Cytoreductive surgery combined with hyperthermic intraperitoneal chemotherapy for the treatment of primary peritoneal serous carcinoma: Results of a Chinese retrospective study

Abstract

Purpose Primary peritoneal serous carcinoma (PPSC) is a rare condition with a poor survival rate, even after treatment with debulking surgery followed by systemic chemotherapy. This study evaluated the efficacy and safety of cytoreductive surgery (CRS) combined with hyperthermic intraperitoneal chemotherapy (HIPEC) for the treatment of PPSC.

Patients and methods This retrospective study included 22 female patients with primary advanced PPSC (group A, n = 12) or recurrent PPSC (group B, n = 10) treated with 25 CRS + HIPEC procedures. The primary end point was overall survival (OS), and the secondary end points were safety profiles.

Results A total of 25 CRS + HIPEC procedures were performed in these 22 patients. The median OS was 31.0 months (95% confidence interval (CI) 22.3–39.7), and the 1-, 3-, and 5-year survival rates were 100%, 45.5%, and 27.3%, respectively. Subgroup analyses revealed that the median OS was 31.0 months (95% CI 19.8–42.2) for group A vs. 38.5 months (95% CI 9.6–67.4) for group B (P = 0.832, log rank test); 51.5 months (95% CI 34.9–68.1) for peritoneal cancer index (PCI) ≤ 15 vs. 20.3 months (95% CI 12.6–28.0) for PCI > 15 (P = 0.000, log rank test); and 38.5 months (95% CI 22.5–54.5) for completeness of cytoreduction (CC) of 0–1 vs. 23.5 months (95% CI 15.3–31.7) for CC of 2–3 (P = 0.178, log rank test). There were no perioperative deaths. Serious adverse events (SAEs) occurred in two patients (9.1%). A univariate analysis identified PCI ≤ 15 as the only prognostic predicator (hazard ratio (HR) 13.1, 95% CI 2.7–63.4, P = 0.001).

Conclusions CRS + HIPEC could contribute to favourable outcomes for select PPSC patients with acceptable safety profiles.

Keywords:

• Cytoreductive surgery
• hyperthermic intraperitoneal chemotherapy
• primary peritoneal serous carcinoma

Introduction

Swerdlow first described primary peritoneal serous carcinoma (PPSC) in 1959 [1]. PPSC is a rare and heterogeneous tumour that is diagnosed at approximately one tenth the frequency of epithelial ovarian cancer [2]. PPSC is an extra-ovarian malignancy that spreads widely inside the peritoneal cavity, primarily in the omentum and peritoneal surface with intact ovaries or minimal ovarian involvement, and is found predominantly in elderly and post-menopausal women [3,4]. This malignancy is also diagnosed in women many years after an oophorectomy for benign diseases or a prophylactic oophorectomy in women with a hereditary susceptibility to ovarian cancer.

PPSC is histologically, clinically and molecularly similar to stage III/IV ovarian serous carcinoma [5–7], which arises primarily from the peritoneal surface epithelium. The epithelial layers of the ovary and peritoneum share a common embryonic heritage and are derived from the coelomic epithelium during embryonic life [8]. Currently there is no standard treatment for PPSC, but it is traditionally managed in a similar way to stage III/IV ovarian cancer, including maximal surgical debulking of the peritoneal deposits followed by a platinum/taxane-based chemotherapy regimen. The survival rate of patients with PPSC is similar to that in studies by Bloss et al. and Dubernard et al. [4,9] or 2 to 6 months less than that of ovarian cancer patients [10] despite the similarity of treatment with cytoreduction and adjuvant systemic chemotherapy. However, the prognosis for low-grade serous carcinoma of peritoneum (LGSPC) may be better than for low-grade serous carcinoma of the ovary (LGSOC) [11].

A comprehensive treatment strategy that integrates aggressive cytoreductive surgery (CRS) to remove the macroscopic tumour and hyperthermic intraperitoneal chemotherapy (HIPEC) to eradicate the microscopic residual disease has been extensively investigated over the last three decades, and this strategy is the standard treatment for pseudomyxoma peritonei [12], malignant peritoneal mesothelioma [13], and colorectal peritoneal carcinomatosis [14]. This strategy also has survival advantages in patients with limited peritoneal carcinomatosis of gastric [15,16] and pancreatic origin [17]. It was adopted in China for the treatment of gastrointestinal and gynaecological malignancies [16,18,19], and several preclinical studies have been conducted [20,21]. The procedure has also been used to treat selected patients with PPSC. This retrospective study summarises our experience with 22 PPSC patients treated with 25 CRS + HIPEC procedures.

Patients and methods

Selection of patients

This study included 22 consecutive Chinese female patients with primary advanced PPSC (group A, n = 12) and recurrent PPSC (group B, n = 10) between 43 and 70 years old (median, 61) who were treated with 25 CRS + HIPEC procedures from April 2007 to April 2014. The diagnosis of PPSC was based on the recommendations of the Gynecologic Oncology Group [22]: (1) the ovaries were absent, normal in size or enlarged via a benign process, (2) the involvement of extra-ovarian sites was greater than on the ovarian surface of either ovary, (3) the ovaries were not involved with the tumour or exhibited only serosal or cortical implants less than 5 × 5 mm microscopically, and (4) the histopathological and cytological characteristics of the tumour were predominantly of the serous type.

