Neoadjuvant treatment of gastric cancer with peritoneal dissemination
Introduction
In all studies that address the treatment of carcinomatosis for gastric cancer, complete cytoreduction is a requirement for prolonged survival in the management of this condition.1–12 For gastric cancer the proportion of patients who receive a complete cytoreduction is small; even in selected patients less than one-third of patients have resection of all visible cancer.78 Neoadjuvant chemotherapy has been proposed as a treatment modality that would increase the proportion of gastric cancer patients with peritoneal seeding who could receive a complete clearing of the cancer.1213 Unfortunately, systemic neoadjuvant chemotherapy has never significantly downstaged peritoneal seeding and the presence of carcinomatosis is considered by many a contraindication to the neoadjuvant treatment strategy.#
We report a new neoadjuvant treatment modality for gastric cancer with peritoneal seeding; it combines intraperitoneal with systemic chemotherapy in an attempt to eradicate disease on visceral peritoneal surfaces and thereby increase the proportion of patients who may receive complete cytoreduction. It is an attempt to eradicate cancer nodules by multimodality chemotherapy that would enter the cancer nodule from the systemic circulation but also diffuse into the nodule from a chemotherapy solution in the peritoneal cavity. A prospective phase II study was initiated to demonstrate the efficacy of this treatment in the palliation of patients with gastric cancer and carcinomatosis and further explore its toxicities.#
Patients and methods
Patients
Patients were enrolled between April, 2001 and December, 2003. PC was diagnosed by biopsy using laparotomy, laparoscopy, or by the cytologic examination of ascites. The eligibility criteria included: (1) histologically or cytologically proven PC from gastric adenocarcinoma; (2) absence of hematogenous metastasis and remote lymph node metastasis; (3) age 65years or younger; (4) Eastern Clinical Oncology Group scale of performance status 2 or less; (5) adequate bone marrow, liver, cardiac, and renal function; and (6) absence of other severe medical conditions or synchronous malignancy.#
Informed consent according to the institutional guideline was obtained from all patients.#
After the cytological or histological diagnosis of peritoneal dissemination, a peritoneal port system (Bard Port, C.R. Bard Inc., USA) was introduced into the abdominal cavity under local anesthesia, and the tip was placed on the cul-de-sac of Douglas.#
Chemotherapy
For intraperitoneal chemotherapy, 40mg of taxotere and 150mg of carboplatin were introduced over 30min in 1000ml of saline. On the same day, 100mg /m2 of methotrexate and 600mg/m2 of 5-fluorouracil were infused in 100ml of saline over 15min via a peripheral vein. This regimen was repeated weekly for two courses. Before and after NIPS, 500ml of saline was injected into the peritoneal cavity through the port, and fluid was recovered for cytology. In patients with positive cytology before NIPS, peritoneal wash cytology was performed after two courses of NIPS. If the result showed positive cytology, NIPS was added for the other two courses. Then, the test was repeated after four courses of NIPS, and NIPS continued again in the cases of positive results for peritoneal wash cytology.#
Investigation
In the patients with negative cytology before NIPS, endoscopy and CT scan were performed after two courses of NIPS. If no effect was observed on the tumors, patients additionally received two or more courses of NIPS. The number of NIPS chemotherapy treatments depends on the effect on tumors or the status of peritoneal cytology after NIPS, because preoperative cytoreduction by NIPS against PC widely spread in the peritoneal cavity is an indispensable procedure for the complete cytoreduction of PC.#
Staging
The peritoneal stage was determined from The Japanese General Rules for Gastric Cancer Study: metastasis to the adjacent peritoneum (P1), a few metastases to distant peritoneal sites (P2), and numerous metastases to the distant peritoneum (P3).14 Also, the distributions and sizes of peritoneal metastases were recorded at all laparoscopic or surgical interventions. Effects of NIPS were evaluated by comparing the size and number of PC before and after NIPS. In the non-operable patients, the effects of NIPS were evaluated by laparoscopy, or CT scan.#
If the peritoneal wash cytology revealed negative study or the tumors showed partial response after NIPS, laparotomy was done to perform peritonectomy procedure. Patients showing progressive disease did not undergo laparotomy, and patients with positive cytology even after four or six courses of NIPS were treated with chemotherapy.#
Techniques
The technique used to remove peritoneal dissemination was the peritonectomy procedure developed by Sugarbaker and Yonemura.45 For the complete removal of visceral peritoneum bearing cancer, total gastrectomy, subtotal colectomy and partial resection of the small intestine were performed. If the mesentery of the small bowel was affected, nodules were removed without resection of the small bowel.#
