Abstract
Background: To assess impact of surgery and hyperthermic intraperitoneal chemotherapy (HIPEC) in patients treated for a first relapse of ovarian cancer (FROC). Patients and Methods: Patients with a FROC treated with second-line chemotherapy, surgery and HIPEC were retrospectively included from 13 Institutions. Studied parameters were interval free between the end of initial treatment and the first relapse, second-line chemotherapy, peritoneal cancer index and completeness of surgery, HIPEC, mortality and morbidity, pathological results and survival. Results: From 2001 to 2010, 314 patients were included. The main strategy was secondary chemotherapy followed by surgery and HIPEC (269/314-85.6%). Mortality and morbidity rates were respectively 1% and 30.9%. Median follow-up was 50 months, 5-year overall survival was 38.0%, with no difference between platinum-sensitive or -resistant patients and 5-year disease-free survival was 14%. Conclusion: HIPEC allows encouraging survival in the treatment of FROC, better in case of complete surgery, with acceptable mortality and morbidity rates.
Epithelial ovarian cancer is the first cause of death from gynaecological cancer (1). In Europe, the average 5-year survival is 37.6% (2). More than 75% of patients treated for advanced epithelial ovarian cancer experience a relapse during the first five years (3). In cases of relapse, the main prognostic factors are the progression-free interval between the end of initial treatment to diagnosis of relapse, physical status, ascitis, number and size of tumor nodes, and feasibility of a complete secondary cytoreductive surgery (4, 5).
Although primary treatment of ovarian cancer is well-established with the combination of complete cytoreductive surgery and courses of platinum-based intravenous chemotherapy, treatment of relapse is based on multiple novel targeted-therapeutics and the role of secondary cytoreductive surgery is controversial (6). Treatment strategies are led by the description of relapse as platinum-sensitive (recurrence >6 months after the end of initial treatment) or -resistant (recurrence <6 months) (7).
For platinum-sensitive relapse, standard treatment is based on secondary carboplatinum and pegylated liposomal doxorubicin or carboplatinum and paclitaxel intravenous chemotherapy, with a median overall survival of 30 to 33 months and a disease-free survival of 11 to 13 months (8,9). For platinum-resistant relapse, treatment is based on single-agent chemotherapy. Aurelia, a recent trial with single-agent chemotherapy and bevacizumab, led to a median overall survival of 16.6 months and a disease-free survival of 6.7 months (10).
Even if adding secondary cytoreductive surgery to second-line chemotherapy is not standard treatment for first relapse, survival results of secondary cytoreductive surgery in series of more than 100 patients were found to be encouraging, with a median overall survival ranging from 45 to 63 months (5, 11, 12). An international randomized trial, DESKTOP III from the German arbeitsgemeinschaft gynäkologische onkologie group, is ongoing to assess the role of secondary cytoreductive surgery as a frontline treatment for first platinum-sensitive relapse in patients for whom complete surgery before any course of intravenous chemotherapy is feasible, with a good performance status and ascitis less than 500 cc at the time of the relapse (13).
Since peritoneal carcinomatosis is the main form of relapse, intraperitoneal treatments could be assessed. Hyperthermic intraperitoneal chemotherapy (HIPEC), combines complete cytoreductive surgery and heated intraperitoneal chemotherapy. It was initially used in ovarian cancer treatment by Steller in 1999 (14). Several clinical series of HIPEC, in association with the standard courses of intravenous platinum-based chemotherapy for ovarian cancer relapse, were reported, with interesting overall survival results of 28 to 54 months (15). Two-thirds of HIPEC performed for ovarian cancer are performed with cisplatin alone or in combination with mitomycin, mitoxantrone or doxorubicin (16). Recently, Mulier et al., assessed the better potential survival benefit of adding HIPEC to the treatment of ovarian cancer at different time points of the disease, and suggested that survival benefit may be the largest in the settings of secondary secondary cytoreductive surgery for relapse (17).
