Document baj2dykp61ra5x6YdNpZ3vLD3

Published Ahead of Print on January 10, 2011 as 10.1200/JCO.2010.30.2554 The latest version is at http://jco.ascopubs.org/cgi/doi/10.1200/JCO.2010.30.2554 JOURNAL OF CLINICAL ONCOLOGY REVIEW ARTICLE From The Ohio State University Comprehensive Cancer Center, Columbus, OH; MLL Munich Leukemia Laboratory, Munich; and University of Ulm, Ulm, Germany. Submitted May 2, 2010; accepted November 3, 2010; published online ahead of print at www.jco.org on January 10, 2011. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article. Corresponding author: Guido Marcucci, MD, The Ohio State University, 898 Biomedical Research Tower, 460 W 12th Ave, Columbus, OH 43210; e-mail: guido.marcucci@osumc.edu. 2011 by American Society of Clinical Oncology 0732-183X/10/2899-1/$20.00 DOI: 10.1200/JCO.2010.30.2554 Molecular Genetics of Adult Acute Myeloid Leukemia: Prognostic and Therapeutic Implications Guido Marcucci, Torsten Haferlach, and Hartmut Dohner ABSTRACT Molecular analyses of leukemic blasts from patients with acute myeloid leukemia (AML) have revealed a striking heterogeneity with regard to the presence of acquired gene mutations and changes in gene and microRNA expression. Multiple submicroscopic genetic alterations with prognostic significance have been discovered. Application of gene- and microRNA profiling has identified genome-wide expression signatures that separate cytogenetic and molecular subsets of patients with AML into previously unrecognized biologic and/or prognostic subgroups. These and similar future findings are likely to have a major impact on the clinical management of AML because many of the identified genetic alterations not only represent independent prognosticators, but also may constitute targets for specific therapeutic intervention. In this report, we review genetic findings in AML and discuss their clinical implications. J Clin Oncol 28. 2011 by American Society of Clinical Oncology INTRODUCTION Acute myeloid leukemia (AML) is a genetically heterogeneous clonal disorder characterized by the accumulation of somatically acquired genetic alterations in hematopoietic progenitor cells that alter normal mechanisms of self-renewal, proliferation, and differentiation.1 In recent years, gene mutations and deregulated expression of genes and noncoding RNAs (ie, microRNAs) have been identified, providing insights into the mechanisms of leukemogenesis and unraveling the enormous molecular genetic heterogeneity within distinct cytogenetically defined subsets of AML, in particular the large group of cytogenetically normal (CN) AML.2-5 With progress in genomics technology, such as gene- and microRNA-expression profiling, genome-wide single nucleotide polymorphism based mapping arrays, next-generation sequence techniques, and functional genomics, systematic characterization of cancer genomes has now become feasible, and the identification of disease-relevant mutations and related gene and microRNA profiles will be largely facilitated. From a clinical perspective, there are at least three important aspects with respect to the genetic changes in AML.6 First, the current WHO classification reflects the fact that an increasing number of cases of AML can be categorized on the basis of their underlying genetic defects that define distinct clinicopathologic entities.7 Second, it has become clear that specific chromosome abnormalities and molec- ular genetic changes are among the most important prognostic markers and therefore may be used for stratification of patients with AML to risk-adapted therapeutic strategies. Finally, novel therapies are being developed that target some of the identified genetic defects. It is therefore anticipated that these genetic markers will acquire a predictive value, that is, the ability to predict differential efficacy of a therapy. Thus it is hoped that genetic markers will allow us to optimize the treatment of distinct subtypes of AML. RECURRENT GENE MUTATIONS IN AML Somatically acquired mutations have been identified in several genes (Table 1). In general, these mutations are found most frequently in CN-AML but are associated with other cytogenetic subgroups as well. GENE MUTATIONS IN AML: CLINICAL PRACTICE To date, only diagnosis of NPM1, CEBPA, and FLT3 mutations has entered clinical practice and affects diagnosis, risk assessment, and also guidance of therapy.6,7 On the basis of characteristic clinical, pathologic, and biologic features, AML with NPM1 mutation and AML with CEBPA mutation have been incorporated as provisional entities in the 2008 WHO classification of AML.7 Although FLT3 mutations are not considered to define a distinct entity, 2011 by American Society of Clinical Oncology Information downloaded from jco.ascopubs.org and provided by at HAM-TMC LIBRARY on February 2, 2011 from Copyright 2011 American Soc1ie9t2y.6o8f .C3l0in.2ical Oncology. All rights reserved. Copyright 2011 by American Society of Clinical Oncology 1 Marcucci, Haferlach, and Do hner Gene Mutated genes NPM1 CEBPA FLT3 ITD TKD IDH1, IDH2 KIT WT1 RUNX1 MLL-PTD Table 1. Recurrent Molecular Genetic Abnormalities in Adult AML: Biologic Features and Clinical Significance Biologic Features Clinical Significance Nuclear-cytoplasmic shuttling phosphoprotein with pleiotropic functions Mutations lead to abnormal cytoplasmic localization of the protein that can be diagnosed by immunohistochemistry on bone marrow sections Mutations found in 25%-35% of AML (CN-AML, 45%-64%; del(9q) outside a complex karyotype, 35%-40%; trisomy 8, approximately 15%) Associated with FLT3-ITD (approximately 40%), FLT3 TKD (10%-15%), and IDH mutations (approximately 25%) Leukemic blasts show high CD33 and absent to low CD34 expression Associated with myelomonocytic or monocytic morphology Higher prevalence in female sex Master regulatory transcription factor in hematopoiesis Mutations predominantly in CN-AML (10%-18%), and AML with del(9q) occurring outside a complex karyotype (approximately 40%) Mutations mostly biallelic (N-terminal and C-terminal); only double-mutant cases may define the entity Member of the class III receptor tyrosine kinase family; FLT3 and its ligand play an important role in proliferation, survival, and differentiation of hematopoietic progenitor cells FLT3-ITD found in approximately 20% of all AML (normal karyotype: 28%-34%) Insertion site of the ITD commonly in the JM domain; in approximately 30% in the tyrosine kinase 1 domain Homozygous mutations result from mitotic recombination leading to partial UPD FLT3 TKD point mutations found in 5%-10% of all AML [CN, 11%-14%; inv(16)/t(16;16), 14%-24%] Cytosolic (IDH1) and mitochondrial (IDH2) metabolic enzymes catalyzing oxidative decarboxylation of isocitrate to -ketoglutarate; involved in cellular defense of oxidative damage Mutations commonly located at residue 132 (IDH1) and residues 140 and 172 (IDH2); lead to a neomorphic enzyme activity and to accumulation of a putative oncogenic metabolite, R(2)-2-hydroxyglutarate (2HG) Mutations found in approximately 16% of unselected AML (IDH1, 8%; IDH2, 8%); association with CN-AML (IDH1, 10%-16%; IDH2, 10%-19%) IDH1 R132 (and possibly also IDH2 R140) mutation associated with molecular low-risk AML (mutated NPM1 without FLT3-ITD) IDH2 R172 mutation rarely found in combination with other known gene mutations and associated with distinctive microarray gene- and microRNA-expression profiles Member of the class III receptor tyrosine kinase family; KIT and its ligand stem cell factor have a key role in survival, proliferation, differentiation, and functional activation of hematopoietic progenitor cells Mutations mostly found in CBF-AML (25%-30%) Transcription factor implicated in regulation of apoptosis, proliferation, and differentiation of hematopoietic progenitor cells Mutations found in 10%-13% of CN-AML Transcription factor required for definitive hematopoiesis Mutations found in 5%-13% of AML; association with trisomy 13, trisomy 21, and normal karyotype Mutations associated with MLL-PTD; inverse correlation with presence of CEBPA and NPM1 mutations Association with undifferentiated morphology (FAB M0) DNA binding protein that regulates gene expression in hematopoiesis possibly through epigenetic mechanisms MLL-PTD found in 5%-11% of normal karyotype AML, and up to 90% of AML with trisomy 11 (continued on following page) Provisional disease entity (WHO 2008) In younger adult patients, mutated NPM1 without FLT3-ITD predicts for achievement of CR and favorable RFS and OS In younger adults with mutated NPM1 without FLT3ITD, standard induction therapy followed by repetitive cycles of HiDAC is a reasonable treatment option; patients may not benefit from allogeneic HSCT in first CR Favorable impact of NPM1 mutations in older patients Older patients with NPM1-mutated AML may benefit from intensive conventional chemotherapy Provisional disease entity (WHO 2008) Mutations associated with higher CR rate and favorable RFS and OS Only double, not single, CEBPA mutations predict for favorable outcome In younger adults, standard induction therapy followed by repetitive cycles of HiDAC is a reasonable treatment option for AML with CEBPA mutation FLT3-ITD associated with inferior outcome, in