yulin wang Yongwen Huang ^*

• State Key Laboratory of Oncology in South China, Sun Yat-sen University Cancer Center (SYSUCC), Guanghzou, China

The current clinical data regarding the re-administration of PARPi maintenance therapy in platinum sensitive recurrent ovarian cancer (PSROC) is limited. This study aims to investigate the efficacy and associated factors of PARPi re-maintenance therapy in PSROC patients in China.In this study, there were 201 patients with PSROC who had received maintenance therapy previously and achieved complete or partial response after platinum-based chemotherapy upon recurrence. The re-maintenance therapy group (Re-PARPi) and chemotherapy alone group (Chem-A) were categorized based on whether PARPi was reused after recurrence chemotherapy. A propensity-score matching (PSM) analysis was conducted between re-maintenance therapy group (Re-PARPi-P) and chemotherapy alone group(Chem-A-P)to adjust for imbalanced risk factors. The efficacy was evaluated via progression-free survival (PFS) and prognostic factors were also analyzed.In the PSM subgroup, the median PFS (mPFS) of Re-PARPi-P group (44 cases) and Chem-A-P (44 cases) group were 10.0 months and 6.5 months (HR 1.64, P=0.041) respectively, confirming that re-maintenance therapy was superior to relapse chemotherapy alone. The mPFS was 10.8 months in all patients in the Re-PARPi group (51 cases), with 11.0 months in BRCAm group and 10.2 months in BRCAwt group (P=0.806). Intervals of more than 6 months between two PARPi therapies might improve the efficacy of PARPi re-treatment (mPFS 11.2 months vs. 7.8 months, HR 3.94, P=0.005). Age, BRCA status, number of previous treatment lines, CA125 level prior to re-administration, and other factors were not significantly related to the efficacy of re-maintenance therapy. Patients with a frameshift mutation (p. Ile1824Aspfs3) in the C-terminal domain of BRCA1 germline gene had significantly better efficacy with PARPi re-treatment compared to other groups. Only nonsense mutation (p.Gln1037, p.Cys328, p.Leu1072) occur in BRCA germline gene with re-treatment with PARPi might be suboptimal. The incidence of PARRi re-treatment interruption was 3.9%.PARPi re-maintenance therapy in PSROC might improve prognosis compared to chemotherapy alone, regardless of their genetic mutation status. Patients with re-maintenance therapy might benefit if the interval between the use of PARP inhibitors exceeded 6 months. The structural domains of BRCA mutations with different sensitivity to PARPi might serve as a promising biomarker for optimizing treatment. Re-treatment with PARPi was well-tolerated.