Randomized controlled trials evaluating cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC) in prevention and therapy of peritoneal metastasis: a systematic review

Abstract

Background

Cytoreductive surgery (CRS) combined with hyperthermic intraperitoneal chemotherapy (HIPEC) is increasingly used to prevent or treat peritoneal metastases (PM) in selected indications. The objective of this article was to review published, recruiting or planned randomized controlled trials (RCTs) evaluating CRS and HIPEC versus standard of care. Comparator was systemic chemotherapy and/or CRS alone.

Content

Systematic review according to PRISMA guidelines. Electronic searches for published RCT using PubMed (from 1980 to November 2016) and for ongoing RCT in the United States and European clinical databases (until November 2016). Current update on ongoing trials from the 10th PSOGI meeting in November 2016 in Washington DC. Fourteen RCTs on CRS and HIPEC were excluded for various reasons.

Summary

Thirty-eight trials designed for randomizing 7,303 patients were identified: 11 in colorectal cancer (6 for prevention of PM, n=1,107 patients; 5 for therapy, n=781), 10 in ovarian cancer (5 in frontline therapy, n=438 patients; 5 for treating recurrence, n=1,062) and 17 in gastric cancer (14 for prevention of PM, n=3,659 patients; 3 for therapy, n=256). Results of 9 RCTs have been published: 1 in colorectal cancer (105 patients), 1 in ovarian cancer (130 patients) and 7 in gastric cancer (together 669 patients). Five RCTs have completed recruitment and follow-up is ongoing. There is a clear trend in recent trial design from therapeutic to preventive indications.

Outlook

The number of published RCT evaluating CRS and HIPEC in prevention or therapy of PM is relatively small. There is some evidence that CRS and HIPEC improve survival in recurrent colorectal origin, evidence in ovarian and gastric cancer remains debated. A large number of studies is ongoing that might deliver additional evidence. Trial design and interpretation of results remain difficult because of multiple methodological challenges.

Introduction

Peritoneal metastasis (PM) is a relatively common way of cancer spreading, has a dismal prognosis and remains an unmet medical need [1]. Systemic palliative chemotherapy (SPC) is the guideline-recommended treatment in this situation and its use is supported by numerous randomized controlled trials (RCTs) in various cancer types. However, only few RCT on SPC have considered the presence of PM as an inclusion or exclusion criteria, and even less have defined regression of PM as a main outcome criteria. This is in part due to the absence of accurate radiological method to assess therapy response in PM, in particular for small-volumetric disease without ascites formation [2].

The few data available to date suggest that efficacy of SPC on PM might be less than on parenchmymatous metastases (liver, lung, etc.), at least in colorectal cancer [3]. Intraperitoneal chemotherapy (IPC) has been suggested as an alternative or additional form of therapy for PM because of their potential to increase many fold the intraperitoneal drug concentration, the therapeutic index between local and systemic dose and therefore the efficacy of chemotherapy [4]. In ovarian cancer for example, seven RCTs have demonstrated the efficacy of IPC in combination with systemic chemotherapy [5]. However, IPC is hampered by local toxicity, frequent catheter-linked complications and pharmacological limitations such as poor drug repartition and by limited tissue uptake [6]. This might explain why the use of IPC in clinical practice remains limited [7].

Combing complete cytoreductive surgery (CRS) with intraoperative hyperthermic IntraPeritoneal Chemotherapy (HIPEC) has been proposed 30 years ago [8] to address these challenges, based on three theoretical principles:

• (1)to reduce tumor nodes to a quasi-microscopic size in order to meet poor drug penetration,
• (2)to administer chemotherapy into the peritoneal cavity using an open approach (Coliseum technique) or recirculation systems (closed HIPEC) to improve homogeneity of drug repartition,
• (3)to heat up the chemotherapy solution in order to enhance cytotoxicity.

In the meantime, morbidity and mortality of CRS and HIPEC have decreased as a consequence of better surgical techniques and perioperative care, and this combined procedure is increasingly used [9]. Numerous retrospective studies and data from prospective registries report on a survival benefit after CRS and HIPEC in selected indications. In colorectal cancer, for example, the overall 5-year survival is reported to be as high as 51 % after optimal CRS and HIPEC combined with intravenous systemic chemotherapy (so-called bidirectional chemotherapy) [10, 11]. On the basis of these encouraging results, indications for CRS and HIPEC have been gradually extended to PM from other organs than the colorectum, including the appendix [12, 13], the small bowel [14], the stomach [15] and the ovary [16]. However, the vast majority of studies, even when large populations are analysed, are non-comparative and the level of evidence for CRS and HIPEC remains modest [17]. Since CRS and HIPEC remains an aggressive procedure [18], its acceptance outside the surgical oncology community remains low and indications and results of this complex, advanced form of therapy are still the object of considerable debate [19].

The aim of this systematic review is to summarize published ongoing and planned RCT evaluating a possible therapeutic advantage for CRS and HIPEC versus other strategies in prevention and therapy of PM.

Materials and methods

Study design

This is a systematic review according to PRISMA guidelines [20]. The corresponding chart flow is provided in Figure 1.

Figure 1:

Chart flow for the study design.

Clinical trials database search strategy

We performed a systematic search of the US National Institute of Health clinical trials database (www.clinicaltrials.gov) on November 30, 2016 with the search terms (HIPEC AND randomized AND surgery). We identified 44 studies.

