Articles

Hyperthermic intraperitoneal chemotherapy for recurrent

ovarian cancer (CHIPOR): a randomised, open-label,
phase 3 trial
Jean-Marc Classe, Pierre Meeus, Delphine Hudry, Romuald Wernert, François Quenet, Frédéric Marchal, Gilles Houvenaeghel, Anne-Sophie Bats,
Fabrice Lecuru, Gwenaël Ferron, Cécile Brigand, Dominique Berton, Laurence Gladieff, Florence Joly, Isabelle Ray-Coquard,
Sylvaine Durand-Fontanier, Gabriel Liberale, Marc Pocard, Constantin Georgeac, Sébastien Gouy, Jean-Marc Guilloit, Frédéric Guyon,
Cristina Costan, Jean-Marc Rousselet, Lara de Guerké, Naoual Bakrin, Emilie Brument, Elodie Martin, Bernard Asselain, Loïc Campion,
Olivier Glehen for the UNICANCER/CHIPOR Investigators*

Summary
Background Hyperthermic intraperitoneal chemotherapy (HIPEC) at interval cytoreductive surgery for ovarian cancer Lancet Oncol 2024
improves overall survival but its role in recurrent disease is uncertain. We aimed to compare outcomes in patients Published Online
treated with or without HIPEC during surgery for recurrent ovarian cancer. November 13, 2024
[Link]
S1470-2045(24)00531-X
Methods The multicentre, open-label, randomised, phase 3 CHIPOR trial was conducted at 31 sites in France,
See Online/Comment
Belgium, Spain, and Canada, and enrolled patients with first relapse of epithelial ovarian cancer at least 6 months [Link]
after completing platinum-based chemotherapy. Eligible patients were aged 18 years or older with WHO S1470-2045(24)00592-8
performance status of less than 2. After six cycles of platinum-based chemotherapy (and optional bevacizumab), *Investigators are listed in the
patients amenable to complete cytoreductive surgery were randomly assigned centrally in a 1:1 ratio, using a web- appendix (p 1)
based system and a minimisation procedure, during surgery to receive HIPEC (cisplatin 75 mg/m² in 2 L/m² of Institut de Cancérologie de
serum at 41±1°C for 60 min) or not, stratified by centre, completeness of cytoreduction score, platinum-free interval, L’Ouest, Saint Herblain, France
(Prof J-M Classe MD,
and latterly, planned poly(ADP-ribose) polymerase inhibitor use. The primary endpoint was overall survival, D Berton MD, E Martin MSc,
analysed on an intention-to-treat basis in all randomly assigned patients. This ongoing trial is registered with L Campion MD); Nantes
[Link], NCT01376752. Université, INSERM 1307, CNRS
6075, Université d’Angers,
CRCI2NA, Nantes, France
Findings Between May 11, 2011, and May 14, 2021, 415 female patients were randomly assigned (207 HIPEC, 208 no (J-M Classe, L Campion); Centre
HIPEC). At the primary analysis (median follow-up 6·2 years, IQR 4·1–8·1), 268 (65%) patients had died (126 [61%] Léon Bérard and University
of 207 in the HIPEC group; 142 [68%] of 208 in the no-HIPEC group). Overall survival was significantly improved Claude Bernard, Lyon, France
with HIPEC (stratified hazard ratio 0·73, 95% CI 0·56–0·96; p=0·024). Median overall survival was 54·3 months (P Meeus MD,
I Ray-Coquard MD); Centre
(95% CI 41·9–61·7) with HIPEC versus 45·8 months (38·9–54·2) without. Grade 3 or worse adverse events within Oscar Lambret, Lille, France
60 days after surgery occurred in 102 (49%) of 207 patients receiving HIPEC versus 56 (27%) of 208 receiving no (D Hudry MD); Institut de
HIPEC, the most common being anaemia (47 [23%] vs 30 [14%]), hepatotoxicity (23 [11%] vs 18 [9%]), electrolyte Cancérologie de L’Ouest,
disturbance (28 [14%] vs two [1%]), and renal failure (20 [10%] vs three [1%]). There were three deaths within 60 days Angers, France (R Wernert MD);
ICM Val d’Aurelle, Montpellier,
of surgery, all in the no-HIPEC group. France (F Quenet MD); Institut
de Cancérologie de Lorraine,
Interpretation Adding HIPEC to cytoreductive surgery after response to platinum-based chemotherapy at first Vandoeuvre-Lès-Nancy, France
epithelial ovarian cancer recurrence significantly improved overall survival. When treating patients with late first (Prof F Marchal MD); Aix-
Marseille University CNRS,
relapse of high-grade serous or high-grade endometrioid ovarian cancer amenable to complete cytoreductive surgery INSERM, Institut Paoli-
at specialist centres, platinum-based HIPEC should be considered to extend overall survival. Calmettes, Department of
Surgical Oncology, CRCM,
Marseille, France
Funding French National Cancer Institute and French League Against Cancer.
(Prof G Houvenaeghel MD);
Hôpital Européen Georges-
Copyright © 2024 Elsevier Ltd. All rights reserved, including those for text and data mining, AI training, and similar Pompidou, APHP-Centre
technologies. Université Paris Cité, Paris,
France (A-S Bats MD,
Prof F Lecuru MD†); Oncopole
Introduction long-term survival are a prolonged platinum-free interval CLAUDIUS REGAUD, IUCT-
Ovarian cancer is the leading cause of death from and complete surgical resection.2,3 Oncopole, Toulouse, France
gynaecological cancer among women.1 The standard The primary site of ovarian cancer relapse is usually (G Ferron MD, L Gladieff MD);
CHU Hautepierre, Strasbourg,
treatment at primary diagnosis and at first platinum- the peritoneum,4,5 and peritoneal carcinomatosis is a
France (Prof C Brigand MD);
sensitive relapse is platinum-based chemotherapy and common cause of symptoms and death. Although Centre François Baclesse, Caen,
complete cytoreductive surgery, with bevacizumab or postoperative intraperitoneal chemotherapy via a catheter France (Prof F Joly MD,
maintenance poly(ADP-ribose) polymerase (PARP) following primary cytoreductive surgery in patients with J-M Guilloit MD); University
Caen Normandy, Caen, France
inhibitor (or both) depending on clinical and tumour newly diagnosed ovarian cancer resulted in improved (F Joly); CHU Dupuytren, CNRS
characteristics.1,2 The main factors associated with progression-free survival and overall survival, this benefit

