic myeloproliferative disorders. Blood 2005; 106: 2162–2168.
Levine RL, Wadleigh M, Cools J, Ebert BL, Wernig G, Huntly BJ et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis. Cancer Cell 2005; 7: 387–397.
Tefferi A, Gilliland DG . JAK2 in myeloproliferative disorders is not just another kinase. Cell Cycle 2005; 4: 1053–1056.
Scott LM, Tong W, Levine RL, Scott MA, Beer PA, Stratton MR et al. JAK2 exon 12 mutations in polycythemia vera and idiopathic erythrocytosis. N Engl J Med 2007; 356: 459–468.
Pikman Y, Lee BH, Mercher T, McDowell E, Ebert BL, Gozo M et al. MPLW515L is a novel somatic activating mutation in myelofibrosis with myeloid metaplasia. PLoS Med 2006; 3: e270.
Pardanani AD, Levine RL, Lasho T, Pikman Y, Mesa RA, Wadleigh M et al. MPL515 mutations in myeloproliferative and other myeloid disorders: a study of 1182 patients. Blood 2006; 108: 3472–3476.
Kohlmann A, Grossmann V, Klein HU, Schindela S, Weiss T, Kazak B et al. Next-generation sequencing technology reveals a characteristic pattern of molecular mutations in 72.8% of chronic myelomonocytic leukemia by detecting frequent alterations in TET2, CBL, RAS, and RUNX1. J Clin Oncol 2010; 28: 3858–3865.
Gelsi-Boyer V, Trouplin V, Adelaide J, Bonansea J, Cervera N, Carbuccia N et al. Mutations of polycomb-associated gene ASXL1 in myelodysplastic syndromes and chronic myelomonocytic leukaemia. Br J Haematol 2009; 145: 788–800.
Meggendorfer M, Roller A, Haferlach T, Eder C, Dicker F, Grossmann V et al. SRSF2 mutations in 275 cases with chronic myelomonocytic leukemia (CMML). Blood 2012; 120: 3080–3088.
Brecqueville M, Rey J, Bertucci F, Coppin E, Finetti P, Carbuccia N et al. Mutation analysis of ASXL1, CBL, DNMT3A, IDH1, IDH2, JAK2, MPL, NF1, SF3B1, SUZ12, and TET2 in myeloproliferative neoplasms. Genes Chromosomes Cancer 2012; 51: 743–755.
Piazza R, Valletta S, Winkelmann N, Redaelli S, Spinelli R, Pirola A et al. Recurrent SETBP1 mutations in atypical chronic myeloid leukemia. Nat Genet 2013; 45: 18–24.
Hoischen A, van Bon BW, Gilissen C, Arts P, van LB, Steehouwer M et al. De novo mutations of SETBP1 cause Schinzel-Giedion syndrome. Nat Genet 2010; 42: 483–485.
Cristobal I, Blanco FJ, Garcia-Orti L, Marcotegui N, Vicente C, Rifon J et al. SETBP1 overexpression is a novel leukemogenic mechanism that predicts adverse outcome in elderly patients with acute myeloid leukemia. Blood 2010; 115: 615–625.
Albano F, Anelli L, Zagaria A, Coccaro N, Casieri P, Minervini A et al. SETBP1 and miR_4319 dysregulation in primary myelofibrosis progression to acute myeloid leukemia. J Hematol Oncol 2012; 5: 48.
Panagopoulos I, Kerndrup G, Carlsen N, Strombeck B, Isaksson M, Johansson B . Fusion of NUP98 and the SET binding protein 1 (SETBP1) gene in a paediatric acute T cell lymphoblastic leukaemia with t(11;18)(p15;q12). Br J Haematol 2007; 136: 294–296.
Shaffer LG, Tommerup N . ISCN 2013: An international System for Human Cytogenetic Nomenclature. Karger: Basel, New York, 2013.
