Screening and validation for plasma biomarkers of nephrotoxicity based on metabolomics in male rats

摘要: Currently, drug-induced nephrotoxicity is widespread and seriously affects human health. However, the conventional indexes of renal function lack sensitivity, leading to a delay in detection nephrotoxicity. Therefore, we need identify more sensitive for evaluating In this study, used gentamicin (100 mg kg−1), etimicin kg−1) amphotericin B (4 establish injury models rats, collected information using ultra-performance liquid chromatography quadrupole time-of-flight mass spectrometry screening stage. Thirteen metabolites were selected after multivariate statistical integration analyses. Then, conducted trend analysis select 5 biomarkers [thymidine, LysoPC(16:1), LysoPC(18:4), LysoPC(20:5), LysoPC(22:5)] whose content changed consistently at different timepoints drug administration. To verify sensitivity specificity these nephrotoxicity, receiver operating characteristic (ROC) support vector machine (SVM) analyses applied. The area under curve 0.806–0.901 95% confidence interval according ROC analysis. We SVM classified model biomarkers, prediction rate was 95.83%. have strong high accuracy; are initiate prompt treatment.