[Link]/oncology Published online November 13, 2024 [Link] 1

 Articles

UMR 7252, Limoges, France

(S Durand-Fontanier MD); Research in context
Institut Jules Bordet, Hôpitaux
Universitaires de Bruxelles, Evidence before this study Added value of this study
Bruxelles, Belgium We searched PubMed for clinical trial reports published CHIPOR provides the first prospective randomised evidence
(G Liberale MD); Université Paris between Jan 1, 2000, and Dec 31, 2010, with the search terms showing the benefit of HIPEC in patients undergoing complete
Cité, INSERM, U1275 CAP Paris-
Tech, Paris, France
“HIPEC”, “ovarian cancer”, and “randomised”. Our search cytoreductive surgery for the first late relapse (treatment-free
(Prof M Pocard MD); identified no phase 3 trials evaluating hyperthermic interval of at least 6 months since last platinum-based
Department of Digestive, intraperitoneal chemotherapy (HIPEC) in ovarian cancer. When chemotherapy) of ovarian cancer.
Hepatobiliary Surgery and the present trial was designed, intraperitoneal administration
Liver Transplantation, Pitié- Implications of all the available evidence
Salpêtrière Hospital, Assistance
of chemotherapy had shown disappointing results and was
These robust results showing an overall survival benefit from
Publique–Hôpitaux de Paris, associated with substantial toxicity. HIPEC potentially offered a
HIPEC in the late-relapsing ovarian cancer setting add to the
Paris, France (Prof M Pocard); more effective approach and had already shown promise in
Centre Médico-Chirurgical de la recently reported OVHIPEC-1 trial demonstrating significantly
colorectal cancer, but the only available evidence in support of
Sarthe, Le Mans, France improved overall survival with the adoption of HIPEC during
(C Georgeac MD); Department
HIPEC as a treatment approach for ovarian cancer came from
interval cytoreductive surgery as primary treatment for
of Gynecological Surgery, single-arm or retrospective studies and no results from
advanced epithelial ovarian cancer. When treating patients with
INSERM Unit 10–30, Gustave randomised trials were available in any ovarian cancer setting.
Roussy Cancer Campus, late first relapse of serous or high-grade endometrioid ovarian
There was a need for randomised trials to provide a definitive
Villejuif, France (S Gouy MD); cancer amenable to complete cytoreductive surgery, platinum-
University Paris-Saclay, Gif-sur- answer on the role of HIPEC in ovarian cancer. Furthermore,
based HIPEC administered in specialist centres should be
Yvette, France (S Gouy); at the time of trial design, there were no randomised data to
considered following neoadjuvant chemotherapy.
Institut Bergonié, CLCC support surgery for relapsed ovarian cancer.
Bordeaux Nouvelle Aquitaine,
Bordeaux, France (F Guyon MD);
CHU Grenoble Alpes, Grenoble,
France (C Costan MD); CHU was not observed when combined with maintenance perform gynaecological surgery) in France, Belgium,
Dijon, Dijon, France bevacizumab.6 Furthermore, repeated postoperative Spain, and Canada (appendix p 1). The trial was
(J-M Rousselet MD); Hôpital intraperitoneal cycles are associated with increased pain conducted in accordance with the Declaration of
Maisonneuve-Rosemont
and severe toxicity,7 which limits their widespread use. Helsinki, international Good Clinical Practice
CIUSSSEMTL–Université de
Montréal, Montreal, QC, No phase 3 trials have evaluated intraperitoneal standards, and all local laws and regulations concerning
Canada (L de Guerké MD); chemotherapy in the recurrent ovarian cancer setting. clinical trials. The trial protocol, amendments, and
Centre Hospitalier Lyon Sud, Hyperthermic intraperitoneal chemotherapy (HIPEC) other relevant study documentation were approved by
Pierre-Bénite, France
is a single intraoperative procedure that delivers the Comité de Protection des Personnes Ouest IV
(N Bakrin MD,
Prof O Glehen MD); CICLY, chemotherapy in a heated solution directly into the (Nantes, France). All patients provided written
Université Lyon 1, Lyon, France abdominal cavity after cytoreductive surgery.8,9 The aim informed consent before any trial-specific procedures
(N Bakrin, Prof O Glehen); UCGI, is to improve the treatment of peritoneal microscopic or treatment. An independent data safety and
Prodige Intergroup,
UNICANCER, Paris, France
residual disease, adding the effects of thermal shock, monitoring committee reviewed mortality and
(E Brument MSc); ARCAGY- increased chemotherapy penetration at the peritoneal morbidity rates on an ongoing basis, comparing them
GINECO, Paris, France surface, and increased cancer sensitivity to chemotherapy with reference rates for similar surgery for ovarian
(B Asselain MD) by impairing DNA repair and apoptosis induction.10 In cancer without HIPEC. The trial was registered with
†Current affiliation: the OVHIPEC-1 trial, HIPEC improved both overall [Link] (NCT01376752).
Institut Curie, Paris, France
survival and progression-free survival when used as Eligible patients had epithelial ovarian, primary
Correspondence to:
primary treatment for advanced epithelial ovarian cancer peritoneal, or fallopian tube carcinoma with disease
Prof Jean-Marc Classe,
Department of Oncologic during interval cytoreductive surgery.9,11 In the recurrent recurrence at least 6 months after completing first-line
Surgery, Institut de Cancérologie setting, various (predominantly retrospective) cohort platinum-based chemotherapy, limited to the abdomen
de l’Ouest, 44805 Saint Herblain, studies assessing HIPEC provide only low-level with no distant metastases (metastatic pleural effusion
France
evidence,12,13 and guidelines for first late relapse of and inguinal lymphadenopathy were permitted). All
[Link]@
[Link] ovarian cancer recommend against this approach patients received second-line platinum-based doublet
See Online for appendix
outside clinical trials.2,14 The randomised, phase 3 chemotherapy before surgery. If the surgeon considered
CHIPOR trial evaluated platinum-based chemotherapy that complete resection was feasible (based on thoraco-
followed by complete cytoreductive surgery, with or abdomino-pelvic CT scan or MRI, with optional
without HIPEC, for the first relapse of ovarian cancer laparoscopy), surgery was performed 5–12 weeks after
with a treatment-free interval of at least 6 months since completing second-line chemotherapy. Patients were
last platinum-based chemotherapy. required to be aged 18 years or older, with WHO
performance status of less than 2, and have normal renal
Methods function (creatinine ≤1·5 × upper limit of normal,
Study design and participants creatinine clearance >60 mL/min) according to the
This multicentre, open-label, randomised, phase 3 trial Modification of Diet in Renal Disease equation. Key
was done in 31 hospitals and cancer centres with a high exclusion criteria were known hypersensitivity to
level of surgical expertise (national authorisation to cisplatin, prior anti-angiogenic therapy within 8 weeks

2 [Link]/oncology Published online November 13, 2024 [Link]

 Articles

before surgery, non-epithelial ovarian histology, previous performed every 3 months, with additional scans done if
HIPEC for ovarian cancer, uncontrolled infection, and indicated by symptoms or elevated CA-125. Because
anticipated need for more than two gastrointestinal investigational treatment was administered in a single
resections at the time of HIPEC. dose at random assignment, the only reason for
withdrawal from the study was patient or physician
Randomisation and masking decision; patients continued to be followed up unless
During surgery, patients were randomly assigned they withdrew consent. Adverse events were graded
centrally in a 1:1 ratio, using a web-based system according to National Cancer Institute Common
(TenAlea, ALEA Clinical Services, Abcoude, Terminology Criteria for Adverse Events version 4.0.
Netherlands) and a minimisation procedure, to receive Patient-reported outcomes were assessed using the
HIPEC or not at the end of the surgical procedure. European Organisation for Research and Treatment of
Randomisation was stratified by centre, outcome of Cancer (EORTC) Quality of Life Questionnaire core
cytoreductive surgery as measured by completeness of module (QLQ-C30) and a pain visual analogue scale
cytoreduction score (CC0 [no residual tumour] vs CC1 ranging from 0 (no pain) to 10 (maximum pain
[residual tumour <0·25 cm]), interval between imaginable). The pain visual analogue scale was
completing primary therapy and recurrence (6 months completed at baseline and every day during hospitalisation
to ≤12 months vs >12 months to ≤18 months vs in the morning, at noon, and in the evening, and then at
>18 months), and planned PARP inhibitor use after 3 months and 6 months. QLQ-C30 was completed at
chemotherapy (yes vs no; added on Oct 16, 2020, after baseline and at 7 days, 1 month, 6 months, and 1 year
enrolling 498 patients, to reflect evolving standards of after surgery.
care). Initially, patients were also stratified by
carboplatin partner (paclitaxel or pegylated liposomal Outcomes
doxorubicin), but given the global shortage of pegylated The primary outcome measure was overall survival,
liposomal doxorubicin, the protocol was amended on defined as the interval between randomisation and death
July 2, 2012, after enrolling 137 patients, to permit from any cause. Secondary outcome measures (all
alternative effective chemotherapy regimens, and this assessed by the investigators, not centrally reviewed)
stratification factor was removed. There was no included progression-free survival (clinical, radiological,
masking in this open-label trial. or biological progression, second cancer, or death), pain,
health-related quality of life, and safety (particularly renal
Procedures adverse events) within 60 days after surgery. Additional
After six cycles of platinum-based chemotherapy (and secondary endpoints of peritoneal progression-free
optional bevacizumab), eligible patients underwent
surgery through a median laparotomy. During surgery,
patients with incomplete resection (≥0·25 cm residual 517 patients enrolled