particular those cases with high mutant to wild-type allelic ratio; non-JM FLT3-ITD possibly associated with dismal outcome Allogeneic HSCT should be considered in AML with FLT3-ITD Phase II and III clinical trials evaluating FLT3 tyrosine kinase inhibitors underway Prognostic significance controversial Phase II and III clinical trials evaluating FLT3 tyrosine kinase inhibitors underway IDH2 mutations found more frequently in older patients IDH1 and possibly also IDH2 mutations confer higher risk of relapse and inferior OS in molecular low-risk CN-AML (mutated NPM1 without FLT3-ITD) IDH2 R172 mutation associated with lower CR rate and possibly with inferior outcome IDH1 SNP rs11554137 (located in the same exon as the R132 mutation) in one study found to be associated with inferior outcome in molecular highrisk CN-AML (either NPM1 wild-type or FLT3-ITD positive) Mutations associated with inferior outcome in CBF-AML Currently no data available supporting the use of KIT mutational status to guide therapy Clinical trials evaluating KIT inhibitors underway Prognostic significance somewhat controversial; most studies report a negative prognostic impact Postremission therapy with HiDAC may abrogate poor prognosis WT1 SNP rs16754 located in the proximity of the WT1 mutational hotspot in one study found to be associated with inferior outcome in CN-AML Prognostic significance under investigation; first data indicate association with lower CR rate and shorter RFS and OS In initial studies associated with shorter CR duration, inferior RFS and EFS, but not OS In multivariable analyses in general not an independent prognostic factor 2 2011 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY Information downloaded from jco.ascopubs.org and provided by at HAM-TMC LIBRARY on February 2, 2011 from Copyright 2011 American Soc1ie9t2y.6o8f .C3l0in.2ical Oncology. All rights reserved. Prognostic Markers in AML Table 1. Recurrent Molecular Genetic Abnormalities in Adult AML: Biologic Features and Clinical Significance (continued) Gene NRAS KRAS TP53 TET2 ASXL1 JAK2 CBL Deregulated genes BAALC ERG EVI1 MN1 Biologic Features Membrane-associated proteins regulating mechanism of proliferation, differentiation, and apoptosis Mutations found in 9%-14% of CN-AML, in up to 40% of CBF-AML (in particular inv16), and in 25%-30% of AML with inv(3) Mutations found in 5%-17% of CBF-AML Encodes tumor suppressor protein p53, which responds to diverse cellular stresses to regulate target genes that induce cell cycle arrest, apoptosis, senescence, DNA repair, or changes in metabolism Mutations/deletions mostly in AML with complex karyotype (56%-78%) Belongs to a 3-member family of highly conserved genes; protein function unknown, may be involved in epigenetic regulation Mutations found in a wide spectrum of myeloid neoplasms; MDS (20%), MPN (12%), secondary AML (25%); frequency in de novo AML under investigation Putative polycomb group protein; possibly involved in chromatin remodeling Mutations found in MDS, MPN, AML Protein tyrosine kinase, involved in a specific subset of cytokine receptor signaling pathways Mutations common in MPN, rare in AML (mostly in CBF-AML) Adaptor protein for receptor protein-tyrosine kinases, positive regulation of receptor protein-tyrosine kinase ubiquitination Mutations rare in AML (mostly in CBF-AML) Clinical Significance No prognostic significance shown Mutant NRAS may predict sensitivity to cytarabine No prognostic significance shown Associated with inferior outcome (overlap with complex karyotypes) Clinical significance under investigation Clinical significance under investigation Likely not of major clinical significance Likely not of major clinical significance Biologic function unknown High BAALC expression correlates with that of genes expressed in hematopoietic progenitors and those involved in chemoresistance Member of the ETS family of transcription factors ERG rearrangements involved in other cancers (eg, prostate cancer) Highly expressed in CN-, complex karyotype, and megakaryoblastic AML EVI1 protein involved in regulation of transcription factors critical for hematopoiesis (eg, GATA 1, GATA2) and in epigenetic regulation Deregulated expression of EVI1 found in AML with inv(3)(q21q26.2) or t(3; 3)(q21;q26.2); EVI1-RPN1 EVI1 overexpression also found in approximately 10% of unselected AML; association with 7 and t(11q23)/MLL rearrangements MN1 gene has the role of a transcription coregulator Associated with cytogenetic abnormalities in other type of cancer High expression associated with NPM1 wild-type and BAALC and MN1 upregulation In CN-AML associated with miR-126 High BAALC expression associated with inferior CR rate, DFS, and OS High ERG expression associated with inferior CR rate and EFS High ERG expression mostly impacts prognosis in CNAML with molecular low risk (NPM1 mutated without FLT3-ITD) Deregulated EVI1 expression associated with low CR rate and inferior survival; effect most pronounced in cytogenetic intermediate-risk AML Allogeneic HSCT may improve survival of AML with high EVI1 High MN1 expression correlates with lower CR, shorter DFS, and OS Low MN1 expression in one study has been shown to predict response to therapy with ATRA in elderly AML patients Abbreviations: AML, acute myeloid leukemia; ITD, internal tandem duplication; CR, complete remission; RFS, relapse-free survival; OS, overall survival; CN, cytogenetically normal; TKD, tyrosine kinase domain; HiDAC, high-dose cytarabine; HSCT, hematopoietic stem-cell transplantation; JM, juxtamembrane domain; UPD, uniparental disomy; SNP, single nucleotide polymorphism; CBF, core-binding factor; PTD, partial tandem duplication; EFS, event-free survival; FAB, French-American-British; EFS, event-free survival; MDS, myelodysplastic syndromes; MPN, myeloproliferative neoplasms; DFS, disease-free survival; ATRA, all-trans-retinoic acid. they provide important prognostic information (Fig 1). Furthermore, they are now being targeted in clinical trials with tyrosine kinase inhibitors. Therefore, NPM1, CEBPA, and FLT3 mutations are recommended to be analyzed in clinical trials and in routine practice at least in patients with CN-AML who will receive treatment other than low-dose chemotherapy or best supportive care.6 NPM1 Mutations Abnormal cytoplasmic localization of the NPM1 protein shown by immunohistochemical analysis led to the discovery of NPM1 mutations in AML.8 This mislocalization is caused by mutations in exon 12 of the gene. NPM1 mutations are found in approximately one third of adult cases of AML, making it the most frequent mutation known in this disease to date. NPM1 mutations are associated with other recurrent genetic changes, secondary chromosome abnormalities such as 8, 4, del(9q),9 and additional gene mutations, most frequently in FLT32,3,10 and IDH1.11-15 Prognostic significance. NPM1 mutations, in particular the genotype "mutated NPM1 without concurrent FLT3 internal tandem duplication (ITD)," has been associated with achievement of complete remission (CR) and favorable outcome.2,3,10 On the basis of this observation, AML with mutated NPM1 without FLT3-ITD has then recently been allocated to the genetic favorable-risk category of AML, together with core-binding factor (CBF) AML.6 Therapeutic implications. Similar to CBF-AML, standard induction chemotherapy followed by three to four cycles of high-dose cytarabine is a recommended treatment approach for AML with mutated NPM1 without FLT3-ITD.6 In studies from Cancer and Leukemia Group B (CALGB), patients with this genotype also had favorable www.jco.org 2011 by American Society of Clinical Oncology 3 Information downloaded from jco.ascopubs.org and provided by at HAM-TMC LIBRARY on February 2, 2011 from Copyright 2011 American Soc1ie9t2y.6o8f .C3l0in.2ical Oncology. All rights reserved. Marcucci, Haferlach, and Do hner A 1.0 0.8 Mutant NPM1 without FLT3-ITD Mutant CEBPA Other genotypes B 1.0 0.8 Mutant NPM1 without FLT3-ITD Mutant CEBPA Other genotypes Overall Survival (%) Relapse-Free Survival (%) 0.6 0.6 0.4 0.4 0.2 0.2 P < .001 P < .001 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 Time (years) No. at risk Other genotypes 212 90 60 38 27 14 9 5 4 2 0 Mutant NPM1 136 102 83 64 50 34 24 13 8 3 1 without FLT3-ITD Mutant CEBPA 62 40 32 26 17 11 6 4 4 3 0 Time (years) No. at risk Other genotypes 266 153 90 68 39 22 15 9 7 Mutant NPM1 150 123 101 75 56 38 25 14 10 without FLT3-ITD Mutant CEBPA 67 54 39 30 19 13 8 6 5 4 4 3 0 2 0 C 1.0 D 1.0 Relapse-Free Survival (%) Relapse-Free Survival (%) 0.8 0.8 0.6 0.6 0.4 0.4 0.2 Donor No donor P = .71 0 1 2 3 4 5 6 7 8 9 10 No. at risk Donor No donor Time (years) 97 71 60 46 41 28 19 10 7 3 1 38 31 23 18 9 6 5 3 1 0 0 0.2 Donor No donor P = .003 0 1 2 3 4 5 6 7 8 9 10 No. at risk Donor No donor Time (years) 148 57 36 19 13 60 33 24 19 14 86 63 321 221 0 0 Fig 1. Clinical outcome in cytogenetically normal acute myeloid leukemia (AML). (A) Relapse-free survival according to genotype; (B) overall survival according to genotype; (C) relapse-free survival of patients with mutant NPM1 without FLT3 internal tandem duplication (ITD) according to the availability of an HLA-matched related donor; (D) relapse-free survival of patients with other