Cancer Therapy Vol 2, 79-84, 2004

1LTMG Hospital, Sion, Mumbai, India, ²Washington Cancer Institute, Washington, DC, USA

__________________________________________________________________________________

Key Words: Gastrectomy, carcinomatosis, induction chemotherapy, mitomycin C, cisplatin, doxorubicin

Gastric cancers that extend to the serosal surface or that involve lymph nodes are at high risk for resection site recurrence and for peritoneal carcinomatosis. The incidence varies from 20-50% (Gunderson and Sosin, 1982; Wisbek et al, 1986; Landry et al, 1990; Yoo et al, 2000). Systemic chemotherapy has not been found to be effective as an adjuvant treatment to reduce the incidence of local-regional recurrence for patients with peritoneal carcinomatosis. Intraperitoneal chemotherapy in the perioperative period has shown benefit in clinical trials. In this review the theoretical basis for local-regional recurrence.

Yoo et al, (2000) reviewed 2328 patients with gastric cancer who underwent curative resection between 1987-1995. In 508 patients there was documentedevidence of recurrence. A single anatomic site for recurrence was observed in 403 patients and 83 had recurrence at two or more sites. Isolated peritoneal recurrence was noted in 172 patients and was the most frequent single pattern (33.9%). Hematogenous recurrence, the second pattern observed, was seen in 133 cases (26%) of which 75 cases had hepatic metastases. Local-regional recurrence involving the gastric stump, anastomoses, lymph nodes or an adjacent organ, the third observed pattern, was seen in 19.3% of cases. The length of time to recurrence was 27.3 months for local-regional recurrence, 18.1 months for peritoneal recurrence and 14.6 months for haematogenous recurrence. Serosal invasion and lymph node metastases were common risk factors for all patterns of recurrence. These data demonstrate the need for achieving better local-regional control and for prevention of peritoneal seeding.

The rationale for integrating perioperative intraperitoneal chemotherapy into the surgical treatment of gastric cancer was presented by Sugarbaker and coworkers (Sugarbaker et al, 1989). They suggested that three sources of microscopic residual disease could occur after gastrectomy (Figure 1). The first and most obvious cause

*Occurs at resection site, on abraided bowel surfaces and beneath abdominal incision.

Figure 1. The tumor cell entrapment hypothesis suggests three mechanisms for microscopic residual cancer cells in patients having an R-0 gastrectomy.

of contamination of the peritoneal cavity by the cancer cells is serosal invasion by T3 or T4 malignancy. The surgical trauma of cancer resection combined with the natural tendency of the cells to exfoliate result in a positive cytology in these patients (Boku et al, 1990; Bando et al, 1999; Kodera et al, 1999).

A second prominent cause of cancer cell spillage with surgery occurs as a result of transection of lymphatic channels in patients with positive lymph nodes. This is more an issue with multiple nodes involved rather than a few positive perigastric lymph nodes. Fujimura and colleagues, (1997) documented the ability of the reverse transcriptase polymerase chain reaction to identify free gastric cancer cells in the peritoneal cavity. Marutsuka and coworkers established that lymph node positive patients have a high likelihood of cancer cells in the peritoneal cavity after gastrectomy. They concluded that lymph node dissection opened lymphatic channels and spread viable cancer cells into the free peritoneal cavity (Koga et al, 1988).

A third source of cancer cell contamination is blood lost from the cancer specimen into the peritoneal space. Perhaps this is a contributor to the poor prognosis seen when cancer patients require large blood transfusion.

Tumor cells are dislodged at the time of surgery as a result of surgical trauma and are then implanted onto traumatized peritoneal surfaces. Here the implants are entrapped by blood clots and enmeshed in fibrin deposits. They are presumably nourished by the growth factors released during the inflammatory phase of healing. To prevent this sequence of events chemotherapy is given intraoperatively and in the early postoperative phase.

In the operating room the chemotherapy solution is heated to a temperature of 41° C at the point of delivery. The effects of hyperthermia are:

1) Heat greater than 43° C affects cancerous tissues more than the normal tissues.

2) Heat softens the tissues and decreases the interstitial pressure thereby facilitating drug penetration into the tumour.

3) Heat increases the cytotoxicity of selected chemotherapeutic agents.

A temperature profile observed in the operating room with hyperthermic intraoperative intraperitoneal chemotherapy is shown in Figure 2.

The pharmacology of intraperitoneal drug delivery provides strong theoretical support for these treatments. The local exposure of tissues to chemotherapy solution fare greater and the systemic toxicities lower if the drug delivery is intraperitoneal (Figure 3).

These studies of the natural history of gastric cancer suggest that patients with primary disease could be specially selected for adjuvant intraperitoneal chemotherapy.

Figure 2. Temperature profile for heated intraoperative intraperitoneal chemotherapy drugs. Mitomycin C, cisplatin, doxorubicin have been used.