tumour) were withdrawn from the study and those with a

macroscopically complete tumour resection (CC0 or 102 excluded
CC1) were then randomly assigned to receive HIPEC (1-h 25 unresectable or CC >1
7 declined to participate
infusion of cisplatin 75 mg/m² in 2 L/m² of serum at 7 low creatinine clearance
41±1°C) or no HIPEC. The HIPEC administration 4 investigator decision
schedule was based on evidence available at the time.15,16 21 other
38 unknown
HIPEC was administered at the end of the surgical
procedure, either open or closed abdomen technique
according to each centre’s standard practice. Each centre 415 randomly assigned
used the same HIPEC process (temperature and
duration) for all patients across the entire study period.
Hyperhydration was maintained throughout the
207 assigned to receive HIPEC 208 assigned to receive no HIPEC
procedure (see protocol [appendix]). A protocol 207 received allocated 208 received allocated
amendment (June 19, 2018, after enrolling 392 patients) intervention intervention
following results from the OVHIPEC-1 trial11 allowed
prophylactic thiosulfate administration to reduce the risk
Analysis populations Analysis populations
of cisplatin-related renal toxicity. After surgery, patients 207 efficacy 208 efficacy
received standard-of-care maintenance therapy at the 207 safety 208 safety
157 pain visual analogue scale 159 pain visual analogue scale
investigator’s discretion. 151 QLQ-C30 147 QLQ-C30
Postoperative assessment after 7 days, 1 month, and
every 3 months thereafter for 4 years included clinical Figure 1: Trial profile
examination, CA-125 measurement, and adverse event CC=completeness of cytoreduction score. HIPEC=hyperthermic intraperitoneal
reporting. A thoraco-abdomino-pelvic CT scan was chemotherapy. QLQ-C30=Quality of Life Questionnaire core module.

[Link]/oncology Published online November 13, 2024 [Link] 3

 Articles

survival, extraperitoneal progression-­free survival, and

HIPEC (n=207) No HIPEC (n=208) time to first subsequent therapy were prespecified in the
Age, years 62 (55–68) 59 (53–67) statistical analysis plan. Peritoneal progression-free
Histological grade and type survival was defined as the time from randomisation to
High-grade serous 151 (73%) 159 (76%) peritoneal progression (clinical evidence of peritoneal
High-grade endometrioid 8 (4%) 7 (3%) mass or peritoneal effusion with cytological diagnosis, or
Low-grade serous 33 (16%) 26 (13%) measurable peritoneal mass on imaging), with censoring
Other or unknown 15 (7%) 16 (8%) at the date of death or last follow-up in the absence of
BRCA mutation status peritoneal progression. Extraperitoneal progression-free
Positive 50 (24%) 53 (25%) survival was defined as the time from randomisation to
Negative 125 (60%) 124 (60%) extraperitoneal progression (extraperitoneal mass on
Unknown 32 (15%) 31 (15%) clinical examination or imaging), with censoring at the
Front-line chemotherapy for newly diagnosed disease date of death or last follow-up in the absence of
Carboplatin plus paclitaxel 187 (90%) 192 (92%) extraperitoneal progression. Time to first subsequent
Cisplatin plus paclitaxel 2 (1%) 0 therapy was defined as the time from randomisation
Carboplatin plus pegylated liposomal doxorubicin 2 (1%) 3 (1%) to initiation of first subsequent treatment, with
Other 14 (7%) 11 (5%) censoring at death or last follow-up in the absence of
Missing 2 (1%) 2 (1%)
subsequent treatment.
Second-line chemotherapy for recurrent disease immediately before randomisation
Protocol-specified medico-economic and pharma­
Carboplatin plus pegylated liposomal doxorubicin 98 (47%) 98 (47%)
cokinetic outcomes are not reported here, because the
Carboplatin plus paclitaxel 81 (39%) 75 (36%)
analyses have not yet been performed.
Other carboplatin combination 28 (14%) 35 (17%)
Statistical analysis
Second-line chemotherapy cycles completed
The assumed median overall survival in the control
<6 45 (22%) 42 (20%)
group was 29 months, based on the ICON4 trial.17 To
6 141 (68%) 144 (69%)
detect a 12-month improvement (hazard ratio 0·71) with
>6* 21 (10%) 22 (11%)
80% power, two-sided alpha of 5%, recruitment over
Platinum-free interval†
10 years, and 4 years’ follow-up, 202 patients were
6 months to ≤12 months 53 (26%) 54 (26%)
required in each group. Allowing for 10% dropout,
>12 months to ≤18 months 55 (27%) 51 (25%)
444 randomly assigned patients were required to provide
>18 months 99 (48%) 103 (50%)
268 events.
Previous anti-angiogenic therapy‡ 68 (33%) 74 (36%)
An interim analysis was performed after reaching half
Previous bevacizumab in the front-line setting 61 (29%) 69 (33%) (134) of the required number of deaths. According to the
Previous bevacizumab in the recurrent setting (immediately 4 (2%) 4 (2%) Peto–Haybittle method, the threshold for the difference
before randomisation in CHIPOR)
between treatment groups to be considered statistically
Exploratory coelioscopy or laparoscopy 48 (23%) 46 (22%)
significant at the interim analysis (p=0·001) was not met.
Before chemotherapy before randomisation 26 (13%) 26 (13%)
Because all patients were randomly assigned and
During chemotherapy before randomisation 7 (3%) 7 (3%)
treated during surgery, the analysis population for both
After chemotherapy before randomisation 11 (5%) 10 (5%)
efficacy and safety was the intention-to-treat population
Before randomisation (not otherwise specified) 1 (<1%) 3 (1%)
(defined as all randomly assigned patients). Patient-
At or after randomisation or unknown 3 (1%) 0
reported outcomes were analysed in the intention-to-treat
Peritoneal cancer index 5 (3–10) 5 (2–9)
population with a baseline assessment of the relevant
Completeness of cytoreduction score† outcome. Overall survival and progression-free survival
CC0 (no residual disease after surgery) 180 (87%) 180 (87%) were estimated using Kaplan–Meier methodology. For
CC1 (<0·25 cm residual disease after surgery) 27 (13%) 28 (13%) the primary analysis, treatment groups were compared
HIPEC with gastrointestinal tract resection§ 97 (47%) 89 (43%) using a univariable robust Cox regression analysis,
AGO score¶ stratified on the four randomisation stratification factors
Positive 173 (84%) 176 (85%) (centre, completeness of cytoreduction score, platinum-
Negative 33 (16%) 28 (13%) free interval, planned PARP inhibitor use). The stratified
Missing 1 (<1%) 4 (2%) hazard ratio was calculated with corresponding robust
Data are median (IQR) or n (%). AGO=Arbeitsgemeinschaft Gynaekologische Onkologie. HIPEC=hyperthermic 95% CIs. Proportional-hazards assumptions were
intraperitoneal chemotherapy. *In the HIPEC group: 7 cycles in 8 patients, 8 cycles in 3 patients, 9 cycles in 1 patient, checked using Schoenfeld residuals (which indicated
and 10 cycles in 9 patients; in the no-HIPEC group: 7 cycles in 12 patients, 8 cycles in 2 patients, and 10 cycles in
that the proportional hazards assumption was not
8 patients. †Stratification factor. ‡In the HIPEC group, includes pazopanib in 4 patients and nintedanib in 3 patients;
in the no-HIPEC group, includes nintedanib in 3 patients, pazopanib in 1 patient, and sorafenib in 1 patient. §Stomach, violated), and Kaplan–Meier and predicted combined
duodenum, small intestine, colon, rectum. ¶Estimated retrospectively. survival plots (which showed no graphical deviation from
risk proportionality). In addition, numerous sensitivity
Table 1: Baseline (pre-randomisation) characteristics
analyses were performed (for the primary endpoint:

4 [Link]/oncology Published online November 13, 2024 [Link]

