教育講演 / Educational Lecture
The biology and treatment of myelodysplastic syndromes
演題番号 : EL-23
1:Department of Myelodysplastic, Syndrome Center, Columbia University , USA
Myelodysplastic syndromes (MDS) are diseases of the hematopoietic stem cells which predominate in the elderly presenting with variable cytopenias and a dysplastic, generally hypercellular bone marrow. Clinicians have realized for several decades that MDS are a highly heterogeneous group of stem cell bone marrow failure diseases that share some biological characteristics with each other but differ widely in others. Recent advances in next-generation sequencing technologies have now provided a clearer picture of the genetic changes that occur during the initiation and evolution of this disease. Historically, morphology and clinical parameters have allowed the disease to be grouped into multiple subsets. More recently, biologic insights and genetics are clarifying the pathology and demonstrating that the disease is more or less a continuum of multiple mutations that accumulate slowly in a stem cell, giving it a growth advantage that result in eventual clonal dominance. This phenomenon is not specific to MDS, but is a characteristic of all neoplasms1～4). Mutations that confer a selective growth advantage to the cell are the“driver”mutations while mutations that occur randomly during repeated cell proliferation without contributing to tumor evolution are the“passenger”mutations. Thus, by the time that clinical diagnosis is made, there exists a tumor population composed of multiple sub-clones, all of which contain the founding driver mutation(s), but differ in the spectrum of passenger mutations. Mutations can be point mutations, chromosomal deletions, insertions and translocations, as well as epigenetic alterations that result in increased or decreased gene expression. In MDS, the initiating driver mutation has yet to be characterized and indeed there may be a variety of mutations that could initiate the neoplastic process. While specific recurrent chromosomal abnormalities (single and in combinations) have been found to be associated with prognosis with the advent of genome sequencing it is now possible to identify recurrent mutations in individual genes that provide insights into the pathological heterogeneity seen among patients even when they are classified as belonging to the same sub-group. The role of some of these gene mutations in defining the biology of MDS will be discussed below.
The typical MDS patient will present with refractory anemia that is often detected by chance during a routine annual examination. The patient is elderly with a median age of 72 years and may have additional cytopenias. Approximately one-third of these patients will eventually transform to AML, but all others will die as a result of increasing profundity of cytopenias. The only potential cure is an allogeneic stem cell transplant. According to the International Prognostic Scoring System or IPSS, patients who present with a higher than normal blast count in the bone marrow, multiple cytopenias and cytogenetic abnormalities associated with a poor prognosis have a higher risk of transformation than those without. Thus patients are essentially divided into“lower-risk”or“higher-risk”categories and therapeutic decisions are guided by this classification since the goal is to alleviate cytopenias in the former group and arrest the increasing number of blasts in the latter. Unique among hematological malignancies however, is the uncoupling of proliferation from differentiation and apoptosis in lower risk MDS. In these patients the rapidly proliferating clonal cells are able to differentiate along the myeloid lineage; however, they are simultaneously undergoing apoptosis at various stages of maturation in the marrow resulting in variable cytopenias in the peripheral blood5～9). The role of the microenvironment in apoptosis of the MDS marrow has long been appreciated. Here, the marrow stromal cells secrete pro-inflammatory cytokines (extrinsic apoptotic pathway) which in turn activate the intrinsic apoptotic pathway of the hematopoietic cells. Both pathways result in the activation of caspases that trigger cell death by cleavage of specific protein substrates. In higher-risk patients, a subclone of the primary MDS clone is no longer undergoing excessive apoptosis, has lost the ability to differentiate; consequently, immature blasts begin to accumulate in the marrow8, 9). The patient is said to have AML when the blast count reaches 20％, a rather arbitrary boundary.
