A tri-scale in silico framework integrating pharmacovigilance and mechanistic modeling suggests tepotinib-associated acute kidney injury risk

Proactive safety evaluation of molecularly targeted therapies requires generalizable frameworks that integrate real-world evidence with mechanistic insights. As a case in point, tepotinib, a mesenchymal-epithelial transition (MET) inhibitor for non-small cell lung cancer, has a known pulmonary toxicity profile, but its link to acute kidney injury (AKI) and underlying mechanism remain unclear. We devised a tri-scale in silico strategy encompassing population-scale pharmacovigilance signal extraction from the FDA Adverse Event Reporting System (FAERS) (2020-2025) <i>via</i> disproportionality analysis (Reporting Odds Ratio, ROR, and Proportional Reporting Ratio, PRR), network toxicology-based multi-target exploration for shared target and hub gene identification, and mechanistic docking-based molecular plausibility assessment using the Cavity-detection guided Blind Docking (CB-Dock2) platform. Pharmacovigilance analysis identified a significant renal impairment signal (ROR = 20.60). Network toxicology suggested potential nephrotoxicity and revealed 173 shared targets and prioritizing six hub genes (HSP90AA1, AKT1, EGFR, CASP3, TNF, SRC). These genes were shown to be critically involved in PI3K-Akt and MAPK pathways. Molecular docking revealed high-affinity binding between tepotiniband all six hub targets (Vina scores: -8.0 to -10.6 kcal/mol), providing a structural basis for the postulated mechanistic link to AKI. These findings not only highlight the necessity for enhanced renal monitoring in tepotinib-treated patients but, more broadly, establish the FAERS-NetDock Pipeline as a reusable, generalizable and hypothesis-generating framework for evaluating tyrosine kinase inhibitors (TKIs)-induced nephrotoxicity; this framework is immediately applicable to profiling the safety of other TKIs (e.g. crizotinib, capmatinib, savolitinib, afatinib and osimertinib) and is readily adaptable for de-risking a wider spectrum of targeted therapies.