Multi-database pharmacovigilance identifies disproportionate reporting of hepatobiliary events with avacopan: an integrative study with network pharmacology and interpretable machine learning.

Authors

Jing Zhang, Zhiwei Cui, Wenqian Zeng, Fan Zou, Xiyuan Zhang, Liang Shang, De Xie

Published in

Naunyn-Schmiedeberg's archives of pharmacology. Sep 10, 2026. Epub Sep 10, 2026.

Abstract

Avacopan is an oral C5a receptor antagonist approved for severe active granulomatosis with polyangiitis and microscopic polyangiitis. However, its real-world safety profile, particularly hepatobiliary safety signals, remains incompletely characterized. We analyzed avacopan-related adverse event reports from FAERS, JADER, and EudraVigilance. Disproportionality analyses were performed using reporting odds ratio and Bayesian Confidence Propagation Neural Network. Subgroup, sensitivity analysis, time-to-onset, Bradford Hill, network pharmacology, molecular docking, machine learning, and SHAP analyses were used to characterize safety signals, explore potential mechanisms, and identify factors associated with hepatobiliary disorder reporting. In FAERS, 5,293 avacopan-related individual case safety reports comprising 11,901 adverse event terms were identified. Positive signals were detected for infections, gastrointestinal disorders, and hepatobiliary disorders. Liver-related events, including drug-induced liver injury, jaundice, cholestasis, and vanishing bile duct syndrome, were consistently observed across FAERS, JADER, and EudraVigilance. In FAERS, most adverse events occurred early after treatment initiation. A Bradford Hill-informed contextual appraisal provided additional context for the observed DILI reporting signal; however, it did not permit causal inference. Network pharmacology identified 83 overlapping targets, with enrichment mainly involving xenobiotic response, oxidative stress, lipid metabolism, and PI3K-Akt signaling. Molecular docking supported favorable interactions between avacopan and key targets. Among machine learning models, NeuralNet showed the best validation performance for report-level classification, and SHAP analysis identified country as the dominant contributor to model predictions, suggesting substantial influence of regional reporting patterns. This real-world study expands the safety profile of avacopan and highlights hepatobiliary disorders as important signals requiring clinical attention. These findings support strengthened post-marketing surveillance and careful liver monitoring during avacopan therapy.

PMID:
42720689
Bibliographic data and abstract were imported from PubMed on 10 Sep 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.