 Articles

non-stratified, stratified on the four stratification factors, likely to be admitted to the intensive care unit than
adjusted on the four stratification factors, and stratified patients in the no-HIPEC group (table 2). Following
on the four stratification factors with two additional time- surgery, maintenance PARP inhibitor therapy was
dependent variables taken into account; for the secondary initiated in similar proportions of patients in the
endpoints: stratified on the four stratification factors and two treatment groups at a similar interval after surgery
adjusted on the four stratification factors). During follow- (table 2).
up, the occurrence (or not) of two non-baseline events Similar proportions of patients in both groups received
(ovarian cancer relapse, initiation of bevacizumab treatment for progression following surgery in
treatment) could potentially influence overall survival CHIPOR (table 2). Numerically more patients received
(primary endpoint); these events, if they occurred, could bevacizumab in the no-HIPEC group and more
emerge at different times in different patients. A underwent further surgery (without HIPEC) in the
stratified Cox model adjusting for these two variables HIPEC group, although in seven of 19 patients this was
(used as time-dependent variables) explored whether not considered to be “significant resection”. Five patients
HIPEC retained its independent prognostic impact on in each group received HIPEC at progression after
overall survival. Prespecified subgroup analyses of randomised treatment.
efficacy included subgroups defined by age, platinum- At the data cutoff for the final analysis (Jan 8, 2023,
free interval, completeness of cytoreduction score, median follow-up 6·2 years [IQR 4·1–8·1]), 268 patients
peritoneal cancer index, histology, and BRCA mutation had died (126 [61%] of 207 in the HIPEC group; 142 [68%]
status. The two-sided significance threshold was p=0·05 of 208 in the no-HIPEC group). All but six patients
for all analyses. (one in the HIPEC group, five in the no-HIPEC group)
Additional post-hoc exploratory analyses assessed the died from disease progression. The hazard ratio for
impact on renal safety of the protocol amendment
allowing prophylactic thiosulfate and explored efficacy HIPEC No HIPEC
and safety over different enrolment periods. An (n=207) (n=208)
Arbeitsgemeinschaft Gynaekologische Onkologie (AGO) Bowel resection 62 (30%) 52 (25%)
score18 was calculated retrospectively for each patient to Preventive colostomy or ileostomy 20/62 (32%) 10/52 (19%)
assess the resectability of disease despite the lack of HIPEC technique
formal selection for patients with a high probability of Open 154 (74%) NA
complete resection in this trial. Closed 51 (25%) NA
Analyses were done using SAS version 9.4 and Stata Missing 2 (1%) NA
SE 17.0. Duration of surgery, min 337 (272–407) 218 (160–282)
Duration of hospitalisation, days 15 (13–19) 12 (9–16)
Role of the funding source Admission to intensive care unit 142 (69%) 92 (44%)
The funders of the study had no role in study design, Duration of intensive care unit 7 (4–9) 4 (2–7)
data collection, data analysis, data interpretation, or stay, days
writing of the report. The sponsor, R&D UNICANCER, Maintenance therapy before recurrence
had no role in the study design or data interpretation. Its PARP inhibitor 33 (16%) 42 (20%)
representatives collected and analysed the data. All Time from surgery to start of 51 (39–69) 55 (42–87)
versions of the manuscript were prepared by the authors, PARP inhibitor, days
with editorial and writing assistance funded by Duration of PARP inhibitor, 13 (4–30) 15 (8–28)
months
R&D UNICANCER.
Bevacizumab 0 0

Results Treatment at or after progression

following surgery in CHIPOR*
177 (86%) 175 (84%)

Between May 11, 2011 and May 14, 2021, 517 patients were
Chemotherapy 155 (75%) 164 (79%)
enrolled from 31 sites. Of these, 11 (35%) sites enrolled
Bevacizumab 22 (11%) 34 (16%)
ten or more patients, 11 (35%) enrolled five to
PARP inhibitor 43 (21%) 41 (20%)
nine patients, and nine (29%) enrolled fewer than
Time from recurrence to start 9·1 (5·4–24·3) 8·8 (5·4–24·4)
five patients. At surgery, 207 patients were randomly of PARP inhibitor, months
assigned to HIPEC and 208 to no HIPEC (figure 1). Duration of PARP inhibitor, 8·9 (3·4–25·3) 6·9 (3·0–20·4)
Baseline characteristics were similar between treatment months
groups (table 1). Information on gender was not collected HIPEC 5 (2%) 5 (2%)
in this study in patients with ovarian cancer, and laws in Surgery without HIPEC 19 (9%) 9 (4%)
France prevent collection of details on ethnicity unless
Data are n (%), n/N (%), or median (IQR). HIPEC=hyperthermic intraperitoneal
clearly the subject of research. chemotherapy. NA=not applicable. PARP=poly(ADP-ribose) polymerase.
Most patients completed six cycles of chemotherapy *More than 1 type of therapy possible.
before secondary surgery (table 1). Patients receiving
Table 2: Details of surgery and postoperative therapy
HIPEC had a longer duration of surgery and were more

[Link]/oncology Published online November 13, 2024 [Link] 5

 Articles

overall survival (stratified using randomisation 0·63–0·99; figure 2B). Median progression-free
stratification factors) was 0·73 (95% CI 0·56–0·96; survival was 10·2 months (95% CI 9·3–11·9) with
stratified univariable robust Cox p=0·024). Median HIPEC and 9·5 months (8·6–11·6) without HIPEC.
overall survival was 54·3 months (95% CI 41·9–61·7) Analyses of peritoneal progression-free survival and
with HIPEC versus 45·8 months (38·9–54·2) without time to first subsequent therapy also favoured HIPEC;
HIPEC (figure 2A). this was less clear for extraperitoneal progression-free
Progression-free survival events had occurred in 181 survival (figure 2).
(87%) of 207 patients assigned to HIPEC and 185 (89%) Prespecified subgroup analyses showed consistent
of 208 assigned to no HIPEC. The stratified hazard results across all efficacy endpoints for all evaluated
ratio for progression-free survival was 0·79 (95% CI subgroups, with the 95% CIs for the hazard ratio

A Overall survival
100 HIPEC No HIPEC
Events, n (%) 126 (61%) 142 (68%)
HR (95% CI) 0·73 (0·56–0·96); p=0·024
75
Median, months (95% CI) 54·3 (41·9–61·7)
Overall survival (%)

45·8 (38·9–54·2)
5-year rate (95% CI) 46% (39–53) 38% (31–45)
50

25
HIPEC
No HIPEC
0
0 12 24 36 48 60 72 84 96
Number at risk
(number censored)
HIPEC 207 (0) 193 (2) 158 (13) 113 (23) 87 (30) 65 (43) 44 (53) 25 (64) 15 (70)
No HIPEC 208 (0) 194 (1) 150 (14) 106 (23) 75 (34) 49 (46) 28 (53) 13 (59) 7 (62)

B Progression-free survival
100
HIPEC No HIPEC
181 (87%)
Progression-free survival (%)

Events, n (%) 185 (89%)

75 Median, months (95% CI) 10·2 (9·3–11·9) 9·5 (8·6–11·6)
HR (95% CI) 0·79 (0·63–0·99)

50

25

0
0 12 24 36 48 60 72 84 96
Number at risk
(number censored)
HIPEC 207 (0) 87 (0) 45 (1) 26 (6) 19 (11) 14 (13) 11 (16) 6 (21) 5 (21)
No HIPEC 208 (0) 86 (0) 32 (3) 22 (8) 15 (11) 8 (17) 5 (19) 4 (20) 0 (23)

C Peritoneal progression-free survival

100
HIPEC No HIPEC
Events, n (%) 153 (74%) 160 (77%)
Peritoneal progression-free

75 Median, months (95% CI) 12·3 (10·5–15·6) 12·1 (9·7–12·9)

HR (95% CI) 0·74 (0·57–0·94)
survival (%)

50

25

0
0 12 24 36 48 60 72 84 96
Time since randomisation (months)
Number at risk
(number censored)
HIPEC 207 (0) 104 (7) 57 (13) 33 (26) 25 (33) 17 (38) 13 (42) 8 (47) 5 (49)
No HIPEC 208 (0) 99 (10) 40 (20) 28 (27) 19 (33) 10 (41) 5 (45) 4 (46) 0 (48)

(Figure 2 continues on next page)

6 [Link]/oncology Published online November 13, 2024 [Link]

 Articles

D Extraperitoneal progression-free survival

100 HIPEC
No HIPEC
Extraperitoneal progression-free

75
survival (%)

