RESUMO
INTRODUCTION: This study aimed to estimate the toxicities of PARP inhibitors (PARPis), based on randomized controlled trials (RCTs) and the FDA Adverse Event Reporting System (FAERS) database. METHODS: Four electronic databases were searched from inception to 16 April 2024, for RCTs of approved PARPis. The primary and secondary outcomes were grade 3-5 adverse events (AEs) and grade 3-5 hematological AE, respectively. We conducted network meta-analyses to calculate the relative risks (RRs) and 95% confidence intervals (CIs) of outcomes. A disproportionality analysis was conducted to estimate the signals of hematological AEs associated with PARPis from the FAERS database. RESULTS: Overall, 27 RCTs involving 11,067 patients with cancer were included. Olaparib had the best safety profile for any grade 3-5 AEs and hematological AEs among four approved PARPis. Olaparib did not increase the risk of thrombocytopenia (RR: 1.48; 95%CI: 0.64-3.39), but other PARPis did. Furthermore 14,780 hematological AE reports associated with PARPis were identified in the FAERS database, and all PARPis were associated with strong hematological AE signals. Hematological AEs mainly occurred within the first 3 months (80.84%) after PARPi initiation. CONCLUSION: Olaparib had the best safety profile among five PARPis. PARPi-associated hematological AEs mainly occurred within the first 3 months. REGISTRATION: PROSPERO (CRD42022385274).