| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Surgical Oncology, Washington Cancer Institute, 106 Irving Street N.W., Suite 3900, Washington, DC 20010
Correspondence: Address correspondence and reprint requests to: Paul H. Sugarbaker, MD, FACS, FRCS; E-mail: paul.sugarbaker{at}medstar.net.
A recent survey by the American Cancer Society of 957 adults revealed that more than half believed that surgery can cause cancer spread.1 From my personal experience of 30 years of reoperative surgery in patients with gastrointestinal cancer, I am convinced that surgical procedures often disseminate cancer within the abdomen and pelvis. I agree with most Americans that a major problem exists. It is completely possible that patients with a gastrointestinal cancer come to the operating room with a contained process, but, unfortunately, as a result of surgical trauma, they leave the operating theater with cancer spread to the operative site and to the surrounding peritoneal surfaces. Often, the large primary cancer that is the focus of the operative procedure is successfully removed; it is the cancer cells the surgeon does not see that result in the abdominal and pelvic (local-regional) recurrence in a substantial proportion of patients. In colorectal cancer, this surgically induced cancer spread may occur in approximately 10% of patients, 30% of rectal cancer patients, 40% of gastric cancer patients, and 50% of pancreas cancer patients. As might be expected, the narrower the margins of resection and the greater the extent of lymph node positivity, the more likely that iatrogenic cancer dissemination will occur.2
If cancer surgery can induce the spread of cancer cells, then one would expect that some surgeons would induce more dissemination than others. There is no doubt that the mortality rate for large gastrointestinal cancer operations varies in a meaningful way between experienced and less experienced surgeons.3 Data that are more difficult to secure involve the differences in survival accumulated over the 5 to 10 years after a gastrointestinal cancer operation. Some investigators have suggested that the difference between the best and the worst rectal cancer surgeons can be as high as 50%.4 The difference in survival was directly related to the local-regional rate of recurrence.
Added to this problem of surgically induced gastrointestinal cancer spread is the cancer cell contamination that occurs as a part of the natural history of the disease. In 10% to 20% of gastrointestinal cancer patients, the peritoneal space is seeded by cancer at the time of the first symptoms.2 This carcinomatosis at the time of first diagnosis is most common with the mucinous histological types of cancer. Local-regional treatment failure remains a major flaw in the successful management of gastrointestinal cancer as a result of surgically induced cancer dissemination and naturally occurring full-thickness invasion of the bowel wall by the primary cancer.
Some surgeons have been concerned not only with the surgical removal of the primary cancer mass, but also with the intraperitoneal spread of cancer cells. There are surgical options to minimize or even eliminate local-regional cancer dissemination and the eventual progression to terminal carcinomatosis. Meticulous resection techniques have been suggested to be of value.5 Also, systemic chemotherapy may be of some help.6 However, the optimal local-regional control in patients with gastrointestinal cancer may go far beyond improved resection techniques and systemic chemotherapy administration. Intraperitoneal chemotherapy washing, usually combined with hyperthermia, has shown itself to be effective in several clinical situations frequently seen in patients with gastrointestinal cancer. Yu and colleagues7 showed that perioperative intraperitoneal chemotherapy reduced the local-regional recurrence and improved survival in patients with gastric cancer. These data were the result of a prospective and randomized clinical trial. Also, appendiceal malignancy with carcinomatosis has been successfully managed with a combined treatment using cytoreductive surgery and perioperative intraperitoneal chemotherapy.8 In selected patients with carcinomatosis from colorectal cancer, a curative approach has been reported in multi-institutional phase II studies and in a prospective phase III protocol.9,10
The surgeon or the layperson concerned with the management of patients with gastrointestinal cancer is left with a major dilemma. He or she sees local-regional gastrointestinal cancer treatment failure as a major flaw in the successful management of these patients. However, management strategies to improve local-regional control of cancer dissemination at the resection site and on peritoneal surfaces have been repeatedly demonstrated. The obvious question concerns the failure to progressively integrate perioperative intraperitoneal chemotherapy treatments into the standard of care of patients with gastrointestinal cancer.
This failure has, in my opinion, come about in large part not for lack of efficacy. Limited application has resulted from the high morbidity and mortality associated with this "heroic treatment" in the past. This important article from the Istituto Tumori by Younan and colleagues11 shows that the large morbidity and mortality reported in the past is no longer of necessity a part of the combined treatment. A mortality rate of 1% and a bowel complication rate of 11% when primary unprotected anastomoses are performed document the safety of this approach in the prevention or treatment of carcinomatosis. This article, along with others, shows that visceral resections, peritonectomy procedures, and perioperative intraperitoneal chemotherapy can be used to prevent or eradicate local-regional disease with an acceptable morbidity and mortality in patients with gastrointestinal cancer.
Over the last 20 years, numerous modifications in the use of perioperative intraperitoneal chemotherapy, usually hyperthermic perioperative intraperitoneal chemotherapy, have occurred. This decrease in complications has occurred simultaneously with increased efficacy, primarily as a result of improved patient selection. Moderate heat (41.5°C) rather than extreme heat (43°C–44°C) is now used. Also, moderate chemotherapy doses and more uniform distribution of heat and chemotherapy solution have diminished focal damage to the small bowel. The reduction in morbidity and mortality follows the pattern of a learning curve.
In my opinion, the surgical responsibilities for the management of gastrointestinal cancer now go beyond the optimal resection of the primary tumor and modern use of systemic chemotherapy. Selected patients with colorectal cancer, small-bowel adenocarcinoma, gastric cancer, appendiceal malignancy, and peritoneal mesothelioma need to be considered for perioperative intraperitoneal chemotherapy washing. Not only patients who demonstrate low and moderate carcinomatosis should be treated. Those patients who are at high risk for developing local-regional disease need to be managed so that not only the gross visible disease, but also the intra-abdominal cellular component of the disease, is eradicated.
In my opinion, a transition needs to occur. No longer is cytoreductive surgery and intraperitoneal chemotherapy to be considered a research tool. It is the standard of practice that needs to be explored and optimized in patients at high risk for peritoneal dissemination and in patients with limited carcinomatosis. Of course, additional research is necessary to continue to optimize these management strategies. But what should the surgeon do as he or she treats the next patient at high risk for local-regional recurrence? Patients currently under treatment should not be suboptimally managed. This is a surgical responsibility; medical oncology referrals are not adequate. Cancer cells (what the surgeon does not see) are capable of killing the gastrointestinal cancer patient. Treatment options to prevent cancer progression at the resection site and on peritoneal surfaces are available and at reasonable risk.
Received for publication July 15, 2005. Accepted for publication July 25, 2005.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
