Summary The PRDM16 (1p36) gene is rearranged in acute myeloid leukaemia (AML) and myelodysplastic syndrome (MDS) with t(1;3)(p36;q21), sharing characteristics with AML and MDS with MECOM (3q26.2) translocations. We used fluorescence in situ hybridization to study 39 haematological malignancies with translocations involving PRDM16 to assess the precise breakpoint on 1p36 and the identity of the partner locus. Reverse‐transcription polymerase chain reaction (PCR) was performed in selected cases in order to confirm the partner locus. PRDM16 expression studies were performed on bone marrow samples of patients, normal controls and CD34 + cells using TaqMan real‐time quantitative PCR. PRDM16 was rearranged with the RPN1 (3q21) locus in 30 cases and with other loci in nine cases. The diagnosis was AML or MDS in most cases, except for two cases of lymphoid proliferation. We identified novel translocation partners of PRDM16 , including the transcription factors ETV6 and IKZF1 . Translocations involving PRDM16 lead to its overexpression irrespective of the consequence of the rearrangement (fusion gene or promoter swap). Survival data suggest that patients with AML/MDS and PRDM16 translocations have a poor prognosis despite a simple karyotype and a median age of 65 years. There seems to be an over‐representation of late‐onset therapy‐related myeloid malignancies.
The case of a patient presenting with a myeloproliferative disorder (MPD) characterized by a t(8;22) (p12;q11) translocation was investigated. The rearrangement resulted in the production of BCR-FGFR1 and FGFR1-BCR chimeric transcripts after in-frame fusions of BCR exon 4 with FGFR1 exon 9 and FGFR1 exon 8 with BCR exon 5, respectively. The four previously reported patients with such translocation presented with an atypical chronic myeloid leukemia (CML) without Philadelphia chromosome. In addition to a myeloproliferation, the patient had a B cell proliferation. The phenotypic characterization of the lymphoid cells in the bone marrow showed a continuum of maturation from blast B cells to polyclonal lymphocytes. In the blood, B cells showed a complete polyclonal maturation. The BCR-FGFR1 gene fusion was detected by dual-color fluorescence in situ hybridization in both CD19- and CD19+ populations. In contrast to the other FGFR1-MPDs that show myeloid and T cell proliferation, we propose that this t(8;22) MPD is a myeloid and B cell disease, and potentially a novel type of hematological disease. Although the FGFR1-MPD is rare, its study provides interesting clues to the understanding of hematopoietic stem cell biology and oncogene activation.
Myeloproliferative neoplasms (MPNs) are clonal disorders characterized by myeloproliferation and predisposition to bone marrow fibrosis. Philadelphia-negative classical MPNs comprise essential thro...
BCR-ABL and its chromosomal counterpart Philadelphia chromosome (Ph) are considered as the hallmark of chronic myeloid leukemia (CML) [1]. The expression of BCR-ABL is not restricted to CML and is ...
A4 JAK2 V617F is a mutation found in the majority of patients with polycythemia vera (PV), and a significant number of patients with essential thrombocytopemia (ET) and chronic idiopathic myelofibrosis (CIMF).The discovery of this mutation has profoundly modified the diagnosis of Ph- MyeloProliferative Disorders (MPD), but its clinical significance still remains undefined.Using different technologies, characterized by various sensitivity and sensibility, it has been reported that the incidence of this mutation is ranging from 65% to 97% in PV, 25 to 57% in ET and 35 to 57% in CIMF. Those variations are mainly due to the heterogeneity and to the poor sensitivity of the assay used, highlighting the needs of a standardized, accurate and sensitive assay. The aim of this study was to validate the performances of a new allele specific RQPCR genotyping assay and to compare with the traditional PCR sequencing strategy.This assay benefit from the plasmid technology which allows the precise calibration and normalisation of RQ-PCR results and is highly reproducible. By generating standard curves based on the known concentration of plasmid dilutions, this technology allows precise measurement of the JAK2 wild type (WT) and V617F alleles copy number in human cell samples and, without the need of subsequent handling such as capillary electrophoresis or melting curve analysis, directly provides a normalized value which is independent of the cell count in each assayed sample. Normalized results can be compared between RQ-PCR systems and testing sites.For accuracy testing and determination of detection limit we used serial dilutions of V617F homozygous HEL cell line and genomic DNA mix into non mutated K562 cell or genomic DNA. Our data show a reproducible baseline threshold of 512 V617F homozygous cells in 5 10-6 WT cells (10-4) and 6.4 pg of positive genomic DNA (2.5 10-4, 48 copies) with very good correlations toward serial dilution in each case (R² = 0.99).The greater sensitivity of this assay allowed identifying 17% more mutated patients as compared with PCR sequencing. Healthy subjects and secondary polyglobuly patient samples always resulted WT (>99.9%). In patient samples VF allele range from 5 to 96% in positive cases and is always inferior to 0.3% in negative. No significant differences were observed in JAK2 V617F % range between the different diagnostic groups, but our results show differences in median value ranging from 80% for PV to 25% for ET and IMF. Results obtained on shared samples (n=32) in two different labs show a very high correlation (R²=0.98).The sensitivity and the dynamic range of the JAK2 V617F assay are compatible with its use for highly precise and sensitive quantification at diagnosis and during MRD follow-up. Validation of an equivalent assay for the quantification of the mutation on RNA is ongoing.
