In this retrospective study, we found that without considering the BRCA1/2 mutation status, both PFS and OS after neoadjuvant chemotherapy followed by interval debulking surgery were similar to survivals with primary surgery followed by chemotherapy, which was consistent with the conclusions of previous randomized controlled trials[8, 15]. In BRCAmut patients, NAC-IDS significantly shortened PFS, however, it had no effect on OS. For BRCAwt patients, NAC did not significantly affect the prognosis. To our knowledge, this is the first retrospective study to date to discuss the effect of NAC on prognosis based on the BRCA mutational status.
The proportion of patients with BRCA1/2 mutations was 35.1%, which is much higher than that previously reported in the literature[16–18]. This could be related to the fact that all patients included in the study were at an advanced stage. Both BRCA1 and BRCA2 are tumor suppressor genes, and their functional proteins play an important role in DNA homologous recombinant double-stranded break repair. In the absence of functional BRCA1/2 gene, tumor cells are more aggressive and therefore more prone to distant metastasis, leading to an increased proportion of patients with FIGO III and IV stage[19]. According to previous reports, the highest proportion of BRCA1/2 mutation in Chinese ovarian cancer patients was 28.5%. However, if stage III and IV patients were separately counted, the proportion of patients with BRCA1/2 mutation was significantly increased in these studies[16, 18].
The primary evaluation criteria for the initial treatment of advanced ovarian cancer patients is whether satisfactory cytoreductive surgery can be achieved in PDS, especially R0 resection, which could significantly prolong survival[20]. At present, clinical practice guidelines and expert consensus suggest that NAC is recommended for FIGO stage III to IV patients with poor physical status and unable to tolerate surgery, and for patients in whom it is difficult to achieve satisfactory tumor cytoreductive surgery (R0 and R1)[20–22]. Several studies have confirmed that the serum CA-125 level was an important predictor of surgical outcome, defined as successful cytoreductive surgery with a residual tumor ≤ 1 cm, and is a significant factor in decision-making regarding the proper selection for PDS or NAC[23–25]. In our study, the largest primary lesion at diagnosis was similar in both the NAC and PDS groups regardless of BRCA mutation, but the CA-125 levels in patients receiving NAC were significantly higher than those in the PDS group, which is one of the important reasons for choosing NAC. In addition, NAC can significantly increase the proportion of R0 resection and reduce the amount of bleeding during surgery, consistent with the conclusions of previous reports[6, 15].
Although R0 resection rates were significantly improved in both BRCAmut and BRCAwt patients receiving NAC, the OS did not improve, nor did PFS in BRCAwt patients. In contrast, in BRCAmut patients, NAC significantly shortened PFS; further multivariate analysis found that NAC-IDS, FIGO IV stage, and R1/R2 resection were risk factors for poor PFS. Compared with BRCAwt patients, BRCAmut patients previously showed a significantly increased peritoneal tumor load[26], were associated with nodular peritoneal disease pattern[27], and showed increased sensitivity to NAC[28]. In our study, the proportion of BRCAmut patients who achieved partial response after neoadjuvant chemotherapy was 75.0%, which was significantly higher than that of BRCAwt patients (50%). The above reasons could cause more peritoneal lesions to become invisible after NAC and, thus, cannot be completely resected; therefore, R0 resection may not be truly achieved; instead, the complete response (CR) status was achieved through chemotherapy, which may be the main reason for reduced PFS in BRCAmut patients.
BRCA1/2 genes are tumor suppressors, that play an important role in DNA damage repair and normal cell growth. BRCA1/2 mutations can inhibit the repair of DNA damage, resulting in homologous recombination deficiency, which eventually lead to carcinogenesis[29]. In BRCA1/2 mutated tumor cells, DNA double-stranded repair function is lost; since PARP inhibitors can block single-stranded DNA repair, this results in a "synthetic lethal" effect that leads to the death of tumor cells[9]. Therefore, PARP inhibitors have excellent maintenance treatment effects in BRCAmut patients, which can significantly prolong the PFS and OS[30, 31]. In the SOLO-2 clinical trial of olaparib[32], maintenance therapy with olaparib in BRCAmut platinum-sensitive relapse patients extended the relapse time by 24.7 months and reduced the risk of recurrence or death by 70%. In addition, the objective response rate in BRCAmut patients with platinum-sensitive relapses above three lines was about 70% when treated with PARP inhibitors[12]. The evaluation of PFS in our study, although excluding patients with PARP inhibitors for maintenance therapy, there were a large number of patients treated with PARP inhibitors in the posterior line of treatment. This is probably the main reason for the subsequent overall survival consistency.
This retrospective study had several limitations. First, several studies have suggested that surgery is appropriate after three cycles of NAC[6, 8, 21]. Patients with 1–3 cycles were included in our study; however, patients with one cycle of chemotherapy, were not suitable if PR was not achieved, which may affect the prognosis to some extent. In addition, we did not collect details of patients using PARP inhibitors in the posterior line; therefor, no further analysis of OS was performed to exclude the effect of PARP inhibitors.