Introduction

Ovarian cancer is the third commonest and most lethal malignancy gynecological malignancy (Jemal et al. 2007). Epithelial ovarian tumors account for majority (>70%) of all ovarian cancers. It typically presents with vague gastrointestinal and constitutional symptoms of abdominal bloating, distension, weight loss, and fatigue (Goff et al. 2000). Owing to the heterogeneity of these clinical symptoms, early diagnosis is often delayed. Late presentation results in the majority of patients being diagnosed with advanced disease (Stage III/IV). The 5-year survival rate of patients with advanced ovarian cancer is <25% (Ozols 2005). In the final stages of this disease, patients suffer from severe anorexia, dyspnea and pain from malignant bowel obstruction, ascites, and pleural effusion as a result of the extensive burden of tumor.

Epithelial ovarian tumor arises from the serosal lining of the ovary. This covering of the ovary communicates with the serosal lining of the abdominopelvic cavity, and is known as the peritoneum. Tumor growth results in the exfoliation of malignant cells into the peritoneal fluid. They circulate freely and typically implant in the pelvis and subdiaphramatic recesses owing to gravity and the incumbent position. This spread of tumor within the peritoneum is termed peritoneal carcinomatosis (Sugarbaker 1996). Intraoperatively, it is characterized by the extensive presence of macroscopic whitish tumor nodules of variable sizes and consistency that may coalesce to form plaques or masses within the abdominopelvic cavity. Tumor dissemination from the peritoneal cavity into the pleural cavity may also occur through the lymphatic lacunae present within the diaphragmatic peritoneum (Abu-Hijleh et al. 1995; Carmignani et al. 2003). This results in severe pleural effusion which compromises lung and cardiac functions. In the past, peritoneal carcinomatosis was regarded as a terminal condition and patients were treated symptomatically. However, as this disease is largely confined to the peritoneal surfaces, it is now considered a loco-regional disease.

The feasibility of hyperthermic intraperitoneal chemotherapy (HIPEC) as a treatment for peritoneal carcinomatosis was first demonstrated by Spratt et al. (1980) in the early 1980s. Its development continued under Dr. Sugarbaker from the Washington Cancer Institute in the mid-1990s who advocated this combined procedure of surgical resections and hyperthermic chemoperfusion to achieve complete intraoperative cytoreduction. This procedure involves cytoreductive surgery (CRS) with peritonectomy procedures aimed at resecting peritoneal surfaces with tumor implants and visceral dissections with a maximal surgical effort to remove as much tumor as possible macroscopically, followed by direct instillation of heated chemotherapy which together serves as a synergistic medium to enhance cytotoxicity to address microscopic residual disease (Witkamp et al. 2001). The results of this treatment have been shown to be beneficial for patients with peritoneal carcinomatosis from appendiceal cancer (Yan et al. 2007a), colorectal cancer (Yan et al. 2006), and peritoneal mesothelioma (Yan et al. 2007b).

This study aims to provide a collective review of the current evidence available for the combined regimen of CRS and HIPEC for ovarian cancer peritoneal carcinomatosis.

Methods

Literature search strategy

A literature search was conducted using the MEDLINE (1966 to August May 2009), PubMed (January 1980 to May 2009) and EMBASE (1974 to May 2009) databases. The reference lists of articles identified were manually searched to locate other articles of relevance. The search was limited to English language articles and to humans. The search terms used to locate studies were ‘Ovarian cancer’, ‘Intraperitoneal’, and ‘Cytoreductive surgery’. All relevant articles identified were assessed with application of inclusion and exclusion criteria.

Selection criteria

The selection criteria were as follows: all studies >10 patients, adopting the combined CRS and HIPEC treatment with a diagnosis of advanced (Stage III/IV) or recurrent ovarian cancer. Studies which included the results of other gastrointestinal and pelvic malignancies were included if the results of the ovarian cancer subjects were analyzed separately and clearly reported. We excluded studies reporting the pharmacokinetic data (Phase I studies). Where multiple publications from the same institution were identified, only the most recent update with the largest number of patients or longer follow-up group was included. CRS consisted of peritonectomy procedures (anterior parietal peritonectomy, omentectomy ± splenectomy, right and left subphrenic peritonectomy, pelvic peritonectomy, and lesser omentectomy with stripping of the omental bursa ± cholecystectomy) and visceral resections (rectosigmoidectomy, right colectomy, total abdominal colectomy, hysterectomy, and small bowel resection) (Sugarbaker 1995). The type and extent of peritonectomy procedures were not uniformly performed in all the studies included. HIPEC was administered intraoperatively after CRS. Studies were selected for evaluation if they were level I evidence: randomized controlled trials (RCTs); level II evidence: nonrandomized controlled clinical trials or well-designed cohort studies; level III evidence: observational studies, as described by the US Preventive Services Task Force.