genotypes, excluding the mutant CEBPA genotype, according to the availability of an HLA-matched related donor. "Other genotypes" is defined as the FLT3-ITD genotype and the triple-negative genotype consisting of wild-type NPM1 and CEBPA without FLT3-ITD. The data in (D) suggest that molecular high-risk AML (eg, AML with FLT3-ITD) may benefit from allogeneic hematopoietic stem-cell transplantation in first complete remission. Tick marks represent patients whose data were censored at the last time they were known to be alive and in complete remission (A, C, D) or whose data were censored at the last time they were known to be alive (B). Adapted with permission from Schlenk et al.10 outcome when treated with autologous hematopoietic stem-cell transplantation (HSCT).3 Patients with mutated NPM1 without FLT3-ITD may not be considered candidates for allogeneic HSCT in first CR, unless they have a very low risk for transplant-related mor- bidity and mortality or new transplantation strategies are evaluated within a clinical trial.6,10 Of note, NPM1 mutations also seem to predict better outcome in older patients,16-18 even extending to patients older than 70 years.17 Thus it is also important to assess for NPM1 mutations in older patients to identify those individuals who are likely to benefit from intensive conventional chemotherapy. The German-Austrian AML Study Group previously reported that AML with mutated NPM1 without FLT3-ITD benefit from alltrans-retinoic acid (ATRA) given in combination with conventional chemotherapy.16 This finding was not confirmed in a study by the British Medical Research Council.19 Differences in the trial designs may account for these discrepant data. Results from ongoing trials 4 2011 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY Information downloaded from jco.ascopubs.org and provided by at HAM-TMC LIBRARY on February 2, 2011 from Copyright 2011 American Soc1ie9t2y.6o8f .C3l0in.2ical Oncology. All rights reserved. Prognostic Markers in AML need to be awaited to determine whether ATRA will have a role in the molecular therapy of AML with NPM1 mutation. CEBPA Mutations Two major types of heterozygous CEBPA mutations have been identified in AML.20 Nonsense mutations affecting the N-terminal region result in a truncated CEBPA isoform with dominant-negative properties, and in-frame mutations in the C-terminal basic regionleucine zipper domain result in CEBPA proteins with decreased DNAbinding or dimerization activity. In approximately two thirds of cases, C- and N-terminal mutations are biallelic mutations (also called double mutations), with the majority (approximately 90%) of them being compound heterozygous (C-terminal on one allele and N-terminal on the other) and the rest being homozygous (either both C- or N-terminal), likely through the mechanism of uniparental disomy. CEBPA mutations are predominantly found in CN-AML and in cases with 9q deletion (Table 1).2 Prognostic significance. Among CN-AML, CEBPA mutations have consistently been associated with a relatively favorable outcome, similar to that of AML with mutated NPM1 without FLT3-ITD.2,3,10 Importantly, recent studies show that only double, but not single, CEBPA mutations predict for this favorable outcome, whereas AML with a single CEBPA mutation is associated with survival similar to that of AML with wild-type CEBPA.20 This finding is biologically substantiated by a discrete gene-expression signature of doublemutated cases21 and by mouse experiments modeling the mutations by knock-in mutagenesis.22 Therefore, with regard to disease classification and risk stratification, only AML with double CEBPA mutations likely define a distinct entity with favorable prognosis. However, whether the biologic and clinical significance of double CEBPA mutations related to lack of at least one normally functioning allele remains to be elucidated. Therapeutic implications. Therapeutic recommendations are similar to those for AML with mutated NPM1 without FLT3-ITD, that is, standard induction chemotherapy followed by three to four cycles of high-dose cytarabine.6 Also, AML with double CEBPA mutations may not benefit from allogeneic HSCT; however, this statement is currently not substantiated by data, but rather by the assumption that in general, patients with favorable-risk AML do not seem to benefit from this approach in first CR. Because of the low incidence of the mutation, effects of novel antileukemic agents and of allogeneic HSCT in this subset of AML can only be evaluated in intergroup trials or in retrospective large meta-analyses. FLT3 Mutations Mutations that result in the constitutive activation of FLT3 have been identified in two functional domains of the receptor, the juxtamembrane (JM) domain and the tyrosine kinase domain (TKD).1 FLT3-ITD are found in approximately 20% of unselected cases of AML and mainly cluster in the JM domain; however, recently it was shown that approximately 30% of ITDs do not insert in the JM but in the TK1 domain of the receptor.23 The activation loop in the carboxyterminal lobe of the TKD is affected by point mutations, small insertions, or deletions, mainly at codons 835 and 836, in 5% to 10% of cases of AML.1,2 Rare point mutations or insertions have been reported at other codons in the TKD. In vitro studies and results from global gene-expression profiling (GEP) revealed that there are similarities but also important differences in signal transduction properties between FLT3-ITD and FLT3 TKD mutations that may explain differences in clinical phenotypes.1 Prognostic significance. Prognosis of CN-AML with FLT3-ITD is significantly inferior compared with CN-AML without the mutation when treated with current standard chemotherapy.2,3,10 There is evidence that outcome is related to the ratio of mutated versus wild-type allele in that a high burden of mutated allele predicts for inferior survival.3 Furthermore, a recent study suggests that patients with AML with non-JM ITD do significantly worse than those with AML with JM domain ITD.24 The prognostic relevance of FLT3 TKD mutations remains controversial.2,3 Therapeutic implications. Therapy of AML with FLT3-ITD is a clinical dilemma. Given the poor results after standard chemotherapy, new modalities need to be evaluated. There is accumulating evidence that allogeneic HSCT in general is an attractive option for patients who are at high risk of relapse. A beneficial effect of allogeneic HSCT has been shown for the entire group of patients with intermediate-risk cytogenetics and for CN-AML with unfavorable genotypes that includes most AML with FLT3-ITD.6,10 Thus, although evidence from prospective trials is lacking (Fig 1), allogeneic HSCT should be considered in patients with FLT3-ITD-positive AML. Randomized phase III trials evaluating FLT3 inhibitors in combination with chemotherapy as a frontline approach to patients with AML with activating FLT3 mutations are underway. GENE MUTATIONS IN AML: INVESTIGATIONAL STUDIES All other gene mutations identified in AML have not yet entered clinical practice and remain investigational. KIT Mutations KIT mutations are found in 25% to 30% of cases of CBF-AML and are rare in other AML subsets.25 In most studies, KIT mutations have been associated with inferior outcome. Of note, KIT is not only mutated, but is also expressed at significantly higher levels in CBF-AML compared with other AML subsets.26,27 At present, there are no data supporting the use of KIT mutational status to guide therapy. Clinical trials are currently underway evaluating KIT inhibitors in CBF-AML. IDH1/IDH2 Mutations Mutations of IDH1 and IDH2 were first reported in gliomas and were reported only more recently in AML.5,11-15 Interestingly, of the three types of mutations that have been hitherto reported, those affecting the arginine residues on IDH1 codon 132 and IDH2 codon 172 have been found both in brain tumors and AML, whereas those affecting the arginine residue on IDH2 codon 172 are private to AML. The functional differences in the mutant proteins remain to be fully elucidated. The aggregate frequency of these two mutations in AML is relatively high, with approximately 15% to 20% of all patients with AML and 25% to 30% of patients with CN-AML harboring either IDH1 or IDH2 mutations.11,12 Of note, IDH proteins bring a new class of mutated proteins in leukemogenesis, that is, metabolic enzymes. Both IDH1 and IDH2 mutants cause loss of the physiologic enzyme function and create a novel ability of the enzymes to convert -ketoglutarate into 2-hydroxyglutarate, a putative oncogenic metabolite.28 Initial studies from larger and homogeneous cohorts of patients indicate that IDH1 and possibly also IDH2 R140 mutations are www.jco.org 2011 by American Society of Clinical Oncology 5 Information downloaded from jco.ascopubs.org and provided by at HAM-TMC LIBRARY on February 2, 2011 from Copyright 2011 American Soc1ie9t2y.6o8f .C3l0in.2ical Oncology. All rights reserved. Marcucci, Haferlach, and Do hner A 1.0 B 1.0 Disease-Free Survival (probability) Overall Survival (probability) 0.8 0.8 0.6 0.6 0.4 0.4 0.2 NPM1 mutated, FLT3-ITD negative, IDH1/IDH2 wt (n = 34) NPM1 mutated, FLT3-ITD negative, IDH1 mutated (n = 12) P = .046 0 1234 Time (years) 5 