Metabolomics of Chronic Myelogenous Leukemia Based on GC-MS / 中国实验血液学杂志

ABSTRACT

<p><b>OBJECTIVE</b>To investigate the metabolism characteristics, to search for potential biomarkers associated with disease and to explore related metabolic pathways by analyzing the plasma metabolic profile of patients with chronic myelogenous leukemia (CML) through metabolomies.</p><p><b>METHODS</b>Twenty-six newly diagnosed CML patients in the First Affilated Hospital of Soochow University from February 2015 to April 2015, 26 allogeneic hematopoietic stem cell donors as healthy controls and 26 patients treated with tyrosine kinase inhibitors (TKI) to obtain the best efficacy as post-treatment controls were enrolled in this study. All the metabolites of plasma were extracted by Gas Chromatography-Mass Spectrometer(GC-MS) to collect metabolic fingerprint. Multivariate pattern recognition analysis and t test were combined to screen out the metabolic biomarkers at different time points. The receiver operating characteristic curve analysis was used to evaluate the clinical efficacy of metabolites, and the metabolic pathway analysis was performed.</p><p><b>RESULTS</b>Significantly different metabolite expression mode was found seen between CML and healthy control groups. Six changed metabolites in CML were confirmed by multivariate and variate statistical analyses. Compared with the healthy controls, the levels of tetradecanoic acid and glycerol were decreased, the lactic acid, myo-inositol, d-galactose and glycine in CML patients also increased (all VIP>1,P<0.05, AUC>0.7). The plasma metabolites in CML patients after treatment with tyrosine kinase inhibitors (TKI) showed a recovery trend toward to normal levels. The plasma metabolic pathways of CML were mainly related with galactose, pyruvate, glycerolipid, inositol phosphate and glycine, serine and threonine metabolism (all impact value>0.10).</p><p><b>CONCLUSION</b>Significant changes in plasma metabolite levels were found in CML patients. Metabolomics combined with multivariate pattern recognition analysis may be a new tool to assist diagnosis.</p>