Comparative neuropsychiatric safety signals of tacrolimus versus cyclosporine in solid organ transplantation: ten-year FAERS pharmacovigilance study

Introduction: Neuropsychiatric toxicity is a recognized complication of calcineurin inhibitor (CNI) therapy; however, large-scale comparative assessments across tacrolimus immediate-release (IR), once-daily LCP-tacrolimus (LCPT), and cyclosporine remain limited. Methods: We performed a decade-long pharmacovigilance study using the FDA Adverse Event Reporting System (FAERS) database from 2015 to 2025 to characterize the spectrum, seriousness, temporal trends, and drug-specific reporting signals of neuropsychiatric adverse events (AEs) in transplant recipients in whom tacrolimus IR, LCPT, or cyclosporine was listed as a suspect agent. Events were categorized as neurological, psychiatric, or combined neuropsychiatric. Disproportionality was assessed using reporting odds ratios (RORs) and proportional reporting ratios (PRRs). Results: A total of 5,437 neuropsychiatric AE reports were identified, involving tacrolimus IR (71.9%), cyclosporine (23.7%), and LCPT (4.5%). Most reports were neurological (74.1%), followed by psychiatric (17.1%) and combined neuropsychiatric events (8.8%). Serious outcomes were frequently reported (92.5%), including hospitalization (52.1%) and reported death (16.2%). Combined neuropsychiatric events had the highest hospitalization rate (62.1%); however, such outcomes should be interpreted in the context of complex clinical illness and treatment exposure rather than as directly attributable to CNI-related neuropsychiatric toxicity alone. Tremor demonstrated elevated reporting disproportionality for tacrolimus IR (ROR 1.97; PRR 1.78) and LCPT (ROR 2.42; PRR 1.97). In contrast, cyclosporine showed no tremor signal but demonstrated elevated disproportionality for encephalopathy (ROR 1.73; PRR 1.45), insomnia (ROR 1.87; PRR 1.76), and anxiety (ROR 1.50; PRR 1.48). Reporting increased from 2015 to 2019 and remained elevated through 2025. Conclusion: These findings suggest distinct drug-specific neuropsychiatric reporting patterns among CNIs in FAERS. However, given the inherent limitations of spontaneous reporting systems and differences in real-world utilization across CNIs, these findings should be interpreted as hypothesis-generating pharmacovigilance signals rather than evidence of causation, incidence, or comparative risk. They highlight the potential clinical relevance of neuropsychiatric tolerability when individualizing CNI selection, monitoring, and formulation strategies, and support further integration of pharmacovigilance data with pharmacokinetic and registry-level evidence to improve individualized risk assessment.