Data extraction and critical appraisal

The studies were independently and critically appraised using a standard protocol. Data extracted include the methodology, quality criteria, perioperative variables, and the morbidity and mortality outcomes. All data were extracted and tabulated from the relevant articles’ texts, tables, and figures.

The patient group treated in each study was classified according to the indications for treatment for CRS and HIPEC at five time points in the natural history of ovarian cancer as described by the consensus of the Peritoneal Surface Oncology Group: (1) at the time of primary treatment where optimal cytoreduction is achieved, (2) at the time of interval debulking, (3) as a consolidation therapy following complete pathological response following initial therapy as confirmed by a second-look laparotomy, (4) at the time of first recurrence, and (5) as salvage therapy (Helm et al. 2008).

Discrepancies were resolved by discussion and consensus. Following tabulation of the results, study design, year of publication, number of patients, criteria used to define CRS, completeness of cytoreduction, HIPEC protocol, treatment outcomes, treatment-related morbidity and mortality, and prognostic factors associated with outcomes were synthesized. Meta-analysis was inappropriate because of the heterogeneous nature of the included studies and the lack of a comparative arm in most studies. Clinical effectiveness was synthesized through a narrative review with full tabulation of results of all included studies.

Results

The search revealed a total of 132 abstracts from which 19 studies employed a combined regime of CRS and HIPEC in the treatment of patients with ovarian cancer peritoneal carcinomatosis. These were retrieved and appraised in this review. None of the studies reviewed were randomized trials. The level of evidence in the studies reviewed were mostly class II or class III (nonrandomized comparative studies and observational studies). In addition, the patient cohort treated was of a heterogeneous group. Most studies included patients with either advanced or recurrent ovarian cancer who have undergone previous surgery and chemotherapy. A significant proportion of the patients within these studies also had documented chemoresistance and had undergone multiple treatments (time point 5). The full details are listed in Tables 1 and 2.

Table 1 Patient characteristics from 19 studies comprising of 895 patients
Full size table
Table 2 Time-point of use of HIPEC and protocol reported in 19 studies
Full size table

Time-point of use of HIPEC

All 19 studies reviewed performed HIPEC as part of the combined treatment with cytoreduction. However, the HIPEC protocol varied in each institution. The most commonly used chemoperfusate was Cisplatin. The median intraabdominal temperature was 42°C with a range of 38–48°C. The median duration of infusion was 90 min with a range of 60–120 min. Eleven out of 19 studies employed the use of CRS and HIPEC as primary treatment for a proportion of their patients with advanced ovarian cancer. Eleven out of 19 studies employed the use of CRS and HIPEC as salvage therapy for patients with recurrent or persistent ovarian cancer peritoneal carcinomatosis, indicating that this group of patients have failed conventional treatment. The full details are listed in Table 2.

Perioperative mortality and morbidity results

In total, 895 patients from 19 different studies were reviewed. The mortality rate associated with treatment ranged from 0 to 10%. The median duration of operation and HIPEC treatment ranged from 4 to 10 h. The median length of hospital stay ranged from 8 to 25 days. Morbidity was analyzed and reported according to the National Cancer Institute, common toxicity criteria. Briefly, grade I postoperative complication was where the diagnosis was established but no intervention was required for resolution. Grade II postoperative complication was where medical treatments were required for resolution. Grade III postoperative complications required an invasive intervention such as a radiological intervention were required for resolution. Grade IV postoperative complication required urgent definitive intervention such as returning to the operating room or ICU were required for resolution. Grade I morbidity ranged from 6 to 70%, grade II morbidity ranged from 3 to 50%, grade III morbidity ranged from 0 to 40% and grade IV morbidity ranged from 0 to 15%. Common postoperative complications include ileus, anastomotic leakage, bleeding, wound infection, toxicity, pleural effusion, infections, fistula, transient hepatitis, and thrombocytopenia. The full details are listed in Table 3.

