Abstract: Objective To compare the differences in adverse reaction signals among three poly adenosine diphosphate ribose polymerase (PARP) inhibitors, namely olaparib, niraparib and rucaparib based on the United States of America Food and Drug Administration Adverse Event Reporting System (FAERS) database, and to evaluate their safety characteristics and clinical significance. Methods Original report data were extracted from the FAERS database from the first quarter of 2020 to the fourth quarter of 2024, with the keywords of the active ingredients and trade names of olaparib, niraparib and rucaparib. Signals were detected by using proportional imbalance analysis and the Bayesian confidence propagation neural network method, and the robustness of the results was verified by combining multiple sensitivity analysis. Results This study included 19 780 reports of adverse events (8 562 for olaparib, 6 327 for niraparib, and 4 891 for rucaparib). In terms of gender distribution, niraparib had the highest proportion of female patients. The majority of patients were aged 45–64 years. Regarding serious adverse events, olaparib was associated with the highest mortality rate. Multi-model analysis revealed that hematological toxicity was the core concern for olaparib; rucaparib showed the most prominent risk of abnormal liver function; and niraparib was specifically associated with abnormal liver function. Analysis by system organ classes showed that the proportion of hematologic disorders for olaparib was significantly higher than that for niraparib and rucaparib. The proportion of hepatobiliary abnormalities for niraparib was approximately 3.3 times that for olaparib, and it had the highest number of abnormal signals. Neurological symptoms and cardiovascular events with olaparib, as well as respiratory events related to rucaparib and olaparib, all demonstrated a certain tendency toward drug-specific organ damage. Furthermore, all three drugs exhibited distinct drug-related characteristics in terms of gastrointestinal/general disorders and administration site conditions. Sensitivity analysis, after controlling for concomitant medication interference, showed a slight decrease in the signal strength for olaparib-associated anemia, niraparib-associated abnormal liver function, and rucaparib-associated acute kidney injury, confirming their biological plausibility. In contrast, the signals of acute kidney injury caused by olaparib and abnormal liver function with rucaparib were significantly attenuated. Conclusion The toxicity differences of PARP inhibitors may be related to drug metabolic characteristics and target selectivity. Clinical attention should be paid to the hematological toxicity of olaparib, the hepatobiliary damage of niraparib, and the renal injury risk of rucaparib. Enhanced monitoring of target organs is recommended for elderly patients and those receiving combination therapies.