Haferlach C, Rieder H, Lillington DM, Dastugue N, Hagemeijer A, Harbott J et al. Proposals for standardized protocols for cytogenetic analyses of acute leukemias, chronic lymphocytic leukemia, chronic myeloid leukemia, chronic myeloproliferative disorders, and myelodysplastic syndromes. Genes Chromosomes Cancer 2007; 46: 494–499.
Schnittger S, Schoch C, Dugas M, Kern W, Staib P, Wuchter C et al. Analysis of FLT3 length mutations in 1003 patients with acute myeloid leukemia: correlation to cytogenetics, FAB subtype, and prognosis in the AMLCG study and usefulness as a marker for the detection of minimal residual disease. Blood 2002; 100: 59–66.
Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE, Gerasimova A, Bork P et al. A method and server for predicting damaging missense mutations. Nat Methods 2010; 7: 248–249.
Kumar P, Henikoff S, Ng PC . Predicting the effects of coding non-synonymous variants on protein function using the SIFT algorithm. Nat Protoc 2009; 4: 1073–1081.
Schwarz JM, Rodelsperger C, Schuelke M, Seelow D . MutationTaster evaluates disease-causing potential of sequence alterations. Nat Methods 2010; 7: 575–576.
Schnittger S, Bacher U, Kern W, Schroder M, Haferlach T, Schoch C . Report on two novel nucleotide exchanges in the JAK2 pseudokinase domain: D620E and E627E. Leukemia 2006; 20: 2195–2197.
Grand FH, Hidalgo-Curtis CE, Ernst T, Zoi K, Zoi C, McGuire C et al. Frequent CBL mutations associated with 11q acquired uniparental disomy in myeloproliferative neoplasms. Blood 2009; 113: 6182–6192.
Schnittger S, Eder C, Jeromin S, Alpermann T, Fasan A, Grossmann V et al. ASXL1 exon 12 mutations are frequent in AML with intermediate risk karyotype and are independently associated with an adverse outcome. Leukemia 2013; 27: 82–91.
Schnittger S, Bacher U, Haferlach C, Geer T, Muller P, Mittermuller J et al. Detection of JAK2 exon 12 mutations in 15 patients with JAK2V617F negative polycythemia vera. Haematologica 2009; 94: 414–418.
Schnittger S, Bacher U, Haferlach C, Dengler R, Krober A, Kern W et al. Detection of an MPLW515 mutation in a case with features of both essential thrombocythemia and refractory anemia with ringed sideroblasts and thrombocytosis. Leukemia 2008; 22: 453–455.
Grossmann V, Schnittger S, Kohlmann A, Eder C, Roller A, Dicker F et al. A novel hierarchical prognostic model of AML solely based on molecular mutations. Blood 2012; 120: 2963–2972.
Thiele J, Kwasnicka H, Orazi A, Tefferi A, Birgegard G . Polycythemia vera. In: Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H et al. (eds). WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues 4th edn. International Agency for Research on Cancer (IARC): Lyon, p 40–43 2008.
Tefferi A . Novel mutations and their functional and clinical relevance in myeloproliferative neoplasms: JAK2, MPL, TET2, ASXL1, CBL, IDH and IKZF1. Leukemia 2010; 24: 1128–1138.
Jeromin S, Haferlach T, Grossmann V, Alpermann T, Kowarsch A, Haferlach C et al. High frequencies of SF3B1 and JAK2 mutations in refractory anemia with ring sideroblasts associated with marked thrombocytosis strengthen the assignment to the category of myelodysplastic/myeloproliferative neoplasms. Haematologica 2012; 98: e15–e17.
Damm F, Itzykson R, Kosmider O, Droin N, Renneville A, Chesnais V et al. SETBP1 mutations in 658 patients with myelodysplastic syndromes, chronic myelomonocytic leukemia and secondary acute myeloid leukemias. Leukemia 2013; e-pub ahead of print 5 February 2013; doi:10.1038/lieu.2013.35.
Kanagal-Shamanna R, Bueso-Ramos CE, Barkoh B, Lu G, Wang S, Garcia-Manero G et al. Myeloid neoplasms with isolated isochromosome 17q represent a clinicopathologic entity associated with myelodysplastic/myeloproliferative features, a high risk of leukemic transformation, and wild-type TP53. Cancer 2012; 118: 2879–2888.
