Search

Acquired aplastic anemia (AA) is a bone marrow failure resulting from an autoimmune T-cell attack against hematopoietic stem and progenitor cells (HSPCs). Several escape mechanisms within the HSPC compartment have been reported: i) adaptive alterations causing loss of high-risk HLA allele of AA development [deletion or uniparental disomy (UPD) of chromosome 6p including HLA cluster region, and loss-of-function (LOF) mutations in HLA gene], ii) clonal expansion of paroxysmal nocturnal hemoglobinuria (PNH) clone resulting from PIGAmutation, and iii) presence of clonal hematopoiesis of indeterminate potential (CHIP) mutations (i.e., BCOR/ BCORL1, DNMT3A, ASXL1). Multiple types of alterations often co-occur, however due to the inherent limitations of conventional methods, it is not known if i) these alterations are in the same or independent clones, ii) what is the mutational hierarchy, and iii) what is the cell of origin. We addressed these questions using single-cell and bulk genomic sequencing.

Background:Infantile hemangioma (IH) is the most frequent vascular tumor of early childhood. Recently, propranolol, a nonselective ß1- and ß2-adrenoceptor inhibitor, was introduced into the therapy of severe proliferating IH with excellent results. However, the underlying mechanism of action of propranolol is still unclear.Methods:We performed immunohistochemistry for cluster of differentiation 31 (CD31), D2-40, glucose transporter-1 (GLUT-1), and Ki67 in order to characterize 21 vascular anomalies (nine IH, seven venous malformations (VMs), and five lymphatic malformations (LMs)). Furthermore, we analyzed the expression of ß1-, ß2-, and ß3-adrenoceptor mRNA in these specimens as well as in hemangioma-derived stem cells by quantitative real-time PCR (qPCR).Results:We show that the expression of ß1-adrenoceptor mRNA is 10.7-fold higher in IH independent of the proliferative or regressive phase as well as 2.5-fold higher in hemangioma-derived stem cells as compared with ß2-adrenoceptor mRNA. In LM, the expression of ß2-adrenoceptor mRNA was ninefold higher than that of ß1-adrenoceptor mRNA. VM showed low expression levels of all ß-adrenoceptor mRNAs, and ß3-adrenoceptor mRNA was hardly detectable in any specimens examined.Conclusion:These results provide the first evidence of distinctions between IH and vascular malformations with regard to ß-adrenoceptor subtype mRNA levels.

Pearson bone marrow-pancreas syndrome (PS) is a rare, usually fatal mitochondrial disorder involving the hematopoietic system in early infancy. Due to the diversity of clinical symptoms, the diagnosis can be difficult. The authors describe a boy with severe hypoplastic anemia in whom extensive clinical, biochemical, and morphologic findings led to the diagnosis of PS, and molecular analysis revealed a novel deletion of mitochondrial DNA from nucleotide position 10.371 to 14.607.

No relevant conflicts of interest to declare.

No relevant conflicts of interest to declare.

Introduction

Pediatric myeloid neoplasms are a clinically and genetically heterogeneous group of clonal diseases. Unlike in adults inherited predisposition plays a major role in leukemogenesis. A number of germline mutations are known to increase the risk of somatically occurring myeloid transformation. These include the genes associated with the classical inherited bone marrow failure syndromes, but also transcriptopathies resulting from RUNX1, CEBPAor GATA2deficiency. Herein we review the most significant findings from recent years and our present experience on the clonal landscape of myeloid malignancies in children with a focus on pediatric MDS and JMML. In both disease entities, monosomy 7 is the most frequent chromosomal aberration. It is likely that haploinsufficiency of genes located on chromosome 7 contributes to clonal evolution. Several of the more than 1000 lost genes were postulated as mechanistic drivers. However, in most cases additional somatic mutations not related to chromosome 7 are found at diagnosis, making it difficult to pinpoint the initial oncogenic hit. Driver somatic hits frequently encountered in adults with MDS, affecting TET2, DNMT3Aor the spliceosome genes, do not play a major role in the pathogenesis of childhood disease. Also, considering the young age, environmental effects and age-related pre-malignant clonal mutations can be disregarded. In fact, GATA2germline mutations can be considered the most common "initial hit" in pediatric MDS. In our experience, GATA2deficiency accounts for 37% of primary pediatric MDS with monosomy 7. The prevalence of GATA2mutations in MDS with monosomy 7 increases dramatically in adolescence where they are found in up to 72%. In comparison, 16% of pediatric patients with MDS and trisomy 8 have an underlying GATA2deficiency. The somatic mutational landscape in childhood MDS is associated with cytogenetic subgroups, i.e. SETBP1and ASXL1mutations are overrepresented in cases with monosomy 7. Interestingly, longitudinal NGS analysis and single colony sequencing indicate that SETBP1lesions precede the development of ASXL1mutations. We identified the genes SETBP1, ASXL1, NRAS, KRAS, RUNX1, PTPN11and BCOR/ BCORLas the most frequent targets of recurrent mutations in a large cohort of pediatric patients with primary MDS. The biological significance of these genomic changes is currently not understood and warrants further studies. In contrast to pediatric MDS, signaling pathways involved in JMML have been better defined during recent years. Somatic aberrations in PTPN11, NRAS, KRAS, CBL, and NF1genes lead to a constitutively overactive RAS pathway in leukemic cells. Additional activating mutations were identified in some JMML patients, including the RAS pathway gene RRAS, and concomitant mutations in SETBP1and JAK3. The latter were postulated to develop as subclonal events during disease progression and to confer poor prognosis. However most recently, very rare clones with SETBP1mutation were reported to be present at diagnosis in as much as 30% of JMML patients. This might indicate that multiple clones with different clonal composition arise during initial stages of disease, and therapy-driven selection pressure results in the expansion of these subclones. It is not clear how the somatic mutations interact with the epigenetic landscape in JMML clones, especially with regard to promoter hypermethylation, and what are the definitive biological consequences. In summary, a number of germline and somatic mutations have been identified which can initiate the development of abnormal hematopoiesis in children. Premalignant stem cells undergo evolutionary selection and produce somatically altered malignant clones resulting in myeloid transformation.

No relevant conflicts of interest to declare.

Wagner: CORD:USE: Membership on an entity's Board of Directors or advisory committees; VidaCord: Membership on an entity's Board of Directors or advisory committees.

