Preoperative intraperitoneal oxaliplatin for unresectable peritoneal carcinomatosis of colorectal origin: a pilot study

Patient selection

All consecutive patients presenting with unresectable PC of colorectal origin between January 2015 and June 2015 were included in a pilot study in the premises of a phase I trial testing intraperitoneal oxaliplatin in the neoadjuvant setting. Inclusion criteria were: histologically-proven colorectal or appendix adenocarcinoma, Peritoneal Carcinomatosis Index (PCI)≥25 with surgically unresectable distribution, at least one line of previous systemic chemotherapy received, ECOG≥1. Exclusion criteria consisted in the presence of extraperitoneal metastases except for parietal invasion, ovarian metastases and retroperitoneal lymph nodes, peripheral sensory neuropathy grade≥2, anesthetic or medical contraindications to surgery and chemotherapy. All patients gave informed consent prior to the trial. The present pilot study was approved by the Ethics Committee of the Montpellier Cancer Institute in accordance with the ethical standards of the Helsinki Declaration of 1975.

Six patients were included in the study during the scheduled time interval. They were initially selected based on radiologic imaging (CT-scan and MRI indicating the presence of probable unresectable peritoneal disease) and clinical and biological data. All patients were evaluated by a senior surgeon (FQ) who programmed surgical exploration followed either by surgical resection and HIPEC in case of resectability or by insertion of intraperitoneal chamber catheters followed with intraperitoneal and systemic chemotherapy. Clinical history, imaging and histological data as well as perioperative data were prospectively recorded. Intraperitoneal infusion was administrated in the medical oncology department with the participation of the designated surgeon (OS, MA). Follow-up data were also prospectively recorded.

Operative technique

All selected patients had previously been operated using a xypho-pubic midline laparotomy, and two of the patients had already undergone a cytoreductive surgery with oxaliplatin-based HIPEC in 2012 and 2013, respectively.

For the patients with PC of appendiceal origin, exploratory laparoscopy was tempted as all preoperative imaging techniques were pointing out to an unresectable disease. Three trocars were inserted on the midline (one 10 mm-trocar, two 5 mm-trocars). Exploration was limited in the pelvic region by the presence of a bulky disease. Two chamber catheters were inserted, the first one placed in the subxyphoidian region toward the right costal margin, and the second in the left suprapubic region. The space for the chambers was created supra-aponevrotically while the catheter was inserted into the abdomen through a fine trocar for percutaneous cystocatheters under camera control.

The rest of the patients had an exploratory laparotomy as the possibility of resection could not be completely ruled out with imagery alone. The exploration was performed according to the standards of classical cytoreductive surgery [27–29]. The chamber catheters insertion respected the same topography than in the laparoscopic approach. The PCI was calculated for all patients and biopsies were performed.

Intraperitoneal and systemic regimens

The administered regimen aimed to intensify systemic chemotherapy with intraperitoneal administration of oxaliplatin at the intravenous dose of 85 mg/m². Unlike other intraperitoneal regimens using cisplatin, paclitaxel or doxorubicin [20, 22, 30], oxaliplatin is considered to be associated with potentially severe abdominal pain [30]. Therefore an intravenous continuous pump perfusion with oxycodone was set up during every cycle at 1 mg/h for the initial 6 h followed with a 0.5 mg/h dose up to 24 h. The doses were adjustable depending on the patients’ chronic treatments.

The instability of oxaliplatin in NaCl 0.9 % solution was described therefore G5 % serum was used [31]. Typically, we started with a 250 mL perfusion of G5 % on each catheter. Oxaliplatin administration in 500 mL of G5 % was then performed simultaneously on each catheter during 60 min, followed with rinsing with another 250 mL G5 % on each device.

The regional therapy was associated with intravenous FOLFIRI or LV5FU2, respectively. Targeted therapies were also associated when possible. The decisions concerning the recommended schemes were validated by the local tumor board after taking into account previous regimens, toxicities and responses. Patients considered as oxaliplatin-resistant during previous regimens were excluded. One patient with mucinous disease was converted from systemic FOLFIRINOX to systemic FOLFIRI+intraperitoneal oxaliplatin+Vectibix. Cycles were administered every 2 weeks and a one-night hospital stay was considered mandatory for every patient.

Patient characteristics

The six patients included in this study were four females and two males with a mean age of 57.5 years (range 23–78). The demographic, clinical and histological characteristics of the patients are shown in Table 1. The median PCI was 29.5 (range 25–39). One patient developed a recurrence of already treated PC associated with a midline parietal metastasis. Two patients directly presented with synchronous PC with poor response to systemic chemotherapy.

Outcomes

The outcomes of all patients are presented in Table 2. One patient was excluded from intraperitoneal and systemic therapy due to failure to perfuse the intraperitoneal chamber catheter. This patient developed an entero-cutaneous fistula that needed operation and eventually presented a favorable evolution.

All other patients were treated with 1 to 4 cycles. Only two patients completed the protocol receiving four cycles each. The main toxicity consisted of abdominal pain (grades 2 and 3) manageable with IV continuous perfusion of either oxycodone or morphine. Grade 2 neutropenia was found in one patient. Grade 3 stomatitis was also encountered in one patient. A grade 3 diarrhea with severe hypokalemia was also recorded in one patient treated with FOLFIRI-VECTIBIX as systemic therapy. No clinically significant hyperglycemia was recorded. The protocol was immediately stopped for this patient who was then administered systemic FOLFOX-VECTIBIX instead, with acceptable tolerance.

