胸腺瘤
化学
重症肌无力
代谢组学
色谱法
高效液相色谱法
尿苷
质谱法
内科学
生物化学
病理
医学
核糖核酸
基因
作者
Yun Li,Xiaoling Zang,Peng Jiao,Xuyan Xue,Xin Zhang,Sha Wang,Ning Song,Zhihua Lv
标识
DOI:10.1080/00032719.2023.2215885
摘要
Thymoma is a rare thymic tumor but is the most common anterior mediastinal tumor. Currently, there are still limitations regarding to its screening and the pathology remains elusive. In this study, nontargeted metabolomic analysis of serum was performed to study the metabolic characteristics of thymoma using ultra-performance liquid chromatography—high resolution mass spectrometry (UPLC-HRMS) in positive and negative ionization modes. The metabolites that were differentially abundant in thymoma were identified by MS/MS and matching to chemical standards. Forty-one metabolic features differentially abundant in thymoma were identified in the corresponding age groups, including uridine, dipeptides, arachidonic acid (AA)/8,11,14,17-eicosatetraenoic acid, lactic acid and its dimer, acylcarnitines, lysophosphatidylcholines (LPCs), and ceramide 1-phosphate (Cer1P) (d18:0/3:0). Among these metabolites, lactic acid and 9-decenoylcarnitine also had significant abundance changes between thymoma patients with and without myasthenia gravis (MG). Uridine, lactic acid and Cer1P (d18:0/3:0) may also be associated with the aberrant immune status in thymoma. Orthogonal partial least squares—discriminant analysis (oPLS-DA) showed that classification between thymoma and thymic hyperplasia was possible with 94.7% sensitivity, 100% specificity and 96.8% accuracy for all patient samples. The discriminant feature panel included uridine, AA/8,11,14,17-eicosatetraenoic acid, lactic acid dimer, and N-(1-deoxy-1-fructosyl)tryptophan. The distinct serum metabolic characteristics of thymoma identified in this study may offer insights into improvement of the diagnosis and discovery of new therapeutic strategies for thymoma, in addition to providing information for study of the underlying molecular mechanism of thymoma pathogenesis.
科研通智能强力驱动
Strongly Powered by AbleSci AI