All patients signed an informed consent form before treatment, and the institutional review board approved the study protocol. Routine preoperative studies included assessments of performance status; a physical examination; and routine blood and serum biochemistry tests, including liver and renal function evaluations and coagulation studies. Additional examinations included a chest X-ray, contrast-enhanced three-dimensional abdominal-pelvic computed tomography (CT), oral Gastrografin radiography, pulmonary and cardiac function assessments, and an ascites cytology examination.

The inclusion criteria were as follows: (1) between 20 and 75 years of age, (2) Karnofsky performance status score >50, (3) life expectancy >12 weeks, (4) blood profile and serum biochemical results within normal ranges, and (5) cardiovascular, pulmonary and other major organ functions capable of withstanding major surgery. The following primary exclusion criteria were used: (1) age <20 or >75 years, (2) any lung, liver, or prominent retroperitoneal lymph node metastasis identified during preoperative assessment, (3) major organ functions that could not withstand this comprehensive procedure, or (4) a pathological study that did not meet the PPSC criteria.

CRS + HIPEC procedure

The patient was placed in the supine or modified lithotomy position with the gluteal fold at the end of the operating table to allow full access to the perineum during the surgical procedure. Pressurised inflatable insulating protective bags were used to prevent deep venous thrombosis of the lower extremities. The abdominal cavity was opened via a midline incision from the xiphoid to the pubis under general anaesthesia and haemodynamic monitoring. Generous abdominal exposure was essential for detailed abdominal exploration and evaluation of the peritoneal cancer index (PCI), considering the characteristics of ascites and the size and distribution of disseminated tumour nodules according to the principle of Sugarbaker [23]. Optimal CRS was performed to remove the entire peritoneum with tumour involvement and any involved adjacent tissues and organs, following the peritonectomy procedure developed by Sugarbaker [24]. All adhesions were separated or resected if possible in case tumour cells were trapped within the scar tissue. The goal of CRS was the removal of the entire visible tumour via extensive organ resection or peritonectomy. All surgical specimens underwent a pathological examination. Unresectable tumours were cauterised using a ball-tipped electrosurgical instrument at maximal electric power, particularly on the edge of tumour nodules. High-voltage electrosurgery leaves a margin of heat necrosis that is devoid of viable malignant cells. The completeness of CRS (CC) [25] was evaluated before HIPEC, which was implemented using the open Coliseum technique. Cisplatin (120 mg) and docetaxel (120 mg) were administered in 10 cases, cisplatin (120 mg) and mitomycin C (MMC, 30 mg) were administered in six cases, and paclitaxel (120 mg) and lobaplatin (100 mg) were administered in six cases. Each chemotherapy agent was dissolved in 6 L of heated saline. The heated perfusion solution was introduced using an automatic hyperthermia chemotherapy perfusion device (ES-6001, Wuhan E-sea Digital Engineering, Wuhan, China) and infused into the peritoneal cavity at a rate of 500 mL/min through an inflow tube placed under the diaphragm. An outflow tube for perfusion was placed in the Douglas pouch immediately prior to HIPEC. The skin of the abdomen was attached to a retractor ring, and a plastic sheet covered the open surgical field to maintain a stable temperature and protect the surgeon from aerosol exposure. The perfusion solution in the peritoneal cavity was carefully stirred manually to ensure an even distribution of the heated drugs. The temperature of the perfusion solution in the peritoneal cavity was maintained at 43.0 ± 0.5 °C and monitored using a thermometer in real time. The total HIPEC time was 60–90 min, after which the perfusion solution in the abdominal cavity was removed using the suction tube. Gastrointestinal anastomoses or stomata were performed after HIPEC, and drainage tubes were placed. A subcutaneous intraperitoneal chemotherapy port was implanted, and the abdominal wound was closed using a double-layer relaxing suture. The patient was transported to the intensive care unit for post-operative recovery and then to the surgical ward when her condition stabilised.

The extent of CRS was determined using Sugarbaker’s criteria [25] on the completeness of cytoreduction (CC). A score of CC-0 indicated no residual peritoneal disease after CRS, CC-1 indicated less than 2.5 mm of residual disease, CC-2 indicated a residual tumour between 2.5 mm and 2.5 cm, and CC-3 indicates more than 2.5 cm of residual tumour or the presence of a sheet of unresectable tumour nodules.

Study end points and definitions

The primary end point of this study was disease-specific overall survival (OS), which was defined as the time interval from the first treatment to death due to disease or to the last follow-up. (Figure 1). The secondary end points were perioperative serious adverse events (SAEs), which were defined as complications within 30 days of CRS + HIPEC that were directly attributable to the treatment; this definition was based on the NCI Common Terminology Criteria for Adverse Events (CTCAE) version 4.0 [26]. Perioperative death was defined as death within 30 days of CRS + HIPEC and directly attributed to the procedure.