Table 3 Perioperative mortality and morbidity results of 16 studies comprising of 895 patients with peritoneal carcinomatosis from ovarian cancer
Full size table

Survival results

Although the use of CRS and HIPEC was at different time-points during the natural history of ovarian cancer, a tabulation approach to consolidate results of this treatment is presented to provide an indication of the treatment efficacy. The median time of follow-up ranged from 14 to 64 months, the median disease-free survival ranged from 10 to 57 months, the median overall survival ranged from 22 to 64 months, median overall survival for patient with an optimal cytoreduction ranged from 29 to 66 months, overall 3- year survival rate ranged from 35 to 63%, and 5-year survival rate ranged from 12 to 66%. The full details are listed in Table 4.

Table 4 Survival results of cytoreductive surgery and hyperthermic intraperitoneal chemotherapy of 895 patients with peritoneal carcinomatosis from ovarian cancer
Full size table

Discussion

Traditionally, patients with extensive ovarian cancer peritoneal carcinomatosis are often labeled as having terminal disease. Efforts at aggressive treatment are abandoned and treatment is largely palliative. Palliative surgery, where debulk or by-pass procedures are performed, or systemic chemotherapy is administered, both of which are not performed with a curative intent. The true result of whether this improves symptoms or extends survival is largely unknown (Ozols 2005). Presently, despite its availability in specialized surgical oncology institutions, HIPEC has not been advocated as a treatment option to existing ‘curative’ therapy in ovarian cancer.

Evident from our review, the use of the term CRS in the ovarian cancer literature has been shown to indicate varying extent of cytoreduction with residual tumor volume of 1–2 cm. Groups performing CRS using peritonectomy procedures assess the completeness of cytoreduction based on Jacquet and Sugarbaker’s criteria of complete cytoreduction being ≤0.25 cm (Jacquet and Sugarbaker 1996). In addition, HIPEC was also performed at various time-points during the natural history of the disease and thus the treated group within this review comprised patients that have the following: newly diagnosed and treatment-naive advanced ovarian cancer, patients with their first recurrence following primary treatment, patients who have responded to primary treatment and are currently disease free, as well as patients who have recurrent and persistent ovarian cancer that have failed conventional treatment. This heterogeneity precluded the definitive conclusion of the survival results. However, efforts have been made to collectively tabulate and narrate the results of HIPEC in this heterogeneous group of patients with ovarian cancer peritoneal carcinomatosis to demonstrate the efficacy of this treatment.

Interest in the use of intraperitoneal chemotherapy for ovarian cancer was revived following the publication of results of the Gynecologic Oncology Group (GOG-172) phase III trial which compared intravenous chemotherapy with intravenous plus intraperitoneal chemotherapy in primary stage III ovarian cancer (Armstrong et al. 2006); the National Cancer Institute and GOG performed a meta-analysis of this treatment and made an announcement that this should become the standard of care (Trimble and Christian 2008). The treatment employed in this trial compared maximal cytoreduction followed by administration of adjuvant chemotherapy, cisplatin, and paclitaxel via both the intraperitoneal (IP) and intravenous (IV) route versus a group which only had IV chemotherapy. Although there were obvious survival benefits in the IP group, only 40% of patients were able to complete six cycles of chemotherapy due to high rates of IP catheter complications (Walker et al. 2006). Despite the low rates of treatment completion, survival benefits were still achieved, suggesting that limited use of IP chemotherapy is still purposeful. Therefore in clinical practice, this treatment is not routinely administered and has not become standard (Rowan 2009; Walker 2009).

In recurrent ovarian cancer, options for treatment are more varied with patients being subjected to various chemotherapy trials without any consensus for standard of care. Secondary cytoreduction in patients who have had a long disease-free survival may be beneficial as suggested by Bristow et al. (1996) in a meta-analysis. Medically, patients with platinum sensitive disease are treated with carboplatin with or without paclitaxel. In platinum-resistant disease, patients are often subjected to non-platinum agents such as topotecan, pegylated liposomal doxorubicin, gemcitabine, and oral etoposide (Ozols 2002). The overall median survival of secondary cytoreduction in recurrent ovarian cancer is about 40 months (Matsumoto et al. 2006; Salani et al. 2007). With systemic chemotherapy, the median disease-free survival in patients with platinum-sensitive disease was 9 months when treated with gemcitabine plus carboplatin (Pfisterer et al. 2006), 14 months when treated with doxetaxol plus oxaliplatin (Ferrandina et al. 2007). In patients with platinum-resistant disease, doxetaxol plus vinorelbine can achieve a median disease-free survival of 13 months and overall survival was 9 months (Aravantinos et al. 2003).

