| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Surgical Oncology, Roswell Park Cancer Institute and State University of New York at Buffalo, Elm and Carlton Streets, Buffalo, New York 14263
Correspondence: Address correspondence and reprint requests to: John M. Kane III, MD; E-mail: john.kane{at}roswellpark.org.
Appendiceal neoplasms are a rare clinical entity that typify the real estate adage: location is everything. Similar to the pathophysiology of appendicitis, increasing intraluminal pressure from tumor obstruction or mucin production leads to appendiceal rupture with subsequent peritoneal dissemination (PD). By definition, this is considered stage IV disease for gastrointestinal cancers because of the poor prognosis associated with the clinical inability to control tumor progression. But does PD require the same inherent biological aggressiveness as lymphatic or hematogenous spread? The answer to this question is "not always." For example, accidental spillage of an appendiceal adenoma will lead to PD despite a negligible risk for hematogenous metastases. This may also be true for an early-stage colorectal cancer. In this issue of the Annals, there are two articles—one by Stewart et al.1 and one by Sugarbaker et al.2—regarding studies that used cytoreductive surgery (CS) and regional peritoneal therapy for PD from appendiceal neoplasms; they illustrate the evolution of a focused regional approach for, primarily, a locoregional problem. As previously noted by other authors, the treatment of PD from appendiceal neoplasms serves as an excellent example of Dr. Blake Cady’s philosophy: "In the field of surgical oncology, tumor biology is king, patient selection is queen, and technical maneuvers are the prince and princess who try, but usually fail, to usurp the throne."3
Historically, PD from appendiceal neoplasms has been categorized as either a nonmucinous high-grade tumor (similar to colorectal cancer) or as pseudomyxoma peritonei (PMP). Clinically, PMP presents as abundant intra-abdominal mucin secondary to disseminated tumor. Systemic chemotherapy has rarely been effective for this disease as a result of the overall slow growth of PMP in general, the ability of microscopic cells to survive in the peritoneal cavity by diffusion alone (i.e., no established blood supply), and limited drug penetration secondary to the plasma/peritoneal barrier. The majority of appendiceal PMP patients succumb to intra-abdominal complications (obstruction or perforation), malnutrition, or a declining performance status secondary to progressive, symptomatic mucin/tumor. Consequently, palliative surgical debulking initially evolved as the primary treatment modality to maintain a patent, functional gastrointestinal tract. Even after complete debulking of gross disease, microscopic tumor is present that invariably leads to recurrence.4 Attempts to eradicate this residual subclinical disease have led to the development of regional therapy approaches such as CS with intraperitoneal hyperthermic chem-operfusion (IPHC), also known as hyperthermic intraperitoneal chemotherapy, hyperthermic intra-operative intraperitoneal chemotherapy, continuous hyperthermic peritoneal perfusion chemotherapy, intraperitoneal hyperthermic perfusion, or hyperthermic antiblastic peritoneal perfusion.
Biologically, PMP is actually a spectrum of aggressiveness ranging from diffuse peritoneal adenomucinosis (DPAM) to peritoneal mucinous carcinomatosis (PMC or PMCA), with PMCA and intermediate/discordant features in the middle. In the classic pathologic description by Ronnett et al.,5 DPAM consists of scant, bland to low-grade adenomatous-appearing mucinous epithelium with abundant extracellular mucin/associated fibrosis but minimal cytological atypia or mitotic activity. DPAM is often limited to peritoneal surfaces, and nodal metastasis or parenchymal invasion of intra-abdominal organs rarely occurs. In contrast, PMCA consists of more abundant mucinous epithelium with glands and/or signet ring cells and enough atypia to be considered a mucinous carcinoma. Nodal metastases and superficial organ invasion are not uncommon. The pathologic subtype is also strongly associated with clinical prognosis, because the 5-year overall survival for PMP patients with DPAM is 75%, is 50% for PMCA-intermediate/discordant, and is 14% for PMCA.6
In contrast to other PMP studies, the results of Stewart et al.1 in this issue show that, although tumor histology strongly correlated with overall survival on univariate analysis, it was not an independent prognostic factor in the multivariate model. Interestingly, it appears that a higher grade tumor histology was associated with a poor preoperative performance status (which was an independent prognostic factor). CS/IPHC often entails several hours in the operating room, with multiple gastrointestinal resections, and a prolonged postoperative recovery. Complication rates are not nominal secondary to the extent of surgical resection combined with variable degrees of preoperative malnutrition. As a consequence, only patients "fit" for surgery typically undergo this procedure. At first glance, patient selection may appear to be a bias when trying to interpret the data from CS/IPHC studies. But, in reality, some component may actually be a surrogate for selecting the optimal tumor biology for this treatment.
