| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIALS |
From the MD Anderson Cancer Center, Houston, Texas.
Correspondence: Address correspondence to: Robert A. Wolff, MD, MD Anderson Cancer Center, 1515 Holocombe Blvd., Box 426, Houston, TX 77030; Fax: 713-792-2828; E-mail: rwolff{at}notes.mdacc.tmc.edu
Strategies designed to improve surgical outcomes for patients with gastrointestinal malignancies have been under investigation for decades. Beginning as early as the 1980s, reports of large, well-designed clinical trials have established the benefit of postoperative therapy for some organ sites, such as the colon, the rectum, and more recently, the stomach.1–3 Generally, adjuvant therapy in the form of chemotherapy, or chemoradiotherapy, have yielded benefits both in terms of local control and survival. In pancreatic cancer, the advantage of postoperative therapy for patients undergoing resection has remained somewhat controversial, but in general, has been accepted as a reasonable standard practice. However, reports of adjuvant therapy for pancreatic cancer underscore the inherent difficulties of this approach. The logic behind neoadjuvant chemoradiotherapy is strong: treat all patients with potentially resectable disease, observe tumor biology during the time between diagnosis and completion of preoperative therapy, and resect those patients who remain with resectable disease and no overt evidence of metastatic disease. This strategy spares those patients who have aggressive tumors with rapid disease progression and the small minority of patients who demonstrate intolerance to preoperative treatment, thereby predicting high surgical risk. In this issue of Annals of Surgical Oncology, Dr. White and her colleagues4 suggest that preoperative 5-fluorouracil (5-FU)-based chemoradiotherapy has cytotoxic activity, and imply this will improve the outcomes for patients with resectable pancreatic cancer. Moreover, they report that neoadjuvant therapy has the potential to downstage some patients who initially present with unresectable disease. Unfortunately, the data provided to support these claims is suspect. Nevertheless, the article confirms the findings of other studies, and sets the stage for common endpoints in future investigations of this approach.
The article describes two groups of patients, those who present with resectable disease, and those who present with localized, unresectable disease. After neoadjuvant therapy, there seems to be cross-over between these two groups. This is confusing because some patients who were initially deemed resectable developed evidence, by computed tomography (CT), of unresectable disease, and yet were still resected. In addition, some patients who were classified as having unresectable tumors remained unresectable after chemoradiotherapy, but were also resected. Understanding the definition of resectability in this context is challenging. Therefore, it may be best to separate these two groups for purposes of discussion.
Outcomes of Patients Who Present With Potentially Resectable Disease
The article by White et al. reports on 53 patients who presented with potentially resectable disease. Of those, 53% ultimately underwent resection, sparing almost half the group an unnecessary surgical intervention. Two patients died during the course of chemoradiotherapy, and one patient could not tolerate the therapy, thus "failing" the will-to-live test. Ten patients were found to have metastatic disease in the interval from diagnosis to restaging. Of note, 9 of the 53 patients had local tumor progression that met criteria for unresectability, but 2 of these 9 were still resected. At our institution, two separate trials of 5-FU-based chemoradiotherapy led to resection rates between 57% to 60% among patients deemed to have resectable tumors.5,6 In one cooperative group study, the resectability rate was lower, at 45%.7 The Duke experience (White et al. article) falls within this range when 5-FU-based treatment is used. Previous reports of preoperative therapy demonstrate that toxicity remains formidable with hospitalization rates ranging from 11% to 51%, and in this study, it was as high as 52%. Of note, the reported median survivals for resected patients range from 15 months to 25 months. Although the median survival has not been reached in the current report, figure 4 of the White et al. article suggests the survival benefit of preoperative therapy is not likely to be superior to postoperative therapy for those patients who ultimately undergo resection. The authors claim the preoperative therapy led to a treatment effect, with "fibrosis" observed within the surgical specimen. Caution is required in interpreting this, because these tumors can be associated with a significant fibrotic component without prior therapy. That said, it does appear that the margin status of resections performed after chemoradiotherapy was superior to those achieved with initial surgery. The two observed pathologic complete responses support this claim. Unfortunately, whereas a negative surgical margin may be necessary for improved survival, it is obviously not sufficient, because the majority of patients ultimately succumb with distant metastatic disease.
Downstaging Patients With Locally Advanced Disease
In my experience with chemoradiotherapy, downstaging patients with locally advanced (LA) pancreatic cancer remains anecdotal. White et al. suggest that almost 20% of patients who were defined as having LA disease were downstaged sufficiently to undergo resection. A closer look at the group defined as having LA disease discloses two separate patient populations: those with tumor abutting the superior mesenteric artery (SMA) or celiac artery, and those with encasement of these vessels. Although not a surgeon, this suggests to me two distinct anatomic settings in which the ability to obtain a negative margin is at least feasible in the former group, and impossible in the latter. It seems that only those patients with abutting tumors were considered for surgery, and the authors report tumor downstaging in some of them. However, I am not certain whether this means that the overall tumor mass diminished with the therapy, or the portion of the tumor abutting the vessel vanished. Reading between the lines suggests the surgeons were willing to explore those patients who I describe as having marginally resectable tumors. Typically, these are tumors that are abutting the SMA, but not encasing it. In support of this, of the 58 patients who were deemed as having unresectable disease, 39 remained so after chemoradiotherapy, yet 8 of these patients underwent resection of their tumors.
Where does all this leave us? Preoperative 5FU-based chemoradiotherapy clearly identifies poor surgical candidates through observation of tumor biology and observation of patient tolerance. It may enhance the margin status of patients undergoing resection. These are worthwhile efforts. However, the Duke experience does not support neoadjuvant therapy as a way to improve survival for patients with potentially resectable disease, and I remain skeptical as to its ability to downstage patients. The White et al. article does confirm the findings of other groups and identifies the endpoints (resectability rate, hospitalization rate, and survival) for future trials that use other radiosensitizers, such as gemcitabine,8 or novel therapies designed to prevent or impede the onset of distant metastatic disease.9 In the future, consistent use of staging modalities, such as thin-cut dynamic phase CT scans, with or without laparoscopy or peritoneal washings, will be preferred.
Received for publication September 10, 2001. Accepted for publication October 2, 2001.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  J. M. Pipas, R. J. Barth Jr., B. Zaki, M. J. Tsapakos, A. A. Suriawinata, M. A. Bettmann, J. M. Cates, G. H. Ripple, J. E. Sutton, S. R. Gordon, et al. Docetaxel/Gemcitabine Followed by Gemcitabine and External Beam Radiotherapy in Patients With Pancreatic Adenocarcinoma Ann. Surg. Oncol., December 1, 2005; 12(12): 995 - 1004. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