It is important to point-out that clinical series of HIPEC dedicated to ovarian cancer relapse are characterized by a small number of patients, retrospective data collection, mono-Institutional setting, and heterogeneous inclusion criteria and treatment characteristics, making it difficult to draw any conclusion on HIPEC efficiency (14). In 2013, we published a multi-centre retrospective study of 566 patients with ovarian cancer treated with secondary cytoreductive surgery and HIPEC (18). The cohort was a mixed series, with patients treated for primary ovarian cancer or a relapse, from the first to the third relapse. The current cohort focuses on a specific selection of patients with a first relapse. For this new analysis, we performed a central review of pathological reports of secondary cytoreductive surgery in order to assess the impact of pathological response to chemotherapy and we updated patient information about overall (OS) and disease-free (DFS) survival. Our main objective was survival assessment. Our secondary objective was determination of morbidity, mortality and prognostic factors.
Patients and Methods
Patients. A retrospective observational multi-Institutional study from prospective databases was conducted. All patients met the following inclusion criteria: a first relapse of epithelial ovarian cancer, with two exclusion criteria: two previous HIPEC and use of bevacizumab. Patients were treated with second-line chemotherapy, secondary cytoreductive surgery and HIPEC.
The current series was extracted from a previously published French cohort (18).
According to Markman's classification, two groups were distinguished: i) Those with chemoresistant epithelial ovarian cancer, including relapse after a progression-free interval (PFI) of less than 6 months after the end of initial treatment or persistent disease following optimal standard first-line therapy using combined cytoreductive surgery and systemic chemotherapy, and ii) those with chemosensitive epithelial ovarian cancer relapse after a disease-free interval of more than 6 months after the end of initial treatment (7). The definition of complete surgery was no residual disease.
Studied parameters. Studied parameters were: age, American Society of Anesthesiologist score (19), performans status, residual at initial surgery, and extent of peritoneal carcinomatosis at the time of the first relapse, which was assessed intraoperatively with Sugarbaker's Peritoneal Cancer Index (PCI) (20). For the descriptive analysis, two subgroups of PCI were created: those with <8, and those with ≥8. PCI cuts where chosen according to a Mayo Clinic findcut (http://www2.sas.com/proceedings/sugi28/261-28.pdf).
Information recorded about HIPEC included the date, type (open or closed method), drug used, simultaneous resection of metastases, pathological residual at the time of HIPEC (absence of residual, microscopic residuals, macroscopic residuals) and lymph nodes (presence/absence) and completeness of the cytoreductive surgery (CC score). The CC score was determined by the surgeon at the end of the procedure and classified into three categories: CC-0 indicated that no macroscopic residual cancer remained, CC-1 indicated residual nodule from 0.25 mm to 2.5 mm in size, and CC-2 indicated that the diameter of residual nodules was greater than 2.5 mm (20). Treatment with postoperative systemic chemotherapy was also recorded.
Patients' characteristics at the time of relapse.
Postoperative mortality was assessed within 30 days and postoperative morbidity within 90 days after surgery.
Follow-up data included the vital status of the patient, the relapse status and date of the last event, and new relapse or death. Particular focus was placed on survival of patients treated with a cisplatinum-based HIPEC in order to build hypotheses for further trials with HIPEC based on cisplatinum alone.
A central reviewing of surgical and pathological reports was performed by the same author (NB).
The study and the constitution of patient cohort were approved by institutional scientific board and national committee for data protection (910306 v 1, December 12 2011).
Statistical methods. Descriptive analyses of patients' initial characteristics were performed on a patient basis (one record per patient). Categorical variables were described in terms of frequency and percentages. Continuous variables were described with median, minimum and maximum distributions.
Postoperative morbidity and mortality analyses (death within 30 days, or major complications occurring within 90 days after surgery) were performed.