Abstract 533

Abstract 1542

Juvenile myelomonocytic leukemia (JMML) is a myeloproliferative disorder characterized by overproduction of myelomonocytic cells. Activating mutations of PTPN11, which encodes the protein tyrosine phosphatase, Shp2, are found in 35% of JMML patients. Murine bone marrow low density mononuclear cells (LDMNCs) expressing activating Shp2 mutants preferentially undergo myelomonocytic differentiation and expansion despite being subjected to conditions that normally support only mast cell development. Evaluation of hematopoietic cell-specific transcription factor expression using quantitative RT-PCR indicates equal expression of PU.1 in mutant Shp2- and WT Shp2-expressing hematopoietic progenitors; however, GATA2 expression is substantially reduced and c-Jun expression is increased in mutant Shp2-expressing progenitors. Based on these findings, we hypothesized that mutant Shp2-induced Ras hyperactivation produces constitutive c-Jun expression permitting, as a co-activator of PU.1, excessive monocytic differentiation and reduced GATA2 expression. As a corollary, we hypothesized that ectopic expression of GATA2, but not of GATA2 lacking the C-terminal zinc finger (amino acids 330 – 407, DGATA2), which is needed to compete with c-Jun for binding to PU.1, would normalize mutant Shp2-induced monocytic differentiation and expansion. To examine this hypothesis, we utilized retroviral co-transduction of bone marrow Lin-enriched cells to generate six experimental groups: pMIEG3-Shp-2WT plus pCD4 (empty vector); pMIEG3-Shp2WT plus pCD4-GATA2; pMIEG3-Shp2WT plus pCD4-ΔGATA2; pMIEG3-Shp-2E76K plus pCD4; pMIEG3-Shp-2E76K plus pCD4-GATA2; pMIEG3-Shp-2E76K plus pCD4-ΔGATA2. Cells were stained with anti-human CD4 conjugated to allophycocyanin (APC), sorted for EGFP+APC+ cells, and plated into progenitor assays. Colonies were scored for colony forming unit (CFU)-granulocyte-macrophage (GM), monocyte (M), granulocyte (G), and granulocyte-erythroid-monocyte-megakaryocyte (GEMM). As predicted, cells co-transduced with activating Shp2 mutant E76K and a secondary empty vector produced significantly more CFU-M than cells expressing WT Shp2. Upon co-transduction with GATA2, the number of CFU-M generated from E76K-expressing cells was significantly reduced. In contrast, cotransduction of ΔGATA2 lacking the C-terminal zinc finger failed to normalize the number of CFU-M produced by E76K-expressing cells. Mechanistically, we reasoned that if the c-Jun-PU.1 interaction contributes to the mutant Shp2-induced aberrant monocytic differentiation, we would observe increased expression directed by the macrophage colony-stimulating factor receptor (MCSFR) promoter, originally characterized to be activated by PU.1 and co-activator c-Jun. Consistently, MCSFR promoter-directed luciferase expression exhibited higher levels in NIH3T3 cells expressing Shp2E76K compared to those expressing WT Shp2. Furthermore, co-immunoprecipitation assays using nuclear extracts demonstrate increased c-Jun-PU.1 complexes in mutant Shp2-expressing hematopoietic progenitors, while chromatin immunoprecipitation assays demonstrate an increased association of the c-Jun-PU.1 complex at the MCSFR promoter. Moreover, CD34+ JMML progenitors express significantly higher levels of c-JUN than CD34+ cord blood progenitors from healthy newborns, substantiating the disease relevance of these mechanistic findings. These data suggest that one fundamental sequela of Ras hyperactivation in mutant Shp2-expressing cells is elevated c-Jun expression and an increased c-Jun-PU.1 interaction promoting monocytic differentiation and, thus, overproduction of myelomonocytic cells, the hallmark of JMML. These findings imply that pharmaceutical interventions that inhibit c-Jun expression or function could theoretically inhibit mutant Shp2-induced monocytic differentiation and, thus, serve as a novel approach for treatment of JMML.

Juvenile myelomonocytic leukemia (JMML) is a myeloproliferative disorder characterized by overproduction of myelomonocytic cells. Activating mutations of PTPN11, which encodes the protein tyrosine phosphatase, Shp2, are found in 35% of JMML patients. Murine bone marrow low density mononuclear cells (LDMNCs) expressing activating Shp2 mutants preferentially undergo myelomonocytic differentiation and expansion despite being subjected to conditions that normally support only mast cell development. Evaluation of hematopoietic cell-specific transcription factor expression using quantitative RT-PCR indicates equal expression of PU.1in mutant Shp2- and WT Shp2-expressing hematopoietic progenitors; however, GATA2expression is substantially reduced and c-Junexpression is increased in mutant Shp2-expressing progenitors. Based on these findings, we hypothesized that mutant Shp2-induced Ras hyperactivation produces constitutive c-Jun expression permitting, as a co-activator of PU.1, excessive monocytic differentiation and reduced GATA2 expression. As a corollary, we hypothesized that ectopic expression of GATA2, but not of GATA2 lacking the C-terminal zinc finger (amino acids 330 – 407, DGATA2), which is needed to compete with c-Jun for binding to PU.1, would normalize mutant Shp2-induced monocytic differentiation and expansion. To examine this hypothesis, we utilized retroviral co-transduction of bone marrow Lin-enriched cells to generate six experimental groups:

Approximately 75% of patients with juvenile myelomonocytic leukemia (JMML) harbour mutations in PTPN11, NF1 and RAS genes. The remaining cases presumably carry somatic mutations in other genes in the RAS pathway. BRAF plays a central role in this pathway between RAS and downstream molecules including MEK and ERK. BRAF mutations frequently occur in cancer. Recently, BRAF mutations were found in leukemia. Besides that, germline BRAF mutations cause cardio-facio-cutaneous syndrome, which shares many features with Noonan syndrome (NS). NS predisposes to a myeloproliferative disease resembling JMML. In 65 JMML patients screening for V600E mutations in exon 15 of the BRAF gene was performed from mononuclear cells. In a subset of patients, without RAS or PTPN11 mutations, and no clinical signs of NF1, the entire coding sequence of BRAF was analyzed. Sequence analysis was performed by direct, bidirectional sequencing of purified polymerase chain reaction products. In none of the 65 cases a V600E mutation of the BRAF gene was found. In a subset of patients in which the entire coding sequence of BRAF was analyzed, no mutations were identified either. Mutant proteins of the RAS-RAF-MEK-ERK pathway play an important role in the pathogenesis of JMML, resulting in GM-CSF hypersensitivity. In about 75% of the JMML cases these mutations affect RAS, NF1 or PTPN11 genes. The hypothesis for this study was that BRAF might play an important role in JMML as it is an important downstream effector of RAS. Our data show that apparently BRAF mutations do not play a role in JMML. Therefore, additional analysis of genes of the RAS pathway will be necessary to identify genetic aberrations in cases without known mutations.