Genetic aberrations of lower risk MDS; RPS14 haploinsufficiency and TP53 stabilization
Ribosome protein (RP) encoding genes are essential for the biogenesis and function of the ribosomal machinery. The highly coordinated process begins in the nucleolus where a 47S polycistronic RNA molecule is transcribed and cleaved into smaller molecules; together with multiple protein components, 2 ribosome sub-units are generated with final completion into the mature ribosome occurring in the cytoplasm. The maturation process will ultimately involve interaction of more than 80 ribosomal proteins with the RNA molecules. Recurrent heterozygous mutations in ribosomal proteins leading to aberrant ribosomal biogenesis have been documented in several congenital anemia diseases. It was surprising however when it was found that the erythroid deficiency of MDS patients with 5q－ syndrome was due to the haploinsufficiency of RPS14 on chromosome 510). Additional studies in lower risk MDS detected small deletions in a number of ribosomal proteins in CD34＋ cells, suggesting that disruption of ribosome biogenesis may be a characteristic feature of the disease11). However, it is hard to postulate that such mutations confer a growth advantage to the cell and therefore to clonal expansion. Nevertheless, anemia and increased apoptosis of the MDS cell can be hypothesized to result from increase in free heme which is toxic to the cells. A decrease in mature ribosome formation in rapidly proliferating erythroid progenitor cells would result in the accumulation of free heme to toxic levels.
Recently, the phenomenon of“ribosomal stress”due to aberrant ribosomal biogenesis has been found to result in major disruption of p53 signaling in a murine model of 5q－ syndrome12). Mice carrying a deletion in the 5q－ syntenic portion of chromosome 18 developed dysplastic macrocytic anemia with high expression of p53 and increased apoptosis in the bone marrow. This phenotype was rescued by cross-breeding with p53 deficient mice. Accumulation of RPs due to the ineffective ribosome biogenesis was found to result in binding and sequestering of MDM2, an E3 ligase required for proteasome degradation of p53. Increased p53 results in cell cycle arrest and subsequent apoptosis and thus may also contribute to the biology seen in lower risk MDS.
The role of miRNA in MDS
Micro RNAs are short 22 base pair strands of non-protein coding RNA that control translation in a complex and highly nuanced manner. Targeting mRNA through short complementary sequences results in degradation of mRNA and translation inhibition13). The presence of complementary sequence in multiple coding genes to a single miRNA together with the estimate that several thousand miRNA are present in the genome greatly increases the complexity of gene expression. An addition layer of control relates to the finding that miRNAs can also alter epigenetic signals at gene promoter sites 14). The role of miRNA in the evolution of MDS is currently under intense investigation. While studies have tried to identify miRNAs that differentiate between low risk and high risk disease, the results have not been consistent, due in part to the heterogeneity of the patient populations. The most convincing demonstration of the importance of miRNA to MDS pathology has emerged from studies in murine models15, 16). Surprisingly, when the gene for Dicer I, an enzyme essential for generation of mature miRNA from its precursor molecule, was targeted to stromal osteo-progenitors in adult mice, a percentage of the animals developed hematopoietic myelodysplasia that slowly transformed to AML. Thus, a defect in generation of miRNA in a stromal cell generated a disease in hematopoietic cells that was very similar to human MDS pathology. However, these studies further showed that the expression of the Sbds gene (named for the congenital mutation in patients with Schwachman-Bodian-Diamond disease which causes bone marrow failure) was reduced. This gene disrupts the processing of the RP proteins required for generation of the 60S subunit. The exact mechanism linking Dicer to Sbds and ribogenesis is not, as yet, clear. A second murine study examined the effect of reduced expression of 2 miRNAs that map to the deleted region of chromosome 5 in patients with 5q－ syndrome. When mir-145 and mir-146a expression was inhibited in hematopoietic stem cells, some animals developed a 5q－ syndrome-like disease and ultimately died of AML. Thus miRNA activity appears to be intimately linked to the evolution of MDS but the mechanism remains to be more fully elucidated.
Gene mutations in MDS: Cause and/or Effect?
It is now recognized that mutations are most likely accumulating in a“pre-overt disease cell”but that only those mutations that confer some degree of growth advantage to the clone will be the drivers of the disease phenotype. The heterogeneity found in MDS has long been thought to be due to the presence of variable genetic lesions. Recurrent abnormalities have been found to associate with risk to transformation, but to date there is no evidence which can pinpoint the sequence of mutational accumulation. Genetic lesions can include those which modify chromosome structure (deletions, amplifications, polyploidy), protein structure (point mutations, small deletions or amplifications), and increase or decrease gene expression (mutations in promoter, enhancer, and repressor sequences, activating and inhibitory mutations and mutations in epigenetic modifiers). While recurrent mutations in known oncogenes and suppressor genes have been documented, none are common to all subtypes and none are specific to MDS.
