Discussion
At present, a variety of PARP inhibitors have been used in the clinical treatment of ovarian cancer patients, covering several stages of treatment, which has become an epoch-making breakthrough in the history of ovarian cancer treatment. To our knowledge, this is the first study on the therapeutic effect of PARPi monotherapy compared with platinum-based chemotherapy, and the impact to subsequent platinum-containing chemotherapy and survival after the progression of PARPi in BRCA1/2 mutant patients with secondary platinum-sensitive relapse.
PARPi monotherapy for relapsed ovarian cancer with BRCA1/2 mutations has been proven clinically and has been validated in several clinical studies. In BRCA1/2 mutant patients with platinum-sensitive relapse who received at least two lines of platinum-based chemotherapy, the ORR ranged from 56.0% with niraparib to 80.0% with rucaparib15-19. However, ovarian cancer patients with BRCA1/2 mutations were inherently more sensitive to platinum-based chemotherapy than patients with wild-type ovarian cancer20, and the benefits of using PARPi inhibitors versus platinum-based chemotherapy at the same relapse stage were still uncertain. In our study, for patients with secondary platinum-sensitive relapse, the ORR was 77.4% and 84.0%, and the median PFS was 8.6 and 11.1 months, respectively. The therapeutic effect of PARPi monotherapy and platinum-based chemotherapy was equivalent.
The mechanisms of PARPi and platinum-based chemotherapy are both related to DNA damage repair, and the drug resistance mechanisms of PARPi include alterations in DNA damage repair, reactivation of HR, and replication fork protection, etc8. Theoretically, PARPi resistance may lead to subsequent platinum-based chemotherapy resistance. In Joo Ern’s study21, BRCA1/2 mutation patients who received 3-11 lines of platinum-based chemotherapy before Olaparib were included. After Olaparib resistance, the ORR of platinum-based chemotherapy was 40% (19/48), and the median PFS was 22 weeks, suggesting that there was still a partial response to platinum-based chemotherapy after PARPi resistance. Another study found that both platinum and non-platinum chemotherapy had a response rate after resistance of PARPi maintenance therapy, with median PFS of 7.0 and 8.5 months, respectively22. In our study, the PFI was significantly prolonged after PARPi monotherapy, and the median TFST of subsequent platinum-based chemotherapy after PARPi resistance was 7.5 months, consistent with patients in control group with platinum-sensitive relapse after third-line platinum-based chemotherapy. Further survival analysis confirmed that PRS of patients in the study group was similar to platinum-sensitive relapse patients, and superior to platinum-resistant relapse patients after third-line platinum chemotherapy. This result verified to some extent that although platinum-based chemotherapy had cross-resistance with PARPi, it does not negatively affect the efficacy of platinum-based chemotherapy after the progression of PARPi monotherapy.
PFS is currently the most widely used primary endpoint in clinical trials of PARP inhibitors. Most clinical studies have been conducted in recent years, so data on OS are still limited. In a phase 2 trial of Olaparib maintenance therapy, for patients with BRCA1/2 mutations, median PFS was significantly longer than placebo group (11.2 vs 4.3 months)23, but the OS benefit (29.8 vs 27.8 months) was not statistically significant24. In the Olaparib SOLO-2 trial, maintenance treatment with Olaparib extended the median OS by 12.9 months compared with placebo in patients with relapsed platinum-sensitive BRCA1/2-mutant ovarian cancer6. In the NOVA study, in the same cohort as the SOLO2 trial, maintenance therapy with Niraparib provided a 15.5-month benefit for PFS25, but no benefit for OS (43.6 vs 41.6 months)26. Results from the SOLO1 trial showed that Olaparib maintenance therapy extended the median PFS by 42 months compared to placebo in women with newly diagnosed advanced BRCA1/2 mutant ovarian cancer, although OS data are not yet available, it is expected that OS will benefit from a large increase in PFS as well27. For other PARPi, as well as for patients with other indications, the benefits of OS remain to be determined. In our study, patients with secondary platinum-sensitive relapse were selected and the results showed that the benefit of subsequent platinum-based chemotherapy was not negatively affected after the progression of PARPi monotherapy, and there was still a significant extension of PRS, which reduced the risk of death by 65%.
Factors affecting survival of ovarian cancer patients included tumor histology, FIGO stage, BRCA mutation status, ascites, and whether no residual lesions could be achieved after primary debulking surgery28. In a study with up to 10 years of follow-up, the initial survival advantage in patients with BRCA1/2 mutations may reflect a higher initial sensitivity to chemotherapy, but this response does not predict long-term survival, the strongest predictor of long-term survival was no residual lesions at resection29. In our study, we found that the factors affecting PRS included R0 resection and PARPi monotherapy, after incorporating FIGO stage for multivariate analysis, PARPi monotherapy was the independent prognostic factor, which also reflected the superior therapeutic effect of PARPi monotherapy compared with R0 resection.
To a certain extent, our research has significant advantages. First, this is the first study to evaluate benefits of PARPi monotherapy versus platinum-based chemotherapy at the same relapse stage of BRCA1/2 mutant patients. In addition, the four-line treatment information was collected, at a stage with little evidence for treatment, besides, data of patients who did not receive PARPi treatment before as control group were difficult to obtain. However, the most significant limitation of our retrospective study was the limited number of patients. BRCA1/2 mutations account for less than 30% of ovarian cancer patients30, and those who did not met the criteria for secondary platinum-sensitive relapse were excluded, as were those on maintenance therapy with PARPi or bevacizumab. Although the results of the analysis in our study were significantly different, further studies with large samples should be necessary. The findings of this study were applied only to a specific subset of the ovarian cancer patient population, not to all patients in general.