Schwaab J, Ernst T, Erben P, Rinke J, Schnittger S, Strobel P et al. Activating CBL mutations are associated with a distinct MDS/MPN phenotype. Ann Hematol 2012; 91: 1713–1720.
Schnittger S, Bacher U, Alpermann T, Reiter A, Ulke M, Dicker F et al. Use of CBL exon 8 and 9 mutations in diagnosis of myeloproliferative neoplasms and myelodysplastic/myeloproliferative disorders: an analysis of 636 cases. Haematologica 2012; 97: 1890–1894.
Sanada M, Suzuki T, Shih LY, Otsu M, Kato M, Yamazaki S et al. Gain-of-function of mutated C-CBL tumour suppressor in myeloid neoplasms. Nature 2009; 460: 904–908.
Jankowska AM, Makishima H, Tiu RV, Szpurka H, Huang Y, Traina F et al. Mutational spectrum analysis of chronic myelomonocytic leukemia includes genes associated with epigenetic regulation: UTX, EZH2, and DNMT3A. Blood 2011; 118: 3932–3941.
Gelsi-Boyer V, Trouplin V, Roquain J, Adelaide J, Carbuccia N, Esterni B et al. ASXL1 mutation is associated with poor prognosis and acute transformation in chronic myelomonocytic leukaemia. Br J Haematol 2010; 151: 365–375.
Butler JS, Dent SY . The role of chromatin modifiers in normal and malignant hematopoiesis. Blood 2013; 121: 3076–3084.
Abdel-Wahab O, Adli M, Saunders L, Gao JG, Shih A, Pandey S et al. ASXL1 mutations promote myeloid transformation through inhibition of PRC2-mediated gene repression. Blood 2011; 118: 405a.
Oakley K, Han Y, Vishwakarma BA, Chu S, Bhatia R, Gudmundsson KO et al. Setbp1 promotes the self-renewal of murine myeloid progenitors via activation of Hoxa9 and Hoxa10. Blood 2012; 119: 6099–6108.
Makishima H, Yoshida K, Nguyen N, Sanada M, Okuno Y, Ng KP et al. Somatic mutations in Schinzel-Giedion syndrome gene SETBP1 determine progression in myeloid malignancies. ASH Annual Meeting Abstracts 2012; 120: 2.
TH, WK, CH and SS are part owners of the MLL Munich Leukemia Laboratory GmbH. MM, UB and TA are employed by the MLL Munich Leukemia Laboratory GmbH. CG-P declares no conflict of interest.
MM investigated molecular mutations, analyzed the data and wrote the manuscript; UB contributed to cytomorphologic analysis and classification of cases and wrote the manuscript; TA collected and documented clinical data and compiled statistical analyses; CG-P originally detected SETBP1 gene mutations and discussed with us this study; WK was involved in statistical analyses; CH was responsible for cytogenetics; TH was responsible for cytomorphologic analysis and was involved in the collection of clinical data; SS was the principal investigator of the study and wrote the manuscript. All authors read and contributed to the final version of the manuscript.
About this article
Cite this article
Meggendorfer, M., Bacher, U., Alpermann, T. et al. SETBP1 mutations occur in 9% of MDS/MPN and in 4% of MPN cases and are strongly associated with atypical CML, monosomy 7, isochromosome i(17)(q10), ASXL1 and CBL mutations. Leukemia 27, 1852–1860 (2013). https://doi.org/10.1038/leu.2013.133
- molecular marker
- monosomy 7
Pathogenic Mutations and Atypical Flow Cytometric Findings Characterize the Majority of Unclassifiable Myelodysplastic/Myeloproliferative Neoplasms
American Journal of Clinical Pathology (2021)
Hematology/Oncology Clinics of North America (2021)
International Journal of Molecular Sciences (2020)
Best Practice & Research Clinical Haematology (2020)