0.2 NPM1 mutated, FLT3-ITD negative, IDH1/IDH2 wt (n = 38) NPM1 mutated, FLT3-ITD negative, IDH1 mutated (n = 14) P = .14 0 1234 Time (years) 5 Fig 2. Clinical outcome in cytogenetically normal acute myeloid leukemia (CN-AML) according to IDH1 and IDH2 mutational status. (A) Outcome of the CN-AML subset with NPM1 mutation and without FLT3 internal tandem duplication (ITD) according to IDH1 mutational status; (B) outcome of patients with CN-AML according to IDH2 R172 mutational status. wt, wild type. Adapted with permission from Marcucci et al.11 significantly associated with NPM1 mutations and predict worse outcome for patients with mutated NPM1 without FLT3-ITD.11-14 Interestingly, the distinct R172 IDH2 mutation is rarely associated (Fig 2) with any of the other known prognostic mutations and seems to confer lower probability of achieving CR and possibly also inferior outcome.11-13 WT1 Mutations WT1 mutations are found in 10% to 13% of CN-AML.29-31 In some studies, WT1 mutations have been associated with inferior outcome,29,30 whereas in the largest study reported by the GermanAustrian AML Study Group,31 WT1 mutations did not affect outcome. Differences in postremission therapy may account for these discrepant results. In one study, WT1 single nucleotide polymorphism rs16754 located in the proximity of the WT1 mutational hotspot was found to be associated with inferior outcome in CN-AML.32 RUNX1 Mutations RUNX1 is deregulated in AML by chromosomal translocations and by mutations clustering in the Runt domain of the gene.2 RUNX1 mutations have been associated with undifferentiated (M0) morphology and with specific chromosome aberrations, such as trisomy 21 and trisomy 13. In a study of 470 Chinese adult patients with AML, RUNX1 mutations were found in 13.2% of cases.33 Mutations were associated with lower CR rate and with inferior survival. MLL Mutations Partial tandem duplications (PTD) of MLL are found in 5% to 11% of patients with CN-AML and frequently in those with AML with trisomy 11.2,3 MLL-PTD have been shown to contribute to leukemogenesis through DNA hypermethylation and epigenetic silencing of tumor suppressor genes, pointing to a potential role of DNA methyltransferase and/or histone deacetylase inhibitors in the treatment of this AML subset.3 MLL-PTD have been associated with inferior CR duration and relapse-free survival, although more recent studies show no prognostic impact in patients with CN-AML intensively treated with autologous HSCT or four cycles of consolidations.2,3 Other Gene Mutations Selected biologic and clinical features of other mutations (eg, NRAS,34 TP53,35 TET2,36,37 and ASXL1,38) are summarized in Table 1. These mutations either occur at very low frequency, do not seem to significantly contribute to risk stratification, or have so far been less well studied and their prognostic significance remains to be fully defined.39,40 Nonetheless, it will be necessary to evaluate these biomarkers in future clinical trials to study potential differential treatment effects of new antileukemic agents in subsets of AML defined by the various gene mutations. EXPRESSION OF SINGLE GENES WITH PROGNOSTIC RELEVANCE In addition to structural genetic aberrations, changes in expression of specific genes seem to impact prognosis of molecular subsets of patients with AML. The BAALC (brain and acute leukemia, cytoplasmic) gene, localized on chromosome band 8q22.3, was first identified to have higher expression in AML with trisomy 8; however, a wide range of expression was described in CN-AML.3 Baldus et al41 initially demonstrated that higher BAALC expression levels predicted lower CR rates as well as inferior disease-free and overall survival (OS), which was confirmed by subsequent studies.42,43 The MN1 (meningioma [disrupted in balanced translocation] 1) gene was first identified as a fusion partner of the TEL gene in reciprocal translocations in AML. Recent studies have shown that MN1 overexpression is associated with poor response to induction chemotherapy and higher relapse rate and worse OS in AML.44,45 Interestingly, in one study, low MN1 expression was correlated with better response to ATRA in elderly patients with nonacute promyelocytic leukemia (APL) AML, thereby suggesting that MN1 expression is not only a prognostic but also a predictive marker for response to treatment.44 In patients with CN-AML younger than 60 years, higher MN1 expression was significantly correlated with NPM1 wild-type status and increased BAALC expression.45 ERG (v-ets erythroblastosis virus E26 oncogene homolog, avian) mapping to band 21q22 is involved in various chromosomal rearrangements. With gene-expression analysis based on oligonucleotide arrays, amplification of the ERG locus with consecutive gene overexpression was observed in AML with complex aberrant karyotypes.3 Subsequently, the adverse prognostic significance of 6 2011 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY Information downloaded from jco.ascopubs.org and provided by at HAM-TMC LIBRARY on February 2, 2011 from Copyright 2011 American Soc1ie9t2y.6o8f .C3l0in.2ical Oncology. All rights reserved. Prognostic Markers in AML high blood ERG levels was established in CN-AML.3,46 Interestingly, high ERG expression levels mostly impacted outcome of low molecular risk CN-AML (mutated NPM1 without FLT3-ITD).46 In a subsequent GEP study investigating ERG, MN1, and BAALC transcript levels, ERG expression was the strongest negative prognostic factor and provided prognostic information in addition to established parameters (eg, FLT3-ITD).40 Deregulated expression of EVI1 (ecotropic viral integration site 1) is found in virtually all AML, with inv(3)(q21q26.2/t(3,3)(q21; q26.2) leading to rearrangement of the EVI1 and RPN1 genes. EVI1 overexpression is not restricted to these cases but was found as well in approximately 8% of unselected AML and has been shown to predict poor outcome.47 A subsequent validation study confirmed that high EVI1 expression predicts poor outcome, particularly in the cytogenetic intermediate-risk group. Patients with high EVI1 expression who received allogeneic HSCT in first CR had significantly better 5-year relapse-free survival and OS.48 GENOME-WIDE GEP A decade ago, Golub et al49 demonstrated that acute leukemia subtypes can be classified on the basis of distinctive signatures identified by genome-wide GEP. Since then, multiple studies confirmed that GEP has the potential to predict specific cytogenetic or molecular alterations (Table 2) in AML and to diagnose specific subtypes by unique patterns of expressed genes ("class prediction").54,55 Beyond that, this technology is able to add new prognostic information to established cytogenetic or single gene molecular prognosticators ("class discovery"; Table 3).56,57 The robustness of gene-expression signatures derived from patients with AML at diagnosis is supported by reasonably comparable results attained across different laboratories, the use of different microarray profiling platforms, and validation analyses using different techniques such as quantitative reverse-transcriptase polymerase chain reaction.59-61 However, at this time, GEP remains an investigational tool for research studies in AML that so far has not found entrance in clinical practice and decision making. Prediction of Cytogenetic Alterations AML with t(15;17)/PML-RARA and CBF-AML with t(8;21)/ RUNX1-RUNX1T1 and inv(16)/CBFB-MYH11 were clearly discriminated from other cytogenetic subgroups and were predictable with gene-expression analysis by independent research groups.27,62 Discrimination accuracies up to 100% were achieved, which emphasizes the biologic uniqueness of the respective subtypes in concordance with the distinct morphologic and genetic features.62 Rearrangements of the MLL gene on 11q23 occur both in AML and acute lymphoblastic leukemia (ALL) and are generally correlated with an adverse prognosis. Kohlmann et al63 performed unsupervised and supervised data analysis algorithms in ALL and AML cases with 11q23/MLL rearrangements and demonstrated the segregation of cases according to myeloid or lymphoid lineage. Complex karyotype cases were clearly separated from other cytogenetic AML subtypes by distinct gene-expression signatures characterized by upregulation of genes with a role in DNA repair such as RAD21.64,65 Prediction of Molecular Alterations by Specific Patterns of Genes Expressed NPM1-mutated AML was found to be associated with highly specific underlying gene-expression signatures characterized by activation of distinct HOX cluster genes (a homeodomain-containing family of transcription factors) and by involvement of genes with a function in signaling and apoptosis (Table 2).50 The signatures for NPM1 mutations allowed the prediction of this molecular alteration with the highest accuracy and thus clearly defined the biologic singularity of this AML subtype.51 Similarly, CEBPA mutations have been associated with distinct GEPs. In a recent study by Wouters et al,21 a 19-probe set signature predictive of CEBPA mutations was derived. This classifier showed a high specificity (99%) but a limited sensitivity (67%) in cross-validation. Misclassification resulted almost entirely from CEBPA single-mutated