Overall survival time was defined as the time from the first HIPEC to the date of death, the date of last follow-up, or the cut-off date (February 8 2013), whichever came first. Peri-operative deaths were not excluded from the survival analysis. Relapse-free survival was defined as the time from first HIPEC until relapse or last follow-up. Deceased patients were censored at the date of death. Patients with relapse at the same date of HIPEC or patients treated with CC-2 resections were considered as having immediate relapse and were not included in the relapse-free survival analysis. PCI cutoffs were those chosen for the main cohort (18).
Mortality at 30 days and morbidity at 90 days postoperatively.
Median survival times and survival rates were computed using the Kaplan–Meier method. Hazard ratios and p-values were obtained with the Cox proportional hazards model. Influence of baseline risk factors was assessed using univariate and multivariate Cox proportional hazards model adjusted for institution. Multivariate overall survival was also adjusted for age. Factors with significance of 0.20 or less and with fewer than 25% of missing data were introduced into multivariate model, p-value less than 0.15 were retained in the final model. SAS statistical software (V9.2, SAS Institute Inc., Cary, North Carolina, USA 27513) was used for all analyses.
Results
From January 2001 to December 2010, a total of 314 patients were identified from 13 Institutions.
Patients' characteristics. At the time of initial treatment, surgery was complete in 33.8% (101/314) patients, there was macroscopic residual disease in 66.2% (98/314) patients and this information was unknown for 15 patients.
Patient's characteristics at the time of first relapse are listed in Table I. Among the 314 patients, there were 166 patients with an early relapse i.e. <6 months after the end of initial treatment, and 148 late relapse, at more than 6 months. At the time of the relapse, the mean PCI of patients treated for a chemoresistant relapse (occurring at less than 6 months after the end of initial treatment) was not different from the mean PCI of patients treated for a chemosensitive relapse (>6 months).
Treatment strategy was secondary chemotherapy followed by secondary surgery and HIPEC for 85.6% (269/314) of patients, the remaining patients were treated with secondary surgery and HIPEC before secondary chemotherapy. A total of 236 patients had HIPEC with cisplatin with/without other chemotherapy, among this group 116 patients had HIPEC with cisplatin alone.
Prognostic factors for overall survival.
Table II summarizes postoperative mortality and morbidity results. The postoperative mortality rate was 1% (3/314) and the rate of grade III complication was 30.9% (97/314), with a re-laparotomy for morbidity in 8.9% (28/314) cases. Postoperative morbidity rate was significantly higher in case of PCI up to 9 (p=0.001). Postoperative mortality rate was not different in cases of PCI up to 9 (1 vs. 2%, p=0.59).
The central review of pathological reports was performed for the group of 204 patients treated with second-line chemotherapy followed by secondary surgery and HIPEC in order to assess the pathological response to second-line chemotherapy. Among the group of 204 patients, 46% (95/204) were classified as having chemoresistant disease (pathological complete response, n=15; microscopic residual, n=23; macroscopic residual, n=57) and 53% (109/204) chemosensitive (pathological complete response, n=12; microscopic residual, n=19; macroscopic residual, n=78). Residual disease distribution was not different between chemoresistant and chemosensitive cases.
Survival. Median follow-up was 50 months. A total of 155/314 patients were alive at the end of the follow-up.
Prognostic factors for Recurrence-free survival.
The 5-year OS was 38.0% and 5-year DFS was 14%. Univariate and multivariate analysis are presented in Tables III-IV. Chemosensitivity was favourably linked with relapse-free survival (median DFS of 13 months for chemosensitive and 14 for chemoresistant patients; p=0.013) but not with OS (median OS of 42 months for chemosensitive and 51 for chemoresistant cases; p=0.38) (Figure 1). Considering the group of 116 patients who underwent HIPEC with cisplatin alone, the median OS was 45 months.
CC-score was linked both with OS (median OS of 54 months for CC0 and 36 for CC1-2; p=0.0007) and relapse-free survival (median RFS of 15 months for CC0 and 10 for CC1-2; p=0.018) (Figure 2).