50
HIPEC No HIPEC
25 Events, n (%) 84 (41%) 72 (35%)
Median, months (95% CI) 87·8 (48·5–NE) 86·9 (71·7–NE)
HR (95% CI) 0·83 (0·59–1·17)
0
0 12 24 36 48 60 72 84 96
Number at risk
(number censored)
HIPEC 207 (0) 155 (7) 117 (24) 80 (51) 58 (70) 46 (81) 33 (94) 20 (106) 12 (111)
No HIPEC 208 (0) 158 (9) 108 (40) 73 (72) 51 (89) 36 (104) 17 (122) 8 (130) 1 (135)

E Time to subsequent therapy

100
Proportion of patients receiving first
subsequent therapy (%)

75

50 No HIPEC
HIPEC
Events, n (%) 177 (86%) 175 (84%)
25 Median, months (95% CI) 11·5 (10·8–14·4) 11·4 (9·4–13·2)
HR (95% CI) 0·65 (0·51–0·83)

0
0 12 24 36 48 60 72 84 96
Time since randomisation (months)
Number at risk
(number censored)
HIPEC 207 (0) 100 (1) 53 (3) 28 (11) 20 (16) 14 (18) 11 (21) 7 (25) 6 (25)
No HIPEC 208 (0) 96 (5) 34 (12) 22 (17) 15 (20) 7 (26) 5 (28) 3 (30) 1 (32)

Figure 2: Efficacy by treatment group

(A) Overall survival. (B) Progression-free survival. (C) Peritoneal progression-free survival. (D) Extraperitoneal progression-free survival. (E) Time to first subsequent
therapy. HR is the stratified HR including all four factors (centre, outcome of cytoreductive surgery, interval between completing primary therapy and recurrence, and
planned PARP inhibitor use after chemotherapy). HIPEC=hyperthermic intraperitoneal chemotherapy. HR=hazard ratio. NE=not estimable.

estimates overlapping those of the overall population Within 60 days after surgery, the most common grade 3
(figure 3, appendix pp 3–4). Interaction p values indicated or worse adverse events were anaemia, electrolyte
no significant association between treatment effect and disturbance, hepatotoxicity, and renal failure in the
any of the tumour or patient characteristics analysed, HIPEC group and anaemia in the no-HIPEC group
although a numerically larger effect of HIPEC was seen (table 3). Red blood cell transfusions were required by
in the CC1 subgroup for all endpoints. Additionally, in the 35 (17%) of 207 patients in the HIPEC group and 19 (9%)
subgroup with BRCA mutation (of whom 72 [70%] of of 208 in the no-HIPEC group (more than two units in
103 received a PARP inhibitor), overall survival and 16 [8%] vs two [1%]). Three patients (all in the no-HIPEC
progression-free survival (global, peritoneal, and group) died within 60 days after surgery (one patient
extraperitoneal) appeared to numerically favour the from peritonitis on day 3, two patients from
no-HIPEC group. haemoperitoneum on days 5 and 14). There were
Sensitivity analyses of overall survival showed effects three additional deaths not attributed by the investigator
consistent with the main analysis (appendix p 2). to disease progression (one from unknown cause almost
Additional exploratory post-hoc analyses examined 5 years after surgery in the HIPEC group; two in the
differences between patients enrolled in the first 5 years no-HIPEC group >1 year after randomisation from
of the trial (2011–16) and those enrolled later (2017–21). haemorrhage in one patient and intercurrent disease in
Consistent with evolving standards of care, the proportion one patient).
of patients receiving a PARP inhibitor was higher in the Grade 3 or worse renal failure (comprising acute
latter period (94 [57%] of 164 vs 65 [26%] of 251 enrolled in kidney injury, acute renal failure, and renal toxicity)
2011–16). However, the treatment effect of HIPEC on occurred in 20 (10%) of 207 patients in the HIPEC group
overall survival and progression-free survival persisted versus three (1%) of 208 patients in the no-HIPEC group.
across both time periods. In post-hoc exploratory analyses assessing the impact of

[Link]/oncology Published online November 13, 2024 [Link] 7

 Articles

Number of events/ Stratified HR

Discussion
number of patients (95% CI) To the best of our knowledge, CHIPOR is the largest
HIPEC No HIPEC randomised trial to prospectively evaluate HIPEC in
Age
patients being treated for recurrent ovarian cancer. After
≥65 years 52/88 45/73 0·65 (0·36–1·18) 6·2 years’ median follow-up, CHIPOR demonstrated a
<65 years 74/119 97/135 0·85 (0·60–1·21) statistically significant overall survival benefit with
Histology HIPEC, with median overall survival of 54 months in the
High-grade serous 91/151 107/159 0·77 (0·56–1·05) HIPEC group and 46 months in the no-HIPEC group.
Other 34/55 33/46 0·56 (0·23–1·37) Progression-free survival was also improved with HIPEC,
Median PCI apparently driven by improved peritoneal progression-
>5 58/84 70/87 0·38 (0·22–0·66) free survival. Grade 3 or worse adverse events were twice
≤5 55/103 65/100 0·95 (0·61–1·46) as common with HIPEC, mainly due to anaemia and
Completeness of cytoreduction score
renal failure. These toxicities had no detrimental impact
CC1 20/27 24/28 0·07 (0·01–0·44)
on patient-reported outcomes. This is consistent with the
CC0 106/180 118/180 0·84 (0·64–1·12)
findings from a small study that focused on toxicity and
Platinum-free interval
6–12 months 40/53 41/54 0·63 (0·38–1·04)
quality of life (assessed using FACT-O) in patients
12–18 months 34/55 33/51 0·94 (0·56–1·60)
receiving neoadjuvant chemotherapy and cytoreductive
>18 months 52/99 68/103 0·70 (0·47–1·05) surgery with HIPEC.19 Following the protocol amendment
BRCA status to allow prophylactic thiosulfate, the incidence of severe
Mutated 26/50 29/53 1·36 (0·59–3·14) renal failure was dramatically reduced. The proportion of
Not mutated 74/125 85/124 0·71 (0·48–1·03) patients undergoing bowel tract resection was similar in
All patients 126/207 142/208 0·73 (0·56–0·96) the two groups, but the use of a protective stoma was
0·25 0·50 1 2 3
more common with HIPEC, suggesting concern among
surgeons about fistula risk in the event of gastrointestinal
Favours HIPEC Favours no HIPEC reconstruction during HIPEC. Admission to the
Figure 3: Subgroup analyses of overall survival
intensive care unit was more common in the HIPEC
The size of each square is proportional to the subgroup size. Interaction p values (with no adjustment for multiple group but this reflects the standard precautionary
testing) indicated no significant association between overall survival treatment effect and any of the tumour or practice, regardless of any complications, at the time
patient characteristics tested. Missing data in 4 patients (1 HIPEC, 3 no HIPEC) for histology, 41 (20 HIPEC, 21 no when the trial was initiated. Admission to the intensive
HIPEC) for PCI, and 63 (32 HIPEC, 31 no HIPEC) for BRCA. Among 103 patients with a BRCA mutation, 72 received a
PARP inhibitor. HIPEC=hyperthermic intraperitoneal chemotherapy. HR=hazard ratio. PARP=poly(ADP-ribose) care unit is more common in France than in some other
polymerase. PCI=peritoneal cancer index. health-care systems. There were three postoperative
deaths within the first 60 days in the no-HIPEC group
the thiosulfate protocol amendment on renal safety, and none in the HIPEC group.
grade 3 or worse renal failure was reported in 19 (12%) of The CHIPOR trial enrolled a highly chemosensitive
156 patients receiving HIPEC before the amendment population. Approximately half of the patients had a
(ongoing in 14 patients) versus one (2%) of 51 treated platinum-free interval of more than 18 months, three-
after the amendment. Corresponding incidences in the quarters had high-grade serous histology, and a quarter
no-HIPEC group were two (1%) of 154 and one (2%) of had a BRCA mutation. Neither of the recently validated
54. Likewise, in the HIPEC group the median duration of scoring systems (AGO,18 iMODEL20) was used to select
intensive care unit stay decreased from 7 days (IQR 4–9) patients with a high probability of complete resection.
to 6 days (3–7) following the thiosulfate amendment, and Instead, selection was based on clinical assessment, CT
the median duration of hospitalisation decreased from scan, CA-125 level, and optional laparoscopy. The 87%
16 days (13–20) to 14 days (10–16), whereas the duration rate of CC0 resection is higher than in recent randomised
of intensive care unit stay and hospitalisation remained trials evaluating surgery for first relapse of ovarian cancer
stable in the no-HIPEC group. (76% in DESKTOP III,21 67% in GOG-213,22 77% in
At baseline, 314 (76%) of 415 patients completed the SOC-123), presumably because patients with residual
pain visual analogue scale, decreasing to 89 (21%) by disease of more than 0·25 cm were not randomly
day 14. Pain visual analogue scale scores in both assigned. Furthermore, most patients (79%) received at
treatment groups increased immediately after surgery least six cycles of chemotherapy before surgery, likely
and then decreased gradually, with no differences explaining the low median peritoneal cancer index at
between treatment groups (appendix p 5). Overall, 294 surgery. In addition, all patients were enrolled at centres
(70%) patients completed the QLQ-C30 questionnaire at with a high level of surgical expertise (national
inclusion, 207 (50%) at surgery, 215 (52%) postoperatively, authorisation to perform gynaecological surgery) and
145 (34%) at 6 months after surgery, and 82 (19%) at experienced in using HIPEC. Retrospective review of
12 months after surgery. EORTC QLQ-C30 scores patient records suggested that less than 15% of patients
revealed no meaningful differences between treatment would have been considered to have a negative AGO
groups (appendix pp 6–8). score at baseline (representing high risk of incomplete

