Intraperitoneal mitomycin C improves survival compared to cytoreductive surgery alone in an experimental model of high-grade pseudomyxoma peritonei

Sixty-five locally bred Rowett nude male rats with median weight 209 g (160–258) were used. The animals were maintained under specific pathogen-free conditions and food and water were supplied ad libitum. Housing and all procedures involving the animals were performed in accordance with protocols approved by the Animal Care and Use Committee (Approval ID #3158). Three groups of rats were defined by treatment regimens combining the anti-tumor treatment modalities CRS, IPEC and hyperthermia: The cytoreductive surgery group (CRSG), CRS without IPEC; the normothermic group (NG), CRS and IPEC at 35 ºC which is equal to the body core temperature of the rats; the hyperthermic group (HG), CRS and IPEC at 41 ºC which is a temperature commonly used in HIPEC in patients.

The patient-derived xenograft model was originally classified as peritoneal mucinous carcinomatosis (PMCA) according to Ronnett's classification [12]. The PMCA-3 orthotopic tumor model was previously characterized with respect to growth pattern in mice, microscopic features and immunohistochemical profile [13]. By the Peritoneal Surface Oncology Group International classification [1], the PMCA-3 model is a high-grade PMP with signet ring cell differentiation. Mucinous tumor tissue was harvested from the peritoneal cavity of locally bred female BALB/c (nu/nu) mice and 500 µl was immediately injected into the peritoneal cavity of 4–5 week-old rats. The treatment procedures of CRS and IPEC were performed after median 21 days (19–24).

All treatment procedures were conducted under sterile conditions. General anesthesia was initialized by isoflurane and maintained by subcutaneous injections of a mixture of tiletamine, zolazepam, xylazine and butorphanol [14]. A subcutaneous injection of dalteparine 70 IU/kg (Fragmin, Pfizer, Limoges, France) was administered immediately before the treatment procedure. A midline incision allowed examination of the peritoneal cavity and assessment of tumor distribution. The macroscopic appearance of the tumor model was similar to that in patients suffering from PMP with free mucin and solid lesions of soft consistency, most frequently located on the larger omentum, splenic surface and splenic hilum, liver surface and liver hilum, gonadal fat pads and parietal peritoneum. Peritoneal cancer index (PCI) was determined in seven locations by lesion size: PCI = 0, no tumor; PCI = 1, tumor < 2 mm; PCI = 3, tumor > 2 mm. Overall PCI was calculated as the sum of PCI in the individual locations as outlined in Table 1 with a maximum score of 21, modified from calculation of PCI in animal experiments previously described in [15, 16]. Free mucin was collected using a syringe. During CRS, a bipolar coagulation device was used for resection of peritoneal tumor and organs. The aim of CRS was complete removal of all tumors on the peritoneal surface, but diffuse tumor distribution, including in liver hilum, prevented complete resection of tumor lesions. Intestinal resections were not performed. Residual tumor lesions with a diameter up to 2.5 mm were accepted, which is in line with what is accepted in the treatment of patients with PMP [7]. The weight of mucin and solid lesions removed at surgery was registered. After CRS, the peritoneal cavity was rinsed with saline 0.9% and completeness of cytoreduction (CC) score was determined according to the maximum size of residual tumor: CC = 0, no tumor; CC = 1, tumor < 2.5 mm; CC = 2, tumor > 2.5 mm [6]. Thereafter, the abdominal wall of the animals in the CRSG was closed with two layers with continuous polyfilament suture 3 − 0. The animals in the NG and the HG were prepared for IPEC and placed in a custom-made rack with the abdominal wall elevated to imitate the coliseum position [17‐19].

The peritoneal perfusion system consisted of silicon tubes, an intravenous infusion tube and a container holding the peritoneal perfusion fluid (PPF). A peristaltic pump provided a perfusion flow rate of 40 ml/min and heating of the PPF was achieved by inserting the inflow tube into a temperature regulated water bath, as previously described [17]. Peritoneal perfusion was initiated with 150 ml saline 0.9%. When the desired temperature was reached, a single dose of MMC 1.5 mg (Medac, Hamburg, Germany) was added, whereupon IPEC was conducted for 90 min with an open peritoneal cavity. The mean inflow and outflow temperatures were registered at 10-min intervals by a dual input digital thermometer (TMD90, Wavetek Meterman, Long Branch, New Jersey, USA) calibrated at The National Institute of Technology (Oslo, Norway). In parallel, the rectal temperature was registered using a digital patient thermometer. The MMC concentration in the PPF was monitored by collecting 20-µl samples at 10, 30, 60 and 90 min. Samples were diluted 1:10 and stored at -70 ºC until analysis by high performance liquid chromatography [14]. In both groups the positions of the inflow and outflow catheters were reversed after 45 min, primarily in order to reduce the risk of thermal injury to the tissue exposed to the slightly higher temperature of the PPF at the inflow site in the HG but also to provide a more uniform intraperitoneal distribution of MMC. In the HG, ice and water were used for external cooling of the animals during IPEC to prevent body core overheating. After IPEC, the peritoneal cavity was rinsed with 250 ml saline 0.9% at temperature 35 ºC, whereupon the abdominal wall was closed as described above.

Five animals died during or immediately after treatment; three animals (one in each treatment group) died of bleeding. Two additional animals died in the HG, probably because of overheating. Survival analyses are presented for the animals that completed the treatment procedures (20 animals in each treatment group); including all animals based on intention-to-treat did not influence conclusions. All surgical procedures and administration of IPEC were conducted by one investigator who has extensive experience in clinical CRS and HIPEC (O.S.).

For postoperative analgesia, carprofen (Rimadyl^®, Orion Pharma, Animal Health, Oslo, Norway) 5 mg/kg was administered subcutaneously once daily on days 0–3. Body weight was registered daily during the first week. Experienced animal research technicians that were blinded for treatment group allocation monitored animal welfare and decided when the individual rats were to be terminated. Humane end points were abdominal distention as a sign of tumor and ascites, signs of pain, inactivity, malnutrition and jaundice. Animals were sacrificed under general anesthesia (isoflurane) by intracardial injection of pentobarbital. Rats with no sign of intraabdominal tumor were terminated on day 110. Autopsy was performed in all animals and the weight of ascites and solid tumor was registered.

Statistical analyses were performed using the Statistical Package for the Social Sciences^® program, version 18.0 (SPSS GmBH, Chicago, Illinois, USA). Comparisons of non-categorical parameters [overall PCI, body weight, relative weight loss, amount of solid tumor, mucin and ascites, and area under the time-concentration curve (AUC)] were conducted by two-sided Mann–Whitney test with exact significance and Kruskal–Wallis test with asymptotic significance as appropriate. Comparison of one categorical parameter (splenectomy yes/no) was conducted by Chi-square test, linear-by-linear association. Survival curves were calculated from the date of CRS and IPEC using the Kaplan–Meier product-limit method and differences between groups were analyzed using the log rank test. P values < 0.05 were considered significant.