Upfront HIPEC and bevacizumab-containing adjuvant chemotherapy in advanced epithelial ovarian cancer.

Abstract Introduction: In advanced epithelial ovarian cancer patients, the standard of care is primary debulking surgery, followed by first-line chemotherapy often with bevacizumab addiction. In this context, some experiences have shown that a comprehensive treatment approach to surgery combined with hyperthermic intraperitoneal chemotherapy (HIPEC) could improve the prognosis. Objective: This is a study aimed to explore the feasibility of primary debulking surgery and HIPEC upfront followed by first-line therapy with bevacizumab. Study Design: Phase II monocentric, open label, non-randomised and single-arm study. Forty patients affected by advanced ovarian cancer submitted to primary debulking surgery with HIPEC were enrolled in the study. After surgery, all patients underwent systemic chemotherapy with bevacizumab addiction. Results: Complete cytoreduction (RT = 0) was achieved in all cases. Treatment-related early complications were observed in 23 patients and in 15 cases were G1–G2. Major complications were reported in 8 patients. No postoperative death was recorded. Subsequent chemotherapy was administered in all cases. Median time between surgery and first cycle of chemotherapy was 42 days (range 30–76). Concomitant bevacizumab was administered in 34 patients (85%). Maintenance with bevacizumab was feasible in 33 patients (82.5%) and its withdrawal was necessary for 1 patient (2.5%) due to G3 hypertension. Conclusion: Our data suggest that HIPEC can be safely introduced in the upfront therapy of advanced ovarian cancer.


Introduction
In advanced epithelial ovarian cancer (AEOC) patients, the standard of care is primary debulking surgery (PDS), aimed to reach completeness of cytoreduction, followed by first-line chemotherapy with carboplatin and paclitaxel [1]. For patients in stage IIIC/IV that are not ideal candidates for PDS, the treatment that has recently gained acceptance is a combination of neoadjuvant chemotherapy followed by interval debulking surgery (IDS) [2][3][4][5].
Modifications to the carboplatin-paclitaxel regimen, including extension of front-line agents, high-dose chemotherapy, immunotherapy, biological therapy and single-agent paclitaxel, have made limited improvements to survival rates. At present, the only treatment that has prolonged progression-free survival (PFS) is the addiction of anti-VEGF monoclonal antibody bevacizumab to carboplatin-paclitaxel, as reported in the ICON7 and GOG-0218 phase 3 studies [6,7]. Therefore, there is still a significant unmet need in the first-line therapy for ovarian cancer.
In this context, some experiences have shown that a comprehensive treatment approach of surgery combined with hyperthermic intraperitoneal chemotherapy (HIPEC) could be able to improve the prognosis of AEOC as reported in few clinical studies [8][9][10][11][12][13][14][15][16][17]. Despite the established rationale and these encouraging results, a certain degree of skepticism still surrounds HIPEC in AEOC, involving inherent potential morbidity and the paucity of randomized data confirming its theoretical advantage. Moreover, the literature on HIPEC safety and efficacy in combination with bevacizumab is very less and therefore needs to be better investigated. Based on these considerations, we conducted a phase II monocentric, open label, non-randomized and single-arm study aimed to explore the feasibility of PDS and HIPEC upfront followed by first-line therapy with bevacizumab, as GOG-0218 schedule [7].

Materials and methods
This prospective phase II monocentric, open label, non-randomised and single-arm study was conducted at Division of Gynecologic Oncology of Agostino Gemelli Foundation University Hospital in Rome, from February 2015 to February 2016 and was approved by our IRB (protocol n. 0115/2015). Informed consent was obtained from all subjects. The diagnosis was obtained at frozen section during surgery.
All enrolled patients underwent pre-operative evaluation by CT scan, pelvic ultrasound and tumor markers. Major criteria to abort PDS were the Poorest Eastern Cooperative Oncology Group performance status (i.e., ECOG-PS >2) and/ or higher American Society of Anesthesiology score (i.e., ASA >2). According to previously published data [18], all patients were submitted to Staging-LPS in order to evaluate and quantify peritoneal dissemination of the tumor through a scoring system (i.e., PIV) [19], and only patients with a score <8 were included in the final analysis.
The exclusion criteria were as follows: FIGO stage less than IIIB; coexistence of other oncologic disease; body mass index (BMI) > 30 kg/m 2 ; active infection or general conditions that could interfere with treatment (vasculopathy, autoimmune disorders and diabetes); refusal to sign the informed consent form; previous recipient of chemotherapy treatment; distant (extra-abdominal) unresectable metastases and bowel obstruction.
The patients who met inclusion criteria, and that were considered suitable for PDS at Staging-LPS, underwent mono/bilateral adnexectomy or peritoneal biopsy to confirm the diagnosis of ovarian cancer at frozen section. If the diagnosis of ovarian cancer was confirmed, the patient was submitted to PDS with the aim to achieve complete cytoreduction (RT ¼ 0). The completeness of cytoreduction (CC) was assessed using a score ranging from 0 to 3 (CC-0 indicates no residual tumor; CC-1 indicates nodules <0.25 mm; CC-2 indicates nodules between 0.25 and 2.5 cm in diameter and CC-3 indicates nodules >2.5 cm). After completion of cytoreduction, four drains were positioned in the four abdominal quadrants. HIPEC perfusion was performed with closed technique, and the abdomen was carefully reexplored after HIPEC completion. All patients received intraperitoneal cisplatin 75 mg/m 2 at the temperature of 41.5 C for 60 min immediately after PDS. All patients underwent systemic adjuvant chemotherapy with bevacizumab according to international guidelines [1]. Physical examination, thoracic/ abdominal CT scan and Ca 125 serum level assessment were all performed every 3 months during the first 2 years and every 6 months thereafter. Primary platinum-free interval (PFI) was defined as the time elapsed between the end of carboplatin treatment and first recurrence. Data are given as median and range. Categorical variables are reported as absolute values and percentage.