One of the most common mutations in MDS is found in the epigenetic modifying gene TET217). The gene product converts methylated cytosine to hydroxymethylacytosine and results in cytosine demethylation in the 5’ promoter region. While the mutations are found in 20％ of MDS it is also found in CMML, secondary AML and myeloproliferative neoplasms. The contribution of the mutated gene to the etiology of MDS is not clear and there are conflicting studies as to whether there is any relationship to disease prognosis. Other epigenetic modifying genes of the promoter CpG dinucleotides have also been found in5─10％ of MDS patients18～21). DNMT3A adds a methyl group to cytosine and mutations which tended to occur in amino acid R882 of DNMT3 were originally identified in AML18). The same mutations in MDS are associated with higher risk disease. It is not clear whether the mutations are activating or inhibiting gene expression in vivo and thus the mechanism of action is not clear. IDH1/2 are enzymes that convert isocitrate to α-ketoglutarate. Mutations in these genes cause a further chemical conversion of α-ketoglutarate, thus reducing the levels of this essential co-factor of multiple enzymes including TET2; ultimately the genomic DNA undergoes hypermethylation. Other epigenetic modifying genes, polycomb family gene members ASXL1 and EZH2, have also been found to be mutated in 5─20％ of MDS cases and have been shown to be linked to patients with poor prognosis22, 23).
Mutations in known oncogenes are found in patients with MDS and in general are associated with higher risk disease and poor prognosis24). RUNX1, ETV6 and EVI1 are transcription factors important in normal hematopoiesis and mutations have been identified in varying numbers of MDS patients (2─15％). Mutations in TP53, the tumor suppressor gene, have been identified in 15％ of patients with MDS and are associated with higher risk disease, complex karyotype and resistance to therapy25). TP53 mutations are thought to increase genomic instability in these patients leading to additional mutations and inhibition of differentiation. The role of genomic instability in the evolution of MDS has long been hypothesized since recurrent karyotypic abnormalities were found to contribute to prognosis. The more recent findings that the DNA of MDS patients with normal karyotypes often harbor small copy number alterations (hence increased genomic instability) and that these patients tended to have a poorer prognosis, emphasizes the biologic contribution and clinical importance of genomic instability to MDS evolution26, 27).
The recent reports that heterozygous mutations are present in genes that comprise the essential components of the RNA-splicing machinery (the spliceosome) in a large majority of MDS patients have been unexpected and exciting28～30). Mutations in four genes (SRSF2, SF3B1, U2AF1 and ZRSR2) in particular have been shown to be mutually exclusive in MDS, with 65％ of patients with RARS or RCMD-RS presenting with a mutation in SF3B1 and clustered in exons 12─15. Interestingly, these mutations do not affect structural amino acids and the mechanism behind the association with ringed sideroblasts is unclear. Mutations in U2AF1 and SRSF2 are more frequently found in higher risk MDS and CMML. It is somewhat difficult to explain how mutations in an essential and general genetic function can result in specific disease phenotype. Identification of target genes with aberrant splicing is currently under intensive investigation. The prognostic implication of spliceosome mutations is also not clear at present. Mutations in SF3B1 and U2AF1 have been shown to have minimal influence on survival, while SRSF2 mutations were predictive of poorer survival and increased transformation to AML. Thus, both the clinical significance of spliceosome mutations and the contribution to MDS evolution is, at present, unclear.
