Multi-Omics Identified Factor V Leiden as a Novel Predictor of 90-Day Mortality in Artificial Liver-Supported ACLF: A Translational Study.

This study employed integrated transcriptomic and proteomic analyses to identify clinically actionable prognostic biomarkers in patients with acute-on-chronic liver failure (ACLF) supported by an artificial liver support system (ALSS). For parallel transcriptomic and proteomic characterization, peripheral blood mononuclear cells (PBMCs) were extracted from 15 patients with hepatitis B virus (HBV)-associated ACLF and 4 healthy subjects. Clinical utility was assessed using an independent validation cohort comprising 70 patients with ACLF and 60 age and sex-matched healthy controls. A precision prognostic tool for mortality prediction was developed and validated using Lasso regression, logistic regression, and receiver operating characteristic (ROC) curve analysis. Multi-omics screening revealed 17 molecules with concordant RNA-protein alterations, among which coagulation factor V (F5) and related proteins showed dynamic expression during ALSS treatment. Validation in an independent cohort confirmed significantly elevated mRNA and plasma levels of factor V Leiden (FVL) in ACLF patients versus controls. Lasso and multivariate logistic regression identified baseline plasma FVL and its changes during treatment as independent predictors of 90-day mortality. The model combining baseline FVL and its dynamic changes achieved an AUC of 0.977 for mortality prediction. FVL represents a robust prognostic biomarker in ACLF. Integration of baseline quantification and treatment-responsive dynamics enables precise prediction of mortality risk.