cases, whereas CEBPA double-mutated cases were predictable with the highest accuracy. Furthermore, gene expression was variable depending on the occurrence of a double- or single-mutation status. These data reinforce the hypothesis that double-mutated CEBPA likely defines a distinct biologic and clinical AML entity. Another study on CN-AML compared gene-expression signatures of CEBPA-mutated and unmutated cases and reported more than 93% prediction accuracy for predicting CEBPA status.51 Frequently selected genes for classification included HOXA and HOXB cluster genes, which were downregulated in patients with CEBPA mutatation. CEBPAmutated AML were further characterized by distinct gene-expression signatures comprising upregulation of erythroid-specific genes, including GATA1 and EPOR.52 Bullinger et al53 revealed a novel gene signature that showed overlaps to FLT3-ITD positive AML and strong correlations to FLT3 pathway activation. When compared with so-called classical FLT3ITD, this FLT3-ITD derived signature revealed an even stronger impact on clinical outcomes.53 Beyond that, gene-expression signatures are linked to expression levels of distinct genes that are prognostically relevant in CN-AML. Langer et al42 identified a specific gene-expression signature in association with high BAALC expression that included overexpression of genes involved in drug resistance (MDR1) and stem cell markers (eg, CD133, CD34, KIT). The same group derived a gene profile associated with MN1 high expression levels, which was positively correlated with expression of BAALC, CD200 (associated with inferior prognosis in AML), and ABCB1 (involved in chemoresistance).45 Identification of Novel Prognostic Classifiers Multiple study groups have focused on the identification of prognostic gene-expression signatures, especially in CN-AML (Table 3).26,53,56,58 Bullinger et al26 initially subdivided CN-AML cases into two clusters corresponding to different prognostic profiles. One cluster was characterized by overexpression of several transcriptional regulators such as GATA2, and the second one was characterized by involvement of genes playing a role for leukocyte differentiation and immune response. Radmacher et al56 validated the prognostic power of this gene signature in an independent cohort, thereby underlining the reproducibility and utility of GEP classifiers across distinct study groups and laboratories. It should be underscored that most of the reported signatures were derived using unselected bone marrow or blood mononuclear cell samples www.jco.org 2011 by American Society of Clinical Oncology 7 Information downloaded from jco.ascopubs.org and provided by at HAM-TMC LIBRARY on February 2, 2011 from Copyright 2011 American Soc1ie9t2y.6o8f .C3l0in.2ical Oncology. All rights reserved. Marcucci, Haferlach, and Do hner Table 2. Specific Patterns of Gene-Expression Signatures Associated With Recurrent Mutated Genes in AML Gene/Mutation of Interest No. of Patients Microarray Platform Results Impact on Prognosis Reference NPM1 148 de novo CN-AML, 60 U133 2.0 plus years; CALGB CEBPA, comparison of Analysis of a cohort of 41 U133 plus 2.0 double- v single- CEBPA mutation cases of mutated cases 598 AML CEBPA 285 AML FLT3-ITD RAS (N-RAS; K-RAS) NPM1 275 AML U133 A U133 A NPM1 CEBPA CEBPA 251 CN-AML U133 2.0 plus 175 CN-AML U133 2.0 plus FLT3-ITD Training set, 65 CN-AML; cDNA Stanford test set, 72 CN-AML NPM1 mutation associated with specific NPM1 mutation constitutes gene-expression and miRNA a marker defining a signatures, comparable in patients biologically homogeneous and 60 years of age, characterized entity in CN-AML; might be by upregulation of HOX genes and treated with specific and/or downregulation of MN1, BAALC, and targeted therapies across ERG. NPM1 mutation status was age groups. predicted in patients 60 years of age with 89.4% accuracy, 91.9% sensitivity, and 86.2% specificity. 28 CEBPA double- and 13 CEBPA CEBPA double-mutated cases single-mutated cases. CEBPA double- have specific gene- mutated cases were associated with expression patterns and a unique gene-expression profile distinct clinical outcomes. (sensitivity 100%, specificity 98%) as CEBPA-mutated AML is well as favorable OS and EFS. In outlined by significant contrast, CEBPA single-mutated AML heterogeneity with did not express a discriminating prognostic significance signature and could not be (double- v single-mutated distinguished from wild-type cases status). regarding clinical outcome. CEBPA had specific gene-expression pattern, predictable with 98% accuracy in a validation set. No data available. For FLT3-ITD no signature possible. For RAS no signature possible. Mutations in NPM1 had 100% positive predictive value and 80.3% negative predictive value by gene-expression analysis. Presence of the mutation was associated with revealed a strong HOX and TALE genespecific signature. Confirmation of favorable prognosis of the NPM1mut/FLT3-ITD negative subgroup. Prediction of underlying mutation status No data available. with high accuracy for NPM1 (95.6%) and CEBPA (93.6%) mutated cases. For NPM1-mutated CN-AML, a consensus 301-probe sets signature was observed, including expression of members of the HOXA and HOXB gene families. CEBPA-mutated patients had unique GEP and miRNA expression profiles; specific genes (eg, GATA1) were upregulated, associated with erythroid differentiation, and others downregulated (eg, homeobox genes). The favorable prognostic impact of CEBPA mutation (as determined by PCR analysis) was confirmed. No new data based on GEP only. An "FLT3 signature" composed by 20 New FLT3-ITD signature genes predicted FLT3-ITD status with could better define clinical 73% sensitivity and 85% specificity. outcome than FLT3-ITD The new FLT3 signature was a status as measured by significant predictor of relapse standard PCR and may (P .003) and unfavorable DFS better define FLT3 pathway (P .001) and OS (P .001) and activation. outperformed the FLT3-ITD status. 17 21 27 50 51 52 53 Abbreviations: AML, acute myeloid leukemia; CN, cytogenetically normal; CALGB, Cancer and Leukemia Group B; miRNA, microRNA; OS, overall survival; EFS, event-free survival; ITD, internal tandem duplication; GEP, gene-expression profile; PCR, polymerase chain reaction; DFS, disease-free survival. comprising subpopulations of nonleukemic cells. This approach is of a relatively simple and immediate application and has provided strong and reproducible signatures. However, to improve accuracy of diagnosis and molecular risk assessment, it is likely that future studies will capitalize on improving techniques to sort for minute populations of malignant blasts and/or leukemia initiating cells and will report only gene-expression signatures that are specific of these malignant cell compartments.66 DEREGULATED MICRORNA EXPRESSION IN AML MicroRNAs are naturally occurring noncoding RNAs that are cleaved from hairpin precursors and hybridize to imprecisely complementary mRNA of protein-coding genes, thereby leading to downregulation of the encoded proteins by RNA degradation or translation inhibition.67 Deregulation of microRNAs and in turn of their target genes have 8 2011 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY Information downloaded from jco.ascopubs.org and provided by at HAM-TMC LIBRARY on February 2, 2011 from Copyright 2011 American Soc1ie9t2y.6o8f .C3l0in.2ical Oncology. All rights reserved. Prognostic Markers in AML Table 3. Gene-Expression Profiles Associated With Prognosis in AML Genes of Interest As Revealed by Gene-Expression Analysis Unsupervised analysis identified clinical outcome predictor based on a 133-gene probe set; validation of the novel gene classifier in training and test group Unsupervised approach identified 16 subgroups based on molecular signatures. Clustering was driven by chromosomal lesions (eg, reciprocal rearrangements), molecular mutations (eg, CEBPA), or aberrant gene-expression (eg, EVI1) Supervised clustering, using genes depicted by Bullinger et al25 Supervised approach, followed by validation group Identification of a new prognostically relevant gene signature in CN-AML by unsupervised principal component analysis, followed by independent test set and validation set No. of Patients 116 AML (including 45 CN-AML) 285 AML 64 CN-AML 60 years; CALGB Cohort of 525 adult AML patients (one third youngest, median 31 years, n 175; compared with one third oldest, median 59 years, n 175) 163 CN-AML plus independent cohort of 79 CN-AML (both AMLCG), and validation in 64 CNAML (CALGB) Microarray Platform cDNA Stanford platform U133A U133 plus 2.0 U133A U133 AB and U113 plus 2.0 Results Identification of new subgroups in AML based on GEP, including two prognostically different groups in CN-AML Clusters driven by cytogenetics and molecular genetics including EVI1 expression; identification of novel clusters in normal karyotype; identification of a new cluster with poor outcome Impact on Prognosis Novel gene expressionbased prognostic predictor with a strong influence in CN-AML; prognostic impact as well in multivariate analysis, including parameters such as s-AML, cytogenetics, and FLT3 mutation status Diagnosis of AML and of several subtypes; prediction