Figure 1. Flow chart of the study procedure. Patients with primary peritoneal serous carcinoma (PPSC) were included in this study and were divided into group A and group B. Patients in group A were treated with CRS + HIPEC at first hospitalization or after 2 cycles of neoadjuvant chemotherapy. Patients in group B were first treated with conventional surgery and post-operative adjuvant chemotherapy. CRS + HIPEC was performed at the time of first recurrence.

Post-operative chemotherapy

Post-operative adjuvant chemotherapy included platinum/taxane-based systemic chemotherapy and perioperative normothermic intraperitoneal chemotherapy (PIC) through the intraperitoneal chemotherapy port every 3–4 weeks. Chemotherapy was initiated 3–4 weeks after CRS + HIPEC.

Follow-up

All of the patients were regularly followed up once every 3 months for the first 2 years, once every 6 months for the third year, and once annually thereafter for detailed monitoring of disease status. Patients who were alive at the time of the most recent analysis were censored at the last follow-up. The follow-up examination protocol included clinical, biochemical, and tumour markers and radiological examinations. The last follow-up occurred on 1 May 2015, and the overall follow-up rate was 100%.

Statistical analysis

Patient information was systematically collected to establish a comprehensive database. Data analysis was conducted using the Statistical Package for Social Sciences version 17.0 (SPSS, IBM, Armonk, NY). Numerical data were directly recorded, and categorical data were recorded in different categories. The Kaplan–Meier method with log-rank test was used to compare survival rates. Subgroup OS comparisons were further stratified by treatment timing (primary advanced PPSC vs. recurrent PPSC), PCI (PCI ≤ 15 vs. PCI > 15), and CC (CC 0–1 vs. CC 2–3). A 2-sided P < 0.05 value was considered statistically significant.

Results

Primary clinicopathological characteristics of the 22 patients

A total of 22 consecutive Chinese female PPSC patients from April 2007 to April 2014 were included in this retrospective study. These 22 patients included primary advanced PPSC (group A, n = 12) and recurrent PPSC (group B, n = 10) patients who were treated with 25 CRS + HIPEC procedures. Six patients in group A received neoadjuvant intravenous or intraperitoneal chemotherapy before CRS + HIPEC, including two patients who were treated with two cycles of neoadjuvant bidirectional intravenous and intraperitoneal chemotherapy with intravenous (IV) cisplatin (100 mg/m²) and paclitaxel (175 mg/m²) and intraperitoneal (IP) MMC (20 mg/m²); one patient who was treated with two cycles of neoadjuvant intraperitoneal chemotherapy with lobaplatin (50 mg/m²) and paclitaxel (175 mg/m²); and three patients who were treated with two cycles of systemic chemotherapy with carboplatin (300 mg/m²) and docetaxel (75 mg/m²). The other six patients underwent CRS + HIPEC as the first-line treatment. Major surgeries for the 12 patients in group A included total abdominal hysterectomy (TAH) and bilateral salpingo-oophorectomy (BSO) in four patients, BSO in six patients, and abdominal visceral organ resection with peritonectomy without TAH + BSO in two patients.

All 10 patients in group B underwent conventional CRS as the first-line treatment, with TAH and BSO as the major surgical modalities. Seven patients received a median of nine (range 4–10) cycles of post-operative intraperitoneal or intravenous chemotherapy. The median time from first treatment to recurrence was 13.8 months (range 3.0–38.5).

Fifteen of the 22 patients received systemic chemotherapy after CRS + HIPEC. Sixteen patients received intraperitoneal chemotherapy, and 12 patients received bidirectional intravenous and intraperitoneal chemotherapy. Three patients did not receive post-operative chemotherapy. Post-operative chemotherapy was initiated 3–4 weeks after CRS + HIPEC when the patient’s condition was stable. The median number of adjuvant chemotherapy cycles (SC and/or PIC) was seven (range 0–26).

The median preoperative CA125 level was 387.9 U/mL (range 7.8–4881.0), which fell to 213.8 U/mL (range 12.6–1970.1) immediately after CRS + HIPEC. Table 1 lists the detailed clinicopathological characteristics.

Table 1. Major clinico-pathologic characteristics of the 22 PPSC patients^a.

Items	Value n (%)
Age in years
Median age (range)	61 (43–70)
<60	10 (45.5)
≥60	12 (54.5)
Median KPS score (range)	70 (60–90)
CRS + HIPEC timing
Upfront	6 (27.3)
Interval	6 (27.3)
Salvage	10 (45.4)
Histopathology differentiation
Well/moderately differentiated	12 (54.5)
Poorly differentiated	10 (45.5)
SAE
Yes	2 (9.1)
No	20 (90.9)
Post-operative chemotherapy cycles
Median post-operative chemotherapy cycles (range)	7 (0–26)
<6	7 (31.8)
≥6	15 (68.2)
Post-operative intraperitoneal chemotherapy cycles	4 (0–7)
Median hospital stay in days (range)	18 (12–39)
Median post-operative hospital stay in days (range)	11 (7–30)
Median ICU stay in h (range)	17 (0–51)
Median gastric tube removal time in days (range)	5 (3–10)
Median follow-up in months (range)	45.9 (13.5–106.3)
CA125 level (median, range) in U/mL
Preoperative	387.9 (7.8–4881.0)
Post-operative	213.8 (12.6–1970.1)
PCI
Median PCI (range)	14 (7–27)
≤15	12 (54.5)
>15	10 (45.5)
CC scores
0–1	15 (68.2)
2–3	7 (31.8)
Number of organs resected
0–1 resection	15 (68.2)
2–4 resections	7 (31.8)
Peritoneal regions resected
1–3 regions	7 (31.8)
4–6 regions	8 (36.4)
7–9 regions	7 (31.8)
Median fluid output volume at surgery in mL (range)
Blood loss	650 (150–1200)
Urine output	1000 (450–3350)
Ascites	1250 (0–4000)
≤1000	10 (45.5)
>1000	12 (54.5)
Median fluid intake volume at surgery (range)
Plasma in mL	500 (0–1550)
RBC (U)^b	3 (0–7)
Cryoprecipitation (U)^c	4 (0–8)
Other fluids (ml)	4000 (2000–7500)
CRS + HIPEC duration (median, range) in h	7.8 (4.5–11.7)
HIPEC duration in min	60–90

