Application of plasma metabolome for monitoring the effect of rivaroxaban in patients with nonvalvular atrial fibrillation

• Published in: PeerJ (San Francisco, CA) CA), 2022-08, Vol.10, p.e13853-e13853, Article e13853
• Main Authors: Zhao, Mindi, Liu, Xiaoyan, Bu, Xiaoxiao, Li, Yao, Wang, Meng, Zhang, Bo, Sun, Wei, Li, Chuanbao
• Format: Article
• Language: English
• Subjects: Anticoagulants; Atrial fibrillation; Atrial Fibrillation - drug therapy; Biomarkers; Bone surgery; Cardiac arrhythmia; Cardiology; Care and treatment; Chromatography; Chromatography, Liquid; Comparative analysis; Discriminant analysis; Fibrillation; Glycosylphosphatidylinositol; Hematology; Humans; Laboratories; Lipid metabolism; Liquid chromatography; Mass spectrometry; Mass spectroscopy; Medical research; Medicine, Experimental; Metabolic Sciences; Metabolites; Metabolome; Metabolomics; Molecular Biology; Molecular weight; Patients; Phosphatidylethanolamine; Physiological aspects; Plasma; Plasma metabolomics; Rivaroxaban; Rivaroxaban - pharmacology; Scientific imaging; Software; Stroke - drug therapy; Tandem Mass Spectrometry; Thromboembolism; Warfarin
• ISSN: 2167-8359; 2167-8359
• DOI: 10.7717/peerj.13853

Rivaroxaban, an oral factor Xa inhibitor, has been used to treating a series of thromboembolic disorders in clinical practice. Measurement of the anticoagulant effect of rivaroxaban is important to avoid serious bleeding events, thus ensuring the safety and efficacy of drug administration. Metabolomics could help to predict differences in the responses among patients by profiling metabolites in biosamples. In this study, plasma metabolomes before and 3 hours after rivaroxaban intake in 150 nonvalvular atrial fibrillation (NVAF) patients and 100 age/gender-matched controls were analyzed by liquid chromatography coupled with mass spectrometry (LC-MS/MS). When compared with controls, a total of thirteen plasma metabolites were differentially expressed in the NVAF patients. Pathway analysis revealed that purine and lipid metabolism were dysregulated. A panel of three metabolites (17a-ethynylestradiol, tryptophyl-glutamate and adenosine) showed good predictive ability to distinguish nonvalvular atrial fibrillation with an area under the receiver operating characteristic curve (AUC) of 1 for the discovery phase and 1 for validation. Under rivaroxaban treatment, a total of seven metabolites changed, the lipid and glycosylphosphatidylinositol biosynthesis pathways were altered and the panel consisting of avocadene, prenyl glucoside and phosphatidylethanolamine showed predictive ability with an AUC of 0.86 for the discovery dataset and 0.82 for the validation. The study showed that plasma metabolomic analyses hold the potential to differentiate nonvalvular atrial fibrillation and can help to monitor the effect of rivaroxaban anticoagulation.