In addition, a systematic search of the EC clinical trials database (https://www.clinicaltrialsregister.eu/) was performed the same day, with the search terms “surgery” AND “hyperthermic intraperitoneal chemotherapy” and “randomized”. We identified six studies. This search was matched with US search and five duplicates removed.

Forty-five RCTs were retrieved with the electronic search procedure. After individual review of the study protocols, 14 RCTs were excluded because they did not meet the inclusion criteria. The list of these trials and the reasons for exclusion are provided in Supplementary Material. Thirty-one RCTs remained for further analysis.

Literature database search strategy

We performed a systematic search of the US National Institute of Health literature database (https://www.ncbi.nlm.nih.gov/pubmed) on Dec 1, 2016. The search terms were (“hyperthermic intraperitoneal chemotherapy” OR “continuous hyperthermic peritoneal perfusion”) AND “randomized”. Time period was Table 1 January 1980–November 2016). No search restrictions were imposed. Ninety-six references were retrieved and matched with the list of RCT generated previously. This search was matched with the clinical trials search and duplicates removed. Seven were found that were not listed in the clinical trial registries [21, 22, 37–41]. Thirty-eight published, ongoing or planned RCT remained for further analysis.

Table 1:

Selection criteria for relevant studies.

Characteristic	Criteria
Inclusion criteria	Prospective RCT comparing the outcome(s) of CRS and HIPEC or CHPP vs. systemic chemotherapy (alone or in combination). Prospective RCT comparing the outcome(s) of CRS and HIPEC or CHPP vs. CRS alone
Exclusion criteria	Comparison of various techniques of CRS and HIPEC (e. g. different pressure, time, temperature, drug etc.) No clinical outcomes reported Review or meta-analysis Conference abstract (exception: progress data on patient inclusion) Other intraperitoneal chemotherapy techniques: EPIC, NIPS, PIPAC, etc.
Patients	Patients with peritoneal metastasis (PM)
Intervention	CRS and HIPEC
Outcome	Selected abstract contained information relevant to the safety and/or efficacy
Language	Only articles in English were included

CRS, cytoreductive surgery; EPIC, early postoperative chemotherapy; HIPEC, hyperthermic intraperitoneal chemotherapy; NIPS, neoadjuvant intraperitoneal-systemic chemotherapy protocol; PIPAC, pressurized intraperitoneal aerosol chemotherapy; PM, peritoneal metastasis; RCT, randomized controlled clinical trial.

10th Peritoneal Surface Malignancy Group International (PSOGI) meeting, Washington November 17–19, 2016

Updates presented by the principal investigators of RCT investigating CRS and HIPEC in colorectal, appendiceal, ovarian and gastric cancer were integrated into this systematic review.

Results

Results of RCT investigating the role of CRS and HIPEC for treating PM are analysed separately depending on the organ of origin (colorectal, ovarian and gastric cancer) in order to take into account the natural history of disease (e. g. the relative incidence of isolated peritoneal metastases [PM]) and tumor biology (aggressiveness of disease).

Colorectal cancer

RCTs evaluating CRS and HIPEC in colorectal cancer are presented in Table 2. These trials are investigating two distinct clinical situations: first, the therapeutic situation (in the presence of PM) and, second, the prophylactic situation (to prevent development of secondary PM in high-risk patients).

Table 2:

Prospective randomized trials evaluating CRS and HIPEC in colorectal cancer.

	Indication	Acronym	Principal investigator	Country	Opened	Status	Accrual needed	Patients recruited (November 2016)	Drug IP	M/M results	Oncological outcome
Prophylactic (patients without peritoneal metastasis)
1	NCT01226394	ProphyloCHIP (PRODIGE 15)	Goere D	France	April 2010	Active	130	130	Oxaliplatin	Presented	N/A
2	NCT02179489	N/A	Sun L	China	October 2014	Recruiting	300	N/A	MMC	N/A	N/A
3	NCT02231086	COLOPEC	Tanis PJ	Netherlands	March 2015	Recruiting	200	181	Oxaliplatin	N/A	N/A
4	NCT02614534	HIPECT4	Arjona-Sanchez	Spain	November 2015	Recruiting	190	N/A	MMC	N/A	N/A
5	NCT02965248	APEC	Cai G	China	November 2016	Recruiting	147	0	Raltitrexed Oxaliplatin	N/A	N/A
6	NCT02974556	PROMENADE	Sammartino P	Italy	March 2017	Not yet recruiting	140	0	Oxaliplatin	N/A	N/A
Total							1107
Therapeutic (patients with peritoneal metastasis)
1	N/A	N/A	Verwaal VJ	Netherlands	February 1995	Closed	105	105	MMC	Published	Published
2	NCT00769405	PRODIGE 7	Quenet F	France	February 2008	Active	264	265	Oxaliplatin	Presented	Exp. 2017
3	NCT01540344	COMBATAC	Piso P	Germany	October 2010	Terminated	60	20	Oxaliplatin	Presented	Exp. 2017
4	NCT01628211	HIPEC	Perrone F	Italy	April 2012	Recruiting	140	N/A	Oxaliplatin	N/A	N/A
5	NCT01815359	ICArUS	Nash G	USA	April 2013	Recruiting	212	94	MMC	Presented	N/A
Total							781

CDDP, cisplatin; MMC, mitomycin C; N/A, not available to date (November 2016); Exp., expected. Green: results published. Yellow: patient accrual completed.