(Very) Severe acquired aplastic anemia ((v)SAA) and myelodysplastic syndrome (MDS) are rare diseases in childhood. (V)SAA is a bone marrow failure syndrome characterized by immune mediated destruction of hematopoietic progenitors. MDS is a malignant clonal stem cell disorder, in which the hypoplastic variant is, in case of absence of a cytogenetic clone, difficult to separate from (v)SAA. Recently, studies provided a “molecular signature” of autoimmunity in adult (v)SAA, by showing oligoclonality of TCR Vß CDR3 region, which is refered to as TCR Vß skewing. We investigated the value of TCR Vß repertoire analysis in pediatric MDS-RC and (v)SAA patients. Peripheral blood and/or bone marrow mononuclear samples of patients with (v)SAA (n=38), MDS-RC (n=28) and 18 controls were analysed with Vß heteroduplex analysis of extracted RNA. Skewing was found in 21/38 (55%) of the (v)SAA patients and in 17/28 (61%) of the RC patients. Seventeen patients with clinical (v)SAA showed no oligoclonality. A significant difference in TCR skewing was found between the (v)SAA + MDS-RC patients as compared to the controls (χ2analysis, p=0.001), but not between MDS-RC and (v)SAA (χ2analysis, p=0.8). In this study paired samples (PB and BM) of 25 cases showed a high correlation between the skewing results in both compartments (Pearson correlation, rr 0.98). In this study TCR Vß repertoire analysis did not discriminate between MDS-RC and (v)SAA. Prospective studies will be necessary to investigate whether there is a role for this molecular tool in pediatric MDS-RC for the identification of a subset of patients that is associated with auto-immunity and therfore could be treated with IST up-front, and whether it can be used for molecular response monitoring.

(Very) Severe acquired aplastic anemia ((v)SAA) and myelodysplastic syndrome (MDS) are rare diseases in childhood. (V)SAA is a bone marrow failure syndrome characterized by immune mediated destruction of hematopoietic progenitors. MDS is a malignant clonal stem cell disorder, in which the hypoplastic variant is, in case of absence of a cytogenetic clone, difficult to separate from (v)SAA. Recently, studies provided a “molecular signature” of autoimmunity in adult (v)SAA, by showing oligoclonality of TCR Vß CDR3 region, which is refered to as TCR Vß skewing. We investigated the value of TCR Vß repertoire analysis in pediatric MDS-RC and (v)SAA patients. Peripheral blood and/or bone marrow mononuclear samples of patients with (v)SAA (n=38), MDS-RC (n=28) and 18 controls were analysed with Vß heteroduplex analysis of extracted RNA. Skewing was found in 21/38 (55%) of the (v)SAA patients and in 17/28 (61%) of the RC patients. Seventeen patients with clinical (v)SAA showed no oligoclonality. A significant difference in TCR skewing was found between the (v)SAA + MDS-RC patients as compared to the controls (χ2 analysis, p=0.001), but not between MDS-RC and (v)SAA (χ2 analysis, p=0.8). In this study paired samples (PB and BM) of 25 cases showed a high correlation between the skewing results in both compartments (Pearson correlation, rr 0.98). In this study TCR Vß repertoire analysis did not discriminate between MDS-RC and (v)SAA. Prospective studies will be necessary to investigate whether there is a role for this molecular tool in pediatric MDS-RC for the identification of a subset of patients that is associated with auto-immunity and therfore could be treated with IST up-front, and whether it can be used for molecular response monitoring.

Juvenile myelomonocytic leukemia (JMML) is an aggressive myeloproliferative disorder of childhood characterized by a high frequency of somatic mutations in NRAS, KRAS2, or PTPN11 genes as well as LOH of the NF1 allele (mainly in patients with type 1 neurofibromatosis). Activation of the Ras pathway is the biochemical consequence of these lesions. Hematopoietic stem cell transplantation (HSCT) is the only known curative treatment, with event-free survival (EFS) rates approaching 50–60%. Leukemia relapse still remains the main cause of treatment failure. Monitoring minimal residual disease (MRD) before and after HSCT is challenging due to the lack of tractable molecular markers. We designed and validated a fluorescently based, allele-specific polymerase chain reaction assay called TaqMAMA that can detect point mutations in eight of the most common RAS or PTPN11 mutations with a sensitivity of 1/1000 copies of the wildtype allele. We analyzed the peripheral blood (PB) and/or bone marrow (BM) in 22 patients for the levels of these mutant alleles at diagnosis and pre-HSCT. Eighteen of these patients had weekly-monthly PB and/or BM samples available after 21 HSCTs (3 children had second transplants). We compared the percentage of mutant DNA with rising levels of autologous cells measured by chimerism analysis after HSCT, which is the current surrogate marker used to detect relapse. Analysis of patients at initial diagnosis and pre-HSCT revealed a broad distribution of the frequency of the mutant alleles, regardless of the use or intensity of pre-HSCT chemotherapy. As shown in the table, levels of disease burden prior to HSCT were not predictive of relapse. After HSCT, the level of the mutant allele rose rapidly in patients who relapsed and correlated well with analysis of chimerism (r=0.86, p<0.01), thus validating the assay. In addition, the concordance rate between 15 simultaneously obtained PB and BM samples was excellent in 13 of the pairs, with a maximum of a 3.7 fold difference at < 1% of the mutant allele. In conclusion, patients with JMML harboring the most common Ras pathway mutations can now be monitored using an allele specific MRD assay. Most importantly, these assays can measure the frequency of mutant alleles in PB and provide a sensitive new strategy for MRD monitoring that can evaluate the molecular response to new therapeutic strategies for JMML. Median levels of mutant allele prior to HSCT Clinical Remission after HSCT Gene Mutation % Mutation with Therapy pre HSCT % Mutation with No Therapy pre HSCT Any Therapy No Therapy n Median n Median n n PTPN11 g226a 7 72.7 4 33.0 3 2 a227g 1 20.5 1 29.3 0 1 g214a 2 66.8 0 1 N/A g1508c 0 1 43.8 N/A 1 NRAS g38a 1 72.2 1 45.6 0 1 g37c 1 38.1 0 0 N/A KRAS g35a 1 9.6 0 1 N/A g38a 2 34.4 0 0 N/A

