PARPis after PARPis in patients with recurrent epithelial ovarian cancer: a single institutional experience

To explore the clinical efficacy of PARPis maintenance treatment rechallenge in patients with recurrent epithelial ovarian cancer (EOC) in China.

We included patients diagnosed with primary EOC who received 2 lines of PARPis maintenance treatment after achieving complete response (CR) or partial response (PR) with the previous chemotherapies. The patients' full medical records were included in this study. Clinical and pathologic variables, treatment modalities, and outcomes were assessed. We investigated the treatment patterns and time to next treatment (TTNT).

A total of 31 ovarian cancer patients in our center were included. Among these patients, 20 of them (64.5%) had BRCA1/2 gene mutations. The median duration of PARPi1 and PARPi2 in the entire cohort was 11.2 months (range: 2.0-30.4 months) and 4.8 months (range: 1.0-16.7 months), respectively. Median TTNT1 and TTNT2 for the entire cohort was 12.4 and 7.7 months, respectively. Patients with BRCA1/2 mutation had a significantly better TTNT1 (median TTNT1: 17.3 vs 10.4 months, P = 0.005) than those without. A non- significantly better TTNT2 was observed in patients with BRCA1/2 mutation than those without (median TTNT2: 8.2 vs 5.0 months, P = 0.890). The association between previous chemotherapy response and TTNT was also analyzed. Patients who had a CR to previous chemotherapy had a significantly better TTNT1 (median TTNT1: 16.4 vs 7.6 months, P = 0.001) and TTNT2 (median TTNT2: 11.1 vs 4.9 months, P = 0.003) than those who had a PR. No grade Ⅲ-IV anemia occurred. Grade III PARPis-related thrombocytopenia was found in only 1 patient (3.2%, 1/31) who received PARPi2 treatment. For patients who developed PARPis-related anemia (n = 9) or thrombocytopenia (n = 7) during PARPi1 treatment, 7 patients (77.8%, 7/9) and 6 patients (85.7%, 6/7) developed anemia or thrombocytopenia again during PARPi2 treatment, respectively.

Patients with PARPis resistant recurrent EOC may derive benefit from PARPis re-treatment, especially for those with complete response to the last chemotherapy. Patients with BRCA1/2 mutation were more likely to benefit from PARPis retreatment than those with wild-type. Anemia and thrombocytopenia were more common in PARPis retreatment patients. A small proportion of patients had a longer benefit from PARPis retreatment than from previous PARPis treatment.