Statistical analysis
The sample size was quantified based on previous studies reporting a pooled rate of postoperative major (G3-G4) complications ranging between 45% and 98% [22] disabling an early (<40 days) start to adjuvant chemotherapy (ICON 7). Based on the minimax 2-stage design by Simon [23], we tested the null hypothesis that the true rate of an early start to the administration of chemotherapy with bevacizumab after PDS and HIPEC could reach clinically relevant alternative of 85%, using an alpha-error of 0.05 and a beta-error of 0.2.
Thus, the first step was planned to include 31 patients; if >25 (80%) women started adjuvant chemotherapy with bevacizumab before 40 days, the study would enroll an additional 5 patients up to a total number of 36 patients. Considering a dropout rate of 10%, at least 40 cases were planned to be enrolled.
PFS was calculated from the date of diagnosis to progression of disease or the date last seen while overall survival (OS) was calculated from the date of diagnosis to the date of death of disease or the date of the last follow-up. Data analysis was performed using the NCSS statistical software program, version 11.0 (NCSS Statistical Software, Kaysville, UT) was used.

Results
Forty patients were prospectively enrolled. Patients' characteristics are shown in Table 1. The details of PDS and HIPEC procedures are shown in Table 2. Median surgical complexity score (SCS) was 3 (range: 2-3). Complete cytoreduction (RT ¼ 0) was achieved for all cases. Median operative time was 480 min (range: 360-740) and median Cisplatin dose was 126.5 (100-148). Median postoperative hospital stay was 8 days (range: 5-30). Diaphragm peritonectomy were performed in 67.5% while diaphragm resection in 7.5% of cases. Splenectomy was performed in 75% and 30% of patients, respectively. Pelvic/lombo aortic lymphadenectomy were performed in 62.5% of cases only when metastatic lymph nodes were detected. Treatment-related early complications were observed in 23 patients and in 15 cases were G1-G2. Major complications consisting of pleural effusion requiring drain and bowel anastomosis dehiscence were reported in five and three patients, respectively. Late complications were mild and related to kidney failure ( Table 2). No postoperative death was recorded.
Subsequent chemotherapy was administered ( Table 3) in 100% of cases (40 patients). Median time between surgery and the first cycle of chemotherapy was 42 days (range 30-76). Concomitant bevacizumab was administered in 34 patients (85%). Maintenance with bevacizumab was feasible in 33 patients (82.5%) and its withdrawal was necessary for 1 patient (2.5%) due to hypertension G3. Six out of 40 patients (15%) were not treated with bevacizumab for the following reasons: four patients experienced proteinuria and kidney failure G2 after HIPEC; one patient developed central venous thrombosis and one patient showed a poor performance status after HIPEC (ECOG 2).
At the time of this analysis, with a median follow-up of 25 months (range 5-40), the progression of disease occurred in seven patients (six peritoneal progressions and one lung/ mediastinum metastasis). At present, 37 patients remain alive (Figure 1).