Considering patients treated with second-line chemotherapy followed by secondary surgery and HIPEC, the remaining pathological residual disease was linked both to OS and DFS. Median OS was not reached for patients with no pathological residual disease, while for patients with microscopic residual disease it was 56 months and for patients with a macroscopic residual disease 39 months (p=0.073). The median DFS was 41 months for patients with no pathological residual disease, 13 months for patients with microscopic residual disease and 10 months for those with a macroscopic residual disease (p=0.0019) (Figure 3).
Overall (a) and recurrence-free (b) survival according to chemoresistance (chemoresistant: relapse diagnosed <6 months after the end of initial treatment; chemosensitive: relapse diagnosed >6 months after the end of initial treatment).
In multivariate analysis, PCI, with a mean value of 8, was the only factor remaining significantly linked with overall survival (Table III).
Overall (a) and recurrence-free (b) survival according to completeness of secondary cytoreduction (CC) score. CC 0: No residual disease; CC: residual nodule from 0.25 mm to 2.5 mm; CC-2: residual nodule >2.5 mm.
Discussion
Our multi-Institutional series of patients treated for a first relapse of ovarian cancer with second-line chemotherapy, secondary cytoreductive surgery and HIPEC, had a median OS of 45 months. Including 314 patients, the current study is the largest multi-Institutional series of patients treated specifically for a first relapse with secondary surgery and HIPEC. Our series represents what can be achieved in expert Center in terms of survival, mortality and morbidity, with HIPEC in the treatment of a first ovarian cancer relapse, useful to build hypothesis for further randomized trials. We found no difference regarding OS between platinum-sensitive and platinum-resistant relapse. The main prognostic factors were completeness of secondary surgery, pathological residual disease after second-line chemotherapy and the PCI at the time of secondary surgery.
Overall (a) and recurrence-free (b) survival according to pathological residual disease at the time of secondary surgery and HIPEC.
In the current series, the mortality rate was 1%. A systematic review of patients treated for an epithelial ovarian cancer with cytoreductive surgery and HIPEC found a mortality rate of between 0 and 17%, falling to 0.9 to 5.8% when considering expert center with high volumes (21). In a review of 23 published series of surgery for advanced ovarian cancer without HIPEC, Gerenstein et al. found a postoperative mortality rate of 2.8% (22). In expert center, the reduction in mortality after surgery for advanced ovarian cancer does not appear to be due only to lower complication rates but also to the ability of high-volume hospitals to rescue patients with complications (23).
In the current series, the rate of grade III complications was 30.9%, with 8.9% re-laparotomy. In a recent series of 43 patients treated with complete secondary cytoreductive surgery and HIPEC for a first relapse, the severe morbidity rate was 34.8% (24). In series of HIPEC, the rate of grade III-IV postoperative morbidity was linked with performance status, extent of carcinomatosis, number of surgical procedures, duration of surgery, and HIPEC characteristics such as the dose of chemotherapy (21). In cases of extensive surgery for advanced ovarian cancer without HIPEC, the rate of severe morbidity ranged between 12 and 52% in expert high-volume Centers (25).
Considering the question of mortality and severe morbidity rates, HIPEC must still not be considered as a highly morbid procedure, but as equivalent to extensive surgery without HIPEC, even in cases of first relapse. Careful patient selection in expert Centres, with an optimal level of postoperative care, must be advocated to avoid undesirable complications of this treatment (21).
In the current study, the median OS was 45 months and median DFS was 14 months. OS of patients treated for a first relapse with complete secondary surgery and HIPEC, in series with more than 50 patients, ranged between 23 and 48 months (26-28). In the absence of a control group, it is impossible to conclude if these results are related to HIPEC or to selection of patients fit for such a heavy treatment. In the current series, a low PCI at the time of the relapse, with a mean value of 8, was the only factor remaining significantly linked with overall survival in multivariate analysis.
