SUMMARY We hypothesized that DNA methylation distributes into specific patterns in cancer cells, which reflect critical biological differences. We therefore examined the methylation profiles of 344 patients with acute myeloid leukemia (AML). Clustering of these patients by methylation data segregated patients into 16 groups. Five of these groups defined new AML subtypes that shared no other known feature. In addition, DNA methylation profiles segregated patients with CEBPA aberrations from other subtypes of leukemia, defined four epigenetically distinct forms of AML with NPM1 mutations, and showed that established AML1-ETO, CBFb-MYH11, and PML-RARA leukemia entities are associated with specific methylation profiles. We report a 15 gene methylation classifier predictive of overall survival in an independent patient cohort (p < 0.001, adjusted for known covariates).

© F e r r a t a S t o r t i F o u n d a t i o nR.G.W. Verhaak et al. between both AML sample populations. Mutational analyses to detect recurrent mutations in AML were performed as previously described. [13][14][15][16] All supervised class prediction analyses were performed with Prediction Analysis for Microarrays (PAM) software version 1.28 in R version 2.1.0. 17 Clinical, cytogenetic and molecular information as well as the gene expression profiles of all primary AML cases is available at the Gene Expression Omnibus (www.ncbi.nlm.nih.gov/geo, accession number GSE6891). Results and DiscussionIn this study of 461 clinically and molecularly well-characterized cases of AML (Table 1), we were able to comprehensively validate the application of GEP to predict therapeutically relevant molecular subtypes in AML.We applied PAM to investigate whether karyotypic and mutational abnormalities with prognostic or therapeutic value in AML were accurately predictable based on GEP. PAM allows the selection of the minimal number of genes required for optimal prediction, which may be beneficial in a diagnostic setting. The AML cohort1 (n=247) was used as training set to derive predictive signatures that were subsequently validated on AML cohort2 (n=214). The deduced expression signatures are available in the Online Supplementary Tables S1-18.The cytogenetic status of all AML patients with favorable risk, i.e. those with t(8;21), t(15;17) or inv(16) abnormalities, was predicted with 100 percent accuracy ( Table 2). In fact, among these predicted AML cases, there were cases with favorable cytogenetics that had previously been missed by routine cytogenetics (4 out of 37 inv(16) and 4 out of 25 t(15;17)). The presence of the translocation-related fusion transcripts in these specific cases was confirmed by real-time quantitative PCR. Thus, GEP is a reliable alternative to discriminate these three AML subtypes, 2,3 which represent approximately 20% of all cases.2,3 Prediction of t(15;17) and inv(16) required only few genes, as seen previously. 8 For the t(8;21) cases, 76 probe sets were needed to correctly classify all samples. However, as few as two probe sets, including one associated with the RUNX1T1 (ETO) gene, were sufficient to accurately classify all but one t(8;21) cases, which is also consistent with earlier studies 8 (Online Supplementary Figure S3). AML cases with mutations in the transcription factor CCAAT/enhancer binding protein α (CEBPA), which are associated with a relatively favorable treatment outcome, were predicted with positive and negative predictive values of 100% and 97% respectively. Six out of 15 CEBPA mutant cases were missed in the validation set (sensitivity 60%; Table 2). Of note, the misclassified cases all carried a single heterozygous CEBPA mutation, whereas samples with biallelic mutations (either homo-or heterozygous) were all correctly recognized (data not shown). In the training cohort, all but two (14/16) samples carried biallelic mutations 14,18 and in cross-validation in the training cohort ...

Somatic mutations in isocitrate dehydrogenase 1 and 2 (IDH1 and IDH2) were recently demonstrated in acute myeloid leukemia (AML), but their prevalence and prognostic impact remain to be explored in large extensively characterized AML series, and also in various other hematologic malignancies. Here, we demonstrate in 893 newly diagnosed cases of AML mutations in the IDH1 (6%) and IDH2 (11%) genes. Moreover, we identified IDH mutations in 2 JAK2 V617F myeloproliferative neoplasias (n ‫؍‬ 96), a single case of acute lymphoblastic leukemia (n ‫؍‬ 96), and none in chronic myeloid leukemias (n ‫؍‬ 81). In AML, IDH1 and IDH2 mutations are more common among AML with normal karyotype and NPM1 mutant genotypes. IDH1 mutation status is an unfavorable prognostic factor as regards survival in a composite genotypic subset lacking IntroductionSomatic mutations in the genes encoding the isocitrate dehydrogenases IDH1 and IDH2 were revealed in more than 70% of World Health Organization grade 2 and 3 astrocytomas, oligodendrogliomas, and glioblastomas. [1][2][3] Mutations in IDH1 and IDH2 were mutually exclusive and affected the arginines on position 132 of IDH1 and position 172 of IDH2. 3 Patients with malignant gliomas with IDH1 or IDH2 mutations showed a better response to therapy than those with wild-type IDH genes. 3 Mutations in these residues of IDH significantly disturb the function of both isocitrate dehydrogenases, as demonstrated by impaired production of nicotinamide adenine dinucleotide phosphate. 3,4 In acute myeloid leukemia (AML), mutant IDH enzyme activity results in accumulation of the cancer-associated metabolite 2-hydroxyglutarate. 5,6 Recently, acquired mutations in the gene encoding IDH1 were identified in 8% 7 and 5.5% 8 of newly diagnosed AML cases. IDH1 mutations were significantly associated with normal karyotype and NPM1 mutations. 7,8 Overall, the IDH1 mutation status did not suggest a relationship with overall survival (OS), but the sample sizes were limited in these studies. 7,8 However, a trend for an adverse effect on OS was suggested in normal karyotype AML with NPM1 wild-type . 7 The prevalence and prognostic value of IDH mutations in AML, as well as other hematologic malignancies, remain to be further established. In this study, we determined the frequencies of both IDH1 and IDH2 mutations in cohorts of AML, acute lymphoblastic leukemia (ALL), chronic myeloid leukemia (CML), and JAK2 V617F myeloproliferative neoplasia (MPN). In a cohort of 893 cases of AML, we investigated their distribution in relationship with cytogenetic and molecular risk categories as well as recurrent gene mutations commonly apparent in AML, and we evaluated the impact of IDH mutations on treatment outcome. MethodsBone marrow aspirates or peripheral blood samples of cohorts of patients with various hematologic malignancies were collected after written informed consent in accordance with the Declaration of Helsinki. All experiments described were approved by the Erasmus University Medical Center Institutional Review Board....

