Drug-induced paronychia: a disproportionality and time-to-onset pharmacovigilance study using FAERS and JADER

Paronychia is a painful periungual inflammation that can significantly impair quality of life and even necessitate modifications to cancer treatment. With the expanding use of epidermal growth factor receptor (EGFR) inhibitors and other targeted therapies, the reporting frequency of drug-induced paronychia in pharmacovigilance databases has increased. However, comprehensive assessment of the relative reporting patterns and temporal characteristics of drug-induced paronychia remains limited.

We performed a disproportionality analysis of adverse event reports retrieved from the US FDA Adverse Event Reporting System (FAERS) and the Japanese Adverse Drug Event Report database (JADER). After case-level deduplication, paronychia was identified using MedDRA preferred term. Four signal detection algorithms were employed (Reporting Odds Ratio, Proportional Reporting Ratio, Bayesian Confidence Propagation Neural Network, and Multi-item Gamma Poisson Shrinker). A drug-paronychia pair was classified as a positive signal only when all four algorithms simultaneously satisfied thresholds (ROR lower 95% CI > 1; PRR ≥ 2 with χ² ≥ 4; BCPNN IC025 > 0; MGPS EBGM05 > 2; each requiring N ≥ 3). Time-to-onset characteristics were analyzed using Weibull distribution parameters to distinguish early failure from random failure patterns.

A total of 3,693 paronychia reports in FAERS and 338 in JADER were analyzed. In FAERS, females comprised 53.1% of reports; patients aged 18-64 years represented 36.6%. The ten drugs with the strongest disproportionality signals (ranked by ROR) were Dacomitinib [ROR 374.77; 95% CI 276.31-508.31], Selumetinib [304.66; 228.55-406.10], Afatinib [296.98; 264.27-333.74], Panitumumab [204.37; 183.10-228.11], Amivantamab [167.61; 124.71-225.26], Necitumumab [151.38; 67.14-341.35], Mobocertinib [117.03; 71.21-192.34], Erdafitinib [59.36; 33.57-104.99], Lapatinib [48.09; 39.71-58.24], and Gefitinib [47.28; 37.09-60.27]. Thirty-two drugs exhibited statistically significant associations with paronychia that were not listed in their FDA-approved labeling, highlighting a critical safety gap. Time-to-onset analysis of 30 drugs with valid TTO data revealed that median onset times ranged from 4 days (Necitumumab) to 575.5 days (Alendronate), with several of the most frequently reported EGFR and MEK inhibitors exhibiting an early-failure pattern (Weibull β < 1), supporting concentrated monitoring during the first weeks of therapy.

This pharmacovigilance assessment identified drugs with notable disproportionality signals for drug-induced paronychia, underscoring the utility of real-world data for early detection of safety signals. The identification of unlabeled paronychia risks in 32 medications highlights critical gaps in current safety information. These findings, derived from spontaneous reporting data, indicate associations rather than causal risk and require further clinical and epidemiologic evaluation. The signals identified may help prioritize pharmacovigilance monitoring and future validation studies.