of prognosis based on GEP; definition of new clusters with a prognostic impact Confirmation of the discriminative gene signature from Bullinger et al25 with respect to OS (P .001) and DFS (P .001); correlation of the respective gene signature to the FLT3-ITD status, but moderate differences in survival also in FLT3 wild-type patients Identification of 41 probe sets regulating aging, including downregulation of tumorsuppressor gene CDKN2A (p16INK4A); confirmation of the prognostic impact only in the intermediate- and adverse-risk group, but not in favorable cytogenetic risk group and in NPM1mut/FLT3-ITD negative group A new score based on 86-probe sets was predictive for OS, EFS, and RFS in the test cohort. Confirmation of prognostic significance in multivariate analysis adjusted for age, FLT3-ITD, and NPM1 mutation status Confirmation of the Bullinger signature by an independent study group; should be further explored for improved prognostic predictions in CN-AML Older AML patients have different GEP; downregulation of CDKN2A (p16INK4A) is an independent prognostic parameter in AML besides cytogenetic/molecular genetic risk groups Definition of a new 86-probeset gene-expression signature predictive for prognosis in CN-AML Reference 26 27 56 57 58 Abbreviations: AML, acute myeloid leukemia; CN, cytogenetically normal; GEP, gene-expression profile; s-AML, secondary AML; CALGB, Cancer and Leukemia group B; OS, overall survival; DFS, disease-free survival; ITD, internal tandem duplication; EFS, event-free survival; AMLCG, German-Austrian AML Study Group. been found to contribute to malignant transformation in several human solid tumors and hematologic malignancies by interfering with critical steps of cell development, differentiation, proliferation, and apoptosis processes.68 Aberrant expression of multiple microRNAs have recently been reported in AML.4 Correlations With Morphologic, Cytogenetic, and Molecular Features Similar to the initial studies showing the GEP capability of distinguishing different types of acute leukemia, Mi et al69 showed that microRNA-expression profiling can separate patients with AML from those with ALL on the basis of a signature that included 21 upregulated and six downregulated microRNAs. Of these microRNAs, four--let7b, miR-128a, miR-128b, and miR-223--were the most informative in separating patients with AML from those with ALL. Furthermore, microRNA-expression profiling has been shown to be able to distinguish distinct cytogenetic subtypes of AML. Although complete concordance could not be demonstrated among the microRNA signatures reported by different studies of favorable cytogenetic risk subsets, upregulation of microRNAs transcribed from genes localized at chromosome band 14q32 in APL with t(15;17) and the downregulation of miR-133a in patients with t(8;21) were shared by at least two studies.69-73 Distinct patterns of microRNA expression have been correlated with trisomy 8, balanced chromosomal 11q23 rearrangements involving the MLL gene, and CN-AML.70,73 In AML with aberrations of 11q23/MLL studied by Li et al,70 microRNAs from a unique polycistronic microRNA cluster--miR-17-92--was found overexpressed, as well as miR-196b, located between the homeobox (HOX) A9 and HOXA10 genes at 7p15. www.jco.org 2011 by American Society of Clinical Oncology 9 Information downloaded from jco.ascopubs.org and provided by at HAM-TMC LIBRARY on February 2, 2011 from Copyright 2011 American Soc1ie9t2y.6o8f .C3l0in.2ical Oncology. All rights reserved. Marcucci, Haferlach, and Do hner MicroRNA-expression signatures have also been correlated with recurrent molecular aberrations in AML. NPM1 mutations associate with upregulation of miR-10a, miR-10b, and miR-196a, all of which reside in the genomic cluster of HOX genes that is found to be consistently overexpressed in this molecular subset.17,74 FLT3-ITD has been reported to be associated with miR-155 upregulation. Recent data from animal models showing that nuclear factor kappa B activation drives miR-155 expression, thereby leading to granulocyte/monocyte expansion perhaps by interfering with the phosphatase activity SH2 domain containing inositol phosphatase 1 (SHP1).75 CEBPA mutations have been associated with an upregulation of members of the miR-181 family in CN-AML.52 Notably, CEBPA mutation-related GEP comprised several upregulated genes involved in erythroid differentiation,65 in agreement with the results by Choong et al76 reporting an increase of miR-181a and miR-181b levels during erythroid differentiation. Consequently, it is reasonable to hypothesize that high expression levels of the members of the miR-181 family contribute to the partial erythroid differentiation of leukemic blasts harboring CEBPA mutations.52 Lastly, miR-181 expression was correlated with FrenchAmerican-British (FAB) morphologic groups in CN-AML, with their higher expression seen in FAB M1 and M2 in comparison with FAB M4 and M5. Also in CN-AML, miR-10a, miR-10b, and miR-196a-1 were correlated with expression of HOX genes.77 This is consistent with a high incidence of NPM1 mutations in CN-AML and the reported NPM1 mutation-associated gene-expression signature known to be related to HOX gene overexpression.3,4 Correlations With Clinical Outcome Recent studies have also shown that changes in microRNA expression can affect clinical outcome in AML. Low expression of miRlet7b and miR-9 was found in patients classified in the favorable-risk group, whereas higher expression of these microRNAs was detected in samples from patients with adverse- or intermediate-risk cytogenetic findings.71 Garzon et al73 reported that, across all cytogenetic subgroups, overexpressed miR-20a, miR-25, miR-191, miR-199a, and miR-199b adversely affected OS. A more recent study identified a microRNA-expression signature with prognostic significance in patients with CN-AML belonging to the molecular high-risk group, defined by the presence of FLT3-ITD, wild-type NPM1 alleles, or both.78 The signature comprised 12 microRNA probes and was associated with event-free survival. Five probes in the signature represented miR-181a and miR-181b; their increased expression was associated with a decreased risk of an event (failure to achieve CR, relapse, or death). Moreover, the genome-wide microRNAexpression profile was integrated with gene-expression signature with the goal of identifying genes regulated by microRNAs whose altered expression contributed to leukemogenesis in molecular high-risk CNAML.78 microRNA-derived gene-expression signature included, among others, genes encoding proteins involved in innate immunity that were previously shown to support proliferation and survival of malignant myeloid blasts.78 These data support a functional relationship between microRNA and gene expression and suggest that downregulation of the members of miR-181 family may contribute to the aggressive leukemia phenotype. On the other hand, high levels of miR-181 expression may reduce the aggressiveness of the disease. The latter was confirmed by a subsequent study showing that upregulated miR-181a predicted favorable outcome in CN-AML.79 In conclusion, it is clear that any complete prognostic assessment has to consider the aberrant genome and transcriptome changes occurring in AML. These results need also to be integrated with recurrent aberrant epigenetic changes (ie, DNA hypermethylation and histone posttranscriptional modifications) that deregulate the expression of genes involved in normal hematopoiesis and define biologic and prognostic epigenetic subsets of AML. Hypermethylated, silenced genes seem to play a relevant role in myeloid malignancies as both prognostic markers and therapeutic targets. Recently, Figueroa et al80 examined the methylation status of a relatively large cohort of patients with AML and reported unique methylation profiles for distinct cytogenetic and molecular AML subsets. Furthermore, a 15-gene methylation classifier was found to be predictive for survival, even after controlling for other molecular prognosticators (CEBPA, NPM1, FLT3 mutations).80 These data, therefore, support the important prognostic role of genome-wide epigenetic profiling and underscore the need for standardized analyses that integrate both genetic and epigenetic information to accurately predict for clinical outcome. To pursue integrated information, however, it is imperative that multicenter studies will collaborate in developing and validating novel bioinformatic algorithms that will be readily applicable to routine diagnostics, prognostication, and risk-stratification of patients with AML. Advanced next-generation studies of nucleic acid sequencing technologies are being currently implemented to improve the amount of informative data that can be collected and analyzed in real time, thereby providing a powerful tool for the clinicians for stratifying individual patients to risk-adapted treatment. Hopefully, these approaches will lead to more effective strategies that will be part of novel individualized therapies for patients with AML. The proof of concept for these approaches is provided by implementation of rapid molecular screening tests such as those that have allowed identification and accrual of FLT3-mutated