Inappropriate expression of EVI1 (ecotropic virus integration-1), in particular splice form EVI1-1D, through chromosome 3q26 lesions or other mechanisms has been implicated in the development of high-risk acute myeloid leukemia (AML). To validate the clinical relevance of EVI1-1D, as well as of the other EVI1 splice forms and the related MDS1/EVI1 (ME) gene, real-time quantitative polymerase chain reaction was performed in 534 untreated adults with de novo AML. EVI1-1D was highly expressed in 6% of cases (n ‫؍‬ 32), whereas 7.8% were EVI1 ؉ (n ‫؍‬ 41) when all splice variants were taken into account. High EVI1 predicted a distinctly worse event-free survival (HR ‫؍‬ 1.9; P ‫؍‬ .002) and disease-free survival (HR ‫؍‬ 2.1, P ‫؍‬ .006) following multivariate analysis. Importantly, we distinguished a subset of EVI1 ؉ cases that lacked expression of ME (EVI1 ϩ ME Ϫ ; n ‫؍‬ 17) from cases that were ME ؉ (EVI1 ϩ ME ϩ ; n ‫؍‬ 24). The atypical EVI1 ϩ ME Ϫ expression pattern exhibited cytogenetically detectable chromosomal 3q26 breakpoints in 8 cases. Fluorescence in situ hybridization revealed 7 more EVI1 ϩ ME Ϫ cases that carried cryptic 3q26 breakpoints, which were not found in the EVI1 ϩ ME ϩ group. EVI1 ϩ ME Ϫ expression predicts an extremely poor prognosis distinguishable from the general EVI1 ϩ AML patients (overall survival

PURPOSE The purpose of this study was to investigate frequency and prognostic significance of high EVI1 expression in acute myeloid leukemia (AML). PATIENTS AND METHODS A diagnostic assay detecting multiple EVI1 splice variants was developed to determine the relative EVI1 expression by single real-time quantitative polymerase chain reaction in 1,382 newly diagnosed adult patients with AML younger than 60 years. Patients were treated on four Dutch-Belgian HOVON (n = 458) and two German-Austrian AML Study Group protocols (n = 924). Results The EVI1 assay was tested in the HOVON cohort and validated in the AMLSG cohort. High EVI1 levels (EVI1(+)) were found with similar frequencies in both cohorts combined, with a 10.7% incidence (148 of 1,382). EVI1(+) independently predicted low complete remission (CR) rate (odds ratio, 0.54; P = .002), adverse relapse-free survival (RFS; hazard ratio [HR], 1.32; P = .05), and event-free survival (EFS; HR, 1.46; P < .001). This adverse prognostic impact was more pronounced in the intermediate cytogenetic risk group (EFS; HR, 1.64; P < .001; and RFS; HR, 1.55; P = .02), and was also apparent in cytogenetically normal AML (EFS; HR, 1.67; P = .008). Besides inv(3)/t(3;3), EVI1(+) was significantly associated with chromosome abnormalities monosomy 7 and t(11q23), conferring prognostic impact within these two cytogenetic subsets. EVI1(+) was virtually absent in favorable-risk AML and AML with NPM1 mutations. Patients with EVI1(+) AML (n = 28) who received allogeneic stem cell transplantation in first CR had significantly better 5-year RFS (33% +/- 10% v 0%). CONCLUSION EVI1 expression in AML is unequally distributed in cytogenetic subtypes. It predicts poor outcome, particularly among intermediate cytogenetic risk AML. Patients with EVI1(+) AML may benefit from allogeneic transplantation in first CR. Pretreatment EVI1 screening should be included in risk stratification.