Discussion
Primary debulking surgery followed by chemotherapy is the cornerstone of AEOC treatment. The addiction of the biological anti-angiogenic agent bevacizumab to standard chemotherapy resulted in a prolongation of PFS, suggesting that the combination of carboplatin, paclitaxel and bevacizumab may become the new standard in the first-line treatment of AEOC [24]. However, survival [2,3] results in ovarian cancer remain largely unsatisfactory. In this context, HIPEC has been proposed as a promising strategy based on several theoretical reasons: (i) i.p. chemotherapy is certainly effective in the management of AEOC, as reported in several randomized clinical trials [7]; (ii) hyperthermia has proved to enhance cytotoxicity of platinum compounds [25] and (iii) starting chemotherapy at surgery virtually avoids any delay in chemotherapy. This last reason could be particularly significant because it has been demonstrated that a delay of 7 days in beginning chemotherapy resulted in an 8.7% increase of mortality in patients with complete surgical debulking [26].
In our study, we have found that PDS with HIPEC is feasible and can be combined with the most active primary therapy presently available in AEOC, i.e., carbo-taxol bevacizumab. Despite the aggressive surgical procedures performed, toxicity was mild and easily managed (20% of G3-G4 morbidity) with a median hospital stay of 8 days. This result is relevant because the risk of increased postoperative complications after primary debulking surgery and   0 Hospital stay (median) (range) 8 (5-30) a https://www.nccn.org/professionals/physician_gls/pdf/ovarian.pdf. b Surgical complexity score (SCS). Ã Pleural effusion (requiring drain). ÃÃ Re-laparotomy (bleeding, bowel anastomosis dehiscence). ÃÃÃ Two kidney failure, two kidney failure. 20 (50.0) Anemia G3 2 (5.0) Thrombocytopenia G3-G4 1 (2.5) Non-hematological toxicity Hypertension G2-G3 3 (7.5) Peripheral neuropathy G2 3 (7.5) Ã Only one cycle for hypertension G3. ÃÃ Four patients did not receive bevacizumab due to kidney failure G2 after PDS þ HIPEC; the other two patients had DVT. carboplatin-paclitaxel-bevacizumab adjuvant chemotherapy was raised as a potential issue in the management of AEOC based also on experiences in colorectal cancer treatment. Interestingly, as reported by Duska et al. [27] the addition of bevacizumab to conventional first-line regimen does not imply an increased risk of readmission or postoperative complications. As the subgroup of patients experiencing multiple readmissions (!2) only accounted for around 3% of the entire population, the use of bevacizumab seems to be detrimental. Furthermore, the paper by Duska et al. [27] is also highly valuable to identify the correct time-interval between primary cytoreductive surgery and adjuvant chemotherapy. In particular, given the observation that patients readmitted within 40 days of surgery had a significantly shorter interval from surgery to chemotherapy initiation (22 versus 32 days, p < .0001), 40 days seems to be the gold-standard time-interval to be respected prior to starting adjuvant carboplatinpaclitaxel-bevacizumab chemotherapy. In our series, the median time to start chemotherapy of 42 days suggests that HIPEC addiction does not influence the ideal time to start chemotherapy.
As far as the combination of HIPEC and bevacizumab in ovarian cancer, our data are consistent with the recent paper by Gouy et al. [28] which demonstrated that bevacizumab maintenance treatment could be safely completed on around one-third of patients, with six cycles of carboplatin-paclitaxelchemotherapy followed by IDS and HIPEC. Interestingly, this percentage is completely in line with results from the GOG-0218 and ICON-7 trial, suggesting that even an aggressive multimodal approach combining neoadjuvant chemotherapy (NACT), IDS and HIPEC does not affect the chance of successfully complete bevacizumab-maintenance therapy without enhanced toxicities [6,7].
At present, there is still no consensus in the actual indication to perform HIPEC in ovarian cancer. Despite several studies seems suggest a benefit of HIPEC treatment in ovarian cancer and new technologies are now available [29], no conclusions can yet be drawn. This is due to several limitations and biases of the studies available, which consist of small single institution and not homogeneous series utilizing different drug dosage/schedule and time of exposure in different clinical settings. Recent results of a randomized phase III study suggested that HIPEC at IDS might improve survival of patients undergoing neoadjuvant chemotherapy [8].
Moreover, while waiting for the conclusion of several other randomized trials currently in progress (HORSE NCT01539785, CHORINE NCT01628380 and MMC 2014 NCT02124421), one RCT [8] and one case-control [13] study suggest a potential role of HIPEC in the improvement of patient prognosis at Interval debulking surgery and recurrent ovarian cancer, respectively. Finally, in a recent systematic review and meta-analysis of 37 studies in ovarian cancer by Huo et al. [22], the combination of HIPEC with cytoreductive surgery plus adjuvant chemotherapy, showed significantly better survival compared with cytoreductive surgery plus adjuvant chemotherapy alone. The improved results were reported both for upfront and recurrent settings.
Furthermore, the benefit of HIPEC would continue for 2-8 years after the procedure.
In conclusion, our data suggest that HIPEC can be safely introduced in the upfront therapy of AEOC consisting of primary debulking surgery and carbo-taxol-bevacizumab chemotherapy. On this basis, phase III randomized studies will now be needed to evaluate the prognostic impact of HIPEC in ovarian cancer management.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
This work was supported by Actagroup srl.