As already mentioned, treatment decisions are guided according to the IPSS classification and the goals of therapy are different for lower versus higher risk disease. It is important to point out right at the start that the IPSS system being currently used universally as the gold standard for risk stratification only reliably predicts the prognosis in ─39％ of the cases, thereby underlining the urgent need for developing more accurate predictors. Nothing is more accurately predictive than observing the biology of the disease in the patients themselves, so our policy with both lower and higher risk patients is to follow the natural history of the disease upon initial diagnosis for as long as possible before instituting therapy for signs of disease progression. By practicing this rather conservative approach and not intervening immediately, we get an idea of the natural history of the individual patient’s disease and whether they are remaining stable or progressing rapidly. Surprisingly, some patients whose bone marrow at diagnosis show a high grade MDS despite only manifesting borderline cytopenias have been followed for months (and occasionally years) without requiring any treatment. There is no reason to suspect that delaying treatment thus affects eventual outcome of treatment adversely, an observation which has gained further support from the recent publication that delay even in cases of acute leukemia is acceptable31). These cases, albeit the exception rather than the rule, once again underscore both the biologic heterogeneity of the disease as well as our inability to accurately predict its natural history based on the prevalent prognostic classification. Whether the introduction of the revised IPSS or future incorporation of molecular markers into the classification will refine the predictive power remains to be tested. The following strategies are currently available to improve cytopenias and arrest leukemic transformation:
Stem cell transplant
The only potentially curative treatment for MDS is a stem cell transplant. Unfortunately, given that the median age of the patients is 70＋ years, it is not an option available for many. However, once a candidate is identified as being able to tolerate high dose chemotherapy and has an available donor, the timing of the transplant becomes an important question to be decided. A study reported by Cutler et al has shown that long-term survival is better for highest-risk disease patients if they are transplanted at diagnosis but should be delayed to a later time for patients, particularly younger ones, in the lower-risk groups32). Such distinctions are particularly important as clinical and biologic parameters are being developed to select curative or palliative therapies for patients based on the natural history of their disease.
This is a perfectly reasonable option, especially for lower risk disease patients as well as elderly individuals with higher risk disease and includes the use of Erythroid Stimulating Agents (ESAs), G-CSF and transfusions. Two equivalent recombinant erythropoietin drugs have been developed for this purpose and both have been used with or without granulocyte colony stimulating factor (G-CSF), to increase hemoglobin levels in a sub-set of MDS patients. The patients likely to respond to ESA are those whose baseline serum erythropoietin level is＜500 and who are receiving less than 4 units of blood a month. This comprises approximately 20％ of MDS cases and should be the first line of therapy for anemic subjects. Median overall survival has been found to be significantly longer in lower risk patients who responded to ESAs33) and had at least a six-month duration of response (54.3 months), compared to patients who never responded or relapsed in less than six months (36.7 months). The conclusion here is that the latter group of patients should be offered alternate therapies sooner. In addition, iron chelation is recommended for individuals who have received at least 20 units of red cell transfusions and whose ferritin level is ＞2,500. Rose et al 34) have demonstrated that chelation improve survival in lower risk MDS, but the mechanisms of this potential survival benefit remains unknown. In their study of 97 lower risk patients chelated for at least 6 months mainly with deferoxamine, median overall survival was 53 months and 124 months in non-chelated and in chelated patients (p＜0.0003). In multivariate Cox analysis, adequate chelation was the strongest independent factor associated with better overall survival.
Disease modifying agents
These include both FDA approved drugs as well as experimental agents. The reason to include experimental trials in this discussion is related to the fact that for the majority of patients belonging to all IPSS groups, there is no satisfactory treatment available. For example, in the lower risk group, the del(5q) patients do well with lenalidomide at least as far as quality of life is concerned while in the higher risk group, 5-azacytidine has been shown to improve survival, but for the rest of the patients, experimental trials are the only possibility offering respite from frequent transfusions. These two strategies are discussed below:
1．Lenalidomide: This thalidomide analog was approved by the FDA in 2004 for use in lower risk MDS patients with del(5q) and transfusion dependent anemia based on a study of 148 patients where complete transfusion independence (TI) was seen in 67％ cases and of 85 patients who could be evaluated, 62 had cytogenetic improvement, and 38 of the 62 had a complete cytogenetic remission35). At the same time, Raza et al showed that among 214 similar patients without del(5q), 26％ achieved TI after a median of 4.8 weeks of treatment with lenalidomide36). A 50％ or greater reduction in transfusion requirement occurred in 37 additional patients, yielding a 43％ overall rate of hematologic improvement. Finally, Ades et al37) showed that in 47 patients with del(5q) and higher-risk MDS 27％ achieved hematologic response, including 7 complete remissions (CR) when treated with lenalidomide. Median CR duration was 11.5 months. The conclusion from these three studies is that lenalidomide has clinically meaningful activity in transfusion-dependent patients with MDS but it is currently FDA approved only for patients with lower risk disease and transfusion dependent anemia who have deletion 5q karyotypic abnormality. The major side effects of the drug relate to myelosuppression.
