Drug-specific differences in fatal adverse events among PARP inhibitors: a FAERS pharmacovigilance study

Poly(ADP-ribose) polymerase inhibitors (PARPi) are widely used in the treatment of multiple malignancies; however, their fatal adverse event profiles remain incompletely characterized in real-world settings. Evidence regarding drug-specific differences in fatal toxicity is limited.

This study aimed to comprehensively evaluate fatal adverse events associated with PARPi, focusing on the overall fatal burden across individual agents, drug-specific fatal toxicity signatures at system organ class (SOC) and preferred term (PT) levels, and the temporal dynamics of fatal events. Adverse event reports associated with olaparib, niraparib, rucaparib, and talazoparib were extracted from the FDA Adverse Event Reporting System (FAERS) database. Fatal and non-fatal cases were compared in terms of demographic and clinical characteristics. The proportion of fatal reports and reporting odds ratios (RORs) were calculated to assess drug-specific differences. SOC distribution and PT-level disproportionality analyses (cases ≥ 30, ROR > 1.5, FDR < 0.05) were performed to identify fatal toxicity signatures. Principal component analysis (PCA) and hierarchical clustering were used to evaluate similarity patterns. Time-to-onset (TTO) was analyzed using violin plots and kernel density estimation.

A total of 25,413 reports were included, comprising 5,930 fatal and 19,483 non-fatal cases. Fatal reports were more common in older patients and in those with prostate and pancreatic cancers. The proportion and reporting odds of fatal events varied across PARPi, with olaparib showing a higher fatal reporting odds (OR = 1.29, 95% CI: 1.03-1.63) compared with talazoparib, whereas niraparib and rucaparib demonstrated significantly lower odds. Hematologic toxicities were consistently enriched across agents, supporting a class effect, while non-hematologic toxicities showed substantial heterogeneity. Olaparib was characterized by rare but highly enriched fatal events (e.g., interstitial lung disease, pneumonitis, thrombosis), whereas niraparib exhibited more frequent but moderately increased risks. PCA revealed clear separation of toxicity profiles among agents. Fatal events generally occurred later than non-fatal events, particularly for olaparib (median TTO: 84 vs 48 days, p < 0.001).

Fatal adverse event profiles differ across PARPi, reflecting both shared class effects and drug-specific heterogeneity. These findings highlight the importance of individualized safety assessment and sustained monitoring during PARPi therapy.