8 [Link]/oncology Published online November 13, 2024 [Link]

 Articles

HIPEC (n=207) No HIPEC (n=208)

Grade 1–2 Grade 3 Grade 4 Grade 5 Grade 1–2 Grade 3 Grade 4 Grade 5
Any 74 (36%) 85 (41%) 17 (8%) 0 82 (39%) 45 (22%) 8 (4%) 3 (1%)
Anaemia 95 (46%) 45 (22%) 2 (1%) 0 73 (35%) 29 (14%) 1 (<1%) 0
Hepatotoxicity* 36 (17%) 19 (9%) 4 (2%) 0 15 (7%) 12 (6%) 6 (3%) 0
Thrombocytopenia 47 (23%) 1 (<1%) 1 (<1%) 0 27 (13%) 0 0 0
Renal failure 39 (19%) 14 (7%) 6 (3%) 0 9 (4%) 3 (1%) 0 0
Electrolyte disturbance† 10 (5%) 24 (12%) 4 (2%) 0 5 (2%) 2 (1%) 0 0
Urinary tract infection‡ 11 (5%) 2 (1%) 0 0 16 (8%) 0 0 0
Lymphopenia 4 (2%) 8 (1%) 0 0 14 (7%) 1 (<1%) 0 0
Leukopenia 14 (7%) 1 (<1%) 0 0 9 (4%) 0 0 0
Pain 10 (5%) 1 (<1%) 0 0 9 (4%) 1 (<1%) 1 (<1%) 0
Hypoalbuminemia 8 (4%) 1 (<1%) 0 0 10 (5%) 2 (1%) 0 0
Intestinal obstruction§ 4 (2%) 1 (<1%) 1 (<1%) 0 4 (2%) 4 (2%) 1 (<1%) 0
Pleural effusion 7 (3%) 4 (2%) 0 0 3 (1%) 0 0 0
Lung disorder¶ 7 (3%) 1 (<1%) 1 (<1%) 0 1 (<1%) 2 (1%) 0 0
Peritoneal haemorrhage 0 5 (2%) 4 (2%) 0 0 1 (<1%) 0 2 (1%)
Sepsis|| 5 (2%) 2 (1%) 1 (<1%) 0 3 (1%) 1 (<1%) 0 0
Thrombosis** 5 (2%) 2 (1%) 0 0 2 (1%) 1 (<1%) 1 (<1%) 0
Fistula†† 2 (1%) 5 (2%) 3 (1%) 0 0 4 (2%) 1 (<1%) 0
Neurological complication‡‡ 6 (3%) 1 (<1%) 0 0 2 (1%) 0 0 0
Vomiting 2 (1%) 3 (1%) 0 0 3 (1%) 1 (<1%) 0 0
Abdominal wall abscess 1 (<1%) 1 (<1%) 0 0 3 (1%) 2 (1%) 1 (<1%) 0
Abdominal pain 2 (1%) 1 (<1%) 0 0 5 (2%) 0 0 0
Bleeding§§ 1 (<1%) 5 (2%) 0 0 0 1 (<1%) 0 0
Acute respiratory failure¶¶ 5 (2%) 0 1 (<1%) 0 0 0 0 0
Gastrointestinal disorder 2 (1%) 0 1 (<1%) 0 2 (1%) 0 0 0
Abdominal abscess 2 (1%) 0 0 0 1 (<1%) 1 (<1%) 0 0
Intra-abdominal fluid collection 0 2 (1%) 0 0 0 1 (<1%) 1 (<1%) 0
Procedural haemorrhage 0 2 (1%) 0 0 0 1 (<1%) 1 (<1%) 0
Lymphocele 0 2 (1%) 0 0 0 1 (<1%) 0 0
Peritonitis 0 1 (<1%) 1 (<1%) 0 0 0 0 1 (<1%)
Procedural other complication (not haemorrhage)|||| 0 3 (1%) 0 0 0 0 0 0
Malnutrition 0 0 0 0 0 2 (1%) 0 0
Data are n (%). Additional grade ≥3 adverse events each occurring in only 1 patient comprised grade 4 multiple organ dysfunction syndrome, grade 4 respiratory failure, and
grade 3 bicytopenia, biloma, chronic kidney disease, congestive cardiomyopathy, dehydration, and fluid retention in the HIPEC group, and grade 3 pancreatitis and splenic
infarction in the no HIPEC group. HIPEC=hyperthermic intraperitoneal chemotherapy. *Includes hepatotoxicity, cholestasis, gamma-glutamyl transferase increased, aspartate
aminotransferase increased, transaminase increased, alanine aminotransferase increased, liver function tests increased, drug-induced liver injury, and hepatic cytolysis.
†Includes hyponatraemia, hypokalaemia, hypomagnesaemia, hyperkalaemia, and electrolyte imbalance. ‡Includes urinary tract infection and pyelonephritis. §Includes
gastrointestinal obstruction, intestinal obstruction, ileus, subileus, distal intestinal obstruction syndrome, and ileus paralytic. ¶Includes hypoxia, pneumothorax, dyspnoea,
and atelectasis. ||Includes infection, device-related infection, skin infection, haemophilus infection, Citrobacter infection, Clostridium difficile infection, Staphylococcus test
positive, pneumonia, subcutaneous abscess, and systemic inflammatory response syndrome. **Includes pulmonary embolism, jugular vein thrombosis, thrombosis, venous
thrombosis, superficial vein thrombosis. ††Includes fistula, anastomotic fistula, colonic fistula, intestinal fistula, arterio-enteric fistula, small intestinal perforation, and
intestinal perforation. ‡‡Includes sensory loss, paraesthesia, peripheral neuropathy, ageusia, and impaired gastric emptying. §§Includes haemorrhage, haematemesis,
melaena, pelvic haematoma, retroperitoneal haematoma, vaginal haemorrhage, and postprocedural haematoma. ¶¶Includes acute pulmonary oedema and acute
respiratory distress syndrome. ||||Includes bladder injury, diaphragmatic injury, and unspecified procedural complication.