^aFor the first CRS + HIPEC of the 22 patients; ^b1U = 200 ml; ^c1U = 25 ml.

KPS, Karnofsky performance status; CRS, cytoreductive surgery; HIPEC, hyperthermic intraperitoneal chemotherapy; SAE, serious adverse events; PCI, peritoneal cancer index; CC, completeness of cytoreduction.

Survival of the total population

Eighteen patients (81.8%) had died and four patients (18.2%) were alive at the median follow-up of 45.9 months (range 13.5–106.3), which rendered the data mature for final analysis. The median OS was 31.0 months (95% CI 22.3–39.7) for the entire group, and the 1-, 3-, and 5-year survival rates were 100%, 45.5%, and 27.3%, respectively (Figure 2A).

Figure 2. The Kaplan–Meier overall survival curve of the patients in this study (A). The Kaplan–Meier overall survival curve by treatment time from first treatment (B), by PCI (C) and by CC (D) for primary advanced and recurrent PPSC patients. mo, months.

The OS comparisons were further stratified based on major clinicopathological factors. The median OS in terms of HIPEC timing was 31.0 months (95% CI 19.8–42.2) for group A vs. 38.5 months (95% CI 9.6–67.4) for group B (P = 0.832, log rank test) (Figure 2B). The median OS was 51.5 months (95% CI 34.9–68.1) for patients with PCI ≤ 15 vs. 20.3 months (95% CI 12.6–28.0) for patients with PCI > 15 (P = 0.000, log rank test) (Figure 2C). The median OS was 38.5 months (95% CI 22.5–54.5) for patients with CC 0-1 vs. 23.5 months (95% CI 15.3–31.7) for patients with CC 2–3 (P = 0.178, log rank test) (Figure 2D).

A univariate analysis identified PCI ≤ 15 as the only prognosis predicator (HR = 13.1, 95% CI 2.7–63.4, P = 0.001). Table 2 lists the complete clinicopathological characteristics, treatment course and survival information on the 22 patients.

Table 2. Major clinico-pathological parameters, treatment course and outcome information of the 22 PPSC^a.