Juvenile myelomonocytic leukemia (JMML) is a malignant hematopoietic disorder of early childhood with myeloproliferative and myelodysplastic properties. The proliferative component is a result of RAS pathway deregulation caused by somatic mutation in the RAS or PTPN11 oncogenes (60% of cases) or by underlying neurofibromatosis type 1 (NF-1) with a germline NF1 gene defect (clinically 11% of cases). To search for potential collaborating genetic abnormalities, we used Affymetrix GeneChip Mapping 50K arrays to analyze over 116,000 single nucleotide polymorphisms (SNPs) across the genome using DNA from bone marrow or peripheral blood granulocytes from 16 children with JMML and normal karyotype (mutant RAS, n=4; mutant PTPN11, n=7; NF-1, n=5). Quantitative evaluation of hybridization intensities at each SNP locus failed to identify recurrent allelic gains or losses in the 16 cases. We were specifically interested in chromosome 7 because monosomy 7 or interstitial/terminal 7q deletions are found in about 30% of JMML cases and it is conceivable that submicroscopic 7q lesions occur in the other cases but remain undetected by standard techniques. However, at the resolution provided by the arrays used here, we saw no evidence for genomic copy number alterations on chromosome 7. Interestingly, evaluation of the SNP allelotypes identified large regions of copy-neutral loss of heterozygosity (LOH) on chromosome 17q, including the NF1 locus, in 4 of the 5 samples from patients with JMML and NF-1. The LOH region spanned a genomic range of approximately 55 Mbp in each case and included more than 1,400 contiguous SNPs. Allelic copy numbers were normal within the homozygous regions, indicating uniparental isodisomy (UPD). Compatible with isodisomy, 17q was normal in the corresponding conventional karyotypes. By contrast, the array data provided no evidence for 17q UPD in any of the 12 JMML cases without NF-1. In all four cases with 17q UPD, the recombination breakpoints appeared to be confined to a 400-kbp region on 17q11.1–17q11.2; however, lack of parental or nonleukemic DNA precluded definitive mapping of the breakpoints. Of note, the 17p chromosomal arm retained heterozygosity in all cases, indicating that the p53 tumor suppressor was not affected by the UPD event. We confirmed 17q disomy in the four NF-1 samples using matrix-based comparative genomic hybridization and are currently verifying homozygosity of multiple microsatellites spaced across chromosome 17. Furthermore, NF1 mutational analysis is under way to show that the individual lesion within this tumor suppressor gene is biallelic in leukemic cells from patients with JMML and NF-1. In summary, we assume that a mitotic recombination event in an early hematopoietic progenitor cell led to UPD involving the NF1 locus. Our observations provide strong confirmatory evidence that it is indeed the NF1 gene that is responsible for the increased incidence of JMML in NF-1 patients. In addition, our data support the emerging role of mitotic recombination as a second hit in leukemogenesis and corroborate the concept that RAS pathway deregulation is central to JMML pathogenesis.

Development of a secondary leukemia after chemotherapeutic treatment for childhood acute leukemia is associated with rearrangements of the MLL gene on chromosome 11q23 and characteristically results in acute myeloid leukemia (AML). We identified the NOTCH co-activator Mastermind-like 2 as MLL fusion partner in two cases of pediatric secondary leukemia. The MLL-MAML2 fusion results from a cryptic inv(11)(q21q23) only detectable by interphase fluorescence in situ hybridization with an MLL split signal probe. With a latency of twenty months and five years after initial diagnosis for MLL negative AML and precursor B acute lymphoblastic leukemia (ALL), respectively, both patients developed a therapy-related T-cell acute lymphoblastic leukemia. MAML2 is the first MLL fusion partner involved in human Notch-signaling and was only recently identified as recurrent translocation fusion partner in a subset of salivary gland tumors. The genomic MLL breakpoint shows similar localization and sequence features described for etoposide induced treatment-related AML. MLL-MAML2 positive cells were detectable up to two years prior to clinical apparent secondary leukemia in one case. The discrepant dynamics of clone expansion quantified by either the genomic fusion sequence or Ig/TCR gene rearrangements as clone specific markers suggests a selection process within the inversion positive population and the need for additional mutation events to promote overt leukemic disease. Whole genome expression profiles demonstrated differential expression of both typical MLL and NOTCH downstream genes, which suggests a modulatory role of the MAML2 transcriptional activation domain in MLL leukemogenesis and lineage assignment induced by the MLL-MAML2 fusion protein. MSM21620813 WSF3401600.

Development of a secondary leukemia after chemotherapeutic treatment for childhood acute leukemia is associated with rearrangements of the MLLgene on chromosome 11q23 and characteristically results in acute myeloid leukemia (AML). We identified the NOTCH co-activator Mastermind-like 2 as MLL fusion partner in two cases of pediatric secondary leukemia. The MLL-MAML2 fusion results from a cryptic inv(11)(q21q23) only detectable by interphase fluorescence in situ hybridization with an MLLsplit signal probe. With a latency of twenty months and five years after initial diagnosis for MLL negative AML and precursor B acute lymphoblastic leukemia (ALL), respectively, both patients developed a therapy-related T-cell acute lymphoblastic leukemia.

Noonan syndrome is characterized by short stature, facial dysmorphism, and cardiac defects. We and our colleagues discovered novel de novo germline KRAS mutations that introduce V14I, T58I, or D153V amino acid substitutions in individuals with NS and a P34R alteration in an individual with cardio-facio-cutaneous (CFC) syndrome, which has overlapping phenotypic features with NS. Recombinant V14I and T58I K-Ras proteins display defective intrinsic GTP hydrolysis and impaired responsiveness to the GTPase activating proteins (GAPs) p120 GAP and neurofibromin. We also found that V14I and T58I K-Ras render primary hematopoietic progenitors hypersensitive to growth factors and deregulate signal transduction in a cell lineage specific manner (Nature Genetics38, 331, 2006). We recently began interrogating the P34R and D153V K-Ras mutant proteins and a novel CFC-associated F156L K-Ras mutant protein. Both P34R and D153V K-Ras display normal levels of intrinsic GTP hydrolysis. In contrast, F156L K-Ras displays defective intrinsic GTP hydrolysis that resembles oncogenic G12D K-Ras. P34R K-Ras is completely resistant to both GAPs, which is intriguing as these data suggest that the ability of GAPs to down-regulate Ras-GTP levels is dispensible for development. In contrast, D153V K-Ras is responsive to both GAPs. We expressed mutant Ras proteins in COS-7 monkey kidney cells to investigate activation of Ras and downstream effectors. Cells expressing P34R, D153V, F156L, and G12D K-Ras demonstrate elevated levels of Ras-GTP and phospho-MEK in basal and serum-starved conditions. We conclude that germline KRAS mutations that cause human disease encode proteins with distinct mechanisms of biochemical activation. Further investigation of this allele series will provide insights about the relative importance of the intrinsic Ras GTPase, GAPs, and other biochemical mechanisms in controlling the growth of different cell lineages.

