| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIAL |
From the MD Anderson Cancer Center, Houston, Texas.
Correspondence: Address correspondence to: Paul F. Mansfield, MD, MD Ancerson Cancer Center, 1515 Holcombe Blvd., 444, Houston, TX; Fax: 713-745-4926; E-mail: pmansfiel{at}mail.mdanderson.org
The article by Oñate-Ocaña et al. in this issue of Annals of Surgical Oncology examines not only the role of laparoscopy in the management of patients with gastric cancer but also proposes a pretreatment staging system based on laparoscopic findings. The role of laparoscopy in fully evaluating patients with gastric cancer is widely accepted. Patients are typically classified as either resectable (no evidence of distant disease) or unresectable. Our group and others have demonstrated significant benefits for patients in whom other staging modalities have indicated resectable disease.1,2 Unnecessary laparotomy is avoided and more cost-effective patient management is realized. The current study, as well as previous ones, has demonstrated a diagnostic accuracy of laparoscopy in excess of 90%.
The appropriate decisions regarding this technique are to whom should it be applied and how much information can be gleaned from the procedure. The authors describe selection criteria of tumors <4 cm or T1 lesions presumably based on endoscopic ultrasound evaluation. These are certainly reasonable criteria and similar to the ones used elsewhere. The incidence of peritoneal disease in patients who are EUS T-1 is not precisely known, but must be quite low. Our practice is to consider laparoscopy in all patients with an EUS T-stage >1, and who are otherwise considered potentially resectable. This is true whether the patient is being considered for neoadjuvant therapy or immediate resection.
It is of concern in this study that only 65% (63 of 96; 61% if all 104 patients are considered) of patients who underwent laparoscopy and were then explored were able to undergo an R0 resection. This included a resectability rate of <50% in laparoscopic stage III patients. Although the purpose of laparoscopy is to detect disease not seen on computed tomography (CT) scan, in this study, it is unclear how three patients had peritoneal disease so extensive that the peritoneal cavity could not be entered without such disease being seen on CT imaging. It would be helpful to know what findings at laparotomy (and not seen at laparoscopy) deemed the patients unresectable because it is stated that there was only one false-negative laparoscopic evaluation. It would be important to understand why patients with a positive laparoscopy underwent laparotomy anyway, but were then not resected. Also two false-positive laparoscopic evaluations were reported which underscores the importance of pathologic confirmation of any laparoscopic findings. One should not simply rely on what one sees.
The authors define a four group staging system: stage I, no serosal involvement; stage II, serosal involvement; stage III, adjacent organ invasion; and stage IV, distant disease found at laparoscopy. The pivotal determination for the proposed staging system is serosal involvement by tumor. The authors correctly predicted T-stage in 53 (84%) of 63 patients in whom a resected specimen was available. Most of these patients in fact had a T-4 lesion. Thirty-seven of the 63 patients (59%) had either a suspected or confirmed T-4 lesion. This is an extraordinarily high percentage of patients with advanced lesions for a potentially resectable patient population. In comparison, in the German Gastric Cancer Study Group report by Siewert et al.,3 only 131 of 1654 (13.6%) resected patients had T-4 lesions. This included 472 patients who did not have an R0 resection. Obviously, for patients with locally far-advanced disease CT and ultrasound should provide considerable information. The ability of the surgeon to predict serosal involvement either with the laparoscope or under direct vision is quite variable.4,5 Possik et al.4 reported the sensitivity and specificity of detecting serosal involvement with laparoscopy to be 88.6% and 75%, respectively, whereas Perng et al.5 reported correct T stage assessment by direct intraoperative examination to be correct in only 55% of cases. Clearly the ability to visually determine T stage accurately is quite variable.
The authors are correct in citing the relatively poor accuracy of current attempts at preoperative evaluation of N-stage. This is further complicated by the shift (and appropriately so) of the pathologic nodal staging from one based on location of nodes to one based on number of nodes. It would also be helpful to know how frequently lymph node biopsy was used in this study and how it impacted the staging. Certainly lymph node biopsy adds to the potential information, but also to the complexity (particularly if one is attempting to biopsy a node around the celiac axis). The authors correctly point out the need in a staging system, which lacks the complete surgical specimen, to obviate nodal evaluation. Perhaps molecular correlates of nodal disease could be substituted in such a system. Plasminogen activator inhibitor 1, c-erbB-2, CD44, and E-cadherin have all been found to further delineate higher risk patients.6 Kurozomi et al.7 reported that 88% of patients with CD4 variant 6 and lymphatic invasion harbored nodal metastases. These may help with identifying and staging patients most accurately in neoadjuvant therapy trials.
The utility of a staging system is dependent on several factors including: ease of application, clinical relevance, and reproducibility. The Oñate-Ocaña et al. report addresses the first two of these factors. Despite the above mentioned concerns, the authors’ principle effort, to propose a staging system which is useful in fully evaluating patients without having the benefit of a resected surgical specimen, is important. We have seen in other diseases the evolution of staging systems that incorporate both clinical and pathologic features. It is critical, given the increasing interest in neoadjuvant therapies, to develop the best staging system possible for the evaluation of patients entering onto neoadjuvant therapy trials. This will permit the most accurate evaluation of these treatment strategies. All staging systems undergo change and modification. This is a good start.
Received for publication July 9, 2001. Accepted for publication July 16, 2001.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