Therapeutic indication (peritoneal metastasis)

Five RCTs have been designed in patients with synchronous, isolated PM of colorectal origin, with a planned total accrual of 781 patients. One trial is now closed and results have been published. Two further trials are active, have closed patient recruitment and are waiting for follow-up data. Two additional RCTs are currently recruiting patients.

The first monocentric Dutch trial was published in 2003 and results were updated in 2008 after 8-year follow-up [22, 23]. In this study, 105 patients with PM of colorectal origin were randomly assigned to palliative systemic chemotherapy (control group) or to CRS and HIPEC followed by palliative systemic chemotherapy (test group). Patients with extraperitoneal metastases were excluded. In the test group, surgery was performed with the aim of complete cytoreduction (CC-0) and included subtotal peritonectomy, omentectomy as well as, if needed, resection of other organs involved. Completeness of cytoreduction was assessed at the end of surgery. Then, HIPEC was administered for 90 min using mitomycin C. Only palliative surgery, no CRS, was allowed in the control group, i. e. bypass surgery or ostomy for relief of bowel obstruction. Safety results showed a higher morbidity and mortality in the test group (CRS and HIPEC), notably a postoperative mortality of 8 %. The most important surgical complications were gastrointestinal leakage and peritonitis in 15 % patients. Also systemic side effects such as Grade 3 or 4 leukopenia were registered in a significant proportion of patients. In contrast, oncological results showed a significant advantage for the test group. After a median follow-up of 21.6 months, 39 % patients in the control group were alive compared with 55 % patients in the test group (HR for death, 0.55; 95 % CI, 0.32–0.95; p=0.03). Median overall survival in the control group was 12.6 months compared with 22.4 months in the test group. Overall, survival was poor (5 months) in patients in whom complete cytoreduction could not be achieved (CC-2) and in patients with extensive peritoneal disease (more than six anatomic regions involved according to PCI). Long-term oncological results were published as minimum follow-up was 6 years for all patients (median, 7.8 years; range, 6–9.6). At this point of time, four patients were still alive in the control group (two with disease and two without disease) versus five patients in the experimental group (two with disease and three without disease). Median disease-specific survival was 12.6 months in the control arm and 22.2 months in the CRS/HIPEC group (p=0.03). Median PFS was 7.7 months in the control group and 12.6 months in the CRS/HIPEC group (p=0.02).

This RCT was the first to demonstrate the positive effect of combined CRS and HIPEC on survival of patients with PM and was pivotal in the further development of this therapeutic strategy, as documented by the high number of citations of this work. However, both the design and the results of this study have been challenged:

• –Although this study showed that a multimodal therapy consisting of CRS, HIPEC and systemic chemotherapy nearly doubled survival compared with systemic chemotherapy alone, it did not show how much of this benefit is achieved by surgery and how much by HIPEC.
• –HIPEC might not contribute to the improved survival but cause additional morbidity [24].
• –Survival figures in the control group (chemotherapy alone) were relatively disappointing and do not match results obtained with modern chemotherapeutic agents such as oxaliplatin, irinotecan, bevacizumab and others.
• –If the aim of therapy was curative, the proportion of long-term survivors without evidence of disease was not different between groups.
• –If the aim of therapy was palliative, no patient-reported outcomes (PROs) are reported that would be significant in this situation. In fact, the long hospital stay (29 days) and the incidence of major complications suggest indirectly that QoL might have been compromised in the test group, at least in the short term [25].

In contrast to the trial of Verwaal described above, the French PRODIGE 7 trial (NCT00769405) does not compare CRS and HIPEC versus palliative systemic chemotherapy but investigates the potential additional benefit of CRS and HIPEC over CRS alone. This larger multicentric study is randomizing patients intraoperatively following complete CRS (CC-0 or CC-1) with or without HIPEC with oxaliplatin. Six months of systemic chemotherapy (physician’s best choice) are scheduled and can be performed before and/or after the HIPEC procedure. The majority of surgical teams selected a program with 3 months of chemotherapy before CRS and HIPEC and 3 months after the procedure. The type of systemic chemotherapy is not predefined, and the oncologist is free to administer the therapy of his choice. Enrolment is now completed and first survival results are expected already in 2017.

The German COMBATAC trial [26] is a smaller trial with an intended accrual of 60 patients and was designed to evaluate the feasibility and efficacy of the combined multidisciplinary treatment regimen consisting of perioperative systemic combination chemotherapy plus cetuximab followed by CRS and bidirectional chemotherapy (HIPEC with intraperitoneal oxaliplatin). The trial was stopped after inclusion of 20 patients. Morbidity and mortality data have been presented. First survival data are expected in 2017.

HIPEC in patients with colon cancer at high risk of peritoneal metastasis (prophylactic indication)

Six RCT totalizing an accrual of 1,107 patients have been initiated to analyze the prophylactic role of HIPEC in patients with colon cancer at high risk of developing PM. One trial has completed the accrual; five trials are currently recruiting patients.

