生物标志物
戊酸
脂肪肝
肝病
肝实质
代谢组学
脂肪酸
化学
医学
内科学
疾病
胃肠病学
病理
色谱法
生物化学
醋酸
作者
S. U. Qadri,Xiaosuo Wang,Collin Tran,Michael Fitzpatrick,Paul Bonnitcha,David Sullivan,Hannele Yki‐Järvinen,John O’Sullivan
出处
期刊:Pathology
[Elsevier]
日期:2024-04-01
卷期号:56 (3): 391-397
标识
DOI:10.1016/j.pathol.2023.10.006
摘要
We have recently determined dimethylguanidino valeric acid (DMGV) to be a novel biomarker of liver injury in non-alcoholic fatty liver disease (NAFLD) and an independent predictor of incident diabetes over a decade in advance. DMGV consists of two stereo-isomers, asymmetric dimethylguanidino valeric acid (ADGV) and symmetric dimethylguanidino valeric acid (SDGV). Here we report, for the first time, the upper limits of normal of both isomers in humans at the accepted 5.56% liver fat threshold for NAFLD, determined using in vivo magnetic resonance spectroscopy. We performed independent and blinded comparative analyses of ADGV and SDGV levels using two different liquid chromatography-tandem mass spectrometry (LC-MS/MS) methods in (A) our laboratory, and (B) the New South Wales Chemical Pathology state laboratory, using unique columns, LC-MS/MS equipment, extraction protocols and normalisation approaches. Despite these differences, each laboratory reported consistent absolute concentrations across a range of liver fat percentages. We next determined the diagnostic performance of SDGV compared to ADGV in a cohort of 268 individuals with liver fat measurements. In derivation-validation analyses we determined rule-in/rule-out thresholds and the concentration of SDGV that provides optimal performance across sensitivity and specificity for the identification of NAFLD. In conclusion, we have herein determined for the first time the true human plasma reference range of both isoforms of an emerging novel biomarker of NAFLD, at the accepted upper normal threshold of liver fat. We have also identified that SDGV is the isoform with the best diagnostic performance and determined the optimal cut-point for its detection of NAFLD.
科研通智能强力驱动
Strongly Powered by AbleSci AI