Comparative safety signals of carboplatin and cisplatin in lung cancer: a pharmacovigilance study based on FAERS.

Background: Platinum-based chemotherapy remains a key component of lung cancer treatment. Although carboplatin and cisplatin are widely used, their real-world adverse-event reporting profiles may differ. This study compared carboplatin- and cisplatin-associated safety signals in lung cancer using the FDA Adverse Event Reporting System (FAERS), with emphasis on organ-specific signals, demographic heterogeneity, and time-to-onset (TTO) patterns. Methods: FAERS data from Q3 2005 to Q1 2026 were retrospectively analyzed. Reports were included when carboplatin or cisplatin was recorded as the primary suspect drug and lung cancer was listed as the indication. Disproportionality analyses were performed using reporting odds ratio (ROR), proportional reporting ratio (PRR), Bayesian Confidence Propagation Neural Network (BCPNN), and Multi-item Gamma Poisson Shrinker (MGPS). Positive Preferred Term (PT) signals were further screened for clinical relevance. TTO was summarized using median and interquartile range (IQR), and Weibull models were fitted to assess temporal failure patterns. Age- and sex-stratified analyses and exclusion-based sensitivity analyses were also conducted. Results: A total of 12,045 eligible case reports were included, comprising 10,159 carboplatin-associated and 1,886 cisplatin-associated reports. Carboplatin-related reports were mainly characterized by hematologic and infection-related signals, including anaemia, neutropenia, pancytopenia, thrombocytopenia, febrile neutropenia, leukopenia, and neutropenic sepsis, together with hepatobiliary and hypersensitivity-related signals. Cisplatin-related reports showed broader renal, vascular, auditory or vestibular, gastrointestinal, and electrolyte-related signals, including renal salt-wasting syndrome. Valid TTO information was available for 4,487 carboplatin-associated and 823 cisplatin-associated reports. Median TTO was 25 days for carboplatin (IQR, 8-67 days) and 21 days for cisplatin (IQR, 8-62 days). More than half of reports with valid TTO occurred within 0-30 days after treatment initiation: 55.0% for carboplatin and 59.7% for cisplatin. Weibull analysis supported an early-failure reporting pattern for both agents. Conclusion: Carboplatin and cisplatin showed distinct but overlapping adverse-event reporting profiles in lung cancer. These findings may support organ-specific and time-stratified safety monitoring, particularly during the first treatment cycle. However, they should be interpreted as pharmacovigilance signals rather than estimates of incidence, absolute risk, or causality, and require further prospective validation.