Diamond-Blackfan anemia (DBA) is a congenital red cell aplasia with marked clinical heterogeneity, an increased risk of malignancy and mutations in ribosomal protein (RP) S19 in 25% of probands. To identify other gene(s) mutated in DBA and investigate their expression and function, we performed a genome-wide screen using a 10,000 single nucleotide polymorphism mapping set (Affymetrix) on a large family comprising 10 informative meioses. We found linkage of the DBA phenotype to regions on chromosome 8q, 10 and 6. The RP gene RPS24, is located in the linked region on chromosome 10; we sequenced exons, intron-exon boundaries and the promoter regions in this family, and found a nonsense mutation (316C>T) in exon 4 of RPS24 in five affected individuals, and a wild type sequence in five unaffected family members. This mutation causes the change of Gln106STOP and is predicted to result in formation of a truncated RPS24 protein. Subsequently, we sequenced DNA from 215 unrelated DBA probands, 30 with RPS19 mutations and 185 without. We found another nonsense mutation in exon 2 in a sporadic case, and a splice site deletion resulting in skipped exon 2 in another proband and in his father; over 200 control individuals did not have any of the above sequence changes, indicating that they are pathogenic mutations. To explore the normal role of RPS24 and consider how its dysfunction might result in DBA we performed real time RT-PCR (rt-PCR) and western blotting experiments on 20 normal human tissues and on lymphoblastoid cell lines from diseased and control individuals. Interestingly, rt-PCR of total human RPS24 and RPS19 mRNA revealed a tissue-specific variation in expression level. We found co-ordinate expression of both genes in the majority of studied tissues. Lymphoblastoid cell lines from both probands with nonsense mutations showed a reduced level of RPS24 mRNA, suggesting degradation of mutated transcripts due to nonsense mediated decay, while the RPS19 mRNA level in these patients was normal or elevated. Western blot experiments revealed a reduction of RPS24 protein in lymphoblastoid cell lines from all three mutated probands compared to control samples. Interestingly, co-ordinate expression of RPS24 and RPS19 protein was found in these patients as well as in other patients with RPS19 mutations or without any mutations, suggesting co-regulation of RP expression. To determine whether recruitment of mRNA to polysomes was impaired in DBA patients, we separated lymphoblast cell line lysates from nine diseased and four control individuals on sucrose gradients. We did not detect any significant difference in the RNA ratio of polysome-bound/free ribosomal subunits between diseased and control samples (p<0.3). It is likely that lymphoblasts, which are not defective in DBA, have sufficient ribosomes to mask an abnormality in translation. In summary, our results suggest that in addition to RPS24 and RPS19, which are mutated in ~ 27% of index cases, other DBA genes are also RP genes or genes involved in ribosome biogenesis or translation, and reinforce the notion that DBA is a ribosomal disease. This study opens new avenues for studying and understanding the pathogenesis of DBA.

Juvenile myelomonocytic leukemia (JMML) is an aggressive myeloproliferative disorder of childhood characterized by a high frequency of somatic mutations in NRAS, KRAS2, or PTPN11 genes as well as LOH of the NF1 allele (mainly in patients with type 1 neurofibromatosis). Activation of the Ras pathway is the biochemical consequence of these lesions. Hematopoietic stem cell transplantation (HSCT) is the only known curative treatment, with event-free survival (EFS) rates approaching 50–60%. Leukemia relapse still remains the main cause of treatment failure. Monitoring minimal residual disease (MRD) before and after HSCT is challenging due to the lack of tractable molecular markers. We designed and validated a fluorescently based, allele-specific polymerase chain reaction assay called TaqMAMA that can detect point mutations in eight of the most common RAS or PTPN11 mutations with a sensitivity of 1/1000 copies of the wildtype allele. We analyzed the peripheral blood (PB) and/or bone marrow (BM) in 22 patients for the levels of these mutant alleles at diagnosis and pre-HSCT. Eighteen of these patients had weekly-monthly PB and/or BM samples available after 21 HSCTs (3 children had second transplants). We compared the percentage of mutant DNA with rising levels of autologous cells measured by chimerism analysis after HSCT, which is the current surrogate marker used to detect relapse. Analysis of patients at initial diagnosis and pre-HSCT revealed a broad distribution of the frequency of the mutant alleles, regardless of the use or intensity of pre-HSCT chemotherapy. As shown in the table, levels of disease burden prior to HSCT were not predictive of relapse. After HSCT, the level of the mutant allele rose rapidly in patients who relapsed and correlated well with analysis of chimerism (r=0.86, p<0.01), thus validating the assay. In addition, the concordance rate between 15 simultaneously obtained PB and BM samples was excellent in 13 of the pairs, with a maximum of a 3.7 fold difference at < 1% of the mutant allele. In conclusion, patients with JMML harboring the most common Ras pathway mutations can now be monitored using an allele specific MRD assay. Most importantly, these assays can measure the frequency of mutant alleles in PB and provide a sensitive new strategy for MRD monitoring that can evaluate the molecular response to new therapeutic strategies for JMML.