One interesting finding is the lack of difference in OS between platinum-sensitive and platinum-resistant relapses. This was not a question of the feasibility of secondary surgery; the median PCI was not different between the two groups. The international definition of relapse as platinum-sensitive or -resistant is based on the arbitrary binary definition of the delay between the end of initial treatment and diagnosis of the relapse (7). The group of platinum-resistant relapses (<6 months) is heterogeneous, including two distinct cases. The first is a patient treated for aggressive ovarian cancer without complete remission despite surgical effort at an expert centre. In this case, it is difficult to distinguish between relapse and persistent disease. This relapse is really refractory to platinum. The second case is a patient treated with incomplete primary surgery in a non-expert center. Early relapse in this situation may correspond to the growth of remaining tumour after incomplete surgery. In this case, the secondary cytoreductive surgery, corresponds to a delayed complete surgery. In the case of primary treatment, the European randomized trial, showed that in the case of a large residual after primary surgery, when complete surgery was feasible, a new debulking surgery significantly lengthened survival (29). In the current study, patients with early or late relapse were fit for complete surgery. To distinguish patients with a true early relapse refractory to platinum, we must add other criteria to the disease-free interval, such as the completeness of primary surgery performed in an expert center.
Residual disease is known to be one of the strongest prognostic factors, even after secondary surgery (5, 29). As in the case of primary treatment, secondary cytoreductive surgery before HIPEC must be complete, CC-0 or CC-1 according to the completeness of the cytoreductive surgery score (20). In the two latest French series, information on pathological residuals at the time of secondary surgery was not known (18, 28). In the current study, a central review of pathological reports after secondary cytoreductive surgery was performed, revealing a significant survival benefit of performing a CC-0 surgery compared to CC-1.
HIPEC is not a standardised technique: the kind of drug, dosing, dilution, duration and temperature vary from one series to another (16). Consequently, it is hard to draw any conclusion from the comparison of published series. Considerable standardization effort is required. Nowadays, in the absence of prospective randomized trials with a good level of evidence, the specific impact on survival of adding HIPEC to complete secondary cytoreductive surgery for patients treated for a first relapse of epithelial ovarian cancer remains questionable.
The mortality rate and severe morbidity rate ultimately make necessary to assess the hypothetical survival advantage of HIPEC through randomized trials. In the current series, patients treated with a cisplatinum-based HIPEC for first ovarian cancer relapse experienced a median OS of 45 months, when in the last Calypso trial, the mean OS of patients treated with pegylated liposomal doxorubicin and carboplatin intravenous chemotherapy was 33 months, with fewer than 20% of the patients requiring a secondary cytoreductive surgery (9). Even if results of HIPEC are encouraging, the real benefit of platinum-based HIPEC must be tested in a randomized setting.
Currently, two prospective European trials are ongoing testing the impact of HIPEC with cisplatinum on first platinum-sensitive relapse of epithelial ovarian cancer. HORSE, an Italian randomized multicentre trial comparing interval surgery plus HIPEC versus interval surgery alone, aimed to assess relapse-free survival (available at http://clinicaltrials.gov/show/NCT01539785). CHIPOR, a French prospective multicentre randomised trial comparing, after six courses of second-line chemotherapy, surgery plus HIPEC versus surgery alone, aimed to assess overall survival (available at http://clinicaltrials.gov/show/NCT01376752). In these two trials, the goal of surgery is to be complete.
In conclusion, in the current series, HIPEC in the treatment of first ovarian cancer relapse allows encouraging survival rates, better in the case of complete secondary surgery, with a high response rate to primary chemotherapy, with acceptable mortality. In order to assess these results, prospective randomized trials are strongly needed and inclusion in ongoing trials strongly recommended.
Footnotes
Conflicts of Interest
The authors declare that there are no conflicts of interest.
- Received April 28, 2015.
- Revision received June 7, 2015.
- Accepted June 9, 2015.
- Copyright© 2015 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