2．Hypomethylating Agents: Both 5-azacytidine (5-aza) and decitabine are FDA approved for MDS, but survival benefit has only been demonstrated for Int-2 or high grade MDS treated with 5-aza. The initial Phase III CALGB study reported by Silverman et al38) compared 5-aza to best supportive care in 191 MDS patients and found that responses were seen in 60％ patients treated with the drug compared with 5％ on the supportive care arm. Median time to transformation was also prolonged for the 5-aza arm (21 versus 13 months). Fenaux et al39)) then showed an approximate doubling of survival in 358 higher risk MDS patients treated with 5-aza compared to one of three conventional care regimens (best supportive care, low dose araC or high dose chemotherapy) extending and confirming the CALGB findings. Finally, Lubbert et al showed in 233 higher risk MDS patients that decitabine produces an overall response rate of ─34％ when compared with supportive care40). Overall survival in this study was not significantly improved, perhaps because treatment was not continued to disease progression.
3．Combination Therapy: Among 36 patients with Int-1, Int-2 and high risk MDS treated with a combination of lenalidomide and 5-aza, Sekeres et al41) reported that the combination was well tolerated with an overall response rate of approximately 72％ with 44％ achieving a complete response and 28％ showing hematologic improvement. Median overall survival in this study was 37＋ months (range, 7─55＋) for CR patients, and 13.6 months for the entire cohort (range: 3─55). Soriano et al42) combined 5-aza with valproic acid (VPA), a histone deacetylase inhibitor and all-transretinoic acid (ATRA) in 53 higher risk MDS and AML patients. Overall response rate was 42％ with a higher rate (52％) in previously untreated older patients and a median remission duration of 26 weeks.
Conclusions and summary
MDS are a unique set of malignant bone marrow stem cell diseases that are marked by monoclonal expansion and excessive apoptosis of hematopoietic cells leading to a cellular marrow paradoxically coupled with cytopenias. The initiating event leading to the clonal expansion remains unknown but recent studies suggest that a key contribution to the pathology comes from the microenvironment. Yet another distinctive feature of MDS is that when hypomethylating agents were used in a variety of cancers, the best responses were seen in MDS signifying a critical role for epigenetics. Finally, at the molecular level, defective ribogenesis and mutations in genes coding for RNA splicing proteins are more frequently associated with MDS than with other malignancies. As far as natural history is concerned, because of the disease heterogeneity, there are patients with lower risk MDS who can live for years with supportive care alone while some with higher risk disease can have a pernicious course with very short survival. This leads to strategies of therapy ranging from low intensity approaches to AML-like chemotherapies with stem cell transplant as being the only potential cure. The hope for the future is that an improved molecular understanding will lead to a more accurate prognostic classification as well as identify potential targets for drug development.
1）Raza A, Galili N. The genetic basis of phenotypic heterogeneity in myelodysplastic syndromes. Nat Rev Cancer. 2012; 12: 849-859.
2）Vogelstein B, Papadopoulos N, Velculescu VE, Zhou S, Diaz LA Jr, Kinzler KW. Cancer genome landscapes. Science. 2013; 339: 1546-1558.
3）Walter MJ, Shen D, Ding L, et al. Clonal architecture of secondary acute myeloid leukemia. N Engl J Med. 2012; 366: 1090-1098.
4）Walter MJ, Shen D, Shao J, et.al. Clonal diversity of recurrently mutated genes in myelodysplastic syndromes. Leukemia. Prepublished on February 27, 2013, as DOI 10.1038/leu.2013.58.
5）Anastasi J, Feng J, Le Beau MM, Larson RA, Rowley JD, Vardiman JW. Cytogenetic clonality in myelodysplastic syndromes studied with fluorescence in situ hybridization: lineage, response to growth factor therapy, and clone expansion. Blood. 1993; 81: 1580-1585.
6）Raza A, Gezer S, Mundle S, et al. Apoptosis in bone marrow biopsy samples involving stromal and hematopoietic cells in 50 patients with myelodysplastic syndromes. Blood. 1995; 86: 268-276.
7）Raza A, Mundle S, Shetty V, et al. Novel insights into the biology of myelodysplastic syndromes: excessive apoptosis and the role of cytokines. Int J Hematol. 1996; 63: 265-278.