Juvenile myelomonocytic leukemia (JMML) is a malignant hematopoietic disorder of early childhood with myeloproliferative and myelodysplastic properties. The proliferative component is a result of RAS pathway deregulation caused by somatic mutation in the RASor PTPN11oncogenes (60% of cases) or by underlying neurofibromatosis type 1 (NF-1) with a germline NF1gene defect (clinically 11% of cases). To search for potential collaborating genetic abnormalities, we used Affymetrix GeneChip Mapping 50K arrays to analyze over 116,000 single nucleotide polymorphisms (SNPs) across the genome using DNA from bone marrow or peripheral blood granulocytes from 16 children with JMML and normal karyotype (mutant RAS, n=4; mutant PTPN11, n=7; NF-1, n=5). Quantitative evaluation of hybridization intensities at each SNP locus failed to identify recurrent allelic gains or losses in the 16 cases. We were specifically interested in chromosome 7 because monosomy 7 or interstitial/terminal 7q deletions are found in about 30% of JMML cases and it is conceivable that submicroscopic 7q lesions occur in the other cases but remain undetected by standard techniques. However, at the resolution provided by the arrays used here, we saw no evidence for genomic copy number alterations on chromosome 7. Interestingly, evaluation of the SNP allelotypes identified large regions of copy-neutral loss of heterozygosity (LOH) on chromosome 17q, including the NF1locus, in 4 of the 5 samples from patients with JMML and NF-1. The LOH region spanned a genomic range of approximately 55 Mbp in each case and included more than 1,400 contiguous SNPs. Allelic copy numbers were normal within the homozygous regions, indicating uniparental isodisomy (UPD). Compatible with isodisomy, 17q was normal in the corresponding conventional karyotypes. By contrast, the array data provided no evidence for 17q UPD in any of the 12 JMML cases without NF-1. In all four cases with 17q UPD, the recombination breakpoints appeared to be confined to a 400-kbp region on 17q11.1–17q11.2; however, lack of parental or nonleukemic DNA precluded definitive mapping of the breakpoints. Of note, the 17p chromosomal arm retained heterozygosity in all cases, indicating that the p53 tumor suppressor was not affected by the UPD event. We confirmed 17q disomy in the four NF-1 samples using matrix-based comparative genomic hybridization and are currently verifying homozygosity of multiple microsatellites spaced across chromosome 17. Furthermore, NF1mutational analysis is under way to show that the individual lesion within this tumor suppressor gene is biallelic in leukemic cells from patients with JMML and NF-1. In summary, we assume that a mitotic recombination event in an early hematopoietic progenitor cell led to UPD involving the NF1locus. Our observations provide strong confirmatory evidence that it is indeed the NF1gene that is responsible for the increased incidence of JMML in NF-1 patients. In addition, our data support the emerging role of mitotic recombination as a second hit in leukemogenesis and corroborate the concept that RAS pathway deregulation is central to JMML pathogenesis.

Noonan syndrome is characterized by short stature, facial dysmorphism, and cardiac defects. We and our colleagues discovered novel de novo germline KRASmutations that introduce V14I, T58I, or D153V amino acid substitutions in individuals with NS and a P34R alteration in an individual with cardio-facio-cutaneous (CFC) syndrome, which has overlapping phenotypic features with NS. Recombinant V14I and T58I K-Ras proteins display defective intrinsic GTP hydrolysis and impaired responsiveness to the GTPase activating proteins (GAPs) p120 GAP and neurofibromin. We also found that V14I and T58I K-Ras render primary hematopoietic progenitors hypersensitive to growth factors and deregulate signal transduction in a cell lineage specific manner (Nature Genetics 38, 331, 2006). We recently began interrogating the P34R and D153V K-Ras mutant proteins and a novel CFC-associated F156L K-Ras mutant protein. Both P34R and D153V K-Ras display normal levels of intrinsic GTP hydrolysis. In contrast, F156L K-Ras displays defective intrinsic GTP hydrolysis that resembles oncogenic G12D K-Ras. P34R K-Ras is completely resistant to both GAPs, which is intriguing as these data suggest that the ability of GAPs to down-regulate Ras-GTP levels is dispensible for development. In contrast, D153V K-Ras is responsive to both GAPs. We expressed mutant Ras proteins in COS-7 monkey kidney cells to investigate activation of Ras and downstream effectors. Cells expressing P34R, D153V, F156L, and G12D K-Ras demonstrate elevated levels of Ras-GTP and phospho-MEK in basal and serum-starved conditions. We conclude that germline KRASmutations that cause human disease encode proteins with distinct mechanisms of biochemical activation. Further investigation of this allele series will provide insights about the relative importance of the intrinsic Ras GTPase, GAPs, and other biochemical mechanisms in controlling the growth of different cell lineages.

Diamond-Blackfan anemia (DBA) is a congenital red cell aplasia with marked clinical heterogeneity, an increased risk of malignancy and mutations in ribosomal protein (RP) S19 in 25% of probands. To identify other gene(s) mutated in DBA and investigate their expression and function, we performed a genome-wide screen using a 10,000 single nucleotide polymorphism mapping set (Affymetrix) on a large family comprising 10 informative meioses. We found linkage of the DBA phenotype to regions on chromosome 8q, 10 and 6. The RP gene RPS24, is located in the linked region on chromosome 10; we sequenced exons, intron-exon boundaries and the promoter regions in this family, and found a nonsense mutation (316C>T) in exon 4 of RPS24in five affected individuals, and a wild type sequence in five unaffected family members. This mutation causes the change of Gln106STOP and is predicted to result in formation of a truncated RPS24 protein. Subsequently, we sequenced DNA from 215 unrelated DBA probands, 30 with RPS19mutations and 185 without. We found another nonsense mutation in exon 2 in a sporadic case, and a splice site deletion resulting in skipped exon 2 in another proband and in his father; over 200 control individuals did not have any of the above sequence changes, indicating that they are pathogenic mutations. To explore the normal role of RPS24and consider how its dysfunction might result in DBA we performed real time RT-PCR (rt-PCR) and western blotting experiments on 20 normal human tissues and on lymphoblastoid cell lines from diseased and control individuals. Interestingly, rt-PCR of total human RPS24 and RPS19 mRNA revealed a tissue-specific variation in expression level. We found co-ordinate expression of both genes in the majority of studied tissues. Lymphoblastoid cell lines from both probands with nonsense mutations showed a reduced level of RPS24 mRNA, suggesting degradation of mutated transcripts due to nonsense mediated decay, while the RPS19 mRNA level in these patients was normal or elevated. Western blot experiments revealed a reduction of RPS24 protein in lymphoblastoid cell lines from all three mutated probands compared to control samples. Interestingly, co-ordinate expression of RPS24 and RPS19 protein was found in these patients as well as in other patients with RPS19mutations or without any mutations, suggesting co-regulation of RP expression. To determine whether recruitment of mRNA to polysomes was impaired in DBA patients, we separated lymphoblast cell line lysates from nine diseased and four control individuals on sucrose gradients. We did not detect any significant difference in the RNA ratio of polysome-bound/free ribosomal subunits between diseased and control samples (p<0.3). It is likely that lymphoblasts, which are not defective in DBA, have sufficient ribosomes to mask an abnormality in translation.

