Background
Peritoneal carcinosis can be a result of intraperitoneal tumor cell spread after surgical treatment of colonic cancer. Tumor cell attachment occurs through blood-or lymphatic vessels or by accidentally opening the colonic specimen [
1]. Serosal invasion of the primary tumor leads in up to 50% of patients to intraperitoneal metastases [
2]. However, even stage 1 and stage 2 of colonic cancer, gastric cancer, uterine cancer or pancreatic cancer may cause peritoneal dissemination [
2]. In gastrointestinal cancer the detection of free intraperitoneal tumor cells, serves as an independent prognostic factor [
3]. Free floating intraperitoneal tumor cells may attach to, degrade and migrate through the extracellular matrix (ECM) [
4]. Particularly if the peritoneum is damaged and the etracellular matrix is exposed tumor cell adhesion accumulates [
5]. In former studies we found a widespread peritoneal carcinosis, with tumor cell adhesion in most peritoneal aeras after intraabdominal instillation of tumor cells (Cell line: DHD/K12/TRb). We could demonstrate that tumor cells predominantly adhere to injured peritoneal areas. However, the goal of the underlying study was not locoregional recurrence but the effect of phospholipids in the complete peritoneal cavity. In addition preliminary studies could show that a phospholipid emulsion is able to significantly reduce intraperitoneal tumor cell adhesion [
3,
6,
7].
Phospholipids are natural constituents of peritoneal fluid secreted by mesothelial cells. These polar phosphoric acid di-esters are capable to form a lubricant layer on the peritoneal surface, which is of paramount importance to prevent adhesion [
19,
20]. Additionally, treatment with phospholipids (e.g. gangliosides) affect integrin function, causing reduced cell motility and adhesion capability after exogenous addition of phospholipids [
8‐
10]. Furthermore other adhesion-preventing substances are known. For example Jeekel et al. evaluated the effects of intra-abdominal treament with Icodextrin, a glucose polymer solution, in a coloncarcinoma CC531 rat model [
11].
As we could demonstrate the positive effect of phospholipids in low dosage in former studies, the aim of the underlying experiment was to compare the influence of rising phospholipid concentrations on the one hand and different tumor cell concentrations on the other hand with special emphasis on possible side effects.
Methods
Animals and anesthesia
In this study a total of 130 female BD-IX rats (mean body weight 200 g +/- 10 g) were operated. The animals were kept under standard laboratory conditions with free access for food and water entire study, which was performed according to the rules of the "Deusche Tierschutzgesetz" (50.203.AC 18, 9/02) and to the guidelines for the use of laboratory animals. The animals were assigned to the following groups of 10 rats (Table
1). The surgical procedure was performed under sterile conditions and general anesthesia by intramuscular injection of ketamine (100 mg/kg Bodyweight BW) (Ketamin 10%, Sanofi-Cefa, Düsseldorf, Germany) and rompun 2% (5 mg/kg BW) (Rompun 2%, Bayer, Leverkusen, Germany).
Table 1
Arrangement of animal groups in experiment A
| 10 rats | 10 rats | 10 rats | 30 days |
Tumor cell culture
Colonic adenocarcinoma induced in syngenic female BD-IX rats is the source of the cell line (DHD/K12/TRb) used in this investigation [
5]. Cells were obtained from the European Collection of Animal Cell Cultures (ecacc, Salisbury, UK). They were cultivated in monolayers in tissue culture flasks (75 cm
2, Falcon, Becton Dickinson, Heidelberg, Germany) in DMEM and Ham's F10 (1:1; GIBCO) supplemented with 10% fetal bovine serum (GIBCO) and gentamycin (0,005%; GIBCO). Cells were incubated at 37°C in a humidified atmosphere of 5% CO
2. They were passaged after treatment with 0,125% trypsin for 2 min. Following centrifugation for 10 min at 200 g, cells were suspended in 20 ml PBS and pelleted. The cell pellet was resuspended in 30 ml complete medium and seeded with a splitting ratio of 1:3. Only cells from three passages were used for the experiments. On the day of operation 2*10
6 cells were suspended in 100 μl complete medium for application for animals in the experimental setting A [
5] (Table
1). In the experimental setting B the cell amount ranged from 1*10
4 to 5*10
5 (Table
2).
Table 2
Arrangement of animal groups in experiment B
| 10 rats/1*104
| 10 rats/1*104
| 30 days |
| 10 rats/5*104
| 10 rats/5*104
| |
| 10 rats/1*105
| 10 rats/1*105
| |
| 10 rats/2.5*105
| 10 rats/2.5*105
| |
| 10 rats/5*105
| 10 rats/5*105
| |
Surgical procedure
All animals underwent a laparotomy via midline incision of 2 cm length. Before closure of the laparotomy wound tumor cells and either normal saline (controls) or phospholipid solutions with a concentration of either 6% or 9% (Fresenius, Bad Homburg, Germany) were instilled into the peritoneal cavity.
Experiment A
The animals in experiment A received a constant amount of 2*10
6 tumor cells according to our former experiment [
5].
Experiment B
In experiment B different numbers of tumor cells (10,000, 50,000, 100,000, 250,000 and 500,000) were administered intraperitoneally with a constant amount of phospholipid emulsion (PL 9%) or normal saline in the control group.
Phospholipids
The phospholipid solution consists of phosphatidylcholine 70% of the total weight, phosphatidylethanolamine 15% of the total weight, neutral lipids 8% of the total weight, sphingomyelin < 3% of the total weight and lysophosphatidylcholine < 3% of the total weight.
