Matrix and Sampling Effects on Quantification of Protein Biomarkers of Drug-Induced Liver Injury

Macrophage colony stimulating factor 1 receptor (MCSFIR), osteopontin (OPN), high-mobility group protein B1 (HMGB1), glutamate dehydrogenase (GLDH), keratin 18 (KI8), and caspase-cleaved keratin 18 (ccK18) are considered promising mechanistic biomarkers for the diagnosis of drug-induced liver injury. Here, we aim to elucidate the impact of the sample matrix and handling on the quantification of these emerging protein biomarkers. We investigated effects such as time from collection to centrifugation during serum (+/- gel) or EDTA plasma preparation on two assay platforms: immunoaffinity liquid chromatography mass spectrometric assays and sandwich immunoassays. Furthermore, we measured GLDH activity with an enzymatic activity assay. Matrix effects were observed particularly for HMGB1 and MCSFIR. HMGB1 levels were higher in serum than in plasma, whereas higher concentrations of MCSFIR were observed in plasma than in serum. A comparison of sample collection to centrifugation time ranging from 15 to 60 min demonstrated increasing levels of HMGB1 in serum, while MCSFIR, OPN, GLDH, and ccK18 concentrations remained stable. Additionally, there was a poor correlation in HMGB1 and ccK18 levels between serum and plasma. Considering the observed matrix effects, we recommend plasma as a matrix of choice and cross-study comparison studies to be limited to those using the same matrix.