It is important for surgeons to keep in mind that there are some variables over which we have no control (tumor biology), whereas others clearly fall under our influence (patient selection and the technical aspects of surgery). Both articles in this issue note remarkably similar and reasonable rates of major perioperative morbidity (38%–40%) and mortality (2%–6%) for CS/IPHC.1,2 But these results are not due to chance. Sugarbaker et al.2 have prospectively and systematically analyzed their outcomes over time and modified the perioperative care to minimize treatment-related toxicity. Factors under the control of the surgeon can also make a difference in clinical outcome/survival. Stewart et al.1 have shown that age at the time of perfusion (patient selection), performance status (patient selection), the completeness of surgical debulking (technical factor), and the duration of IPHC (technical factor) are all independent predictors of overall survival. The only variable not directly attributable to surgical decision making was the interval between diagnosis and treatment (tumor biology). Therefore, 80% of the factors associated with outcome could be influenced by the treating surgeon. Although it is easy to blame a poor clinical outcome on tumor biology (the king), preoperative decisions (the queen) and operative technique (the prince/princess) can still have a major impact in the treatment of many cancers (such as total mesorectal excision and rectal cancer recurrence).
So where does this leave us in the treatment of PD from appendiceal neoplasms? Both articles in this issue come from institutions with significant experience in the use of CS/IPHC for appendiceal PD. They have shown that tumor biology does play a major factor in the outcome from this disease. Therefore, should patients with a bad tumor biology and PD undergo CS/IPHC? The answer is "probably," but one must take this information into account when assessing a potential surgical candidate. This is analogous to determining the relative benefits of performing hepatic resection for colorectal liver metastases according to the number of lesions, their size, the disease-free interval, the primary tumor stage, and so on. The current authors have also shown that the perioperative morbidity and mortality associated with CS/IPHC are acceptable and comparable to those with the surgical treatment of other gastrointestinal malignancies, such as esophageal and pancreatic cancer (with an expected survival that is also similar to these other tumors, even in the potentially curative, nonmetastatic setting).
There are still a lot of unanswered questions regarding the role of CS/IPHC for isolated PD from appendiceal neoplasms and other cancers. What is the optimal temperature and duration of perfusion to maximize response? How much does the hyperthermia add to the treatment? Are there newer chemotherapeutic agents that could be used in place of mitomycin C? What other tumor histologies are amenable to this approach? What is the role, if any, of additional systemic therapy?
Despite the increasing use of CS/IPHC in the United States, Europe, and Asia, combined with multiple published phase I/II studies (including very large numbers of patients, such as the current one from Sugarbaker et al.2), a significant criticism of CS/IPHC continues to be the paucity of phase III randomized trials. This criticism is somewhat unfair in that not all clinical situations are amenable to a practical randomized trial. Where are the randomized phase III studies showing the superiority of hepatic resection for colorectal liver metastases over systemic chemotherapy alone? The role of metastasectomy in the armamentarium of cancer treatment has often evolved on the basis of a comparison of surgical outcome (with an understanding of tumor biology, patient selection, and the importance of minimizing perioperative morbidity/mortality) versus the expected natural history of the disease. Why do the same rules not apply for CS/IPHC? For more common causes of PD, such as colorectal cancer, performing a randomized phase III trial of no surgery versus CS/IPHC would prove to be clinically difficult, because patients are often hesitant to enroll in trials in which surgical therapy is determined by chance. In addition, one of the criticisms of the Verwaal et al.7 randomized phase III colorectal cancer CS/IPHC trial was actually a lack of a selection bias for the surgical arm, because there was no up-front decision that the extent of PD was technically amenable to a complete cytoreduction (which is associated with a better response to IPHC). A CS-alone versus CS/IPHC trial creates an ethical dilemma because colorectal cancer PD is not typically treated with surgical resection. This would subject patients to a therapeutic arm that is not similar to the current standard of care. This very issue may adversely affect accrual to the current National Cancer Institute trial of CS with or without IPHC in patients with PD from low-grade gastrointestinal adenocarcinoma (specifically, colorectal cancer). But PD from appendiceal neoplasms could ultimately salvage this scientific question. Given that the traditional treatment of PMP has previously been CS alone (which is still widely accepted by many surgeons), a randomized CS versus CS/IPHC trial in this patient population, looking at recurrence-free survival and quality of life, might actually accrue patients and ultimately satisfy the remaining critics of CS/IPHC for isolated PD. Maybe once and for all, we can begin to make sense out of what otherwise appears to be a clinical and therapeutic alphabet soup.
Received for publication January 5, 2006. Accepted for publication January 5, 2006.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
