Development of esophageal metaplasia and adenocarcinoma in a rat surgical model without the use of a carcinogen.

In order to establish an animal model for studying the cause and prevention of esophageal adenocarcinoma (EAC) and its frequent precursor, Barrett's esophagus (BE), factors affecting the pathogenic processes were investigated in an esophagoduodenal anastomosis model with rats. Experiments by us and others have shown that surgical treatment produced reflux esophagitis with cell hyperproliferation, but not EAC. Additional treatment with a carcinogen has been shown to be necessary for the development of EAC, squamous cell carcinomas (SCC) or EAC/SCC mixtures. We found that the surgically treated animals developed anemia due possibly to reduced iron absorption. When the operated animals were supplemented with iron, EAC occurred at a high rate (73%) after 30 weeks, and treatment with N'-nitrosonornicotine did not enhance the rate of tumorigenesis. Treatment with carcinogen, however, induced SCC in the group of rats killed after 22 weeks. The results suggest that iron overload, which is known to cause oxidative damage, is an enhancing factor for adenocarcinogenesis. The pathogenesis of EAC in the iron-supplemented, non-carcinogen treated group resembles human esophageal adenocarcinogenesis in many features. All the BE was the specialized type with goblet cells (containing sialomucin or sulfomucin) and columnar cells (containing acid or neutral mucin) as well as an incompletely developed brush border. Almost all of the BE was located at the bottom of the esophagus and was continuous with the duodenal mucosa; dysplasia became more frequent at later time points. All of the cancers were well-differentiated mucinous EAC, and most of the EAC had an adjacent area of BE with dysplasia. The results are consistent with the proposed human sequence for pathogenic events of BE progression to 'BE with dysplasia' and then to EAC. Esophagoduodenal anastomosis and iron treatment in rats produces a high rate of BE and EAC which are morphologically similar to human BE and EAC; this may be a useful animal model to study the development and prevention of EAC in humans.