Table 3: Most common adverse events occurring within 60 days of surgery (grade 1–2 in ≥10% of patients or any grade ≥3)

surgery), had this scoring system been available at the HIPEC. Although the sample sizes were small and the
time of enrolment to CHIPOR. Of note, AGO score was trial was not powered to detect differences in subgroups,
balanced between the two treatment groups. These there appeared to be a more pronounced effect of HIPEC
findings indicate that patients enrolled in CHIPOR in patients with incomplete resection (CC1) across almost
represented a selected population and results may not be all endpoints. This observation could suggest a
applicable to patients with poor response to neoadjuvant therapeutic effect of intraperitoneal heated treatment on
chemotherapy. They also raise the question whether the incomplete surgery, which merits prospective
AGO score could be used to aid selection of patients for investigation. The apparent lack of progression-free

[Link]/oncology Published online November 13, 2024 [Link] 9

 Articles

survival benefit with HIPEC in patients with BRCA longest in patients undergoing secondary complete
mutations is difficult to interpret, particularly given the cytoreductive surgery with HIPEC followed by platinum-
small sample sizes and the lower event rates (which based chemotherapy and olaparib. When the CHIPOR
might be attributable to the better prognosis or potentially trial was designed, it was known that optimal surgery at
later enrolment of these patients). first relapse required complete resection28 but there were
In the overall population, despite the relatively small no available randomised results to inform on the timing
difference in progression-free survival at the median, the of surgery. The DESKTOP III trial, which provided
hazard ratios capturing the magnitude of benefit across evidence that cytoreductive surgery followed by
the whole curve were quite consistent between endpoints. chemotherapy was superior to chemotherapy alone for
Intriguingly, the overall survival benefit emerged at recurrent disease, reported when enrolment to CHIPOR
around 48 months, after approximately half of the was almost complete.21 The GOG-021322 and SOC-123
patients had died. As peritoneal progression is typically randomised phase 3 trials in the recurrent setting also
the main cause of death of patients with recurrent assessed cytoreductive surgery before chemotherapy, and
ovarian cancer, the longer-term benefit of HIPEC might there are still no randomised trial data in the recurrent
be driven by and amplified by its effect on peritoneal setting comparing cytoreductive surgery before and after
progression-free survival. Another hypothesis involves chemotherapy. Consequently, cytoreductive surgery
the heterogeneity of ovarian cancer.24 There might be a following neoadjuvant chemotherapy, as used in
population of patients with less chemosensitive clones, CHIPOR, is still justified and was preferred because it
which grow despite platinum and result in early death minimised the risk of delaying chemotherapy in the
irrespective of treatment. Of note, subgroup analyses event of severe postoperative morbidity with HIPEC.
indicated that patients who did not have high-grade Another major change in the management of ovarian
serous cancer benefited from HIPEC at least as much as cancer is treatment selection according to molecular
those with high-grade serous cancer, as indicated by the profiling, which is now standard. However, when the
more favourable hazard ratios. CHIPOR trial was designed, BRCA mutational status
Overall, there were no quantitative imbalances in was rarely assessed and this information is missing in
maintenance therapy following surgery between 15% of patients treated in CHIPOR. These gaps provide
treatment groups. Relatively few patients in either group several opportunities for further research using the
received maintenance PARP inhibitors (20% in the CHIPOR dataset. For example, application of KELIM
no-HIPEC group and 16% in the HIPEC group) as these scoring29 might help to identify whether CHIPOR
agents were not used routinely until 2018. None received enrolled a heterogeneous population with respect to
maintenance bevacizumab, which became available chemosensitivity and might elucidate whether patients
earlier than PARP inhibitors (perhaps because of with a KELIM score of more than 1 might derive greater
exposure in the front-line setting in around one-third of progression-free survival benefit from HIPEC than those
patients, precluding re-exposure in the platinum- with an unfavourable KELIM score.
sensitive setting, or because it was being reserved for In conclusion, CHIPOR provides robust evidence that
later lines). Likewise, there were no imbalances between HIPEC improves overall survival in patients undergoing
treatment groups in the proportion of patients receiving complete cytoreductive surgery for platinum-sensitive
subsequent systemic therapy for progression or the type recurrent ovarian cancer. Progression-free survival was
of therapy administered, although specific details of also improved, particularly peritoneal progression-free
subsequent therapies were not routinely collected. survival. Toxicity was increased with the use of HIPEC
A potential weakness limiting interpretation is the long and this treatment should be administered in specialist
duration of enrolment and follow-up, necessary for a centres with experience of preventing and managing
survival endpoint in this setting. Since initiation 10 years toxicities and with access to prophylactic thiosulfate.
ago, treatment options, lifestyle changes, and improved Nevertheless, there was no detectable detrimental effect
anaesthesiology have led to dramatic improvements in on quality of life with HIPEC and no increase in pain in
prognosis,25,26 which are reflected in longer overall the immediate postoperative period. Therefore, when
survival (median 46 months in the control group treating patients with late first relapse of high-grade
compared with an assumed 29 months based on serous or high-grade endometrioid ovarian cancer
ICON417). Furthermore, the evolving landscape led to amenable to complete cytoreductive surgery, platinum-
heterogeneous post-surgical management in the based HIPEC should be considered to extend
CHIPOR trial and might raise questions about the overall survival.
relevance of data generated before the PARP inhibitor Contributors
era. However, an analysis of more than 600 patients UNICANCER was responsible for the trial conduct and statistical
undergoing secondary cytoreductive surgery provides analyses (performed by LC and EM). J-MC and LC wrote the manuscript
supported by a medical writer funded by UNICANCER. All authors had
reassurance that these findings can be applied more full access to all the data in the study, participated in manuscript
broadly in the context of BRCA-mutated ovarian cancer development and finalisation, assume responsibility for the accuracy and
treated with PARP inhibitors.27 Overall survival was completeness of the data, and vouch for the trial’s fidelity to the protocol.

10 [Link]/oncology Published online November 13, 2024 [Link]

