Safety recommendations for ALK tyrosine kinase inhibitors in non-small cell lung cancer: evidence from FAERS and CVARDD real-world databases.

Background: Non-small cell lung cancer (NSCLC) remains one of the leading causes of cancer mortality worldwide. The development of anaplastic lymphoma kinase tyrosine kinase inhibitors (ALK-TKIs) has significantly improved survival among patients with ALK-positive NSCLC. However, prolonged treatment and wider clinical use have led to increasing reports of adverse events (AEs). Existing studies have primarily explored individual drugs, with limited comparative evidence across different ALK-TKIs regarding sex-specific safety differences and time-to-onset patterns. Methods: This real-world pharmacovigilance study analyzed data from the U.S. Food and Drug Administration Adverse Event Reporting System (FAERS, Q1 2004-Q2 2025) and the Canadian Vigilance Adverse Reaction Database (CVARDD). Disproportionality analyses were performed using four algorithms-Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Bayesian Confidence Propagation Neural Network (BCPNN), and Multi-Item Gamma Poisson Shrinker (MGPS)-to detect adverse event signals for crizotinib, alectinib, and brigatinib. Sex-stratified risk analyses, cross-database validation, and Weibull time-to-onset modeling were further conducted to assess robustness and temporal patterns of AE occurrence. Results: A total of 18-683 AE reports were identified (crizotinib = 9 030; alectinib = 8 486; brigatinib = 1 167). Distinct toxicity spectra were observed among the three ALK-TKIs. Brigatinib exhibited the strongest hepatotoxic and pulmonary signals, notably hepatic function abnormal (ROR = 13.95) and pleural effusion (ROR = 11.03), indicating a high risk of early liver and respiratory toxicity. Alectinib showed pronounced metabolic and edema-related AEs (oedema, ROR = 9.47; hepatic function abnormal, ROR = 9.21), suggesting a tendency toward fluid retention and hepatobiliary dysfunction. Crizotinib demonstrated a more balanced safety profile but still presented notable risks for pleural effusion (ROR = 8.88) and hepatic function abnormal (ROR = 7.92), both showing early-onset patterns (median TTO = 34.5 days and 14.5 days, respectively). Sex-stratified analyses revealed that males were more prone to renal, cardiac, and respiratory toxicities, whereas females were more likely to develop hepatic and hematologic events. Weibull modeling indicated an "early failure" pattern (β< 1) for all agents, meaning AEs predominantly occurred within the first 12 weeks of therapy. Cross-database validation confirmed consistent risk signal direction and strong reproducibility between FAERS and CVARDD datasets. Conclusions: All three ALK-TKIs demonstrate distinct, generation-dependent safety profiles characterized by early-onset hepatobiliary and pulmonary toxicities, with evident sex-specific differences in organ susceptibility. Intensive safety monitoring during the initial 12 weeks of therapy is essential for preventing severe outcomes.