Abstract
Pemphigus is a rare yet serious autoimmune skin disease. This study aimed to identify the drugs most commonly associated with pemphigus risk using the Federal drug administration adverse event reporting system (FAERS) database. Data on drugs and adverse events between 2004 and the second quarter of 2024 were obtained from the FAERS database. The medical dictionary for regulatory activities (MedDRA) was used to identify pemphigus cases, whereas the DrugBank database was used to determine the generic names of the medications. Four signal quantification methods were used for the analysis: odds ratio, proportional reporting ratio, Bayesian confidence propagation neural network, and empirical Bayesian geometric mean. Additionally, the timing of drug-induced pemphigus was calculated using Weibull shape parameters. Finally, potential factors affecting pemphigus were identified in the FAERS database using univariate and multivariate logistic regression analyses. Of the 7315 adverse event reports linked to pemphigus, diclofenac sodium was involved in 1227 cases, which was the highest number. Disproportionality analysis Disproportionality analysis identified the top 5 drugs that caused adverse reactions: diclofenac sodium, abatacept, tocilizumab, and alendronate sodium. Based on the time to onset of drug-induced pemphigus, the Weibull shape parameter (0.60) indicated that pemphigus induced by this class of drugs was classified as an early failure. Seven independent risk factors related to drugs were identified through multivariable analysis, including diclofenac sodium, abatacept, tocilizumab, alendronate sodium, secukinumab, cetirizine, and chlorhexidine, odds ratio >1 for each. FAERS database was analyzed to identify drugs with strong pemphigus signals, which can be used as a guide for standardizing drug use in clinical practice.