8）Parker JE, Fishlock KL, Mijovic A, Czepulkowski B, Pagliuca A, Mufti GJ. ‘Low-risk’ myelodysplastic syndrome is associated with excessive apoptosis and an increased ratio of pro- versus anti-apoptotic bcl-2-related proteins. Br J Haematol. 1998; 103: 1075-1082.
9）Parker JE, Mufti GJ, Rasool F, Mijovic A, Devereux S, Pagliuca A. The role of apoptosis, proliferation, and the Bcl-2-related proteins in the myelodysplastic syndromes and acute myeloid leukemia secondary to MDS. Blood. 2000; 96: 3932-3938.
10）Ebert BL, Pretz J, Bosco J, et al. Identification of RPS14 as a 5q－ syndrome gene by RNA interference screen. Nature. 2008; 451: 335-339.
11）Pellagatti A, Hellström-Lindberg E, Giagounidis A, et al. Haploinsufficiency of RPS14 in 5q－ syndrome is associated with deregulation of ribosomal- and translation-related genes. Br J Haematol. 2008; 142: 57-64.
12）Barlow JL, Drynan LF, Hewett DR, et al. A p53-dependent mechanism underlies macrocytic anemia in a mouse model of human 5q－ syndrome. Nat Med. 2010; 16: 59-66.
13）Bartel DP. MicroRNAs: target recognition and regulatory functions. Cell. 2009; 136: 215-233.
14）Khraiwesh B, Arif MA, Seumel GI, et al. Transcriptional control of gene expression by microRNAs. Cell. 2012; 140: 111-122.
15）Raaijmakers MH, Mukherjee S, Guo S, et al. Bone progenitor dysfunction induces myelodysplasia and secondary leukaemia. Nature. 2010; 464: 852-857.
16）Starczynowski DT, Kuchenbauer F, Argiropoulos B, et al. Identification of miR-145 and miR-146a as mediators of the 5q－ syndrome phenotype. Nat Med. 2010; 16: 49-58.
17）Ko M, Huang Y, Jankowska AM, et al. Impaired hydroxylation of 5-methylcytosine in myeloid cancers with mutant TET2. Nature. 2010; 468: 839-843.
18）Ley TJ, Ding L, Walter MJ, et al. DNMT3A mutations in acute myeloid leukemia. N Engl J Med. 2010; 363: 2424-2433.
19）Ewalt M, Galili NG, Mumtaz M, et al. DNMT3a mutations in high-risk myelodysplastic syndrome parallel those found in acute myeloid leukemia. Blood Cancer J. 2011; 1: e9.
20）Kosmider O, Gelsi-Boyer V, Slama L, et al. Mutations of IDH1 and IDH2 genes in early and accelerated phases of myelodysplastic syndromes and MDS/myeloproliferative neoplasms. Leukemia. 2010; 24: 1094-1096.
21）Figueroa ME, Abdel-Wahab O, Lu C, et al. Leukemic IDH1 and IDH2 mutations result in a hypermethylation phenotype, disrupt TET2 function, and impair hematopoietic differentiation. Cancer Cell. 2010; 18: 553-567.
22）Boultwood J, Perry J, Pellagatti A, et al. Frequent mutation of the polycomb-associated gene ASXL1 in the myelodysplastic syndromes and in acute myeloid leukemia. Leukemia. 2010; 24: 1062-1065.
23）Nikoloski G, Langemeijer SM, Kuiper RP, et al. Somatic mutations of the histone methyltransferase gene EZH2 in myelodysplastic syndromes. Nat Genet. 2010; 42: 665-667.
24）Bejar R, Stevenson K, Abdel-Wahab O, et al. Clinical effect of point mutations in myelodysplastic syndromes. N Engl J Med. 2011; 364: 2496-2506.
25）Kita-Sasai Y, Horiike S, Misawa S, et al. International prognostic scoring system and TP53 mutations are independent prognostic indicators for patients with myelodysplastic syndrome. Br J Haematol. 2001; 115: 309-312.
26）Gondek LP, Tiu R, O'Keefe CL, Sekeres MA, Theil KS, Maciejewski JP. Chromosomal lesions and uniparental disomy detected by SNP arrays in MDS, MDS/MPD, and MDS-derived AML. Blood. 2008; 111: 1534-1542.