The French Prophylochip (or PRODIGE 15) trial (NCT01226394) investigates a potential favorable effect of HIPEC in colorectal patients at risk of developing colorectal PM. Intended accrual was 130 patients. Inclusion criteria at the time point of colectomy are resected minimal synchronous PM, ovarian metastases and/or tumor perforation. All patients are then treated for 6 months with adjuvant systemic chemotherapy, in the rule FOLFOX-4. Then, in the presence of peritoneal recurrence, patients receive best choice therapy and were excluded of the study. In the absence of recurrence, patients are randomized into surveillance alone, or exploratory laparotomy and HIPEC. The end point is 3 years of disease-free survival. The study is now closed for inclusion and first survival results are expected at the beginning of 2019.

The Dutch COLOPEC trial [27] investigates the oncological effectiveness of adjuvant HIPEC using oxaliplatin, following a curative resection of a T4 or intraabdominally perforated colon carcinoma in preventing the development of PM. Intended accrual was 174 but has been up scaled to 200 patients. Patients who underwent curative resection for a UICC category T4 (or perforated) colon cancer are randomized to receive adjuvant HIPEC performed simultaneously with colectomy, or as a staged procedure (less than 10 days or 5–8 weeks postoperatively). Standard adjuvant systemic chemotherapy is given within 3 weeks after HIPEC in both study groups. Secondary aims are to determine the incidence of PM in pT4 and perforated colon cancer as determined by laparoscopy as a new gold standard, to identify molecular parameters indicating high risk of developing PM and to determine treatment-related morbidity of open and laparoscopic adjuvant HIPEC. Presence or absence of peritoneal recurrence is evaluated 18 months postoperatively by CEA and CT thorax/abdomen; in the absence of radiological signs of recurrence, a diagnostic laparoscopy is performed. The study has already recruited 90 % of the intended patients and is expected to be completed in 2017. Oncological results are expected at the beginning of 2019.

In China, a further RCT (NCT 02179489) has been opened in October 2014 and is currently recruiting 300 patients at risk of developing colorectal PM. No information is available on the number of patients included to date. Patients with pT4 colon cancer or metastases to ovary will be randomized to receive mitomycin C-based HIPEC or not. The primary end point is 3 years of disease-free survival.

Ovarian cancer

Ten RCTs evaluating the contribution of HIPEC in ovarian cancer are published or ongoing (Table 3). These RCTs examine the outcome of CRS and HIPEC in two situations: front-line therapy and therapy of intraperitoneal recurrence.

Table 3:

Prospective randomized trials evaluating CRS and HIPEC in ovarian cancer.

	Indication	Acronym	Principal investigator	Country	Opened	Status	Accrual needed	Patients recruited (November 2016)	Drug IP	M/M results	Oncological outcome
	Front-line
1	NCT01091636	N/A	Park SY	Korea	March 2010	Active	170	185	CDDP	N/A	Exp. 2017
2	NCT02681432	HIPECOVA	Villarejo Campos	Spain	January 2012	Recruiting	94	N/A	Paclitaxel	N/A	N/A
3	NCT02328716	CARCINOHIPEC	Campos PC	Spain	February 2012	Recruiting	32	N/A	CDDP	N/A	N/A
4	NCT01628380	CHORINE	Ansaloni L	Italy	June 2012	Recruiting	94	41	CDDP+Paclitaxel	N/A	N/A
5	NCT02124421	N/A	Sardi A	USA	April 2014	Recruiting	48	N/A	Paclitaxel	N/A	N/A
							438
	Recurrence
1	N/A	N/A	Spiliotis	Greece	2006	Closed	120	120	Various	N/A	Published [21]
2	NCT00426257	OVHIPEC	van Driel WJ	Netherlands	February 2007	Active	242	245	CDDP	N/A	Exp. 2017
3	NCT01376752	CHIPOR	Classe JM	France	April 2011	Recruiting	444	243	CDDP	N/A	N/A
4	NCT01539785	HORSE	Scambia G	Italy	September 2012	Recruiting	158	103	CDDP	N/A	N/A
5	NCT01767675	N/A	Chi D	USA	January 2013	Recruiting	98	44	Carboplatin	N/A	N/A
	Total						1062

CDDP, cisplatin; N/A, not available to date (November 2016); Exp., expected. Green: results published. Yellow: patient accrual completed.

Front-line therapy

So far, there is no published RCT evaluating the contribution of HIPEC in front-line line treatment of ovarian cancer. Five RCTs are ongoing in this indication, totalizing 438 patients. The most advanced Phase II–III randomized study from Korea has completed enrolment of 185 patients for a needed accrual of 170 [28]. Aim of this study is to assess progression-free survival rate after CRS as a front-line therapy, combined or not with HIPEC and followed by intravenous chemotherapy. Only patients with optimal cytoreduction were included and randomization was intraoperative. Drug administered intraperitoneally was cisplatin (75 mg/m² body surface). Oncological results are expected within 2 years from now.

Four further RCTs are currently recruiting patients in this indication, two in Spain (NCT02681432 and NCT02328716), one in Italy (NCT01628380) and one in Baltimore, USA (NCT02124421). All four studies are relatively small sized, recruiting for more than 4 years, and did not reach the intended accrual of patients. The CHORINE study from Italy, for example, has enrolled so far 41 patients out of 94 needed, and the recruitment period has been extended by 2 years. This study includes women with ovarian cancer who responded to three cycles systemic chemotherapy (carboplatin/taxol). After complete CRS (CC-0 or CC-1) could be achieved, patients are randomized intraoperatively to receive or not a combination HIPEC with cisplatin 100 mg/m² and paclitaxel 175 mg/m² body surface. Since the primary outcome criteria is progression-free survival over a follow-up period of 2 years, no oncological results are expected before 2020.