Evaluation of peritoneal carcinosis
After intervals of 30 days, the animals were sacrificed by inhalation of a lethal dose of isoflurane. The abdomen was opened by bilateral paramedian incisions for complete exploration. The extent of peritoneal carcinosis (mm
2) was measured using a digitizer board and calculation by costum-made software on a personal computer [
12]. After subtle resection, the tumor volume (ml) was measured by water displacement. Furthermore a modified peritoneal cancer index (PCI), as described by Sugarbaker et al. [
13], was determined. The original PCI was adapted concerning tumor size and areas in rats; Tumor size < 2 mm (LS-1); 2.1–5 mm (LS-2), > 5 mm or confluence (LS-3). Four areas (liver, spleen, colon, and diaphragma) were added to refine the original PCI because former animal studies showed metastases in these areas as well and evaluation confirmed more detailed information about the tumor cell dissemination. Therefore the maximum score was 51 from up to 3*17 areas. Exploration and evaluation were carried out by an independent, blinded observer [
5,
14].
Statistical analysis
All data are expressed as means +/- standard error of the mean. Statistical analysis was performed by a two-way ANOVA with pairewise comparison.
Discussion
The invasion of malignant tumor cells is determined by the capability to migrate, adhere to and degrade ECM components [
15]. This study was focused on the adhesion of tumor cells and on the preventive influence of phospholipids.
We used an established experimental model of peritoneal carcinosis after intraabdominal instillation of tumor cells (Cell line: DHD/K12/TRb) into female BD-IX rats [
5,
16]. All animals in this experiment developed peritoneal metastases. The effects could reliably be described by the area of attachment, the tumor volume and the PCI.
The idea of inhibiting intraperitoneal tumor nidation is not new. Several intraperitoneal therapies have been tested. Hagiwara et al. examined the anti-adherent effect of dextran sulphate after tumor implantation in mice [
17]. A prolonged survival in the treatment group after inoculation of melanoma cells was described. Moreover, intraabdominal tumor growth of CC531 adenocarcinoma cells in rats undergoing laparoscopy could be diminished using low-molecular-weight heparin in combination of intraperitoneal lavage und subcutaneous injection [
18]. Ogaswara et al. [
19] have described the inhibition of tumor invasion and growth testing intraperitoneal applied antioxidants like epigallocatechin gallate (EGCG) in a colon 26-L5 adenocarcinoma mice model. Treating carcinomatosis of colorectal cancer by cytoreduction and hyperthermic intraperitoneal chemotherapy have been studied by Zoetmulder et al. [
20]. They calculated the survival after cytoreduction and perfusing the abdomen with mitomycin C (35 mg/m
2) at 40 degrees C to 41 degrees C for 90 minutes. They observed a survival benefit for colorectal cancer patients with peritoneal carcinomatosis [
21]. Regarding angiogenesis as another key step in tumor growth, invasion and metastasis, antiangiogenic is an additional approach for antitumor treatment. Thus Nestler et al. [
22] have investigated the effect of angiostatin on the growth of CC531 colon carcinoma cells in vitro and in a laparoscopic animal model of peritoneal carcinomatosis. They found a significantly diminished intraperitoneal tumor growth in rats after intraperitoneal application of 20 mg angiostatin.
Tumor cells are known to stimulate peritoneal fibrosis, creating a congenial environment for peritoneal metastasis [
10,
23]. Phospholipids are capable of forming a remarkably resistant lubricant layer on the peritoneal surface [
24‐
27]. We suggest that the ability of phospholipids to cover peritoneal defects with exposed extracellular matrix subsequently inhibits tumor cell attachment [
5]. Intraperitoneal phospholipids have already been used to prevent postoperative adhesion. They showed a significant reduction of adhesion formation [
10,
25,
26]. The authors found no adverse side effects and no impairment of healing of anastomosis, laparotomy wounds and liver incisions after intraabdominal treatment with phospholipids [
10,
26]. We examined the effect of intraperitoneal phospholipids in view of the tumor cell adhesion [
5]. In former studies we could reliable demonstrate a significant reduction of tumor cell adhesion in case of a constant tumor cell concentration (1*10
6 tumor cells) [
5,
14]. With respect to the idea using phospholipids as an adjuvant intraperitoneal therapy in case of standard operations we wanted to examine the effect in case of lower tumor cell concentrations.
In
experiment A we could demonstrate a significant reduction of peritoneal dissemination as measured by all evaluation methods after treatment with 6%- and 9%-phospholipids. In former studies -using the same tumor cell concentrations but lower phospholipid concentrations (1,5% und 3%)- we found a reduction of peritoneal dissemination, too. However, the difference between control-group and treatment-group was not as distinctive as in this case [
5]. In
experiment B we diversified the number of tumor cells to mimic the clinical situation, that during the resection of a colon cancer only very few tumor cells were released.
There was a remarkable reduction of peritoneal dissemination especially in low tumor cell concentrations, supporting our theory of treating every patient with a gastrointestinal tumor at the end of the operation with phospholipids to avoid adhesions and peritoneal tumor dissemination. The absence of statistical significance in group 5*104 evaluating PCI and "area of tumor adhesion" possibly can be traced back to a low falling number because of the dead of an animal in this group. Valuating the results of the underlaying study we have to stress that the extent and not the incidence of peritoneal carcinomatosis was reduced. One also has to be aware that our results derived from a rat model attending only one kind of gastrointestinal cancer.
This in vivo model of intraperitoneal application of phospholipids to decrease the tumor cell adhesion describes the effect from a macroscopic and clinical point of view. Former studies could exclude a cytotoxic effect of the used phospholipids [
25,
28,
29]. However, various studies could demonstrate alterations in the adhesive properties of tumor cells depending on the degree of differentiation indicating a change of integrin expression [
30‐
33]. According to this further studies are currently in progress to evaluate the effect of phospholipids on the structure of cell membranes.
Competing interests
The auhor(s) declare that they have no competing interests.