Real-world evidence of PARPi-related MDS/AML risk in breast cancer patients: An international collaborative network analysis.

Abstract

535 Background: Poly (ADP-ribose) polymerase (PARP) inhibitors have emerged as a significant therapeutic advance in breast cancer treatment. However, concerns about therapy-related myeloid neoplasms, specifically myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML), necessitate thorough investigation of their safety profile in real-world settings. Methods: Using the TriNetX Global Collaborative Network, we conducted a retrospective analysis comparing breast cancer patients treated with PARP inhibitors (PARPi) versus conventional chemotherapy only (anthracyclines and taxanes). Our primary analysis utilized propensity score matching (1,702 patients per group) accounting for age and race, to evaluate the risk of developing MDS/AML. Hazard ratio (HR) was used to compare the incidence of MDS/AML between the matched cohorts. Secondary analyses included an unmatched Cox proportional hazards model in a larger cohort (1,826 vs 36,257 patients), comparison between different PARP inhibitors, and assessment of mortality risk factors. Results: In the propensity score-matched analysis, PARPi treated patients demonstrated a statistically significant higher risk of developing MDS/AML versus chemotherapy only cohort (16 cases versus 10, HR=5.25; 95% CI: 1.96-13.92; p<0.0001). Treatment patterns differed notably, with PARPi-treated patients receiving more carboplatin (HR=1.73; 95% CI: 1.42-2.10) but less anthracycline therapy (HR=0.25; 95% CI: 0.20-0.31). The unmatched Cox regression analysis confirmed these findings with a higher risk of developing AML/MSD in the PARPi cohort (HR=3.47; 95% CI: 1.87-6.27) and identified age (HR=1.03; 95% CI: 1.02-1.05) and platinum therapy (HR=2.12; 95% CI: 1.26-3.59) as independent risk factors. Within the triple negative group, the data remains statistically significant (HR=3.14; 95% CI: 1.459- 6.757 ). No significant differences in MDS/AML risk were observed between olaparib and talazoparib. While mortality was comparable between groups, prior platinum exposure emerged as a significant mortality risk factor (HR=2.39; 95% CI: 1.09-5.09). Conclusions: Our findings indicate a significantly increased risk of therapy-related myeloid neoplasms with PARP inhibitor treatment compared to conventional chemotherapy, particularly in the context of previous platinum exposure in breast cancer patients. These results underscore the importance of careful patient selection and monitoring during PARP inhibitor therapy, while highlighting the need for extended follow-up studies to fully characterize long-term safety profiles.