Therapy of intraperitoneal recurrence

Six RCTs totalizing 1110 women have been performed or are ongoing in this recurrent ovarian cancer, but the results of only 1 single study have been published so far. This monocentric randomized study from Greece [21] examined 120 women with advanced ovarian cancer (FIGO stage IIIc and IV) who experienced disease recurrence after initial treatment with conservative or debulking surgery and systemic chemotherapy. Eligible women were randomized preoperatively by computer assignment to CRS/HIPEC (test group) or to CRS alone followed by systemic chemotherapy (control group). IPC was cisplatin plus paclitaxel for platinum-sensitive disease or doxorubicin plus paclitaxel or mitomycin for platinum-resistant disease. Overall survival was much better in the CRS/HIPEC group than in the control group (26.7 months vs. 13.4 months, p=0.006). Completeness of cytoreduction and PCI score were associated with survival. This study was the first one to demonstrate in a random manner the additional effect of HIPEC on complete CRS, which is an important validation for this strategy. However, this study has several limitations. The trial was not registered in an official trial database. No size sample calculation was provided a priori. Statistical methods applied were incompletely reported. Treatment-related morbidity and mortality were not reported. Control and test groups were not completely comparable:

• –Due to preoperative randomization, more patients in the test group had complete cytoreduction than in the control group (65 % vs. 55 %).
• –The experimental group (HIPEC) had more patients with earlier disease (FIGO IIIc) (68 %) compared with the control group (60 %).

In platinum-sensitive disease, survival was 26.8 months in the HIPEC group versus 15.2 months in the non-HIPEC group (p=0.035), but in platinum-resistant disease, survival did not differ statistically between groups (26.6 months in the HIPEC group vs. 10.2 months in the non-HIPEC group; p value not reported). These various factors make the interpretation of the trial results somewhat challenging. Extrapolation to other patients is difficult [29].

The most advanced ongoing RCT (OVHIPEC) in recurrent ovarian cancer was opened for recruitment in 2007 by the Dutch Gynecological Oncology Group [30]. Women with ovarian cancer FIGO III were randomized into secondary CRS with or without HIPEC. Inclusion criteria were women eligible for interval debulking for the following two reasons:

• –primary debulking surgery not feasible due to tumor extension or general condition (patients treated by primary chemotherapy).

or

• –incomplete primary debulking with residual disease >1 cm.
• –After 9 years, thus study has recruited all patients needed (245 patients for an accrual needed of 240) and survival results are expected in 2017.

The CHIPOR study [31], which opened in April 2011 in France with an intended accrual of 444 patients, is a multicentric study designed to evaluate the effectiveness of adjoining HIPEC to CRS as treatment for the first recurrence of ovarian cancer in patients with platinum-sensitive ovarian cancer. Only patients with complete CRS are included into this trial (intraoperative randomization). By November 2016, 243 patients (55 %) were recruited, inclusion rate being lower than planned. Obviously, there is some difficulty in recruiting patients for this trial after the orphan-drug designation granted in October 2014 by the European Medicine Agency to olaparib for use as monotherapy for the maintenance treatment of women with relapsed platinum-sensitive epithelial ovarian with BRCA mutations who have already responded to platinum-based chemotherapy, as the trial includes only first recurrence cases [32].

Two further RCTs are currently recruiting patients in recurrent ovarian cancer, including the HORSE study (Rome, Italy) where 2/3 of the accrual has already been achieved [33] and a smaller randomized Phase II trial from the United States (MSKK, New York) [34]. In contrast to most trials above applying CDDP (cisplatin) intraperitoneally, the latter study investigates the efficacy and the safety of intraperitoneal carboplatin.

Gastric cancer

Published and ongoing RCTs evaluating the contribution of HIPEC in peritoneal metastasis of gastric cancer (PMGC) are summarized in Table 4. Together, 17 trials intend to examine the outcome after CRS and HIPEC in 3,915 randomized patients. The majority of these RCT (n=14) are designed for preventing the development of PMGC in patients at risk; only three RCT examine the outcome of CRS and HIPEC after intraperitonel recurrence.

Table 4:

Prospective randomized trials evaluating CRS and HIPEC in gastric cancer.

