Post-marketing safety concerns with foscarbidopa/foslevodopa: A pharmacovigilance study with disproportionality...

Foscarbidopa/foslevodopa, a 24-hour subcutaneous infusion of levodopa/carbidopa prodrugs, is approved for advanced Parkinson disease with motor fluctuations. Due to limited post-marketing data, the association between this novel formulation and related adverse events (AEs), particularly infusion-related or neurological complications, remains underexplored. This study aims to identify potential safety signals of foscarbidopa/foslevodopa-associated AEs using data mining of the FDA Adverse Event Reporting System (FAERS). To quantify safety risk signals of foscarbidopa/foslevodopa, disproportionality analyzes were conducted using the reporting odds ratio and Bayesian Confidence Propagation Neural Network methods, based on FAERS data from the last quarter of 2024 (2024Q4) to the second quarter of 2025 (2025Q2). Signal thresholds were strictly defined, and sensitivity analyses were performed to verify the robustness of results; subgroup analyses were conducted to explore sex differences and severity-related characteristics of AEs. The FAERS recorded 7527 foscarbidopa/foslevodopa-related AEs affecting 1914 patients. A total of 1616 patients experienced serious AEs, including 265 deaths. The most frequently reported AEs were on and off phenomenon (n = 223), fall (n = 195), hallucination (n = 156), and dyskinesia (n = 132). Potential safety signals were detected, such as on and off phenomenon, hallucination, and infusion site infection. Notably, FAERS data are subject to spontaneous reporting bias, missing information, and lack of denominator exposure data, so the above results only reflect statistical associations rather than definite causal relationships. FAERS data identify potential clinically significant associative AEs with foscarbidopa/foslevodopa, such as infusion site reactions and infections and hallucination. These findings support the need for enhanced monitoring of high-risk patients (e.g., mental and behavioral disorders, skin disorders) and reinforce the importance of real-world surveillance for novel Parkinson disease therapies. All findings are hypothesis-generating and require further clinical validation to confirm causal relationships.