Peritoneum covering the diaphragm was removed by electrosurgical dissection between the peritoneum and the diaphragmatic muscle. If necessary the entire peritoneum covering both the right and left hemidiaphragm was removed. Pelvic peritonectomy was completed by stripping the pelvic peritoneum covering the bladder; the cul-de-sac was completely removed with the rectosigmoid colon. An ileosigmoid anastomosis or permanent ileostomy was performed. In females, the uterus was removed with the pelvic peritoneum combined with bilateral salpingo-oophorectomy. The goal of peritonectomy was the complete removal of all visible cancer nodules.#
Complications were prospectively identified and verified by a chart review. All complications related to NIPS chemotherapy and peritonectomy were recorded.#
Outcome data were obtained from medical records and patient interview. Statistical analyses were performed by using SPSS software (SPSS Inc., Chicago, USA). Survival curves are calculated by the Kaplan–Meier method.#
Results
Clinical characteristics of the 61 patients are listed in Table 1. The average age was 45.6years. All 61 patients had P3 dissemination. Ascites was present in 33 patients. Twenty-eight patients had primary gastric cancer and the remaining 33 patients had recurrent PC.#
Prior to NIPS chemotherapy, peritoneal fluid cytology was positive in 39 patients and reverted to negative cytology after treatment in 22. In 33 patients with ascites, all peritoneal fluid disappeared in 12 patients after NIPS chemotherapy. Thirty-eight patients showed partial response after NIPS (Table 2). After NIPS chemotherapy, 30 patients had an operative intervention, and the other 31 patients did not undergo surgery due to the progression of disease (N=19) and refusal of operation (N=12). P3 status changed to be P2 in 2 patients.#
Table 3 indicates the operation interventions. Total gastrectomy was performed in 20 primary cases, subtotal colectomy in 21, total hysterectomy in combination with bilateral salpingo-oophorectomy in 13, resection of small bowel in 13 and resection of small bowel mesentery in 14. Left and right subdiaphragmatic peritonectomy was performed in 15 and 13 patients. Pelvic peritonectomy was performed in 17 patients. Fulguration of peritoneal nodules was used as an adjunctive surgical technique in 23 patients. Complete cytoreduction was achieved in 14 (50%) of 30 patients. In 32 patients with recurrent cancer, complete cytoreduction was performed in only one patient (3.1%), and 11 patients received incomplete cytoreduction. On the contrary, complete cytoreduction was done in 13 (46%) of 28 primary cases.#
In all patients, bone marrow suppression of grade 3 or 4 was recorded in 6 patients and diarrhea in 2 patients, respectively. Bone marrow suppression developed after 3 courses in three, 5 courses in two and 6 courses in one patient.#
Complications developed in 7 patients after cytoreductive surgery with peritonectomy. Pneumonia developed in 2 patients and renal failure occurred in one patient. There were 2 instances of anastomotic leakage, and 2 of abdominal abscess. The overall operative mortality rate was 3.3% (1/30), and the cause of death was multiple organ failure due to sepsis from abdominal abscess.#
Twenty-six patients were alive at the time of analysis. The survival distribution for all patients is shown in Fig. 1. Median survival time (MST) of all patients was 14.4months, with a 1-year survival of 67%. MST of the 30 patients who had surgical intervention was 18.0months, and that of patients who did not receive the operation was 9.6months (Fig. 2). There was a significant survival difference between the two groups (P=0.031, Z=4.64). Patients who received a complete resection had a median survival of 20.4months, and survival of patients who received incomplete cytoreduction was 15months (Fig. 2).#
There was no survival difference between primary and recurrent cases. MSTs of primary and recurrent cases after cytoreduction were 14.4 and 15.6months, and those after no operation cases were 9.6 and 8.4 months, respectively.#
Discussion
Chemotherapy for PC
In the treatment of cancer arising from gastrointestinal tract, PC has long been considered as a fatal disease, and most surgeons have considered PC as a non-surgical condition. Recently, some investigators have reported limited success with new treatment strategies which include intraperitoneal chemo-hyperthermia, early postoperative intraperitoneal chemotherapy and peritonectomy.3–9 These studies suggest that intraperitoneal chemotherapy can deliver a high dose intensity into the peritoneal cavity. Unfortunately, drug penetration deep to the peritoneal surface is limited. Effective diffusion distance into tissues is reported to range from 100μm to 1000μm.1516 Accordingly, any superiority of intraperitoneal chemotherapy over intravenous delivery in PC will be limited to those patients with very small tumor volumes at the time when regional treatment is initiated. Hyperthermia augmented the penetration distance of anticancer drugs up to 2000μm.16 Furthermore, hyperthermia also alters permeability of tumor cell membranes to enhance uptake of chemotherapeutic drugs.17 In addition, hyperthermia shows a synergistic cytotoxicity on cancer cells when combined with chemotherapeutic drugs such as Mitomycin C, Etoposide and CDDP.18 However, intraperitoneal hyperthermia combined with chemotherapy is effective only for prophylaxis of peritoneal recurrence and small nodular type of established PC.19 Accordingly, peritoneal nodules larger than 2mm cannot be eradicated by IP chemotherapy and hyperthermia.#
Surgical treatment for PC