	Indication	Acronym	Principal investigator	Country	Opened	Status	Accrual needed	Patients recruited (November 2016)	Drug IP	M/M results	Oncological outcome
	Prophylactic (patients without peritoneal metastasis)
1	N/A	N/A	Ikeguchi	Japan	1980	Closed	174	174	MMC	Published [38]	Published [38]
2	N/A	N/A	Koga	Japan	1980	Closed	47	47	MMC	Published [39]	Published [39]
3	N/A	N/A	Hamazoe	Japan	1983	Closed	82	82	MMC	Published [37]	Published [37]
4	N/A	N/A	Fujimoto	Japan	1987	Closed	141	141	MMC	Published [40]	Published [40]
5	N/A	N/A	Yonemura	Japan	N/A	Closed	139	139	MMC	Published [41]	Published [41]
6	NCT02396498	N/A	Zhang H	China	April 2014	Recruiting	270	N/A	CDDP	N/A	N/A
7	NCT01882933	GASTRICHIP	Glehen O	France	May 2014	Recruiting	321	150	Oxaliplatin	N/A	N/A
8	NCT02240524	HIPEC	Cui S	China	July 2014	Recruiting	582	N/A	Paclitaxel	N/A	N/A
9	NCT02381847	N/A	Guan W	China	January 2015	Recruiting	60	N/A	CDDP	N/A	N/A
10	NCT02356276	HIPEC-01	Fang Z	China	February 2015	Recruiting	584	N/A	Paclitaxel	N/A	N/A
11	NCT02969122	Cy-Plus	Ji J	China	November 2016	Recruiting	59	N/A	Docetaxel	N/A	N/A
12	NCT02960061	N/A	Wang HS	China	November 2016	Not yet recruiting	640	0	Paclitaxel	N/A	N/A
13	NCT02528110	N/A	Xiong B	China	July 2016	Not yet recruiting	500	0	Paclitaxel	N/A	N/A
14	NCT01683864	HIPEC_Stomach	Königsrainer	Germany	January 2017	Not yet recruiting	60	0	MMC+CDDP	N/A	N/A
	3659
	Therapeutic (patients with peritoneal metastasis)
1	NCT00454519	N/A	Li Y	China	March 2007	Closed	60	68	MMC+CDDP	Published [44]	Published [44]
2	NCT00941655	GYMSSA	Rudloff U	USA	July 2009	Closed	16	16	Oxaliplatin	Published [54]	Published [54]
3	NCT02158988	GASTRIPEC	Rau B	Germany	January 2014	Recruiting	180	65	MMC+CDDP	N/A	N/A
							256

MMC, mitomycin C; CDDP, cisplatin; N/A, not available to date (November 2016). Green: results published.

Prophylactic indication (prevention of metachronous peritoneal metastasis) in gastric cancer

Peritoneal recurrence eventually in 40–60 % of patients after radical surgery (gastrectomy and extended (D2) lymphadenectomy). Risk factors for such recurrence are serosa invasion, positive peritoneal cytology, lymph node involvement and signet ring histology. Thus, there is a rationale to offer preventive treatment to gastric cancer patients [35]. This is confirmed by the large number of ongoing trials, in particular in China, where the incidence of gastric cancer is particularly high [36].

In the 1980s, five Japanese studies analysed the efficacy of HIPEC after gastrectomy for gastric cancer with serosal invasion and no synchronous PM [37–41]. The denomination CHPP (continuous hyperthermic peritoneal perfusion) was used at that time since a standard terminology (HIPEC) was not introduced before the PSOGI congress in Milan in 2006. Hyperthermic chemoperfusion was performed with a solution of mitomycin C with a closed circulation technique. All five studies favored the experimental group (CHPP) with a survival gain between 10 % and 20 %.

Nine RCTs with a planned total of 3,076 participants are currently investigating a potential role of HIPEC for preventing development of PM after surgery for gastric cancer. These trials use various substances for IPC, including platin-based drugs (cisplatin or oxaliplatin), mitomycin C or taxanes (paclitaxel or docetaxel).

The most advanced of these preventive trials is the French GASTRICHIP study [42], with 150 out of 321 patients recruited to date. GASTRICHIP is a prospective, open, randomized multicenter phase III clinical study with two arms that aims to evaluate the effects of HIPEC with oxaliplatin on patients with gastric cancer involving the serosa and/or lymph node involvement and/or with positive cytology at peritoneal washing, treated with perioperative systemic chemotherapy and D1–D2 curative gastrectomy. There is an intraoperative randomization in patients where complete curative gastrectomy was possible. Primary end point is overall survival from the date of surgery to death or to the end of follow-up (5 years). Secondary end points are 3- and 5-year recurrence-free survival, site of recurrence, morbidity and quality of life. An ancillary study will compare the incidence of positive peritoneal cytology pre- and post-gastrectomy in two arms of the study and assess its impact on 5-year survival [43]. Survival results of this trial are not expected within the next 5 years.

The largest trial in the prophylactic indication has been initiated 3 years ago at Guangzhou Medical University in Guangzhou, China, with an intended accrual of 582 patients [44]. Primary outcome criteria are overall 5-year survival. In contrast to the GASTRICHIP trial where oxaliplatin is administered, this study is applying paclitaxel intraperitoneally. Estimated study completion date is July 2019. No information on the recruitment is available, but results cannot be reasonably expected before several years.

A second RCT in the prophylactic indication is currently being initiated at the Sun Yat-sen University in the same city of Guangzhou, China [45]. The intended accrual is 640 patients. The investigators intend to find out if, after neoadjuvant chemotherapy, combining HIPEC with gastrectomy and D2 resection could improve survival for patients with advanced gastric cancer, while comparing with the traditional approach of D2 resection. Intraperitoneal drug is paclitaxel. The trial is not yet recruiting.