Mutations in ribosomal protein S19 (RPS19) gene have been found in 25% of patients affected with Diamond-Blackfan anemia (DBA), a congenital erythroblastopenia. We have previously shown that several RPS19 mutant proteins (V15F, InsAG36, W33stop, Y48stop, R56stop, M75stop, R94stop, 274del31, InsG238, G127Q and L131P) exhibit decreased levels of protein expression and do not localize to the nucleolus like the wild type protein in transfected Cos-7 cells. In contrast, other mutants (W52C, T55M, R56Q, R62W, 24Del18, G120S) exhibit normal levels of protein expression and normal nucleolar localization. We hypothesized that decreased levels of expression of mutant proteins such as V15F, G127Q, and L131P may be due to proteosomal degradation. In order to validate our hypothesis, we analyzed the effects of two proteasome inhibitors (MG132 and lactacystin) on mutant RPS19 protein expression levels and their subcellular localization. Following treatment with proteosome inhibitors, the mutant proteins with missense mutations (V15F, G127Q and L131P) were expressed at levels similar to that of wild type protein and localized in the nucleolus. Similarly, proteasome inhibitors also restored the expression levels and normal subcellular localization to RPS19 with non-sense mutations (InsG238, R94stop, and 274del31) that resulted in the translation of RPS19 protein with at least 80 aminoacids. In marked contrast, proteosome inhibitors failed to restore the expression levels of RPS19 with the non-sense mutants that led to synthesis of shortened proteins (InsAG36, W33stop, Y48stop, R56stop, M75stop). Even in the presence of proteosome inhibitors we noted a dramatic decrease in the levels of expression of these mutant proteins and proteins expressed were localized in the cytoplasm. Our findings imply an important role for proteosomal degradation pathway in regulating the expression levels of RPS19. They further suggest that proteasome inhibitors could be considered as a potential treatment for some steroid resistant DBA affected individuals with specific RPS19 mutations.

Mutations in ribosomal protein S19 (RPS19) gene have been found in 25% of patients affected with Diamond-Blackfan anemia (DBA), a congenital erythroblastopenia. We have previously shown that several RPS19 mutant proteins (V15F, InsAG36, W33stop, Y48stop, R56stop, M75stop, R94stop, 274del31, InsG238, G127Q and L131P) exhibit decreased levels of protein expression and do not localize to the nucleolus like the wild type protein in transfected Cos-7 cells. In contrast, other mutants (W52C, T55M, R56Q, R62W, 24Del18, G120S) exhibit normal levels of protein expression and normal nucleolar localization. We hypothesized that decreased levels of expression of mutant proteins such as V15F, G127Q, and L131P may be due to proteosomal degradation. In order to validate our hypothesis, we analyzed the effects of two proteasome inhibitors (MG132 and lactacystin) on mutant RPS19 protein expression levels and their subcellular localization. Following treatment with proteosome inhibitors, the mutant proteins with missense mutations (V15F, G127Q and L131P) were expressed at levels similar to that of wild type protein and localized in the nucleolus. Similarly, proteasome inhibitors also restored the expression levels and normal subcellular localization to RPS19 with non-sense mutations (InsG238, R94stop, and 274del31) that resulted in the translation of RPS19 protein with at least 80 aminoacids. In marked contrast, proteosome inhibitors failed to restore the expression levels of RPS19 with the non-sense mutants that led to synthesis of shortened proteins (InsAG36, W33stop, Y48stop, R56stop, M75stop). Even in the presence of proteosome inhibitors we noted a dramatic decrease in the levels of expression of these mutant proteins and proteins expressed were localized in the cytoplasm. Our findings imply an important role for proteosomal degradation pathway in regulating the expression levels of RPS19. They further suggest that proteasome inhibitors could be considered as a potential treatment for some steroid resistant DBA affected individuals with specific RPS19 mutations.

MDS in children is a rare disorder characterized by dysplasia and defined genetic abnormalities. In most patients (pts) MDS arises without known predisposing conditions (primary MDS). Here, we report the results of 55 males and 30 females with advanced primary MDS enrolled in the prospective EWOG-MDS trial 97. Data were analysed according to the most advanced FAB-type prior to SCT: 32 pts were classified as RAEB, 40 as RAEB-t and 13 as myelodysplasia-related AML (MDR-AML). Median age at diagnosis was 9.5 yrs (0.1–17.6) and median time from diagnosis of advanced MDS to SCT 4 mo (0.5–31). Cytogenetics revealed monosomy 7 in 32 pts, trisomy 8 in 7, a complex karyotype in 9 and other abnormalities in 9; karyotype was normal in 26 pts and unknown in 2. 31 pts had received AML-like therapy prior to SCT. All pts were given an unmanipulated graft after condititioning with busulfan 16 mg/kg, cyclophosphamide 120 mg/kg and melphalan 140 mg/m. Source of stem cells was bone marrow in 56 pts, peripheral blood in 25, cord blood in 2 and unknown in 2. 36 pts were transplanted from an HLA-identical relative (MFD), 49 pts from an HLA-identical or 1-antigen disparate unrelated donor (UD). GVHD prophylaxis consisted of CSA alone for MFD, whereas recipients of a UD graft generally received CSA, methotrexate and anti-lymphocyte globulin. Two pts suffered graft failure. The cumulative incidence of grade II-IV acute GVHD and chronic GVHD was 40% (SE 5%) and 25% (SE 5%), respectively. 18 pts suffered transplant-related mortality (TRM), the cumulative incidence of TRM in pts grafted from a MFD or UD being 14 and 25%, respectively (p=n.s.). Presence of acute GVHD II-IV (p<0.01), spleen size at SCT ≥ 1 cm below the costal margin (p=0.03) and age ≥ 12 years (p=0.04) predicted an increased risk of TRM. 17 patients relapsed at a median of 12 mo after SCT (1–107). The 5-year probability of leukemia recurrence was 29%, with no difference between MFD and UD transplants. While the highest FAB type prior to SCT predicted relapse with a cumulative incidence rate increasing from RAEB (13%, SE 10%) to RAEB-t (24%, SE 11%) and MDR-AML (69%, SE 20%) (p=0.03), the use of intensive chemotherapy prior to SCT or blast percentage at SCT did not. With a median observation time after SCT of 29 months (6–107) the EFS at 5-years was 62% (SE 12%) and 39% (SE 11%) for pts given SCT from a MFD or an UD, respectively (p=n.s). These results indicate that a large proportion of pts with advanced MDS can be rescued by SCT. Disease recurrence remains the main cause of treatment failure. Intensive chemotherapy prior to SCT should not be routinely employed.