patients with AML to multi-institutional clinical trials testing the therapeutic value of tyrosine kinase inhibitors within less than 48 hours from their initial presentation (eg, CALGB 10603). AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a "U" are those for which no compensation was received; those relationships marked with a "C" were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors. Employment or Leadership Position: Torsten Haferlach, Munich Leukemia Laboratory (MLL). (C) Consultant or Advisory Role: None Stock Ownership: None Honoraria: None Research Funding: Torsten Haferlach, ROCHE Diagnostics Expert Testimony: None Other Remuneration: None AUTHOR CONTRIBUTIONS Conception and design: Guido Marcucci, Torsten Haferlach, Hartmut Dohner Financial support: Guido Marcucci, Torsten Haferlach, Hartmut Dhner 10 2011 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY Information downloaded from jco.ascopubs.org and provided by at HAM-TMC LIBRARY on February 2, 2011 from Copyright 2011 American Soc1ie9t2y.6o8f .C3l0in.2ical Oncology. All rights reserved. Prognostic Markers in AML Administrative support: Guido Marcucci, Torsten Haferlach, Hartmut Dhner Provision of study materials or patients: Guido Marcucci, Torsten Haferlach, Hartmut Dhner Collection and assembly of data: Guido Marcucci, Torsten Haferlach, Hartmut Dohner Data analysis and interpretation: Guido Marcucci, Torsten Haferlach, Hartmut Dhner Manuscript writing: Guido Marcucci, Torsten Haferlach, Hartmut Dohner Final approval of manuscript: Guido Marcucci, Torsten Haferlach, Hartmut Dohner REFERENCES 1. Fro hling S, Scholl C, Gilliland DG, et al: Genetics of myeloid malignancies: Pathogenetic and clinical implications. J Clin Oncol 23:6285-6295, 2005 2. Do hner K, Do hner H: Molecular characterization of acute myeloid leukemia. Haematologica 93: 976-982, 2008 3. Mro zek K, Marcucci G, Paschka P, et al: Clinical relevance of mutations and gene-expression changes in adult acute myeloid leukemia with normal cytogenetics: Are we ready for a prognostically prioritized molecular classification? Blood 109:431448, 2007 4. Mro zek K, Radmacher MD, Bloomfield CD, et al: Molecular signatures in acute myeloid leukemia. Curr Opin Hematol 16:64-69, 2009 5. Mardis ER, Ding L, Dooling DJ, et al: Recurring mutations found by sequencing an acute myeloid leukemia genome. N Engl J Med 361:10581066, 2009 6. Do hner H, Estey EH, Amadori S, et al: Diagnosis and management of acute myeloid leukemia in adults: Recommendations from an international expert panel, on behalf of the European LeukemiaNet. Blood 115:453-474, 2010 7. Swerdlow SH, Campo E, Harris NL, et al: WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France, IARC Press, 2008 8. Falini B, Mecucci C, Tiacci E, et al: Cytoplasmic nucleophosmin in acute myelogenous leukemia with a normal karyotype. N Engl J Med 352:254-266, 2005 9. Haferlach C, Mecucci C, Schnittger S, et al: AML with mutated NPM1 carrying a normal or aberrant karyotype show overlapping biologic, pathologic, immunophenotypic, and prognostic features. Blood 114:3024-3032, 2009 10. Schlenk RF, Do hner K, Krauter J, et al: Mutations and treatment outcome in cytogenetically normal acute myeloid leukemia. N Engl J Med 358: 1909-1918, 2008 11. Marcucci G, Maharry K, Wu YZ, et al: IDH1 and IDH2 gene mutations identify novel molecular subsets within de novo cytogenetically normal acute myeloid leukemia: A Cancer and Leukemia Group B Study. J Clin Oncol 28:2348-2355, 2010 12. Paschka P, Schlenk RF, Gaidzik VI, et al: IDH1 and IDH2 mutations are frequent genetic alterations in acute myeloid leukemia (AML) and confer adverse prognosis in cytogenetically normal AML with NPM1 mutation without FLT3-ITD. J Clin Oncol 28:3636-3643, 2010 13. Boissel N, Nibourel O, Renneville A, et al: Prognostic impact of isocitrate dehydrogenase enzyme isoforms 1 (IDH1) and 2 (IDH2) mutations in acute myeloid leukemia: A study by the Acute Leukemia French Association (ALFA) group. J Clin Oncol 28:3717-3723, 2010 14. Wagner K, Damm F, Go hring G, et al: Impact of IDH1 R132 mutations and an IDH1 single nucleotide polymorphism in cytogenetically normal acute myeloid leukemia: SNP rs11554137 is an adverse prognostic factor. J Clin Oncol 28:2356-2364, 2010 15. Schnittger S, Haferlach C, Ulke M, et al: IDH1 mutations are detected in 6.6% of 1414 AML patients and are associated with intermediate risk karyotype and unfavorable prognosis in adults younger than 60 years and unmutated NPM1 status. Blood 10.1182/blood-2010-02-267955 [epub ahead of print on August 30, 2010] 16. Schlenk RF, Do hner K, Kneba M, et al: Gene mutations and response to treatment with all-trans retinoic acid in elderly patients with acute myeloid leukemia: Results from AMLSG trial AML HD98B. Haematologica 94:54-60, 2009 17. Becker H, Marcucci G, Maharry K, et al: Favorable prognostic impact of NPM1 mutations in older patients with cytogenetically normal de novo acute myeloid leukemia and associated gene- and microRNA-expression signatures: A Cancer and Leukemia Group B study. J Clin Oncol 28:596-604, 2010 18. Bu chner T, Berdel WE, Haferlach C, et al: Age-related risk profile and chemotherapy dose response in acute myeloid leukemia: A study by the German Acute Myeloid Leukemia Cooperative Group. J Clin Oncol 27:61-69, 2009 19. Burnett AK, Hills RK, Green C, et al: The impact on outcome of the addition of all-trans retinoic acid to intensive chemotherapy in younger patients with nonacute promyelocytic acute myeloid leukemia: Overall results and results in genotypic subgroups defined by mutations in NPM1, FLT3, and CEBPA. Blood 115:948-956, 2010 20. Koschmieder S, Halmos B, Levantini E, et al: Dysregulation of the C/EBPalpha differentiation pathway in human cancer. J Clin Oncol 27:619-628, 2009 21. Wouters BJ, Lo wenberg B, ErpelinckVerschueren CAJ, et al: Double CEBPA mutations, but not single CEBPA mutations, define a subgroup of acute myeloid leukemia with a distinctive gene expression profile that is uniquely associated with a favorable outcome. Blood 113:3088-3091, 2009 22. Bereshchenko O, Mancini E, Moore S, et al: Hematopoietic stem cell expansion precedes the generation of committed myeloid leukemia-initiating cells in C/EBPalpha mutant AML. Cancer Cell 16: 390-400, 2009 23. Breitenbuecher F, Schnittger S, Grundler R, et al: Identification of a novel type of ITD mutations located in nonjuxtamembrane domains of the FLT3 tyrosine kinase receptor. Blood 113:4074-4077, 2009 24. Kayser S, Schlenk RF, Londono MC, et al: Insertion of FLT3 internal tandem duplication in the tyrosine kinase domain-1 is associated with resistance to chemotherapy and inferior outcome. Blood 114:2386-2392, 2009 25. Paschka P: Core binding factor acute myeloid leukemia. Semin Oncol 35:410-417, 2008 26. Bullinger L, Do hner K, Bair E, et al: Use of gene-expression profiling to identify prognostic subclasses in adult acute myeloid leukemia. N Engl J Med 350:1605-1616, 2004 27. Valk PJM, Verhaak RGW, Beijen MA, et al: Prognostically useful gene-expression profiles in acute myeloid leukemia. N Engl J Med 350:16171628, 2004 28. Ward PS, Patel J, Wise DR, et al: The common feature of leukemia-associated IDH1 and IDH2 mutations is a neomorphic enzyme activity converting alpha-ketoglutarate to 2-hydroxyglutarate. Cancer Cell 17:225-234, 2010 29. Paschka P, Marcucci G, Ruppert AS, et al: Wilms' tumor 1 gene mutations independently predict poor outcome in adults with cytogenetically normal acute myeloid leukemia: A cancer and leukemia group B study. J Clin Oncol 26:4595-4602, 2008 30. Virappane P, Gale R, Hills R, et al: Mutation of the Wilms' tumor 1 gene is a poor prognostic factor associated with chemotherapy resistance in normal karyotype acute myeloid leukemia: The United Kingdom Medical Research Council Adult Leukaemia Working Party. J Clin Oncol 26:5429-5435, 2008 31. Gaidzik VI, Schlenk RF, Moschny S, et al: Prognostic impact of WT1 mutations in cytogenetically normal acute myeloid leukemia: A study of the German-Austrian AML Study Group. Blood 113: 4505-4511, 2009 32. Damm F, Heuser M, Morgan M, et al: Single nucleotide polymorphism in the mutational hotspot of WT1 predicts a favorable outcome in cytogenetically normal acute myeloid leukemia. J Clin Oncol 28:578-585, 2010 33. Tang JL, Hou HA, Chen CY, et al: AML1/ RUNX1 mutations in 470 adult patients with de novo acute myeloid leukemia: Prognostic implication and interaction with other gene alterations. Blood 114: 5352-5361, 2009 34. Neubauer A, Maharry K, Mro zek K, et al: Patients with acute myeloid leukemia and RAS mutations benefit most from postremission highdose cytarabine: A Cancer and Leukemia Group B study. J Clin Oncol 26:4603-4609, 2008 35. Haferlach C, Dicker F, Herholz H, et al: Mutations of the TP53 gene in acute myeloid leukemia are strongly associated with a complex aberrant karyotype. Leukemia 22:1539-1541, 2008 36. Delhommeau F, Dupont S, Della Valle V, et al: Mutation in TET2 in myeloid cancers. N Engl J Med 360:2289-2301, 2009 37. Langemeijer SMC, Kuiper RP, Berends M, et al: Acquired mutations in TET2 are common in myelodysplastic syndromes. Nat Genet 41:838-842, 2009 38. Abdel-Wahab O, Mullally A, Hedvat C, et al: Genetic characterization of