A further RCT aiming at determining the efficacy of HIPEC in the treatment of locally advanced gastric cancer has been recently initiated at Wuhan University, China [46]. Patients are randomized into HIPEC group and control group. In the HIPEC group, the patients undergo radical gastrectomy with D2 lymphadenectomy and HIPEC with paclitaxel and 5-FU. Patients in the control group undergo radical gastrectomy with D2 lymphadenectomy. Patients in both groups receive six cycles of postoperative chemotherapy (SOX or XELOX) and are followed up for 5 years or until death. Accrual needed is 500 patients and the trial is not yet recruiting.

Two further, smaller trials have been recently opened in China in the prophylactic indication in gastric cancer [47, 48].

Therapy of recurrent gastric cancer

Only three trials with a total accrual needed of 264 patients are investigating a potential beneficial effect of HIPEC in the treatment of PMGC.

Two of these trials are small and have already been published:

• (1)Finals results of the first study (from China) were published 5 years ago [49]. This randomized phase III study with a limited number of patients (n=68) evaluated the efficacy and safety of adding HIPEC to CRS for the treatment of PMGC. The primary end point was overall survival, and the secondary end points were safety profiles. Patients were randomized into CRS alone (n=34) or CRS+HIPEC (n=34) receiving a combination of CDDP 120 mg and MMC 30 mg intraperitoneally. The completeness of cytoreduction was identical in both groups (58.8 %). There was no significant difference in mortality and morbidity (11.7 % vs. 14.7 % when HIPEC was added). Median survival was significantly increased in the group who received HIPEC (11 months), compared to 6.5 months in the surgery alone group (p=0.046).
• (2)In contrast to the first study, the second RCT did not compare CRS and HIPEC versus surgery alone, but the comparator was systemic chemotherapy [50]. This preliminary US multicentric study compared seven patients treated by intravenous chemotherapy (FOFIRINOX) alone with nine patients who were treated by intravenous FOLFIRINOX associated with CRS and HIPEC with oxaliplatin. Patients had gastric cancer metastatic to liver and lung and PM. Median overall survival was 11.3 months in the test group (CRS and HIPEC plus systemic chemotherapy) versus 4.3 months in the control group (chemotherapy alone). This study has methodological limitations. First, this is a small sized, preliminary study and the size sample was not large enough to draw solid conclusions; enrolment to complete a larger planned trial was discontinued due to slow accrual. Second, two patients of the surgery group had liver resection for metastasis and two patients of the chemotherapy group refused the allocated treatment after randomization. Interestingly, survival in the test group was similar with the Chinese study and median survival after CRS and HIPEC remained below 1 year in both studies, a disappointing result.

The only ongoing RCT identified in the therapeutic indication in gastric cancer is the German GASTRIPEC trial [51]. At least 65 patients have been recruited so far (November 2016) for an intended accrual of 180 [52]. This study is analyzing perioperative chemotherapy and CRS with HIPEC versus without HIPEC in gastric cancer patients with limited PM. All patients receive preoperative chemotherapy. Patients with negative or unknown HER-2 status are administered EOX combination chemotherapy (Epirubicin, Oxaliplatin and Capecitabine). Patients with positive HER-2 status are treated with Cisplatin, Capecitabine and Trastuzumab (CCT). Neoadjuvant chemotherapy is then followed by gastrectomy and surgical cytoreduction. After intraoperative randomization, the tested group receive additional HIPEC with MMC and CDDP; this group may also get a HIPEC intervention without CRS. In the postoperative period, all patients receive three cycles of chemotherapy within 4–12 weeks and are followed up for 30 months. If tumor progression is documented, then the protocol is terminated. The primary outcome measure is overall survival at 2.5 years. Survival results of this trial are not expected before 2023.

Other histologies

Patients with PM from Fallopian cancer or patients with primary peritoneal cancer were usually included in ovarian cancer studies. There is a single RCT in appendiceal cancer (NCT01580410) but this study examined two HIPEC protocols and did not compare CRS and HIPEC versus standard therapy. No RCT published, ongoing or planned was identified that evaluates CRS and HIPEC in mesothelioma or pseudomyxoma peritonei (PMP).

Discussion

A large number of published, ongoing and planned RCT are evaluating a possible therapeutic advantage for CRS and HIPEC versus other strategies in prevention and therapy of PM: Together, these RCT are randomizing 7,303 patients. They include 11 RCTs in colorectal cancer (6 for prevention of PM, n=1,107 patients; 5 for therapy, n=781), 10 in ovarian cancer (5 in frontline therapy, n=438 patients; 5 for treating recurrence, n=1,062) and 17 in gastric cancer (14 for prevention of PM, n=3,659 patients; 3 for therapy, n=256).

However, only 9 RCTs have been published so far: 1 in colorectal cancer (105 patients), 1 in ovarian cancer (130 patients) and 7 trials in gastric cancer (16–174 patients). Thus, high-level evidence on CRS and HIPEC is still scarce. There is a large consensus that CRS followed by HIPEC for PM shows some improvement in survival for selected patients with colorectal cancer, whereas evidence supporting CRS and HIPEC for other types of cancer remains limited [53].

Therefore, further RCTs are needed to compare best available systemic therapy with and without CRS and HIPEC, to determine optimal therapy parameters and the effect of each therapy component. Five RCTs have already completed recruitment and oncological results will be available soon. Due to numerous additional ongoing trials, the grade of evidence of CRS and HIPEC is expected to improve in the next decade. However, the level of evidence available and the expected evolution of this evidence are heterogeneous depending on the organ of origin.

