Tandem Mass Tag-Based Quantitative Proteomic Profiling Identifies Novel Serum Biomarkers of Drug-Induced Liver Injury in Humans
Ravindra KC, Vaidya VS, Wang Z, Federspiel JD, Virgen-Slane R, Everley RA, Grove JI, Stephens C, Ocana MF, Robles-Díaz M, Isabel Lucena M, Andrade RJ, Atallah E, Gerbes AL, Weber S, Cortez-Pinto H, Fowell AJ, Hussaini H, Bjornsson ES, Patel J, Stirnimann G, Verma S, Elsharkawy AM, Griffiths WJH, Hyde C, Dear JW, Aithal GP, Ramaiah SK. Tandem mass tag-based quantitative proteomic profiling identifies candidate serum biomarkers of drug-induced liver injury in humans. Nat Commun. 2023 Mar 3;14(1):1215. doi: 10.1038/s41467-023-36858-6. PMID: 36869085; PMCID: PMC9984368.
Diagnosis of drug-induced liver injury (DILI) and its distinction from other liver diseases are significant challenges in drug development and clinical practice. In this study we identify, confirm, and replicate the biomarker performance characteristics of candidate proteins in a cohort of patients with DILI at onset (DO; n = 133) and follow-up (DF; n = 120), acute non-DILI at onset (NDO; n = 63) and follow-up (NDF; n = 42), and healthy volunteers (HV; n = 104). Area under the receiver operating characteristic curve (AUC) for cytoplasmic aconitate hydratase; argininosuccinate synthase; carbamoylphosphate synthase, fumarylacetoacetase; fructose-1,6-bisphosphatase 1 (FBP1) across cohorts achieves near complete separation (range: 0.94 - 0.99) of DO and HV. In addition, we show that FBP1 alone or in combination with glutathione S-transferase A1 and leukocyte cell derived chemotaxin 2 could potentially assist in clinicla diagnosis by distinguishing NDO from DO (AUC range: 0.65 – 0.78), but further technical and clinical validation of these candidate biomarkers is needed.
Targeted IS-PRM (SureQuant) assays were developed using heavy stable isotope labeled peptides spiked into serum samples and analyzed on an Exploris 480 mass spectrometer. Identification of the heavy peptide triggered acquisition of the endogenous light peptide. 2-5 transitions per peptide were used to confirm identification and compared to the stable isotope labeled standard. The top ranked transition was used for quantification. Sample names and details are anonymized due to patient privacy policies.