Figure 3. Pharmacokinetic study of intraperitoneal 5-fluorouracil 1000 mg in 2 liters 1.5% dextrose peritoneal dialysis solution. The intraperitoneal concentration is shown as circles and the plasma concentration as squares. The concentration difference over time peritoneal fluid to plasma is 250:1.

Patients for treatment must have complete (R0) resection. If persistent disease exists at any site, the intraperitoneal chemotherapy treatment cannot confer a survival advantage. As a result of radical surgery there must be complete clearance of the primary tumor and involved lymph nodes for proper use of these treatments.

Incomplete containment of the cancer as a result of microscopic residual disease may be unavoidable as a result of the surgical event. Patients with this small volume of cancer recently seeded on peritoneal surfaces may be the ideal patients for perioperative intraperitoneal chemotherapy. However, the timing of the chemotherapy (perioperative) and the route of administration (intraperitoneal) are absolute requirements for benefit in this group of patients. Multiple cycles of intraperitoneal chemotherapy initiated weeks after the gastric cancer surgery showed no benefit.

Clinical studies to support the use of perioperative intraperitoneal chemotherapy as an adjuvant to gastric cancer have steadily accumulated over a decade. The published information is shown in Table 1. Eight studies show a significant advantage or an advantageous trend for patients treated with perioperative intraperitoneal chemotherapy (Figure 4). Most of these studies used hyperthermic intraperitoneal chemotherapy (Koga et al, 1988; Hamazoe et al, 1994; Yonemura et al, 1995; 2001; Ikeguchi et al, 1995; Fujimoto et al, 1999; Hirose et al, 1999). A single study used early postoperative intraperitoneal chemotherapy (Yu et al, 1998). Two studies of intraperitoneal chemotherapy for gastric cancer did not show benefit as an adjuvant treatment. Schiessel and coworkers used adjuvant intraperitoneal cisplatin in a multicenter trial in 64 randomized patients. The treatment was initiated within 4 weeks of surgery; none of the patients had perioperative treatment. There were no survival advantages (Schiessel et al, 1989). Sautner and colleagues reported a similar negative study (Sautner et al, 1994).

Figure 4. A statistical summary of 8 trials testing perioperative intraperitoneal chemotherapy.

Table 1. Eight reports of adjuvant treatment with perioperative intraperitoneal chemotherapy in gastric cancer patients having an R-0 resection (negative margins of excision and absence of disseminated disease).

NA = not available; NS = not significant.

Currently, there is a large theoretical basis and a moderate support from clinical studies to suggest that perioperative intraperitoneal chemotherapy is an important part of a program in management of gastric cancer. However, to date this innovation in patient management has only been adopted at a small number of institutions in the United States, Korea, and Japan. Certainly, it does not represent a standard of practice. It may emerge as a standard of practice if further clinical data can be obtained in the future that shows similar benefit to that presented in this manuscript.

The need is further phase III trials in patients with gastric cancer. Also, a trial must be performed in Western patients with gastric malignancy. Before this can occur as a multi-institutional effort with adequate number of randomized patients, standardization of these perioperative treatments must occur. The group conducting the trial will need to agree on the timing (between 30 and 120 minutes), the heat (between 39 and 43° C), the drugs (mitomycin C, cisplatin, doxorubicin, VP16), open versus closed technology, heated intraoperative chemotherapy versus early postoperative intraperitoneal chemotherapy versus both, and drugs for early postoperative intraperitoneal treatments if used (5-fluorouracil or taxol). A great deal of thought and some further pharmacokinetic and dose escalation studies may be necessary. Also, the patient eligibility requirements will be controversial. Should only stage III patients be entered? Should patients be entered prior to an exploration of the abdomen or would the randomization be intraoperatively after the completion of the gastrectomy? Should patients with early carcinomatosis such as P1 or P2 peritoneal seeding receive treatment? What about patients that have ovarian involvement; should these patients enter the trial? Should cytology, both before and after gastric cancer resection, be required? Should patients with positive cytology be included or excluded from the adjuvant study?

Not only should the perioperative chemotherapy treatments and eligibility treatments be definitely determined, the surgical procedure needs to be well defined too. Most likely, on the basis of the positive result of Yu and colleagues, a D2 gastrectomy should be recommended (Yu et al, 1998).

All these and many other questions will need to be resolved before a multi-institutional trial of perioperative intraperitoneal chemotherapy in patients with resectable gastric cancer can proceed. A workshop to define these parameters and to produce a workable protocol needs to be a high priority goal for the future.

Lymph node positive and serosal invasive gastric cancer have a high incidence of microscopic residual disease following gastrectomy. This results in local and peritoneal surface recurrence. This surgical treatment failure can be reduced by perioperative intraperitoneal chemotherapy. Further studies are necessary to confirm these benefits.