No.	Age (years)	P/R	Histopathology	CRS^b	HIPEC	PCI	CC	Survival (months)	Comments
1	64	R	Serous adenocarcinoma, poorly differentiated	Resection of right diaphragmatic copula peritoneum, left/right paracolic sulci peritoneum, pelvic peritoneum, ileectomy, resection of ileocecum	Lobaplatin 100 mg, PTX 120 mg, 43 °C, 60 min	12	1	47.0, SWT
2	61	P	Serous papillary adenocarcinoma, poorly differentiated	Resection of right diaphragmatic copula peritoneum, right paracolic sulci peritoneum, pelvic peritoneum, appendectomy, rectectomy and TAH + BSO	Lobaplatin 100 mg, PTX 120 mg, 43 °C, 60 min	7	0	13.8, D
3	66	R	Serous papillary adenocarcinoma, moderately differentiated	Resection of right diaphragmatic copula peritoneum, left/right paracolic sulci peritoneum, pelvic peritoneum	Lobaplatin 100 mg, PTX 120 mg, 43 °C, 60 min	25	2	20.3, D
4	56	R	Serous cystadenocarcinoma, well differentiated	Resection of left/right paracolic sulci peritoneum, pelvic peritoneum	DDP 120 mg, MMC 30 mg, 43 °C, 60 min	25	2	13.5, D
5	62	P	Serous papillary adenocarcinoma moderately differentiated	Resection of right diaphragmatic copula peritoneum, left/right paracolic sulci peritoneum, pelvic peritoneum	DDP 120 mg, MMC 30 mg, 43 °C, 90 min	8	1	76.5, SWT
6	62	P	Serous papillary adenocarcinoma, poorly differentiated	Resection of pelvic peritoneum, bilateral fossa iliac peritoneum and BSO	DDP 120 mg, MMC 30 mg, 43 °C, 90 min	9	2	74.0, D
7	52	P	Serous papillary adenocarcinoma, well differentiated	Resection of right/left diaphragmatic copula peritoneum, right/left paracolic sulci peritoneum, pelvic peritoneum, anterior wall peritoneum and BSO	DDP 120 mg, TXT 120 mg, 43 °C, 60 min	26	1	24.3, D
8	70	R	Serous papillary adenocarcinoma, poorly differentiated	Resection of right diaphragmatic copula peritoneum, left/right paracolic sulci, pelvic peritoneum and bilateral fossa iliac peritoneum, splenectomy	Lobaplatin 100 mg, PTX 120 mg, 43 °C, 60 min	13	1	51.5, SWOT
9	68	P	Serous cystadenocarcinoma, well differentiated	Resection of right diaphragmatic copula peritoneum, pelvic peritoneum, bilateral fossa iliac peritoneum, cholecystotomy and TAH + BSO	DDP 120 mg, MMC 30 mg, 43 °C, 90 min	11	0	41.6, D	SAE 1
10	46	P	Serous papillary adenocarcinoma, poorly differentiated	Resection of right diaphragmatic copula peritoneum, pelvic peritoneum and BSO	DDP 120 mg, TXT 120 mg, 43 °C, 60 min	17	1	31.0, D
11	66	P	Serous adenocarcinoma, poorly differentiated	Resection of right diaphragmatic copula peritoneum, pelvic peritoneum, left hemicolectomy and BSO	DDP 120 mg, TXT 120 mg, 43 °C, 60 min	14	1	31.0, D
12	70	P	Serous papillary adenocarcinoma, well differentiated	Resection of right/left diaphragmatic copula peritoneum, right/left paracolic sulci peritoneum, pelvic peritoneum and BSO	DDP 120 mg, TXT 120 mg, 43 °C, 60 min	16	1	15.0, D
13	50	R	Serous papillary adenocarcinoma, poorly differentiated	Resection of right diaphragmatic copula peritoneum, left/right paracolic sulci peritoneum, pelvic peritoneum, anterior wall peritoneum, ileectomy	DDP 120 mg, TXT 120 mg, 43 °C, 90 min	11	1	65.3, D	Twice CRS + HIPEC
14	43	R	Serous adenocarcinoma, well differentiated	Resection of pelvic peritoneum, left/right paracolic sulci peritoneum, anterior wall peritoneum	DDP 120 mg, TXT 120 mg, 43 °C, 60 min	18	1	18.5, D
15	58	P	Serous adenocarcinoma, moderately differentiated	Resection of right/left diaphragmatic copula peritoneum, left/right paracolic sulci peritoneum, pelvic peritoneum and BSO	DDP 120 mg, MMC 30 mg, 43 °C, 60 min	20	3	23.5, D	SAE 2
16	55	R	Serous papillary adenocarcinoma, moderately differentiated	Resection of left/right paracolic sulci peritoneum, pelvic peritoneum	DDP 120 mg, TXT 120 mg, 43 °C, 60 min	22	3	28.5, D
17	61	P	Serous adenocarcinoma, well differentiated	Resection of pelvic peritoneum, bilateral fossa iliac peritoneum	DDP 120 mg, TXT 120 mg, 43 °C, 60 min	27	2	16.1, D
18	69	R	Serous adenocarcinoma, poorly differentiated	Transverse colectomy	DDP 120 mg, MMC 30 mg, 43 °C, 60 min	13	1	38.5, D	Twice CRS + HIPEC
19	43	R	Serous papillary adenocarcinoma, moderately differentiated	Resection of right diaphragmatic copula peritoneum, right paracolic sulci peritoneum	DDP 120 mg, TXT 120 mg, 43 °C, 60 min	12	1	13.5, SWT
20	44	R	Serous adenocarcinoma, poorly differentiated	Resection of right paracolic sulci peritoneum, pelvic peritoneum	DDP 120 mg, TXT 120 mg, 43 °C, 90 min	14	1	39.7, D
21	50	P	Serous adenocarcinoma, moderately differentiated	Resection of right diaphragmatic copula peritoneum, left/right paracolic sulci peritoneum, pelvic peritoneum, right hemicolectomy, rectectomy and TAH + BSO	Lobaplatin 100 mg, PTX 120 mg, 43 °C, 60 min	18	2	34.3, D
22	61	P	Serous papillary adenocarcinoma, poorly differentiated	Resection of right diaphragmatic copula peritoneum, left/right paracolic sulci peritoneum, pelvic peritoneum, right hemicolectomy, cholecystotomy and TAH + BSO	Lobaplatin 100 mg, PTX 120 mg, 43 °C, 60 min	10	1	40.4, D	Twice CRS + HIPEC

^aFor the first CRS + HIPEC of 22 patients; ^bAll 22 patients underwent liver round ligament resection, greater and lesser omentectomy and mesenteric fulguration during CRS.

PPSC, primary peritoneal serous carcinoma; PCI, peritoneal cancer index; CRS, cytoreductive surgery; HIPEC, hyperthermic intraperitoneal chemotherapy; CC, completeness of cytoreduction; R, recurrent; P, primary; TAH, total abdominal hysterectomy; BSO, bilateral salpingo-oophorectomy; SAE, serious adverse events; SWT, survival with tumour; SWOT, survival without tumour; D, dead; DDP, cisplatin; MMC, mitomycin C; TXT, docetaxel; PTX, paclitaxel; SAE 1, pneumonia and septicaemia; SAE 2, anastomotic haemorrhage.