In colorectal cancer, a first trial (PRODIGE-7, NCT00769405) will answer in the short term the question of how much survival benefit is derived from CRS and how much from HIPEC, as patients have been randomly assigned to HIPEC or no HIPEC after CRS. However, the mean PCI of included patient is high, near 20. For that reason, the tested advantage in the HIPEC group could be limited because recent publication all confirmed that in case of PM of colon cancer, the survival is related to the PCI in case of CC0 and HIPEC procedure. About 2 years later, the Prophylochip trial (NCT01226394) is expected to provide evidence on the role of CRS and HIPEC with oxaliplatin for preventing metachronous PM in colorectal cancer.

In primary advanced and recurrent ovarian cancer, evidence for CRS/HIPEC is still inconsistent at the present point of time. The only published RCT has methodologic flaws and results from case control studies and prospective cohorts are conflicting. A possible explanation for the conflicting data is that confounding factors such as quality of CRS, adequate radiological staging of PM, platin-sensitivity, etc.) have not been considered properly in the study designs. Several well-designed RCTs are in progress that might demonstrate better outcome after combined CRS, HIPEC and systemic chemotherapy, compared with current standard of care (CRS with systemic chemotherapy). In the frontline therapy, a Korean RCT (NCT01091636) has recruited all 185 patients needed and oncological results are pending. In the recurrent situation, a Dutch trial (OVHIPEC, NCT00426257) has also recruited all 245 patients needed and follow-up data will be provided soon. These data will be complemented by the large multicentric French CHIPOR trial and by the data from the Italian HORSE trial. Thus, evidence supporting or not CRS and HIPEC in primary and recurrent ovarian cancer is expected to progress significantly within the next 3–5 years.

In gastric cancer, the evidence level is progressing slowly. Two small RCTs have been published in the recurrent situation. Both trials reported an increased but short life expectancy after CRS and HIPEC (median survival below 1 year). Beyond these small-sized studies, the level of evidence of CRS and HIPEC in gastric cancer is not expected to evolve rapidly. Currently, there is only a single RCT from Germany (GASTRIPEC, NCT02158988) recruiting in the recurrent indication and this trial is recruiting slower than expected. In the preventive setting in gastric cancer, five Japanese trials are in favor of HIPEC and a major effort from China can be welcomed with seven out of nine ongoing or planned RCTs, three of them with a very large cohort size. However, due to the recent initiation of the latter trials, to the large accrual of patients needed to achieve statistical significance and to the long follow-up required (5 years), these studies are not expected to contribute information in the short term. First survival data will probably be delivered by the French GASTRICHIP trial (NCT01882933) that has already recruited half of patients needed.

Thus, a large number of RCT are ongoing that might deliver additional evidence on CRS and HIPEC in colorectal, ovarian and gastric cancer. Over the last few years, there is a strong trend in trial design from therapeutic towards preventive indications.

In the future, RCT evaluating CRS and HIPEC will continue to face methodological and practical challenges, starting with the choice of the intraperitoneal drug, the dose and the temperature and the peroperative randomization difficulty. Next difficulty will remain the choice of the adequate comparator (CRS alone vs. combination chemotherapy) as a control. It might be difficult to define precisely the control arm (physician’s best choice?). The experience shows that the time needed for designing, getting approval and completing a RCT on CRS and HIPEC in PM is around 10 years. After a decade of progress, the chemotherapy selected for the control arm might no longer be considered standard at the time point of publication. In the preventive setting, poor accuracy of CT scan for low-volumetric disease remains a significant problem. Inadequacy of RECIST 1.1 criteria for early detection of PM might threaten proper assessment of progression-free survival – the primary outcome criteria in the majority of running HIPEC trials. Inadequate determination of the major outcome criteria might indeed challenge future results of preventive HIPEC studies. The risk–benefit analysis of the trials will remain difficult since only a small proportion of patients will be cured by CRS, HIPEC and systemic chemotherapy. For the majority of patients, comparative quality of life data will be required in the palliative setting. Fortunately, many RCTs have now integrated PROs as secondary aims. Further bias such as subjectivity of completeness of cytoreduction reports, surgeon-dependent complication rates, absence of blinding, etc. might also compromise validity of results. In rare diseases, such as peritoneal mesothelioma or PMP, successful design and conduct of RCT might remain inaccessible. Last but not least, patients might be reluctant to participate in a RCT in which two radically different therapy options, and different risks, are decided by chance.

In summary, the number of published RCT evaluating CRS and HIPEC in prevention or therapy of PM is limited. There is some evidence that CRS and HIPEC improves survival in recurrent colorectal origin, evidence in ovarian and gastric cancer remains debated. A large number of studies is ongoing that are expected to deliver additional evidence. There is a clear trend in HIPEC trial design from therapeutic towards preventive indications. Trial design and interpretation of results will remain difficult because of multiple methodological challenges inherent to disease and therapy.

Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.

Research funding: None declared.

Employment or leadership: M. Pocard is expert consultant for GAMIDA company. M. Pocard had commercial relation with ROCHE, SANOFI, ETHICON company.

Honorarium: None declared.

Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.

Supplementary Material

The online version of this article (DOI: 10.1515/pp-2016-0027) offers supplementary material, available to authorized users.