6531 Background: The PRDM16 gene on chromosome 1p36 is reported to be rearranged in sparse cases of acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) with t(1;3)(p36;q21). Methods: We report the largest series to date of 39 cases of hematological malignancies with PRDM16 alterations out of a series of 120 cases with 1p36 rearrangements screened by fluorescence in situ hybridization (FISH). Results: PRDM16 was found to be rearranged with the RPN1 locus (3q21) in 30 cases and with other loci in 9 cases. The PRDM16 rearrangements are not restricted to myeloid malignancies, as we characterized two cases of lymphoid proliferation with translocations involving PRDM16. We identified novel translocation partners of PRDM16, including transcription factors ETV6 and IKZF1. This is the first report of a translocation involving IKZF1 in a myeloid malignancy. Translocations involving PRDM16 are original in that they lead to its over-expression through 2 different mechanisms (transcriptional upregulation by promoter switch or formation of a chimeric gene). Survival data interestingly suggest that patients with AML/MDS and PRDM16 translocations have a poor prognosis whatever the partner gene, RPN1 versus others. The 32 patients with available survival data had a poor prognosis, as the median overall survival (OS) was 18 months [95% CI, 6 to 31 months] and 5-year OS was 25.7% [95% CI, 8.4-43.0%]. Conclusions: Our study confirms that AML/MDS with PRDM16 translocations share numerous characteristics with AML/MDS associated with translocations involving EVI1 (3q26), which is another transcription factor with a PR domain. If this poor prognosis is confirmed, we propose the addition of a “PRDM16”-entity in the WHO classification of AML/MDS, as is already the case for AML with EVI1 translocations. This new entity could later be broadened to a “PR domain gene rearrangements” subgroup to encompass the AML with EVI1 translocations.
Click to increase image sizeClick to decrease image size Potential conflict of interestDisclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal.