Special analysis of four patients with long survival

By the last follow-up, four patients (18.2%) had survived more than 50 months from the first treatment. The first patient was a 62-year-old primary case with poorly differentiated serous papillary adenocarcinoma. She had a temporary implanted pacemaker because of a complete left bundle branch block before surgery. BSO was performed during CRS + HIPEC, with a PCI of nine and a CC of two. She experienced intermittent recurrence five times after CRS + HIPEC, but platinum/taxane-based systemic chemotherapy controlled the massive malignant ascites and increased the serum CA125 level at every treatment. The total number of chemotherapy treatments after CRS + HIPEC was 26. The patient died of tumour recurrence and massive upper digestive bleeding with an OS of 74.0 months. The second patient was a 70-year-old recurrent case with poorly differentiated serous papillary adenocarcinoma who first underwent TAH + BSO. She underwent CRS + HIPEC 30 months after the first surgery, with a PCI of 13 and a CC of one. She received four cycles of intraperitoneal chemotherapy and two cycles of intravenous chemotherapy after CRS + HIPEC. She is still living tumour-free with an OS of 51.5 months. The third patient was a 62-year-old primary advanced case who had moderately differentiated serous papillary adenocarcinoma. She was treated with two cycles of neoadjuvant intravenous and intraperitoneal bidirectional chemotherapy before CRS + HIPEC, followed by seven cycles of platinum/taxane-based intravenous chemotherapy and two cycles of intraperitoneal chemotherapy. She experienced four episodes of incomplete intestinal obstruction that were relieved with conservative treatment. She underwent hyperthermic intrathoracic chemotherapy with cisplatin (120 mg) and docetaxel (120 mg) 2 years after the first CRS + HIPEC because of pleural metastasis. She is still alive but with a tumour, with an OS longer than 76.5 months. The fourth patient was a 50-year-old recurrent case who had poorly differentiated serous papillary adenocarcinoma and underwent TAH + BS + O as the first surgery. She underwent two CRS + HIPEC procedures thereafter, with a PCI of 11 and a CC of one during the first CRS + HIPEC procedure. Six cycles of intraperitoneal chemotherapy were performed after the first procedure, and another CRS + HIPEC was performed 1 year later. The patient eventually died of liver metastasis with malignant obstructive jaundice, with an OS of 65.3 months.

Serious adverse events

There were no perioperative deaths. SAEs (grades 3–5) occurred in two patients (9.1%). The first patient developed pneumonia and septicaemia on post-operative day 3, with a body temperature of 39.5 °C and a peripheral blood WBC count of 25.5 × 10⁹/L. A sputum culture confirmed the infection as methicillin-resistant staphylococcus aureus sensitive to vancomycin. She gradually recovered 2 weeks later with the use of effective antibiotics, parenteral nutrition support, and albumin infusion, and she survived 41.6 months after the procedure. The second case developed anastomotic haemorrhage with massive bloody stool 12 h after surgery, which was stopped by a blood transfusion, fresh frozen plasma, precipitation and haemostatic agents. The patient did not fully recover and died from deep venous thrombosis and renal failure 3 months later.

Discussion

To the best of our knowledge, our study is the first report from China to present real-life clinical information on CRS + HIPEC for the treatment of PPSC. The median OS in our study was 31.0 months (95% CI 22.3–39.7), and the 1-, 3-, and 5-year survival rates were 100%, 45.5% and 27.3%, respectively. The study also demonstrated no statistically significant difference in OS between group A and group B, which suggests that CRS + HIPEC is equally effective in treating PPSC regardless of whether it is performed as the first treatment or at recurrence.

