Mutations In the Tet Oncogene Family Member 2 (TET2) Gene Refine the New European LeukemiaNet Risk Classification of Primary, Cytogenetically Normal Acute Myeloid Leukemia (CN-AML) In Adults: A Cancer and Leukemia Group B (CALGB) Study

• Published in: Blood Vol. 116; no. 21; p. 98
• Main Authors: Metzeler, Klaus H., Maharry, Kati, Radmacher, Michael D., Mrózek, Krzysztof, Margeson, Dean, Becker, Heiko, Curfman, John, Holland, Kelsi B., Schwind, Sebastian, Whitman, Susan P., Wu, Yue-Zhong, Blum, William, Powell, Bayard L., Carter, Thomas H., Wetzler, Meir, Moore, Joseph O., Kolitz, Jonathan E., Baer, Maria R., Carroll, Andrew J., Larson, Richard A., Caligiuri, Michael A., Marcucci, Guido, Bloomfield, Clara D.
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 19-11-2010
• ISSN: 0006-4971; 1528-0020
• DOI: 10.1182/blood.V116.21.98.98

Abstract 98
Mutations in the TET2 gene were recently identified in a variety of myeloid neoplasms including AML. However, the frequency and clinical relevance of TET2 mutations in CN-AML, the largest cytogenetic subgroup of adult AML, have not been well defined. We report here the frequency and spectrum of TET2 mutations, and their associations with clinical and molecular characteristics, treatment outcomes and genome-wide gene- and microRNA (miR)-expression signatures in a relatively large cohort of 427 patients (pts) with primary CN-AML. The pts, aged 18–83 years, were intensively treated on CALGB frontline protocols, and were analyzed centrally for TET2 mutations by PCR and direct sequencing, and for other prognostic gene mutations (FLT3 internal tandem duplications [ITD] and tyrosine kinase domain mutations, MLL partial tandem duplications, and mutations in NPM1, CEBPA, WT1 and IDH1&IDH2). If available, buccal swabs or remission marrow samples were used to determine TET2 germline status. Gene- and miR-expression profiles were derived using microarrays (Affymetrix HG-U133 plus 2.0 and OSUCCC custom miR array v4.0). At least 1 sequence variation in TET2 was found in 104 pts. Frameshift (n=59) and nonsense (n=34) variations were distributed throughout all coding exons, while missense changes (n=37) clustered mainly (28/37) in 2 evolutionarily conserved domains of TET2. The remaining missense variations in 9 pts were located outside the conserved domains, and analysis of available buccal swabs or remission samples showed that these sequence changes were present in the germline. Since it is unclear whether they represent innocent polymorphisms or disease-relevant mutations, these 9 pts were excluded from further analyses. TET2-mutated (TET2-mut) pts were older (P<.001), had higher white blood counts (P=.04), a lower frequency of IDH1 and IDH2 mutations (P<.001), and showed a trend towards higher frequency of CEBPA mutations (P=.07) compared with TET2 wild-type (TET2-wt) pts. The European LeukemiaNet (ELN) recently proposed a standardized reporting system for AML, in which CN-AML pts are assigned to Favorable-risk (Fav; pts with mutated CEBPA and/or mutated NPM1 without FLT3-ITD) or Intermediate-I-risk (Int-I; all remaining CN-AML pts) categories. We assessed the prognostic relevance of TET2 mutations in the context of the Fav (n=199 pts) and Int-I (n=219) ELN categories. TET2 mutations tended to be more frequent in Fav than in Int-I CN-AML pts (27% v 19%, P=.08), even though types and location of mutations were similar in both groups. Within the Fav category, TET2-mut pts had shorter event-free survival (EFS; P<.001), a lower complete remission (CR) rate (P=.007) and shorter disease-free survival (DFS; P=.003; Fig 1), and shorter overall survival (OS; P=.001; Fig 2) compared with TET2-wt pts. In contrast, in the Int-I category, no difference in EFS (P=.45), CR rates (P=1.0), DFS (P=.36; Fig 1) or OS (P=.72; Fig 2) was found between TET2-mut and TET2-wt pts. In multivariable models, TET2 mutations associated with shorter EFS (P=.004; hazard ratio [HR], 1.71), lower CR rate (P=.03; odds ratio, 0.62) and shorter DFS (P=.049; HR, 1.54) only among Fav, but not among Int-I, CN-AML pts. A TET2 mutation-associated gene-expression signature consisting of 213 probe sets (136 named genes) was identified in ELN Fav CN-AML pts and included genes previously implicated in AML pathogenesis, e.g., upregulated CEBPA, APP, NCAM1 and IDH1, and downregulated MLL. In contrast, no signature of differentially expressed genes was identified in Int-I pts. miR profiling revealed distinct TET2 mutation-associated miR-expression signatures in the ELN Fav and Int-I risk groups. Among miRs upregulated in ELN Fav/TET2-mut pts were miR-148a (targeting DNA methyltransferases, highly expressed in refractory chronic lymphocytic leukemia) and miR-24 (stimulating myeloid cell proliferation, blocking granulocytic and erythroid differentiation). In Int-I/TET2-mut pts, one of the upregulated miRs was miR-204 (targeting HOXA10 and MEIS1, downregulated in NPM1-mut AML). We conclude that TET2 mutations are associated with lower remission rates and inferior survival in the ELN Fav category of CN-AML, and may be useful to refine the ELN molecular classification. TET2 mutation-associated gene- and miR-expression signatures, first identified here, may contribute to our understanding of the biology of TET2-mutated CN-AML.
[Display omitted]
No relevant conflicts of interest to declare.