27）Starczynowski DT, Vercauteren S, Telenius A, et al. High-resolution whole genome tiling path array CGH analysis of CD34＋ cells from patients with low-risk myelodysplastic syndromes reveals cryptic copy number alterations and predicts overall and leukemia-free survival. Blood. 2008; 112: 3412-3424.
28）Yoshida K, Sanada M, Shiraishi Y, et al. Frequent pathway mutations of splicing machinery in myelodysplasia. Nature. 2011; 478: 64-69.
29）Papaemmanuil E, Cazzola M, Boultwood J, et al. Chronic Myeloid Disorders Working Group of the International Cancer Genome Consortium. Somatic SF3B1 mutation in myelodysplasia with ring sideroblasts. N Engl J Med. 2011; 365: 1384-1395.
30）Thol F, Kade S, Schlarmann C, et al. Frequency and prognostic impact of mutations in SRSF2, U2AF1, and ZRSR2 in patients with myelodysplastic syndromes. Blood. 2012; 119: 3578-3584.
31）Bertoli S, Bérard E, Huguet F, et al. Time from diagnosis to intensive chemotherapy initiation does not adversely impact the outcome of patients with acute myeloid leukemia. Blood. 2013; 121: 2618-2626.
32）Cutler CS, Lee SJ, Greenberg P, et al. A decision analysis of allogeneic bone marrow transplantation for the myelodysplastic syndromes: delayed transplantation for low-risk myelodysplasia is associated with improved outcome. Blood. 2004; 104: 579-585.
33）Kelaidi C, Park S, Sapena R, et al. Long-term outcome of anemic lower-risk myelodysplastic syndromes without 5q deletion refractory to or relapsing after erythropoiesis-stimulating agents. Leukemia. Prepublished on January 16, 2013, as DOI 10.1038/leu.2013.16.
34）Rose C, Brechignac S, Vassilief D, et al. Does iron chelation therapy improve survival in regularly transfused lower risk MDS patients? A multicenter study by the GFM (Groupe Francophone des Myélodysplasies). Leuk Res. 2010; 34: 864-870.
35）List A, Dewald G, Bennett J, et al. Myelodysplastic Syndrome-003 Study Investigators. Lenalidomide in the myelodysplastic syndrome with chromosome 5q deletion. N Engl J Med. 2006; 355: 1456-1465.
36）Raza A, Reeves JA, Feldman EJ, et al. Phase 2 study of lenalidomide in transfusion-dependent, low-risk, and intermediate-1–risk myelodysplastic syndromes with karyotypes other than deletion 5q. Blood. 2008; 111: 86-93.
37）Adès L, Boehrer S, Prebet T, et al. Efficacy and safety of lenalidomide in intermediate-2 or high-risk myelodysplastic syndromes with 5q deletion: results of a phase 2 study. Blood. 2009; 11: 3947-3952.
38）Silverman LR, Demakos EP, Peterson BL, et al. Randomized controlled trial of azacitidine in patients with the myelodysplastic syndrome: a study of the cancer and leukemia group B. J Clin Oncol. 2002; 20: 2429-2440.
39）Fenaux P, Mufti GJ, Hellstrom-Lindberg E, et al. International Vidaza High-Risk MDS Survival Study Group. Efficacy of azacitidine compared with that of conventional care regimens in the treatment of higher-risk myelodysplastic syndromes: a randomised, open-label, phase III study. Lancet Oncol. 2009; 10: 223-232.
40）Lübbert M, Suciu S, Baila L, et al. Low-dose decitabine versus best supportive care in elderly patients with intermediate- or high-risk myelodysplastic syndrome (MDS) ineligible for intensive chemotherapy: final results of the randomized phase III study of the European Organisation for Research and Treatment of Cancer Leukemia Group and the German MDS Study Group. J Clin Oncol. 2011; 29: 1987-1996.
41）Sekeres MA, Tiu RV, Komrokji R, et al. Phase 2 study of the lenalidomide and azacitidine combination in patients with higher-risk myelodysplastic syndromes. Blood. 2012; 120: 4945-4951.
42）Soriano AO, Yang H, Faderl S, et al. Safety and clinical activity of the combination of 5-azacytidine, valproic acid, and all-trans retinoic acid in acute myeloid leukemia and myelodysplastic syndrome. Blood. 2007; 110: 2302-2308.
Columbia University Medical Center, New York, NY