Several case studies, and small-sample retrospective studies [3,9,27–34], a single-institution large-sample retrospective study [11], and two prospective studies [4,35] have reported median OS ranging from 17.0–101.7 months for PPSC patients (Table 3). A systemic review by Pentheroudakis et al. [10] evaluated 25 clinical trials on PPSC in three time periods: before 1990, when platinum- and alkylator-based chemotherapy were considered standard, during the 1990s, when platinum combinations without taxanes were the standard treatment, and after 1995, when platinum/taxane combinations were the standard therapy. The OS of PPSC after 1995 was relatively longer. Bakrin et al. [31] conducted the only retrospective multi-institutional study of the treatment of PPSC using CRS + HIPEC in 36 patients from nine institutions, and the 1-, 3-, and 5-year OS rates were 93.6%, 71.5% and 57.4%, respectively, calculated from the date of surgery. All patients except one (96%) received a median of six cycles of neoadjuvant platinum-based chemotherapy. This treatment may downstage the disease with a median preoperative PCI of 10 (range 1–26). In contrast, only six patients in our cohort received two cycles of neoadjuvant chemotherapy, and the median PCI was 14 (range 7–27). This difference may be the primary cause of the relatively poorer survival in our study and may emphasize the importance of neoadjuvant chemotherapy. A study by Gershenson et al. [11] include the largest sample size and demonstrated the longest OS, with a median progression-free survival (PFS) of 28.1 months and a median OS of 101.7 months. Several key factors may contribute to the OS differences between this study and other studies. First, this study focused on 350 patients with LGSPC or LGSOC, which is one particular subtype of serous carcinoma that presents a relatively less lethal tumour with reduced biological behaviour. In contrast, other studies included both low- and high-grade serous carcinoma. Second, PFS and OS were calculated from the date of primary surgery (or date of tissue diagnosis) in the Gershenson study, but 90.6% of patients received PPSC treatment, including primary cytoreduction and all subsequent treatments, such as secondary surgery, chemotherapy, and hormone therapy, at one institution. In contrast, 45.5% of the patients in our study were referral patients with recurrent PPSC, and we did not have complete information on the primary surgery and other treatments received during the recurrence-free survival period. This incomplete clinical information may have partially influenced the OS in our study. Third, only 75 of 350 cases were LGSPC, and the other 275 cases were LGSOC. These tumours share many similarities in biological behaviour, but they are not identical in terms of survival impact. Most PPSC patients are diagnosed at advanced stages because of a lack of specific symptoms, and they are frequently misdiagnosed with ovarian cancer at first presentation [32]. The most common presenting symptoms of PPSC are abdominal distension, massive ascites and a marked rise in the serum CA125 level [27]. Massive ascites and a CA125 increase occurred in 21 of 22 patients (95.5%) in our study, with a median volume of 1250 mL and a median preoperative CA125 level of 387.9 U/mL. Therefore, the possibility of PPSC should be considered in post-menopausal women presenting with abdominal distension, massive ascites and an increased CA125 level.

Table 3. The overall survival information from different studies for primary peritoneal serious carcinoma.

										Survival rate (%)
References	Country	N	Median age	Study design	Treatment	Optimal surgery (%)	Median preoperative CA125 (U/mL)	Median OS (months)	Median PFS (months)	1 year	2 years	5 years
Gershenson et al. [11]	USA	350	46.5	Retrospective	CRS + IV CT + HT	NR	146.3	101.7	28.1	NR	NR	NR
Unal et al. [3]	Turkey	79	63.0	Retrospective	CRS + IV/IP CT	49.5^a	954.0	31.1	16.3	93.5	85.7	48.1
Ifthikar et al. [32]	India	10	56.0	Retrospective	CRS + IV CT	75.0^a	Mean 566.0	NR	22.0	NR	NR	18.0 (PFS)
Kawaguchi et al. [29]	Japan	22	67.0	Retrospective	NACT + CRS + IV CT	54.5^a	638.0	26.5	12.7	NR	NR	NR
Liu et al. [27]	China	22	56.0	Retrospective	NACT + CRS + IV CT	81.8^a	90% elevated	21.0	NR	NR	NR	34.4
Zhang et al. [30]	China	24	59.0	Retrospective	CRS + IV CT	12.5^b	95.8% elevated	42.0	NR	NR	63.0	NR
Bloss et al. [4]	USA	36	65.8	Prospective	CRS + IV CT	NR	NR	22.0	11.0	NR	NR	NR
Ben-Baruch et al. [33]	Israel	22	61.0	Retrospective	CRS + IV CT	28.0^b	NR	21.0	DFS 15.0	NR	NR	18.0
Piver et al. [35]	USA	46	62.0	Prospective	CRS + PAC IV CT 25; CRS + TP IV CT 21	PAC 60.0 TP 81.0^a	PAC 1350.0 TP 1280.0	23.1 TP 21.5 PAC 24.0	TP 17.3; PAC 24.0	NR	NR	NR
Halperin et al. [28]	Israel	28	59.8	Retrospective	CRS + IV CT	39.3^a	Mean 5284.0	17.0	NR	NR	NR	NR
Dubernard et al. [9]	France	37	60.0	Retrospective	NACT + CRS + IV CT	89.2^b	833.0	Not reached	3-year DFS 29%	NR	NR	NR
Eltabbakh et al. [34]	USA	75	65.0	Retrospective	CRS + IV CT	65.3^a	1320.0	23.5	NR	NR	NR	26.5
Bakrin et al. [31]	France	36	60.5	Retrospective	NACT + CRS + HIPEC	88.9^c	NR	Not reached	DFS: 16.7	93.6	NR	57.4
This study	China	22	61.0	Retrospective	NACT + CRS + HIPEC + IV/IP CT	68.2^c	387.9	31.0	NR	100	NR	31.8

Optimal surgery: ^aresidual tumour ≤ 1 cm; ^bresidual tumour ≤ 2 cm; ^cresidual tumour ≤ 0.25 cm.

OS, overall survival; PFS, progression-free survival; CRS, cytoreductive surgery; IV, intravenous; IP, intraperitoneal; CT, chemotherapy; HT, hormonal therapy; NACT, neoadjuvant chemotherapy; PAC, cisplatin, doxorubicin, and cyclophosphamide; TP, paclitaxel and cisplatin; NR, not reported.

