Advanced glycation end products (AGEs), derived from the non-enzymatic glycation reaction, are defined as glycotoxins in various diseases including aging, diabetes and kidney injury. Exploring AGEs as potential biomarkers for these diseases holds paramount significance. Nevertheless, the high chemical structural similarity and great heterogeneity among AGEs present a formidable challenge when it comes to the comprehensive, simultaneous, and accurate detection of multiple AGEs in biological samples. In this study, an UPLC/MS/MS method for simultaneous quantification of 20 free AGEs in human serum was firstly established and applied to quantification of clinical samples from individuals with kidney injury. Simple sample preparation method through protein precipitation without derivatization was used. Method performances including imprecision, accuracy, sensitivity, linearity, and carryover were systematically validated. Intra- and inter- imprecision of 20 free AGEs were 1.93-5.94 % and 2.30-8.55 %, respectively. The method accuracy was confirmed with good recoveries ranging from 96.40 % to 103.25 %. The LOD and LOQ were 0.1-3.13 ng/mL and 0.5-6.25 ng/mL, respectively. Additionally, the 20 free AGEs displayed excellent linearity (R2 >0.9974) across a wide linear range (1.56-400 ng/mL). Finally, through simultaneous quantitation of 20 Free AGEs in 100 participants including kidney injury patient and healthy controls, we identified six free AGEs, including N6-carboxyethyl-L-arginine (CEA), N6-carboxymethyl-L-lysine (CML), methylglyoxal-derived hydroimidazolones (MG-H), N6-formyl-lysine, N6-carboxymethyl-L-arginine (CMA), and glyoxal-derived hydroimidazolone (G-H), could well distinguish kidney injury patients and healthy individuals. Among them, the levels of four free AGEs including CML, CEA, MG-H, and G-H strongly correlate with traditionally clinical markers of kidney disease. The high area under the curve (AUC) values (AUC=0.965) in receiver operating characteristic (ROC) curve indicated that these four free AGEs can be served as combined diagnostic biomarkers for the diagnosis of kidney disease.Copyright © 2024 Elsevier B.V. All rights reserved.