An isodicentric X chromosome, with a breakpoint in Xq13 named idic(X)(q13), is a rare but recurrent abnormality found in about 1% of myelodysplastic syndromes (MDS), especially in old females. Most cases with idic(X)(q13) are MDS, but a number of acute myeloid leukemias (AML), usually secondary to MDS, have also been reported, and a few cases of myeloproliferative neoplasms (MPN).1 The idic(X)(q13) is frequently the only cytogenetic abnormality, albeit one or more copies of the idic, beside a normal X chromosome being present in the same cells. This suggests that idic(X)(q13) may be implicated in the early process of the disease. The presence of idic(X)(q13) is specifically associated with MDS in the WHO classification. Its clonal detection by conventional cytogenetics is sufficient to diagnose MDS, in case of cytopenias without morphological dysplasia.2 Conversely, the isochromosome of the short arm of chromosome X named i(X)(p10) is a very rare abnormality. It is observed in some cases of myeloid or lymphoid malignancies, both as a unique abnormality, and as part of a more complex karyotype, in female, and less frequently in male patients. Myelodysplastic syndromes evolution is heterogeneous with transformation to AML in about 30% of cases. An International Prognostic Scoring System (IPSS) is commonly used to predict leukemic evolution of primary untreated adults and to adapt therapy. Revised (R) in 2012,3 this score includes bone marrow blast percentage, hemoglobin level, absolute neutrophil count, platelet count and cytogenetics. There is no definite prognostic impact from Idic(X)(q13), and is considered as another abnormality in the intermediate group of the IPSS-R. Recently, exome and genome-wide sequencing of MDS cases revealed a great diversity of genomic aberrations, with more than 25 recurrent mutations.4 Several genes involved in pre-messenger RNA splicing have been shown to be mutated in half of MDS patients, revealing a new leukemic pathway involving spliceosome dysfunction. To characterize the clinical and genetic profile of myeloid neoplasms associated with idic(X)(q13) and i(X)(p10), we performed a retrospective study, between 2002 and 2017, of 45 patients with myeloid neoplasm who presented these abnormalities. All patients gave their informed consent in agreement with the Helsinki declaration; and the Institutional Ethics Committee at Avicenne Hospital approved this study. The Idic(X)(q13) chromosome was found in 33 cases, and the i(X)(p10) in 12 patients who were all female. Ten cases were associated with other chromosomal abnormalities mainly involving chromosomes five and seven. The analysis was then restricted to 35 patients with an isolated X chromosome abnormality. In the majority of cases (n = 29), the X chromosome abnormality was detected at diagnosis. Isochromosomes could result from transverse, instead of longitudinal misdivision of the centromere; another mechanism could be chromatid exchange involving two homologous chromosomes. In both scenarios, it leads to the loss of the long arm and gain of the short arm of chromosome X, resulting in a state of genetic imbalance. It is possible that the frequency of i(X)(p10) was previously underestimated in the literature. The cytogenetic distinction between del(Xq) and i(Xp) is difficult due to the similarity of Xp- and q-arm banding patterns, extending from the centromere to band Xq24 (Figure 1). A morphological centralized review of the bone marrow smear was performed by two cytologists. The revised diagnosis was MDS (n = 25) [MDS with ring sideroblasts (n = 1), MDS with uni (n = 1) or multilineage dysplasia (n = 10) and with ring sideroblasts (n = 2), MDS with excess blasts −1 (n = 6), MDS with excess blasts 2 (n = 3), MDS with myelofibrosis (n = 2)] or AML with multilineage dysplasia (n = 2), chronic myelomonocytic leukemia (n = 4), atypical chronic leukemia (n = 1) or unclassifiable cytopenia (n = 3). The majority of patients presented marked dysplastic features (>10%) in one or more of the major lineages. No specific feature was observed. Some patients had a history of previous neoplasia or myeloid neoplasm, including four solid tumors (breast cancer n = 3), two AML with normal karyotype and NPM1 mutation, one acute lymphoblastic leukemia, one lymphoma, one chronic myelocytic leukemia on complete remission, two immune thrombocytopenias and two monoclonal gammopathies. Nineteen analyzed patients presented a median of 2.9 mutations in the 27 genes analyzed by Next Generation Sequencing. The most frequently mutated genes were TET2 (74%), SRSF2 (68%) and ASXL1 (47%). Half of patients presented two or more mutations of TET2 (often a frameshift insertion / deletion and a stop codon). Twelve patients presented a hotspot mutation of SRSF2 (p.P95R/L/H) and one out of nine patients had the hotspot c.1934dupG mutation of ASXL1. Among the other mutated genes, we noted mutations of DNMT3A (10.5%), NRAS (10.5%), SF3B1 (10.5%), IDH1 (10.5%) and JAK2 (10.5%). Only one case had a mutation of CSF3R, CBL, IDH2, EZH2, U2AF1, RUNX1 and WT1 respectively. Interestingly, no patient presented a mutation of TP53. In 94% of cases, TET2 mutations co-occurred with other mutations, especially involving SRSF2 or ASXL1 in 11 (57.8%) and 6 cases (31.5%) respectively. Among patients with MDS, we note a frequent association between one or more TET2 mutations and a mutation of SRSF2 (Table S1). We compared the molecular profile of the i(X)(p10) group (n = 7) with the idic(X)(q13) group (n = 12). The TET2 mutations were associated with i(X)(p10) in 43% of cases, and co-occurred with the SRSF2 mutation in 66% of cases. The TET2 mutations were associated with idic(X)(q13) in 91.6% of cases, and co-occurred with SRSF2 mutation in 75% of cases. There was no significant difference between the two subgroups probably due to the small sample size. With a median follow-up of 21 months (2-114 months), 15 patients died and 8 were lost to follow-up. Nine patients with MDS evolved to a higher grade MDS or AML. Among these patients, two received stem cell transplantation and one obtained a complete remission. The other patients received various treatments such as 5-azacytidine (n = 3), a cytarabine based regimen (n = 3) or palliative care (n = 3). They were all deceased at the end of our study. In this series, the iso/isodic chromosome X was the sole abnormality in 35 cases, which suggests a role of these abnormalities in the pathologic process. No gene involved in the myelodysplastic or leukemic process was discovered in the genetic region located in Xq13.5 The mitochondrial iron transporter gene, and ATP-binding cassette transporter (ABCB7), mapped in Xq21, which is implicated in X-linked sideroblastic anemia and spinocerebellar ataxia is lost. Functional ABCB7 protein is required for mitochondrial iron homeostasis and is involved in the transport of iron from mitochondria to the cytosol.5 The ABCB7 mRNA transcript is down-regulated in case of SF3B1 mutations, resulting in an abnormal degradation of this mRNA. In our cohort, only six patients presented an excess of ring sideroblasts and SF3B1 mutations were found in two of these six patients. This observation supports that other mechanisms are implicated in the myelodysplastic features. The TET2 mutations, found in 74% of our cases, were the most frequent mutations. Two or more TET2 mutations were found in half of our patients. Previously, Paulsson et al.5 identified that 4 patients out of 11 with idic(X)(q13) carried this mutation. The TET2 mutations are usually associated with an increase in hematopoietic self-renewal due to modification of DNA methylation. They are frequent in MDS, involving approximately 20% to 30% of patients. The biallelic inactivation of TET2 seems to be an early oncogenic event.4 Two or three mutations of TET2 were observed in 9 patients out of 18, suggesting the presence of a biallelic inactivation. In patients who presented TET2, SRSF2, or ASXL1 mutations, the variant allelic frequency (VAF) of SRSF2 (m = 41%) and ASXL1 (m = 25%) were lower than TET2 VAF (m = 51%), suggesting that TET2 mutation was the first oncogenic event. Note, SRSF2 is a member of the serine/arginine-rich class of splicing factors that is involved in DNA stability. Mutations in this gene occur in 3% to 11% of early-stage MDS, and are more frequent in CMML (about 25%). Also, ASXL1 is a histone-modifying enzyme component of the polycomb repressive complex two, that regulates the stem cell differentiation and gene transcription repression. In the literature, ASXL1 mutations occurred in about 11% to 21% of MDS patients, and 10% to 15% of MPN patients, and are associated with poor prognosis.6 In our cohort of patients, the majority of cases presented mutations associated with myelodysplastic features with a median of 2.9 mutations/patient. There were no differences between the two subgroups of iso and idic patients. These results demonstrate that i(X)(p10) is a specific myeloid abnormality such as idic(X)(q13). The high level of TET2, SRSF2 and ASXL1 mutations in patients with X chromosome abnormalities, is higher than that usually found in myeloid disorders (75% vs 20% to 30% for TET2; 68% vs 15% for SRSF2; 47% vs 11% to 21% for ASXL1).4 The SRSF2 and ASXL1 mutations are usually associated with worse prognosis in MDS patients,6 contrasting with the intermediate risk of i(X)(p10)/idic(X)(q13) in the IPSS-R score. These results could have an impact on treatment strategies in the future. The authors thank L Lebas, C Calbrix (CLCC Henri Becquerel, Rouen) for their technical assistance, and M Munger for the manuscript revision. D.P., P.E. and V.E. equally participated in the study. D.P. and V.E. designed the study, P.E. and J.V. performed molecular analyzes, D.L. and V.E. reviewed the bone marrow smear and the data. All authors contributed to collected data, revised and approved the manuscript content. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
