Potential use of serum microvesicle indicators to distinguish different types of drug-induced liver injury

Drug-induced liver injury (DILI) may occur after the combination therapy of multiple drugs, which makes it difficult to identify the causative drug for liver injury in epidemiological research and clinical practice. Scutellaria baicalensis (SB), a widely used herb in traditional Chinese medicine, has been reported to cause liver injury when combined with antibiotics. However, diagnosing SB-drug induced liver injury (SB-DILI) and its distinction from antibiotic-induced DILI (AILI) is one of the significant challenges in modern clinical practice. There are still no effective indicators to distinguish between SB-DILI and AILI. Therefore, to discover and validate highly relevant indicators, we implemented a mass spectrometry workflow using label-free quantitative proteomics and Parallel Reaction Monitoring (PRM) in serum microvesicles (MVs) from 29 patients with SB-DILI, 28 patients with AILI, and 33 healthy volunteers (HVs). The results showed that a combined analysis of lysozyme (LYZ), component 4-binding protein α (C4BPA), and complement component 1r (C1R) from serum MVs yielded an area under the curve (AUC) ≥ 0.95, indicating that the combination analysis nearly fully distinguished between SB-DILI and AILI in these cohorts. This study underscores the potential of circulating MV indicators in differentiating between SB-DILI and AILI, offering significant implications for clinical diagnosis and therapeutic approaches for DILI. SIGNIFICANCE: Drug-induced liver injury (DILI) poses significant diagnostic challenges in clinical practice, particularly when caused by herbal-antibiotic combination therapies, where identifying the causative agent is critical yet elusive. This study addresses this unmet need by establishing the first serum microvesicle (MV)-based biomarker panel (LYZ, C4BPA, and C1R) capable of distinguishing Scutellaria baicalensis-induced DILI (SB-DILI) from antibiotic-induced DILI (AILI) with high accuracy (AUC ≥ 0.95). Leveraging label-free quantitative proteomics and rigorous validation via parallel reaction monitoring (PRM), our work advances the field of translational proteomics by demonstrating that MV-derived proteins reflect disease-specific pathophysiological processes, such as complement dysregulation and immune activation. Clinically, this panel addresses the rarity of SB-DILI and AILI (incidence <30/100,000) by providing an objective diagnostic tool to guide timely drug discontinuation, thereby reducing liver damage progression and optimizing therapeutic decisions. Methodologically, our workflow-combining MV isolation with targeted proteomics-sets a precedent for biomarker discovery in rare DILI subtypes. While preliminary, these findings lay the groundwork for multicenter validation studies to translate this innovation into clinical practice, ultimately improving precision medicine strategies for hepatotoxicity management.