Neonatal thrombocytopenia occurs in about 1% of all newborns. Inherited forms like 11q- or Jacobsen syndrome are rare. However, they may remain undetected with karyotyping because the deleted regions in 11q often involve small subtelomeric regions. Here we report on the detection of deletions in 11q in two newborns with normal routine karyotypes who were shown to carry subtelomeric deletions in 11q by means of fluorescence in situhybridization (FISH) using a subtelomeric 11q probe (Abbott, Diagnostics, Wiesbaden, Germany). Both children showed thrombocytopenia (18.000/μl and 26.000/μl, respectively) and dysmegakaryopoiesis (absence of normal megakaryocytes and presence of micromegakaryocytes) associated with facial dysmorphism, cardiac defects and psychomotoric retardation. In the second case, the mother and the grandmother also showed mild thrombocytopenia. In both patients, FISH analyses on peripheral blood and bone marrow showed the loss of the telomere-associated region of 11q distal of the MLL gene. In the first patient, the deletion of 11q resulted from an unbalanced complex rearrangement with duplication of 11p. As the source of this chromosomal aberration, a paternal pericentric inversion of chromosome 11 was identified. The partial monosomy 11q and the partial trisomy 11p in the first patient were confirmed by comparative genomic hybridization (CGH) analysis. Array/matrix CGH assisted in determining the breakpoints at 11p15.1 and 11q24.1. No structural aberrations of 11q were found in the mother of the second patient, but further investigations are under way. These findings give further evidence that small subtelomeric deletions of 11q and probably mutations of genes located therein cause thrombocytopenia. Since it can be very difficult to detect these deletions by karyotyping, FISH using a subtelomeric 11q probe seems to be an extremely useful new diagnostic tool. This new method should be applied in children with congenital thrombocytopenia, in particular if they have additional complex dysmorphic features.

Neonatal thrombocytopenia occurs in about 1% of all newborns. Inherited forms like 11q- or Jacobsen syndrome are rare. However, they may remain undetected with karyotyping because the deleted regions in 11q often involve small subtelomeric regions. Here we report on the detection of deletions in 11q in two newborns with normal routine karyotypes who were shown to carry subtelomeric deletions in 11q by means of fluorescence in situ hybridization (FISH) using a subtelomeric 11q probe (Abbott, Diagnostics, Wiesbaden, Germany). Both children showed thrombocytopenia (18.000/μl and 26.000/μl, respectively) and dysmegakaryopoiesis (absence of normal megakaryocytes and presence of micromegakaryocytes) associated with facial dysmorphism, cardiac defects and psychomotoric retardation. In the second case, the mother and the grandmother also showed mild thrombocytopenia. In both patients, FISH analyses on peripheral blood and bone marrow showed the loss of the telomere-associated region of 11q distal of the MLL gene. In the first patient, the deletion of 11q resulted from an unbalanced complex rearrangement with duplication of 11p. As the source of this chromosomal aberration, a paternal pericentric inversion of chromosome 11 was identified. The partial monosomy 11q and the partial trisomy 11p in the first patient were confirmed by comparative genomic hybridization (CGH) analysis. Array/matrix CGH assisted in determining the breakpoints at 11p15.1 and 11q24.1. No structural aberrations of 11q were found in the mother of the second patient, but further investigations are under way. These findings give further evidence that small subtelomeric deletions of 11q and probably mutations of genes located therein cause thrombocytopenia. Since it can be very difficult to detect these deletions by karyotyping, FISH using a subtelomeric 11q probe seems to be an extremely useful new diagnostic tool. This new method should be applied in children with congenital thrombocytopenia, in particular if they have additional complex dysmorphic features.

MDS in children is a rare disorder characterized by dysplasia and defined genetic abnormalities. In most patients (pts) MDS arises without known predisposing conditions (primary MDS). Here, we report the results of 55 males and 30 females with advanced primary MDS enrolled in the prospective EWOG-MDS trial 97. Data were analysed according to the most advanced FAB-type prior to SCT: 32 pts were classified as RAEB, 40 as RAEB-t and 13 as myelodysplasia-related AML (MDR-AML). Median age at diagnosis was 9.5 yrs (0.1–17.6) and median time from diagnosis of advanced MDS to SCT 4 mo (0.5–31). Cytogenetics revealed monosomy 7 in 32 pts, trisomy 8 in 7, a complex karyotype in 9 and other abnormalities in 9; karyotype was normal in 26 pts and unknown in 2. 31 pts had received AML-like therapy prior to SCT. All pts were given an unmanipulated graft after condititioning with busulfan 16 mg/kg, cyclophosphamide 120 mg/kg and melphalan 140 mg/m. Source of stem cells was bone marrow in 56 pts, peripheral blood in 25, cord blood in 2 and unknown in 2. 36 pts were transplanted from an HLA-identical relative (MFD), 49 pts from an HLA-identical or 1-antigen disparate unrelated donor (UD). GVHD prophylaxis consisted of CSA alone for MFD, whereas recipients of a UD graft generally received CSA, methotrexate and anti-lymphocyte globulin. Two pts suffered graft failure. The cumulative incidence of grade II-IV acute GVHD and chronic GVHD was 40% (SE 5%) and 25% (SE 5%), respectively. 18 pts suffered transplant-related mortality (TRM), the cumulative incidence of TRM in pts grafted from a MFD or UD being 14 and 25%, respectively (p=n.s.). Presence of acute GVHD II-IV (p<0.01), spleen size at SCT ≥ 1 cm below the costal margin (p=0.03) and age ≥ 12 years (p=0.04) predicted an increased risk of TRM. 17 patients relapsed at a median of 12 mo after SCT (1–107). The 5-year probability of leukemia recurrence was 29%, with no difference between MFD and UD transplants. While the highest FAB type prior to SCT predicted relapse with a cumulative incidence rate increasing from RAEB (13%, SE 10%) to RAEB-t (24%, SE 11%) and MDR-AML (69%, SE 20%) (p=0.03), the use of intensive chemotherapy prior to SCT or blast percentage at SCT did not. With a median observation time after SCT of 29 months (6–107) the EFS at 5-years was 62% (SE 12%) and 39% (SE 11%) for pts given SCT from a MFD or an UD, respectively (p=n.s). These results indicate that a large proportion of pts with advanced MDS can be rescued by SCT. Disease recurrence remains the main cause of treatment failure. Intensive chemotherapy prior to SCT should not be routinely employed.