Global pharmacovigilance analysis of dermatologic toxicity in EGFR-TKI-treated NSCLC: a cross-database real-world study.

Introduction: Epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) have transformed the treatment of EGFR-mutant non-small-cell lung cancer (NSCLC), significantly improving survival outcomes and quality of life compared with conventional chemotherapy. However, inhibition of EGFR signaling, a key regulator of epidermal homeostasis, frequently results in dermatologic adverse drug reactions (ADRs), which may compromise treatment adherence and negatively affect patient well-being. This study aimed to characterize the real-world reporting patterns of cutaneous ADRs associated with five EGFR-TKIs using large-scale pharmacovigilance databases. Methods: A retrospective pharmacovigilance analysis was performed using the World Health Organization VigiAccess™ database. Reports associated with gefitinib, erlotinib, afatinib, osimertinib, and lazertinib were extracted and evaluated. Dermatologic ADRs included disorders of the skin, subcutaneous tissue, hair, nails, and mucosa. Reporting frequencies and disproportionality analyses were conducted to compare dermatologic toxicity profiles across agents. To validate the findings and assess reproducibility, an independent analysis was additionally performed using the European Medicines Agency (EMA) EudraVigilance database. Results: A total of 96,386 ADR reports were identified, of which 25,383 (29.7%) involved dermatologic manifestations. Rash and pruritus were the most frequently reported cutaneous ADRs across all EGFR-TKIs, although their proportional distribution differed between agents. Erlotinib and afatinib demonstrated the highest proportions of dermatologic reports (30.8% and 43.0%, respectively), whereas osimertinib exhibited a lower relative frequency (11.6%) but a greater proportion of nail and hair disorders within the dermatologic category. Lazertinib displayed a distinct reporting pattern characterized by frequent rash and pruritus signals, while maintaining an overall dermatologic profile consistent with the class effect of EGFR inhibition. Disproportionality analyses revealed statistically significant differences in reporting patterns among individual EGFR-TKIs. Comparable trends were observed in the EudraVigilance database, supporting the reproducibility and external validity of the identified safety signals. Discussion: Large-scale pharmacovigilance data demonstrate that although dermatologic toxicity represents a class effect of EGFR inhibition, substantial differences exist in the spectrum and relative frequency of cutaneous ADRs among individual EGFR-TKIs. Cross-validation using independent pharmacovigilance systems strengthens the reliability of these findings and underscores the importance of agent-specific toxicity surveillance. These results support proactive multidisciplinary management strategies aimed at minimizing treatment interruptions, optimizing therapeutic adherence, and improving patient-centered care in individuals with EGFR-mutant NSCLC.