In patients with acute myeloid leukemia (AML) a combination of methods must be used to classify the disease, make therapeutic decisions, and determine the prognosis. However, this combined approach provides correct therapeutic and prognostic information in only 50 percent of cases. We determined the gene-expression profiles in samples of peripheral blood or bone marrow from 285 patients with AML using Affymetrix U133A GeneChips containing approximately 13,000 unique genes or expression-signature tags. Data analyses were carried out with Omniviz, significance analysis of microarrays, and prediction analysis of microarrays software. Statistical analyses were performed to determine the prognostic significance of cases of AML with specific molecular signatures. Unsupervised cluster analyses identified 16 groups of patients with AML on the basis of molecular signatures. We identified the genes that defined these clusters and determined the minimal numbers of genes needed to identify prognostically important clusters with a high degree of accuracy. The clustering was driven by the presence of chromosomal lesions (e.g., t(8;21), t(15;17), and inv(16)), particular genetic mutations (CEBPA), and abnormal oncogene expression (EVI1). We identified several novel clusters, some consisting of specimens with normal karyotypes. A unique cluster with a distinctive gene-expression signature included cases of AML with a poor treatment outcome. Gene-expression profiling allows a comprehensive classification of AML that includes previously identified genetically defined subgroups and a novel cluster with an adverse prognosis.

Patients with acute myeloid leukemia (AML) often reach complete remission, but relapse rates remain high. Next-generation sequencing enables the detection of molecular minimal residual disease in virtually every patient, but its clinical value for the prediction of relapse has yet to be established.

Substantial heterogeneity within mutant TP53 acute myeloid leukemia (AML) and myelodysplastic syndrome with excess of blast (MDS-EB) precludes the exact assessment of prognostic impact for individual patients. We performed in-depth clinical and molecular analysis of mutant TP53 AML and MDS-EB to dissect the molecular characteristics in detail and determine its impact on survival. We performed next-generation sequencing on 2200 AML/MDS-EB specimens and assessed the TP53 mutant allelic status (mono- or bi-allelic), the number of TP53 mutations, mutant TP53 clone size, concurrent mutations, cytogenetics, and mutant TP53 molecular minimal residual disease and studied the associations of these characteristics with overall survival. TP53 mutations were detected in 230 (10.5%) patients with AML/MDS-EB with a median variant allele frequency of 47%. Bi-allelic mutant TP53 status was observed in 174 (76%) patients. Multiple TP53 mutations were found in 49 (21%) patients. Concurrent mutations were detected in 113 (49%) patients. No significant difference in any of the aforementioned molecular characteristics of mutant TP53 was detected between AML and MDS-EB. Patients with mutant TP53 have a poor outcome (2-year overall survival, 12.8%); however, no survival difference between AML and MDS-EB was observed. Importantly, none of the molecular characteristics were significantly associated with survival in mutant TP53 AML/MDS-EB. In most patients, TP53 mutations remained detectable in complete remission by deep sequencing (73%). Detection of residual mutant TP53 was not associated with survival. Mutant TP53 AML and MDS-EB do not differ with respect to molecular characteristics and survival. Therefore, mutant TP53 AML/MDS-EB should be considered a distinct molecular disease entity.

We examined the gene expression profiles of two independent cohorts of patients with acute myeloid leukemia [n=247 and n=214 (younger than or equal to 60 years)] to study the applicability of gene expression profiling as a single assay in prediction of acute myeloid leukemia-specific molecular subtypes. The favorable cytogenetic acute myeloid leukemia subtypes, i.e., acute myeloid leukemia with t(8;21), t(15;17) or inv(16), were predicted with maximum accuracy (positive and negative predictive value: 100%). Mutations in NPM1 and CEBPA were predicted less accurately (positive predictive value: 66% and 100%, and negative predictive value: 99% and 97% respectively). Various other characteristic molecular acute myeloid leukemia subtypes, i.e., mutant FLT3 and RAS, abnormalities involving 11q23, -5/5q-, -7/7q-, abnormalities involving 3q (abn3q) and t(9;22), could not be correctly predicted using gene expression profiling. In conclusion, gene expression profiling allows accurate prediction of certain acute myeloid leukemia subtypes, e.g. those characterized by expression of chimeric transcription factors. However, detection of mutations affecting signaling molecules and numerical abnormalities still requires alternative molecular methods.

Somatic mutations in the additional sex comb-like 1 (ASXL1) gene have been described in various types of myeloid malignancies, including acute myeloid leukemia. Analysis of novel markers, such as ASXL1 mutations, in independent clinical trials is indispensable before considering them for clinical decision-making. We analyzed 882 well-characterized acute myeloid leukemia cases to determine the prevalence and prognostic impact of ASXL1 exon12 mutations. Truncating ASXL1 mutations were present in 46 cases (5.3%). ASXL1 mutations were inversely associated with FLT3 internal tandem duplications and mutually exclusive with NPM1 mutations. ASXL1 mutations were an unfavorable prognostic factor as regards survival (median overall survival 15.9 months vs. 22.3 months; P=0.019), with a significantly lower complete response rate (61% vs. 79.6%; P=0.004). In multivariate analyses, ASXL1 mutations were independently associated with inferior poor overall survival (HR 1.52, P=0.032). In conclusion, ASXL1 mutations are common mutations in acute myeloid leukemia and indicate a poor therapy outcome.

PURPOSE: The applicability of FLT3-internal tandem duplications ( FLT3-ITD) for assessing measurable residual disease (MRD) in acute myeloid leukemia (AML) in complete remission (CR) has been hampered by patient-specific duplications and potential instability of FLT3-ITD during relapse. Here, we comprehensively investigated the impact of next-generation sequencing (NGS)–based FLT3-ITD MRD detection on treatment outcome in a cohort of patients with newly diagnosed AML in relation to established prognostic factors at diagnosis and other MRD measurements, ie, mutant NPM1 and multiparameter flow cytometry. METHODS: In 161 patients with de novo FLT3-ITD AML, NGS was performed at diagnosis and in CR after intensive remission induction treatment. FLT3-ITD MRD status was correlated with the cumulative incidence of relapse and overall survival (OS). RESULTS: NGS-based FLT3-ITD MRD was present in 47 of 161 (29%) patients with AML. Presence of FLT3-ITD MRD was associated with increased risk of relapse (4-year cumulative incidence of relapse, 75% FLT3-ITD MRD v 33% no FLT3-ITD MRD; P < .001) and inferior OS (4-year OS, 31% FLT3-ITD MRD v 57% no FLT3-ITD MRD; P < .001). In multivariate analysis, detection of FLT3-ITD MRD in CR confers independent prognostic significance for relapse (hazard ratio, 3.55; P < .001) and OS (hazard ratio 2.51; P = .002). Strikingly, FLT3-ITD MRD exceeds the prognostic value of most generally accepted clinical and molecular prognostic factors, including the FLT3-ITD allelic ratio at diagnosis and MRD assessment by NGS-based mutant NPM1 detection or multiparameter flow cytometry. CONCLUSION: NGS-based detection of FLT3-ITD MRD in CR identifies patients with AML with profound risk of relapse and death that outcompetes the significance of most established prognostic factors at diagnosis and during therapy, and furnishes support for FLT3-ITD as a clinically relevant biomarker for dynamic disease risk assessment in AML.