Acute myeloid leukemia (AML) is a life-threatening hematological malignancy. Traditional diagnosis of AML depends on invasive bone marrow biopsies. To recognize the metabolic characteristics related with AML and search for early non-invasive biomarkers of AML, in this work we applied ultra-high performance liquid chromatography quadrupole time-of-flight tandem mass spectrometry (UHPLC-Q-TOFMS)-based metabolomoc method to profile serum metabolites from 55 de novo AML patients and 45 age- and gender-matched healthy subjects and to screen and validate AML biomarkers. We observed AML-related metabolic differences mainly involved in alanine, aspartate and glutamate metabolism; d-Glutamine and d-glutamate metabolism; tryptophan metabolism; taurine and hypotaurine metabolism; and phenylalanine metabolism as well as fatty acid metabolism. A serum metabolite biomarker panel consisting of glutamic acid, kynurenine and oleic acid was defined and validated based on binary logistic regression analysis and receiver operating characteristic curves (ROC) analysis, yielding an area under the ROC curve (AUC) of 0.981 with 0.975 sensitivity and 0.933 specificity in the discovery set and an AUC of 0.973 with 0.933 sensitivity and 0.933 specificity in the validation set. This work demonstrated the UHPLC- MS-based metabolomics as a low invasive potential tool for the detection of AML, and this composite serum metabolite panel exhibited good diagnostic performance for AML in this case-control study and deserved further validation in a large-scale clinical trial. The identified metabolic pathways were also potentially worthy of further studying the pathogenesis of AML.
Keywords: Acute myeloid leukemia; Diagnosis; Metabolic pathway; Metabolomics; UHPLC-Q-TOFMS.
Copyright © 2019 Elsevier B.V. All rights reserved.