| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||
Ann Thorac Surg 1996;62:1465-1466
© 1996 The Society of Thoracic Surgeons
DR THOMAS W. RICE (Cleveland, OH): This presentation addresses the timeless esophageal question: Which came first, the chicken or the egg? Do esophagogastric junctional carcinomas arise from gastric carcinomas originating in the cardia that eventually invade the esophagus or from esophageal carcinomas originating in the dysplastic Barrett's epithelium that eventually invade the stomach?
Iannettoni and associates found that two markers, p53 and SI, are found in similar proportions in both esophagogastric junctional carcinomas associated with Barrett's epithelium and those carcinomas that are not. Because these markers have been found in some patients with dysplastic Barrett's epithelium and Barrett's adenocarcinoma, this led Iannettoni and associates to propose that those carcinomas of the esophagogastric junction not associated with Barrett's epithelium, that express these markers, arose from Barrett's epithelium in which the specialized columnar lining cannot be detected at resection, therefore obscuring the origin of this subset of esophagogastric junctional carcinomas.
The problem with this argument is the assumption that p53 and SI are exclusive markers of Barrett's dysplasia and Barrett's adenocarcinoma. These markers are found in a wide variety of carcinomas, including gastric cancers that arise in areas removed from the gastric cardia. Mutations of the p53 gene have been found in similar proportions in both Barrett's adenocarcinomas and gastric carcinomas that arise away from the cardia. Sucrase isomaltase is expressed by both dysplastic and malignant Barrett's and gastric epithelium. If I may quote from the introduction of a previously published work from this group, "Sucrase isomaltase is expressed in both incomplete intestinal metaplasia in the stomach and gastric adenocarcinoma."
This leads to three questions. Given the similar occurrence of these markers in both gastric and esophageal adenocarcinomas, is not the following hypothesis as likely as that of the presentation: Tumors of the esophagogastric junction not associated with Barrett's epithelium that express p53 and SI arise from true gastric carcinomas that are also known to express these markers?
Second, even if the hypothesis of this presentation is correct, does it allow the assumption that these Barrett's adenocarcinomas arose in the gastric cardia as implied by the title of this presentation, which is "Barrett's Adenocarcinomas Represent a Subset of Tumors Arising in the Cardia"? Is it not just as likely that these tumors arose in the esophagus?
Third, have you considered the opposite approach, the identification of true gastric tumors that invade the esophagus by the investigation of markers specific to gastric tumors, such as bcl-2, or the identification of causative agents of gastric carcinomas, such as Helicobacter pylori?
DR IANNETTONI: Doctor Rice, thank you for reviewing our article. The definition of what constitutes a gastric cardia tumor can be relatively ambiguous. However, our study was limited to tumors that we have previously described as arising in the proximal 1 to 2 cm of the stomach and extending less than 3 cm onto the esophagus. We evaluated 208 adenocarcinomas of the esophagus to find 25 tumors that would fit these criteria.
Your comments about gastric metaplasia and gastric adenocarcinoma also expressing p53 and SI are correct. The tumors that extended on to the stomach greater than 3 cm were described as gastric cancers and were not included in this study. As well, patients who had signet ring carcinomas were excluded from the study, thus limiting us only to adenocarcinomas that arose in the gastric cardia or distal esophagus. Gastric adenocarcinoma could in fact be one of the possibilities for this; however, by our strict criteria we tried to exclude gastric tumors.
In response to your second question, could a cardia adenocarcinoma be an esophageal cancer growing down? This is the premise for our presentation as to whether or not Barrett's epithelium should be treated any differently based on location, or evaluated systematically no matter where it is located. Most of these tumors were less than 3 cm in size. By definition Barrett's esophagus must be at least 3 cm of specialized columnar epithelium proximal to the esophagogastric junction. Many of these patients without proximal extension had biochemical characteristics of Barrett's epithelium and were not allowed the benefit of surveillance endoscopy because of location of the Barrett's mucosa. Four of these cardia tumors did in fact have identifiable Barrett's epithelium less than 2 cm from the esophagogastric junction, and they were all SI positive.
In response to your last question, of trying to use SI as an indicator for Barrett's epithelium, 90% of Barrett's mucosa demonstrates SI positivity if you exclude all those patients with fundic or gastric mucosa and only include the ones who had intestinal-type Barrett's metaplasia. We thought that this was a significant marker. We have not looked at bcl-2 or any other gastric cancer markers in association with these tumors.
Tumors of the gastric cardia and distal esophagus may have identical origin. And arbitrary distinction based on location rather than precise histologic or immunohistochemical analysis is no longer appropriate. Whether these lesions arise from gastric cells or the distal esophageal mucosa should not be the underlying question, rather, it should be whether these markers SI and p53 are useful in diagnosing a premalignant condition involved with Barrett's mucosa. The answer still requires further investigation, but we can say that p53 and SI are expressed in a significant number of tumors arising in Barrett's epithelium.
Related Article
Ann. Thorac. Surg. 1996 62: 1460-1465.
| ||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