 Articles

LC, EM, and EB accessed and verified the data. All authors had final 8 González-Moreno S, González-Bayón LA, Ortega-Pérez G.
responsibility for the decision to submit for publication. Hyperthermic intraperitoneal chemotherapy: rationale and
technique. World J Gastrointest Oncol 2010; 2: 68–75.
Declaration of interests
9 Aronson SL, Lopez-Yurda M, Koole SN, et al. Cytoreductive surgery
J-MC reports payment from GSK and support from MSD. PM reports with or without hyperthermic intraperitoneal chemotherapy in
grants from GSK, and payment from Com&Co Events. FL reports patients with advanced ovarian cancer (OVHIPEC-1): final survival
support for meeting travel from GSK. GF reports consulting fees from analysis of a randomised, controlled, phase 3 trial. Lancet Oncol
Rand Biotech, support for travel and meetings from GSK, MSD, 2023; 24: 1109–18.
AstraZeneca, and Olympus EMEA, and advisory board participation for 10 Helm CW, Randall-Whitis L, Martin R 3rd, et al. Hyperthermic
GSK and MSD. FJ reports honoraria or consultation fees from Pfizer, intraperitoneal chemotherapy in conjunction with surgery for the
Viatris, Ipsen, Astellas, AstraZeneca, Amgen, Bayer, Novartis/3A, MSD, treatment of recurrent ovarian carcinoma. Gynecol Oncol 2007;
GSK, Janssen, and Eisai, and travel support from Eisai, MSD, Ipsen, and 105: 90–96.
Chugai. IR-C reports consulting fees and honoraria from Agenus, 11 van Driel WJ, Koole SN, Sikorska K, et al. Hyperthermic
Blueprint, BMS, PharmaMar, Genmab, Pfizer, AstraZeneca, Roche, intraperitoneal chemotherapy in ovarian cancer. N Engl J Med 2018;
GSK, MSD, Deciphera, Mersana, Merck Serono, Novartis, Amgen, 378: 230–40.
Macrogenics, Tesaro, and Clovis. MP reports honoraria from 12 Classe JM, Asselain B, Campion L, et al. Survival outcomes after
Thermasolutions, support for attending meetings or travel from hyperthermic intraperitoneal chemotherapy for a first ovarian
GAMIDA Tech, IDI Med, and Thermasolutions, and is a past president cancer relapse: a systematic evidence-based review. Cancers (Basel)
of the ISSPP. SG reports honoraria from AstraZeneca, MSD, GSK, and 2021; 14: 172.
M-Eden. CC reports grants or contracts from MSD, Pfizer, Seagen, and 13 Bouchard-Fortier G, Cusimano MC, Fazelzad R, et al. Oncologic
Daiichi Sankyo. LdG reports honoraria from GSK. BA reports consulting outcomes and morbidity following heated intraperitoneal
chemotherapy at cytoreductive surgery for primary epithelial
fees from Roche, AstraZeneca, and Servier, and advisory board
ovarian cancer: a systematic review and meta-analysis.
participation for Gilead, Daiichi, Pierre Fabre, and BMS. OG reports Gynecol Oncol 2020; 158: 218–28.
consulting fees from Gamida, is president of the BIG-RENAPE group,
14 Harter P, Bogner G, Chiva L, et al. Statement of the AGO
and is national coordinator of the RENAPE group. All other authors Kommission Ovar, AGO Study Group, NOGGO, AGO Austria,
declare no competing interests. Swiss AGO, BGOG, CEEGOG, GEICO, and SFOG regarding the
Data sharing use of hyperthermic intraperitoneal chemotherapy (HIPEC) in
From the publication date, UNICANCER will share de-identified epithelial ovarian cancer. Bull Cancer 2024; 111: 277–84.
individual participant data that underlie the results reported under the 15 Di Giorgio A, Naticchioni E, Biacchi D, et al. Cytoreductive surgery
(peritonectomy procedures) combined with hyperthermic
following conditions: the data shared will be limited to those required
intraperitoneal chemotherapy (HIPEC) in the treatment of diffuse
for independent mandated verification of the published results, the
peritoneal carcinomatosis from ovarian cancer. Cancer 2008;
reviewer will need authorisation from UNICANCER for personal access, 113: 315–25.
and data will be transferred only after signing a data access agreement.
16 Lim MC, Kang S, Choi J, et al. Hyperthermic intraperitoneal
A decision concerning the sharing of other study documents will be chemotherapy after extensive cytoreductive surgery in patients with
examined upon request. UNICANCER will consider access to study data primary advanced epithelial ovarian cancer: interim analysis of a
upon written detailed request sent to dpo@[Link]. phase II study. Ann Surg Oncol 2009; 16: 993–1000.
Acknowledgments 17 Parmar MK, Ledermann JA, Colombo N, et al. Paclitaxel plus
This study was conducted with the support of the French National platinum-based chemotherapy versus conventional platinum-based
chemotherapy in women with relapsed ovarian cancer: the ICON4/
Cancer Institute and French League Against Cancer (PHRC10_20-14).
AGO-OVAR-2.2 trial. Lancet 2003; 361: 2099–106.
We thank all patients and their families, the study investigators (listed in
18 Harter P, Sehouli J, Reuss A, et al. Prospective validation study of a
appendix p 1), staff from the participating centres, and members of the
predictive score for operability of recurrent ovarian cancer: the
Independent Data Monitoring Committee (Jean-Pierre LeFranc, Multicenter Intergroup Study DESKTOP II. A project of the AGO
Hugues Bourgeois, Nadine Dohollou, Hélène Simon, Catherine Uzan, Kommission OVAR, AGO Study Group, NOGGO, AGO-Austria,
Alain Lortholary, and Thibault de la Motte Rouge). We also acknowledge and MITO. Int J Gynecol Cancer 2011; 21: 289–95.
Jennifer Kelly (Medi-Kelsey, Ashbourne, UK), for medical writing 19 Menzies AV, Usher EC, Hsu FC, Levine EA, Lentz SS, Kelly MG.
assistance, funded by an unrestricted grant from UNICANCER. HIPEC after neoadjuvant chemotherapy is associated with
References acceptable toxicity and favorable quality of life in newly diagnosed
1 Colombo N, Sessa C, du Bois A, et al. ESMO-ESGO consensus advanced ovarian cancer patients. Gynecol Oncol 2022;
conference recommendations on ovarian cancer: pathology and 167: 234–38.
molecular biology, early and advanced stages, borderline tumours 20 Tian WJ, Chi DS, Sehouli J, et al. A risk model for secondary
and recurrent disease. Ann Oncol 2019; 30: 672–705. cytoreductive surgery in recurrent ovarian cancer: an evidence-
2 González-Martín A, Harter P, Leary A, et al. Newly diagnosed and based proposal for patient selection. Ann Surg Oncol 2012;
relapsed epithelial ovarian cancer: ESMO Clinical Practice 19: 597–604.
Guideline for diagnosis, treatment and follow-up. Ann Oncol 2023; 21 Harter P, Sehouli J, Vergote I, et al. Randomized trial of
34: 833–48. cytoreductive surgery for relapsed ovarian cancer. N Engl J Med
3 Hoppenot C, Eckert MA, Tienda SM, Lengyel E. Who are the long- 2021; 385: 2123–31.
term survivors of high grade serous ovarian cancer? Gynecol Oncol 22 Coleman RL, Spirtos NM, Enserro D, et al. Secondary surgical
2018; 148: 204–12. cytoreduction for recurrent ovarian cancer. N Engl J Med 2019;
4 Tanner EJ, Black DR, Zivanovic O, et al. Patterns of first recurrence 381: 1929–39.
following adjuvant intraperitoneal chemotherapy for stage IIIC 23 Shi T, Zhu J, Feng Y, et al. Secondary cytoreduction followed by
ovarian cancer. Gynecol Oncol 2012; 124: 59–62. chemotherapy versus chemotherapy alone in platinum-sensitive
5 Rose PG, Piver MS, Tsukada Y, Lau TS. Metastatic patterns in relapsed ovarian cancer (SOC-1): a multicentre, open-label,
histologic variants of ovarian cancer. An autopsy study. Cancer 1989; randomised, phase 3 trial. Lancet Oncol 2021; 22: 439–49.
64: 1508–13. 24 Azzalini E, Stanta G, Canzonieri V, Bonin S. Overview of tumor
6 Walker JL, Brady MF, Wenzel L, et al. Randomized trial of heterogeneity in high-grade serous ovarian cancers. Int J Mol Sci
intravenous versus intraperitoneal chemotherapy plus bevacizumab 2023; 24: 15077.
in advanced ovarian carcinoma: an NRG Oncology/Gynecologic 25 Arnold M, Rutherford MJ, Bardot A, et al. Progress in cancer
Oncology Group study. J Clin Oncol 2019; 37: 1380–90. survival, mortality, and incidence in seven high-income countries
7 Jaaback K, Johnson N, Lawrie TA. Intraperitoneal chemotherapy for 1995-2014 (ICBP SURVMARK-2): a population-based study.
the initial management of primary epithelial ovarian cancer. Lancet Oncol 2019; 20: 1493–505.
Cochrane Database Syst Rev 2016; 2016: CD005340.

[Link]/oncology Published online November 13, 2024 [Link] 11

 Articles

26 Tseng JH, Cowan RA, Afonso AM, et al. Perioperative epidural use 29 You B, Robelin P, Tod M, et al. CA-125 ELIMination rate constant K
and survival outcomes in patients undergoing primary debulking (KELIM) is a marker of chemosensitivity in patients with ovarian
surgery for advanced ovarian cancer. Gynecol Oncol 2018; 151: 287–93. cancer: results from the phase II CHIVA trial. Clin Cancer Res 2020;
27 Yang D, Zhang Y, Gong P, et al. Secondary cytoreductive surgery 26: 4625–32.
followed by olaparib tablets as maintenance therapy in patients with
BRCA mutated recurrent ovarian cancer: a multi-center
retrospective study. Eur J Surg Oncol 2024; 50: 107950.
28 du Bois A, Reuss A, Pujade-Lauraine E, Harter P, Ray-Coquard I,
Pfisterer J. Role of surgical outcome as prognostic factor in
advanced epithelial ovarian cancer: a combined exploratory analysis
of 3 prospectively randomized phase 3 multicenter trials: by the
Arbeitsgemeinschaft Gynaekologische Onkologie Studiengruppe
Ovarialkarzinom (AGO-OVAR) and the Groupe d’Investigateurs
Nationaux Pour les Etudes des Cancers de l’Ovaire (GINECO).
Cancer 2009; 115: 1234–44.

12 [Link]/oncology Published online November 13, 2024 [Link]