Many published studies [31,36] have demonstrated that one of the most important factors in the success of CRS + HIPEC is the PCI, which reflects the dissemination of peritoneal carcinomatosis. The median PCI in our study was 14; 12 patients (54.5%) exhibited PCI ≤ 15, and 10 patients (45.5%) had PCI > 15. Patients with PCI ≤ 15 had a median OS of 51.5 months, but patients with a PCI > 15 had a median OS of 20.3 months (P = 0.000). A univariate analysis identified PCI ≤ 15 as the only independent factor for improved survival. Patients with a relatively high PCI benefit less from CRS + HIPEC. Therefore, attention should be paid to improving preoperative examinations to select appropriate patients for CRS + HIPEC. Our institution routinely relies on three major approaches for patient screening: the serum tumour markers of CEA, CA125 and CA19-9, a preoperative CT-PCI evaluation, and laparoscopic exploration. Patients with PCI > 15 should be excluded from CRS + HIPEC, and palliative treatment may be appropriate for these patients.

Many publications have reported that an optimal CRS to achieve CC 0–1 resection is one of the most important prognostic factors affecting survival. A retrospective study by Eltabbakh et al. [34] revealed that debulking surgery was an independent factor for improved survival. A retrospective multi-centre study of 79 PPSC patients from Turkey indicated that optimal CRS was a favourable prognostic factor [3]. Crane et al. [37] investigated the role of secondary CRS in low-grade serous ovarian cancer and peritoneal cancer and demonstrated that a complete CRS had a better PFS. The median PFS for patients with no gross residual disease after CRS was 60.3 months, compared to 10.7 months for patients with gross residual disease (P = 0.008). Fifteen of 22 patients (68.2%) in our study who reached a CC of 0–1 had a median OS of 38.5 months, and patients with a CC of 2–3 had a median OS of only 23.5 months. However, there was no statistically significant difference because of the small sample size. Therefore, every effort should be made to achieve maximal CRS. All 22 patients in our study underwent liver round ligament resection, greater and lesser omentectomy and mesenteric fulguration during CRS. Extensive surgical resection without adjuvant intraperitoneal chemotherapy may be associated with a poor prognosis. This association may be due to the disruption of anatomical planes, which leads to deep abdominal and pelvic dissemination that is intractable to further treatment [38]. Therefore, CRS to reduce visible tumour burden and HIPEC to eradicate micrometastases should be the preferred treatment modality for PPSC.

We performed a long midline incision from the xiphoid to the pubis for full exposure of the entire abdominal cavity. This technique differs from ovarian cancer debulking surgery, which is primarily performed using a lower middle abdominal incision. Extensive CRS means more blood and plasma loss, more gastrointestinal reconstruction, and a longer surgery duration, and the cytotoxic effects and complications of HIPEC increase during this procedure. There were no perioperative deaths in our study, but SAEs occurred in two patients (9.1%), including pneumonia and septicaemia in one patient and anastomotic haemorrhage in one patient. This SAE rate is relatively lower than in previous reports [31]. This difference is largely attributable to the accumulated experience of more than 10 years of concentrated clinical study, and we passed the learning curve for this technically demanding procedure on to other groups [16,18]. A designated team of a surgical oncologists, anaesthetists, and operating room staff performed all CRS + HIPEC procedures to minimise potential complications. All patients in our group required blood, plasma, cryoprecipitation transfusion, and effective antibiotics during and after surgery. We can now achieve rapid patient recovery with a median post-operative hospital stay of approximately 11 days (range 7–30) despite the extensive abdominal and pelvic resection and long surgery time.

Most patients with PPSC had elevated serum CA125 levels, which have been reported by others, with a median level from 566.0 to 1350.0 U/mL (Table 3). Our study is consistent with these reports, which showed a median level of 387.9 U/mL (range 7.8–4881.0). Altaras et al. [39] correlated CA125 values with the clinical status of PPSC, and the CA125 level was considered the most effective serum tumour marker for the diagnosis and follow-up of PPSC. Kawaguchi et al. [29] also indicated that survival was longer for patients whose CA125 levels normalised to 26 U/mL during and after treatment, and the CA125 nadir after CRS may be a prognostic factor for OS and PFS in patients with PPSC.

This study was a retrospective 7-year observational study of patients with a rare condition. The limitations of an observational study are well known and include incomplete data, a long study period, referral bias, inconsistent therapies, changes in detection methods, and other confounding factors [11]. The changes in perfusion time and chemotherapeutics were especially obvious in this study, and no standard approach was used. However, our work and similar studies on CRS + HIPEC for the treatment of limited peritoneal carcinomatosis in gastrointestinal [15,17,40] or gynaecological malignancies [41,42] demonstrated that CRS + HIPEC is safe and could have favourable outcomes for these patients.

Conclusion

Our study provided the first real-life clinical data from China to demonstrate that CRS + HIPEC could have favourable outcomes for select patients with PPSC with acceptable safety profiles. However, more high-level evidence-based clinical studies are needed to confirm and quantify its efficacy.

Disclosure statement

This work was supported by grants for Hubei Province’s Outstanding Medical Academic Leader Program and the Science Fund for Doctorate Mentors by China’s Ministry of Education (No. 20120141110042). The authors declare no conflicts of interest in this work. The authors alone are responsible for the content and writing of the paper.