The patient with the shortest survival initially had a PCI evaluated at 39. After the first cycle, she developed grade 3 asthenia and grade 2 vomiting. Considering the highly-advanced stage of the disease, the tumor board decided to discontinue the aggressive treatment and referred the patient to palliative care while maintaining systemic FOLFOX. She died 3 months later.

The frequencies of occurrence of the toxicities are presented in Table 3. There were no grade 4 toxicities but there were three grade 3 toxicities, all of them non-hematological.

Rationale

To our knowledge, our study is the first pilot trial to test the use of intraperitoneal chemotherapy based on oxaliplatin in the neoadjuvant setting for patients with unresectable peritoneal metastases. It aimed to evaluate the toxicity of intraperitoneal oxaliplatin associated with systemic chemotherapy and targeted molecules in the neoadjuvant treatment of unresectable PC of colorectal and appendicular origin. A single phase I trial was previously published, which assessed a combination of intraperitoneal oxaliplatin and paclitaxel at lower doses in 19 patients with PC after colorectal cancer (n=9), signet ring carcinoma (n=2), gastric cancer (n=2), ovarian cancer (n=2) and other types of tumor (n=4) [30]. Due to the cancer heterogeneity of the selected patients and the combination of intraperitoneal therapies, we could not consider this study as applicable to our therapeutic scheme.

The rationale of using a combination of intraperitoneal and intravenous therapies in this clinical situation is connected with the peritoneal bio-chemistry: drugs administered directly into the peritoneal cavity show a longer half-life and a higher concentration in this space when compared with the intravenous administration [32]. A plasma peritoneal barrier was already described by some authors [33] as preventing the transport of the intravenous drugs into the peritoneal cavity due to the cell-matrix system surrounding the exchange vessels.

Intraperitoneal therapies other than HIPEC have already shown their value in the adjuvant setting for the ovarian cancer [12, 13, 34] and, more recently, in the neoadjuvant setting for gastric cancer patients with surgically-proven PC (the NIPS regimens) [14–21, 23–25, 35–37]. In this latter context, NIPS was shown to obtain complete histological response of peritoneal cytology or small PC, to offer conversion to resectability of the selected patients but also to improve overall survival in patients with an initial PCI≥20 [23].

Even though these results were obtained for different types of cancer, they might be successfully implemented in colorectal carcinoma. A French group has demonstrated the prognostic similarities of PC and liver metastases in colorectal cancer when there was only one metastatic site involved [26]. Bilateral sub-optimally resectable liver metastases are initially treated with hepatic intra-arterial infusion (HAIP) associated with systemic chemotherapy which offers conversion to resectability in 25–50 % of all cases [38, 39]. A similar strategy in the initially unresectable PC of colorectal origin may correspond to the present protocol.

Drug selection

The ideal antineoplastic agent for intraperitoneal therapy was defined as a drug with: documented activity in the malignant disease to be treated, experimental or clinical evidence for concentration-dependent cytotoxicity against the malignant disease to be treated, no toxicity to the peritoneal lining with regional delivery, slow exit from the peritoneal cavity and rapid clearance from the systemic, and significant and rapid hepatic metabolism to a non-toxic metabolite [40, 41].

All these characteristics had already been studied for oxaliplatin (Table 3) which is featured in many lines of systemic chemotherapy in colorectal cancer [1] and which also is in many specialized centers the elective drug for HIPEC in PC of colorectal origin [42, 43]. The suboptimal parameter for this molecule is the AUC peritoneal-plasma ratio, calculated at 3.5–5, which compares unfavorably with other drugs such as cisplatin [12–21] or paclitaxel (1000) (Table 4). Nevertheless, these other drugs were not shown to be efficient in colorectal cancer systemic treatment.

Given our experience with oxaliplatin in the intraperitoneal setting and its efficacy in the systemic treatment of colorectal carcinoma, we chose this molecule for the present study. It is currently known that doses of 460 mg/m² are well tolerated under general anesthesia when associated with hyperthermia. Our dose of choice corresponded to the regular intravenous dose which was supposed to assure a better tolerance, and specifically decrease abdominal pain.

Tolerance and toxicity

In our study, a pilot study in the premises of a phase I trial, safe doses of IP oxaliplatin were used. There were no grade 4 toxicities and some of the grade 3 toxicities were considered to be related to systemic therapy; the stomatitis was imputed to 5-FU administration as it responded well to 5-FU dose reduction, and the diarrhea was potentially related to camptothecin administration. Abdominal pain was frequently reported in our patients (Table 3) but adapted treatments based on oxycodone and morphine were efficient for symptom relief. Hematological toxicities, which are usually associated with various IP therapies, were only encountered as grade 2.

Perspectives

This pilot therapy was restricted to patients with highly-advanced peritoneal disease for which previous systemic chemotherapy lines were considered of little or moderate efficacy. It was conceived as a unique alternative to obtain disease stability or hopefully, conversion to resectability. The current standard option for these patients, in the absence of clinical trials, is a new line of chemotherapy when available, or best supportive care in most of the cases.

The present toxicity results justify carrying on with a phase I trial assessing IP oxaliplatin, with dose escalation. The local outcomes allowed us to adapt our management of the intraperitoneal ports before starting a phase I trial. The results of this pilot study are promising as the patients who completed the four cycles had an overall survival superior to 12 months. Nevertheless no major improvement in disease stage was reported for these patients, which suggests that higher doses might be needed in order to achieve tumor response.