| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
The Barrett Center for Cancer Prevention, Treatment and Research, Department of Surgery, Division of Surgical Oncology, University of Cincinnati College of Medicine, 234 Goodman Street, ML 0772, Cincinnati, Ohio 45219
Correspondence: Address correspondence and reprint requests to: Syed A. Ahmad, MD; E-mail: ahmadsy{at}uc.edu.
Over the last several decades, surgeons have sought to perform less invasive operations while still maintaining equivalent oncological results when compared with more radical procedures. An example of this is the use of breast-conservation surgery for breast cancer,1 transanal excisions of highly selected rectal tumors,2 and sentinel lymphadenectomy for breast cancer and melanoma. Similar trends have been demonstrated with regard to the extent of margin necessary for tumor clearance for operation on rectal tumors, liver metastases, and sarcomas.3 Treatment algorithms incorporating less radical surgery for early-stage ampullary tumors have also been reported, with mixed results.4 Despite this, many authors have advocated ampullectomy for early-stage tumors. Currently, many surgeons will consider ampullectomy for benign disease and T1 tumors <3 cm.
In this issue of the Annals of Surgical Oncology, Roggin et al.5 compare the perioperative and oncological benefits of ampullectomy in the treatment of nonfamilial ampullary neoplasms when compared with more radical surgery (i.e., pancreaticoduodenectomy). The authors also attempt to characterize the accuracy rate of both preoperative and intraoperative biopsies for periampullary neoplasms. The records of 128 patients with nonfamilial periampullary neoplasms treated over a 7-year period at Memorial Sloan-Kettering Cancer Center were reviewed. Overall, 29 patients underwent an ampullectomy and 99 underwent pancreaticoduodenectomy for their periampullary lesions. Pancreaticoduodenectomy was associated with a statistically significant increase in perioperative morbidity and mortality compared with ampullectomy. Preoperative biopsies identified only 76% of all patients with malignant lesions, and intraoperative frozen-section analysis had a sensitivity rate of 75%. The authors also analyzed recurrence and survival curves in patients with invasive periampullary cancers (n = 99). Two-year estimates of recurrence-free survival were 0% for the ampullectomy group (n = 8) and 48% for the pancreaticoduodenectomy group (n = 91). The 2-year disease-specific survival was 58% in the ampullectomy group and 78% in the pancreaticoduodenectomy group. Most recurrences in the ampullectomy group were locoregional. Multivariate analysis of the pancreaticoduodenectomy group demonstrated vascular invasion and T stage to be associated with lymph node involvement. Patients with lymph node involvement had decreased recurrence-free and disease-specific survival.
The obvious weakness of this study is the limited number of patients (n = 8) with cancer who underwent an ampullectomy as definitive treatment. In addition, the authors do not comment on the pathologic T staging of these patients; thus, drawing conclusions regarding the adequacy of ampullectomy for early-stage lesions is not possible. Despite this, the take-home messages are quite important: (1) preoperative and intraoperative biopsies can miss a significant number of patients with cancer, (2) locoregional failure is high in patients undergoing ampullectomy for cancer, (3) clinicopathologic variables are not accurate enough to predict the presence of nodal metastases, and (4) on the basis of the limited number of patients in this study, recurrence-free survival and disease-specific survival seem superior in patients who undergo more radical surgery.
These conclusions are similar to those in other recently published articles. Yoon et al.6 recently published the Seoul National University experience on 201 pancreaticoduodenectomies for ampullary tumors. In this study, 9% of Tis and T1 tumors and 51% of T2 tumors were found to harbor nodal metastases. Furthermore, 30% of tumors <2 cm were found to have lymphatic metastases. An interesting finding in this study was that the authors were able to demonstrate tumor infiltration along the lengths of the common bile duct or pancreatic duct on the average of 7.7 or 6.3 mm, respectively. This risk did not correlate with tumor size, grade, or morphology. This inability to determine the extent of locoregional and nodal disease is echoed in the current article. Roggin et al.5 demonstrated vascular invasion to be significantly related to lymph node involvement. However, this statistical correlation was offset by the fact that even in patients without vascular invasion, 28% harbored nodal metastases. In addition, 42% of T1 tumors were found to have lymphatic metastases. Finally, of the patients undergoing ampullectomy as definitive treatment of their cancer, 38% underwent an R1 resection, thus indicating an underestimation of their locoregional disease.
On the basis of these data, radical surgery should be performed to treat early-stage ampullary cancer. Comparing ampullectomy with other less invasive operations is not valid. For example, in breast cancer, the standard of care is to sample the axilla by using sentinel lymph node biopsy. If this is positive for metastatic disease, then a formal lymph node dissection is usually undertaken. This is currently not possible for ampullary neoplasms, and, as with other malignancies when the basin cannot be accurately assessed, a formal dissection should be undertaken for local control. Also, for tumors for which less invasive operations are used, effective systemic therapy usually exists. Currently, no effective adjuvant therapy exists for ampullary tumors. Thus, on the basis of the inability to assess the status of the lymphatic basin, high rates of locoregional failure, and the lack of effective systemic therapy, radical surgery is the only effective therapy we have to offer patients with early-stage ampullary tumors. Ampullectomy should be reserved for benign disease and in patients who are unable to tolerate extensive surgery.
Received for publication August 16, 2005. Accepted for publication August 22, 2005.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