Given that complete cytoreduction from the surgical procedure is a strong prognostic factor for the overall survival (Bristow et al. 2002; Winter et al. 2007) and that a recent meta-analysis by Bristow and Chi (2006) has demonstrated that there is no role for neoadjuvant platinum-based chemotherapy. Management of ovarian cancer should primarily involve a maximal surgical effort for complete cytoreduction. In situations whereby extensive disease burden have rendered limitations to achievement of a complete resection, neoadjuvant chemotherapy may be considered. The EORTC-GCG/NCIC-CTG randomized trial comparing primary debulking surgery with neoadjuvant chemotherapy in advanced ovarian, fallopian tube, and peritoneal cancer have shown that similar overall survival and progression-free survival outcomes may be achieved compared to standard primary debulking and with a lower morbidity rate (Vergote et al. 2008). However, emphasis must be made that neoadjuvant chemotherapy should not form the basis of selecting a favorable prognostic group of patients who are chemo-responsive to undergo aggressive surgical cytoreduction.

CRS using limited peritonectomy procedures to resect peritoneal implants and HIPEC aims to allow both macroscopic cytoreduction through surgery and cytotoxic cytoreduction through loco-regional administration of heated chemotherapy. Tabulation of the results from these studies show that CRS and HIPEC are associated with significant severe morbidity rates of up to 40% and a mortality rate of 0–10%. However, we would point out that contemporary mortality and morbidity figures from institutions who routinely perform this procedure are low. The treatment related complications is considered acceptable and further large volume peritonectomy units have low morality rates that range from 0 to 2% (Bereder et al. 2009; Look et al. 2004; Raspagliesi et al. 2006). The complication rate of this treatment has been recently compiled through a systematic review by Chua et al. who reviewed the morbidity and mortality results of 24 treatment centres, of which ten centers that were regarded as high volume specialized centres based on the number of procedures performed, showed a major morbidity rate ranging from 12 to 52% and a mortality rate ranging from 0.9 to 5.8% (Chua et al. 2009b). Majority of these procedures in these institutions were performed in patients with high-volume carcinomatosis with surgical morbidity being attributed specifically to the extent of disease rather than the HIPEC procedure.

Notwithstanding the limitations of inferring treatment efficacy through large case series, we demonstrate a long median disease-free survival in patients with ovarian cancer peritoneal carcinomatosis with a range of 10 to 57 months and an overall median survival of 22 to 64 months in a heterogeneous cohort of patients. Although specific conclusion of outcomes for primary advanced ovarian cancer or recurrent ovarian cancer cannot be accurately elucidated from the published reports, a significant proportion of patients have undergone multiple treatments with a proportion of patients who have failed treatment (time point 5). Hence, this review suggests that CRS and HIPEC in patients with recurrent ovarian cancer may be beneficial when compared with results of conventional secondary cytoreduction or systemic chemotherapy reported in the literature. Rufian et al. (2006) in their study reported 19 patients with advanced ovarian cancer and 14 patients who have relapsed for the first time following primary treatment that involved surgery and systemic chemotherapy. The long median survival of 66 months in the patients that had an optimal cytoreduction may further suggest that intervening early with CRS and HIPEC may be useful. The largest experience to date was reported by Bereder et al. (2009), who reported a median overall survival of 46 months in patients with first relapsed ovarian cancer of which a proportion are chemoresistant. In their institutions, the mortality rate was under 1% and the morbidity rates were about 10%.

Perioperative use of IP chemotherapy as HIPEC or early postoperative IP chemotherapy (EPIC) overcomes the limitations of the IP/IV treatment regime that has been shown to be efficacious in the randomized trials of IP chemotherapy in ovarian cancer which, however, is associated with high rates of IP catheter complications. Administration of IP chemotherapy intraoperatively and during the early postoperative period allows free circulation of the chemoperfusate and hence an improved drug distribution without the compromise of any adhesions; heating of the chemoperfusate enhances the synergistic effect of the cytotoxic agent and most importantly, it avoids the need to implant any peritoneal access device and hence, full compliance (Sugarbaker 2007).

Despite the lack of quality (level I evidence) data, the consistent and reproducible results demonstrated in this review suggest that there is an overall survival advantage associated with CRS and HIPEC in this group of patients that are at various stages of their disease process, who have been heavily treated with both surgeries and chemotherapy and with issues of chemoresistance. For this treatment to become generally accepted, the oncology community must commit to a randomized trial (Chua et al. 2009a). Specifically, trials should be set in the context of HIPEC use according to the various time-points as previously proposed (Helm et al. 2008).