A pharmacovigilance study of Bruton's tyrosine kinase inhibitors: a multidimensional analysis based on FAERS and VigiBase.

Introduction: The safety concerns of Bruton's tyrosine kinase inhibitors (BTKis) have garnered significant attention due to their severe adverse reactions. However, no existing studies have utilized VigiBase, the world's largest adverse event reporting system, to conduct post-marketing safety analyses of these agents. Methods: This study extracted data from VigiBase and the FDA Adverse Event Reporting System (FAERS), employing the reporting odds ratio as the primary method and information component as supplementary, to comprehensively evaluate the safety profiles of ibrutinib, acalabrutinib, and zanubrutinib, with a focus on bleeding risks when combined with anticoagulants or antiplatelet drugs. Results: The results revealed that at the system organ class level, ibrutinib had the strongest signal in cardiac disorders; acalabrutinib in blood and lymphatic system disorders; and zanubrutinib in infections and infestations and blood and lymphatic system disorders. Among the top ten standardised medical dictionary for regulatory activities queries (SMQ), the SMQs with the strongest signals were different for each BTKis, but five identical SMQs were in the top ten, namely supraventricular tachyarrhythmias, tumour lysis syndrome, haematopoietic thrombocytopenia, haemorrhage terms (excl laboratory terms), and haemorrhage laboratory terms. At the preferred term level, acalabrutinib exhibited the strongest signal for Richter's syndrome, zanubrutinib for subcutaneous haemorrhage, while ibrutinib displayed divergent signals between databases-Bing-Neel syndrome in VigiBase and haematotympanum in FAERS. Importantly, bleeding risks varied significantly between monotherapy and combination therapy with anticoagulants/antiplatelet agents, underscoring the need for clinical vigilance regarding site-specific haemorrhage risks. Discussion: These results will provide new data to support the use of BTKis and further safety warnings. However, as a hypothesis-generating approach, it does not establish a definitive causal relationship, which will require further research and validation.