TET1, TET2, and TET3 alterations in myeloid malignancies. Blood 114:144147, 2009 39. Nibourel O, Kosmider O, Cheok M, et al: Incidence and prognostic value of TET2 alterations in de novo acute myeloid leukemia achieving complete remission. Blood 116:1132-1135, 2010 40. Carbuccia N, Trouplin V, Gelsi-Boyer V, et al: Mutual exclusion of ASXL1 and NPM1 mutations in a series of acute myeloid leukemias. Leukemia 24:469-473, 2010 41. Baldus CD, Tanner SM, Ruppert AS, et al: BAALC expression predicts clinical outcome of de novo acute myeloid leukemia patients with normal cytogenetics: A Cancer and Leukemia Group B Study. Blood 102:1613-1618, 2003 www.jco.org 2011 by American Society of Clinical Oncology 11 Information downloaded from jco.ascopubs.org and provided by at HAM-TMC LIBRARY on February 2, 2011 from Copyright 2011 American Soc1ie9t2y.6o8f .C3l0in.2ical Oncology. All rights reserved. Marcucci, Haferlach, and Do hner 42. Langer C, Radmacher MD, Ruppert AS, et al: High BAALC expression associates with other molecular prognostic markers, poor outcome, and a distinct gene-expression signature in cytogenetically normal patients younger than 60 years with acute myeloid leukemia: A Cancer and Leukemia Group B (CALGB) study. Blood 111:5371-5379, 2008 43. Metzeler KH, Dufour A, Benthaus T, et al: ERG expression is an independent prognostic factor and allows refined risk stratification in cytogenetically normal acute myeloid leukemia: A comprehensive analysis of ERG, MN1, and BAALC transcript levels using oligonucleotide microarrays. J Clin Oncol 27:5031-5038, 2009 44. Heuser M, Argiropoulos B, Kuchenbauer F, et al: MN1 overexpression induces acute myeloid leukemia in mice and predicts ATRA resistance in patients with AML. Blood 110:1639-1647, 2007 45. Langer C, Marcucci G, Holland KB, et al: Prognostic importance of MN1 transcript levels, and biologic insights from MN1-associated gene and microRNA expression signatures in cytogenetically normal acute myeloid leukemia: A Cancer and Leukemia Group B study. J Clin Oncol 27:3198-3204, 2009 46. Marcucci G, Maharry K, Whitman SP, et al: High expression levels of the ETS-related gene, ERG, predict adverse outcome and improve molecular risk-based classification of cytogenetically normal acute myeloid leukemia: A Cancer and Leukemia Group B Study. J Clin Oncol 25:3337-3343, 2007 47. Lugthart S, van Drunen E, van Norden Y, et al: High EVI1 levels predict adverse outcome in acute myeloid leukemia: Prevalence of EVI1 overexpression and chromosome 3q26 abnormalities underestimated. Blood 111:4329-4337, 2008 48. Gro schel S, Lugthart S, Schlenk RF, et al: High EVI1 expression predicts outcome in younger adult patients with acute myeloid leukemia and is associated with distinct cytogenetic abnormalities. J Clin Oncol 28:2101-2117, 2010 49. Golub TR, Slonim DK, Tamayo P, et al: Molecular classification of cancer: Class discovery and class prediction by gene expression monitoring. Science 286:531-537, 1999 50. Verhaak RG, Goudswaard CS, van Putten W, et al: Mutations in nucleophosmin (NPM1) in acute myeloid leukemia (AML): Association with other gene abnormalities and previously established gene expression signatures and their favorable prognostic significance. Blood 106:3747-3754, 2005 51. Kohlmann A, Bullinger L, Thiede C, et al: Gene expression profiling in AML with normal karyotype can predict mutations for molecular markers and allows novel insights into perturbed biological pathways. Leukemia 24:1216-1220, 2010 52. Marcucci G, Maharry K, Radmacher MD, et al: Prognostic significance of, and gene and microRNA expression signatures associated with, CEBPA mutations in cytogenetically normal acute myeloid leukemia with high-risk molecular features: A Cancer and Leukemia Group B Study. J Clin Oncol 26:50785087, 2008 53. Bullinger L, Do hner K, Kranz R, et al: An FLT3 gene-expression signature predicts clinical outcome in normal karyotype AML. Blood 111:4490-4495, 2008 54. Haferlach T, Kohlmann A, Schnittger S, et al: Global approach to the diagnosis of leukemia using gene expression profiling. Blood 106:1189-1198, 2005 55. Haferlach T, Kohlmann A, Wieczorek L, et al: Clinical utility of microarray-based gene expression profiling in the diagnosis and subclassification of leukemia: Report from the International Microarray Innovations in Leukemia Study Group. J Clin Oncol 28:2529-2537, 2010 56. Radmacher MD, Marcucci G, Ruppert AS, et al: Independent confirmation of a prognostic geneexpression signature in adult acute myeloid leukemia with a normal karyotype: A Cancer and Leukemia Group B study. Blood 108:1677-1683, 2006 57. de Jonge HJ, de Bont ES, Valk PJ, et al: AML at older age: Age-related gene expression profiles reveal a paradoxical down-regulation of p16INK4A mRNA with prognostic significance. Blood 114: 2869-2877, 2009 58. Metzeler KH, Hummel M, Bloomfield CD, et al: An 86 probe set gene expression signature predicts survival in cytogenetically normal acute myeloid leukemia. Blood 112:4193-4201, 2008 59. Staal FJ, Cario G, Cazzaniga G, et al: Consensus guidelines for microarray gene expression analyses in leukemia from three European leukemia networks. Leukemia 20:1385-1392, 2006 60. Kohlmann A, Kipps TJ, Rassenti LZ, et al: An international standardization programme towards the application of gene expression profiling in routine leukaemia diagnostics: The Microarray Innovations in Leukemia study prephase. Br J Haematol 142:802-807, 2008 61. Shi L, Reid LH, Jones WD, et al: The MicroArray Quality Control (MAQC) project shows inter- and intraplatform reproducibility of gene expression measurements. Nat Biotechnol 24:1151-1161, 2006 62. Schoch C, Kohlmann A, Schnittger S, et al: Acute myeloid leukemias with reciprocal rearrangements can be distinguished by specific gene expression profiles. Proc Natl Acad Sci U S A 99:10008-10013, 2002 63. Kohlmann A, Schoch C, Dugas M, et al: New insights into MLL gene rearranged acute leukemias using gene expression profiling: Shared pathways, lineage commitment, and partner genes. Leukemia 19:953-964, 2005 64. Schoch C, Kern W, Kohlmann A, et al: Acute myeloid leukemia with a complex aberrant karyotype is a distinct biological entity characterized by genomic imbalances and a specific gene expression profile. Genes Chromosomes Cancer 43:227-238, 2005 65. Ru cker FG, Bullinger L, Schwaenen C, et al: Disclosure of candidate genes in acute myeloid leukemia with complex karyotypes using microarraybased molecular characterization. J Clin Oncol 24:38873894, 2006 66. Steidl U, Rosenbauer F, Verhaak RG, et al: Essential role of Jun family transcription factors in PU.1 knockdown-induced leukemic stem cells. Nat Genet 38:1269-1277, 2006 67. Bartel DP: MicroRNAs: Target recognition and regulatory functions. Cell 136:215-233, 2009 68. Calin GA, Croce CM: MicroRNA signatures in human cancers. Nat Rev Cancer 6:857-866, 2006 69. Mi S, Lu J, Sun M, et al: MicroRNA expression signatures accurately discriminate acute lymphoblastic leukemia from acute myeloid leukemia. Proc Natl Acad Sci U S A 104:19971-19976, 2007 70. Li Z, Lu J, Sun M, et al: Distinct microRNA expression profiles in acute myeloid leukemia with common translocations. Proc Natl Acad Sci U S A 105:15535-15540, 2008 71. Dixon-McIver A, East P, Mein CA, et al: Distinctive patterns of microRNA expression associated with karyotype in acute myeloid leukaemia. PLoS One 3:e2141, 2008 72. Jongen-Lavrencic M, Sun SM, Dijkstra MK, et al: MicroRNA expression profiling in relation to the genetic heterogeneity of acute myeloid leukemia. Blood 111:5078-5085, 2008 73. Garzon R, Volinia S, Liu C-G, et al: MicroRNA signatures associated with cytogenetics and prognosis in acute myeloid leukemia. Blood 111:31833189, 2008 74. Garzon R, Garofalo M, Martelli MP, et al: Distinctive microRNA signature of acute myeloid leukemia bearing cytoplasmic mutated nucleophosmin. Proc Natl Acad Sci U S A 105:3945-3950, 2008 75. O'Connell RM, Chaudhuri AA, Rao DS, et al: Inositol phosphatase SHIP1 is a primary target of miR-155. Proc Natl Acad Sci U S A 106:7113-7118, 2009 76. Choong ML, Yang HH, McNiece I: MicroRNA expression profiling during human cord bloodderived CD34 cell erythropoiesis. Exp Hematol 35: 551-564, 2007 77. Debernardi S, Skoulakis S, Molloy G, et al: MicroRNA miR-181a correlates with morphological sub-class of acute myeloid leukaemia and the expression of its target genes in global genome-wide analysis. Leukemia 21:912-916, 2007 78. Marcucci G, Radmacher MD, Maharry K, et al: MicroRNA expression in cytogenetically normal acute myeloid leukemia. N Engl J Med 358:19191928, 2008 79. Schwind S, Maharry K, Radmacher MD, et al: Prognostic significance of expression of a single microRNA, miR-181a, in cytogenetically normal acute myeloid leukemia: A Cancer and Leukemia Group B Study. J Clin Oncol 28:5257-5264, 2010 80. Figueroa ME, Lugthart S, Li Y, et al: DNA methylation signatures identify biologically distinct subtypes in acute myeloid leukemia. Cancer Cell 17:13-27, 2010 12 2011 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY Information downloaded from jco.ascopubs.org and provided by at HAM-TMC LIBRARY on February 2, 2011 from Copyright 2011 American Soc1ie9t2y.6o8f .C3l0in.2ical Oncology. All rights reserved.