In the surgical treatment of PC from appendiceal and colon cancer, complete cytoreduction has been determined to be essential for long-term survival. Culliford et al. reported a 5-year survival of 54% for complete cytoreduction and 15% for incomplete cytoreduction.20 Furthermore, Glehen et al. reported that the 2-year survival rate was 79% for patients with macroscopic complete resection and 44.7% for patients without macroscopic incomplete cytoreduction.9 In PC from gastric cancer, Yonemura et al. reported that patients receiving complete cytoreduction had a significantly higher survival than did those with residual disease.19 Although great differences exist in the biological behavior of colon and gastric cancer, macroscopic complete cytoreduction may be the most important surgical requirement in the treatment of PC with a goal of long-term survival in both conditions. Unfortunately, complete cytoreduction is not possible in P3 dissemination even with the most aggressive peritonectomy. Complete cytoreduction cannot be achieved when small bowel and its mesentery are diffusely involved.#
Development of NIPS
To increase the rate of complete cytoreduction by clearing the peritoneum of cancer nodules, NIPS chemotherapy was developed. The present study indicates that NIPS can increase the rate of complete cytoreduction as a result of a reduction of distribution and volume of peritoneal cancer nodules.#
Furthermore, NIPS chemotherapy was able to eradicate free cancer cells in the peritoneal cavity prior to the operation. Free intraperitoneal cancer cells can be detected in 65% of patients with peritoneal dissemination. 21 These cells are viable and may be trapped with the peritoneal wound created by the surgical procedure. Accordingly, the free cancer cells should be eradicated before peritonectomy.#
In vitro chemosensitivity testing is considered a fairy good predictor of clinical chemosensitivity.2223 Tanaka et al. shows results of such testing in clinically obtained samples from primary gastric cancer, using a collagen gel method.23 Carboplatin, taxotere, 5-fluorouracil, CDDP, and Mitomycin C show high chemosensitivity for gastric cancer. From these results, taxotere, carboplatin, and 5-fluorouracil had been selected for NIPS, and methotrexate was used as a biochemical modulator of 5-fluorouracil.24#
According to Cunliffe,25 for intraabdominal metastasis, nutrition can be derived from the peritoneal surface as well as the blood supply.26 In NIPS, the peritoneal cancer nodule is penetrated bidirectionally, not only through intraperitoneal but also intravenous therapy. Generally, intravenous chemotherapy has little effect on PC.1213 Intraperitoneal chemotherapy alone showed a response rate of less than 30%.727–29 The present study showed a response rate of 65% after NIPS chemotherapy.#
Toxicity and survival after NIPS
From the study of the dose limiting toxicities of IP administration of taxorete, Morgan et al. reported that maximum tolerated dose (MTD) was 125mg/m2, and that no grade 3 or 4 toxicities were found after the i.p. administration of lower than a dose of 80mg/m2.28 Furthermore, Fusida et al. reported that there was no hematological toxicities after weekly i.p. administration of 45mg/m2 of taxotere.29#
In the i.p. administration of carboplatin, MTD was determined to be 500mg/m2,30 and 300mg/m2 of i.p. administration of carboplatin was a safe dose in Japanese patients with ovarian cancer.31#
The doses of 40mg/m2 of taxotere and 150mg/m2 of carboplatin which were used in the present study were considered to be safe doses.#
This combined approach of systemic and intraperitoneal chemotherapy was associated with an absent mortality and a very reasonable morbidity. Also, it was effective for ascites. This combination of systemic and local-regional chemotherapy should be considered in patients with carcinomatosis from gastric cancer. It may be even more effective for those patients who have small volumes of disease and symptomatic ascites. Accordingly, the bidirectional chemotherapy may be the preferred strategy for the preoperative chemotherapy of PC and should be considered for phase III studies.#
Despite the extensive chemotherapeutic and surgical interventions the patients who were able to complete the protocol had an acceptable quality of life, and survival improvement.#
Figures and Tables
Table 1
| Age | 45.6±14.4 (range 18–69) |
| Gender | |
| Male | 28 |
| Female | 33 |
| Grade of dissemination | |
| P1 | 0 |
| P2 | 0 |
| P3 | 61 |
| Ascites | |
| No | 28 |
| Yes | 33 |
| Peritoneal lavage cytology | |
| Negative | 22 |
| Positive | 39 |
| Primary or recurrence | |
| Primary | 28 |
| Recurrence | 33 |
| Cycles of NIPS | 3.8 (2–6) |
| Total | 61 |
Table 2
| Surgical exploration | No operation | Total (%) | |
| No change or PDa | 4 | 19 | 23 (38) |
| Partial response | 26 | 12 | 38 (62) |
| Complete response | 0 | 0 | 0 |
| 30 | 21 | 61 |
Table 3
| No operation | 31 |
| Complete resection | 14 |
| Incomplete resection | 16 |
| Operation methods (24 patients) | |
| Total gastrectomy | 20 |
| Subtotal colectomy | 21 |
| Total hysterectomy/bilateral salpingo-oophorectomy | 13 |
| Small bowel resection | 13 |
| Resection of small bowel mesentery | 14 |
| Left subdiaphragmatic peritonectomy | 15 |
| Right subdiaphragmatic peritonectomy | 13 |
| Pelvic peritonectomy | 17 |
| Electrosurgical fulguration | 23 |
References
4. P.H.SugarbakerPeritonectomy proceduresAnn Surg2119952942