Validation of a rodent model of Barrett's esophagus using quantitative gene expression profiling.

BACKGROUND: A rodent model of gastroduodenal-esophageal reflux can result in replacement of squamous esophageal mucosa with intestinal-type columnar mucosa and carcinoma. The validity of this model is debated, as it is unproven whether this mucosa is intestinal metaplasia due to reflux or represents migration of adjacent jejunal mucosa above the anastomosis. The aim of this study was to evaluate the esophageal intestinal-type mucosa in these animals by measuring expression of trefoil factor genes (TFF-1, -2, -3) and comparing it with adjacent jejunum in order to determine its etiology. METHODS: Twenty-five rats underwent esophagojejunostomy at the ligament of Treitz to induce reflux of gastric and duodenal contents. The animals were sacrificed at 16 weeks (n = 14) and 30 weeks (n = 11). After sacrifice, the distal esophagus, jejunum, and colon were obtained. RNA was isolated, reverse transcribed, and messenger RNA (mRNA) expression of TFF-1, -2, and -3 was measured with real-time polymerase chain reaction (PCR). Linear discriminant analysis classified samples based on gene expression. RESULTS: Esophageal intestinal-type mucosa was present at sacrifice in 18 animals. Compared to jejunum, the expression of TFF-1 and TFF-2 mRNA in the intestinal mucosa of the distal esophagus was increased (p = 0.0007 and p < 0.0001, respectively). Expression of TFF-3 was also increased in esophageal intestinal mucosa compared with jejunum (p = 0.0002), but there was significant overlap in expression between these tissues for this gene. Linear discriminant analysis misclassified esophageal intestinal-type mucosa as jejunum in only one case. In no cases was jejunum misclassified as esophageal intestinal-type mucosa. CONCLUSION: The gene expression profile of esophageal intestinal-type mucosa following surgically induced reflux in a rodent model indicates that this represents intestinal metaplasia, not proximal migration of jejunum. This validates this model for studying the pathogenesis of Barrett's esophagus. Use of this model has potential for assessment of the impact of various therapies on the natural history of reflux disease.

Thymidylate synthase polymorphisms and mRNA expression are independent chemotherapy predictive markers in esophageal adenocarcinoma patients.

Thymidylate synthase (TS) is known to have polymorphisms in the 5' and 3' untranslated region (UTR). These polymorphisms have been reported to be associated with high TS expression and chemoresistance to 5-FU. The aim of this study was to examine the prognostic roles of the 5'-UTR and 3'-UTR TS polymorphisms in esophageal adenocarcinoma patients, as well as their relation with TS mRNA expression. Eighty-three patients with esophageal adenocarcinoma were assessed. Thirty-four had received 5-FU containing chemotherapy and 49 were treated with surgery alone. Surgically resected tumor tissues were analyzed for TS genotype and TS mRNA expression using a quantitative real-time RT-PCR method. No survival difference was seen between the patients with 3RG allele (3RG group) and non-3RG group among surgery-alone patients. However, among patients with a history of 5-FU-based chemotherapy, the non-3RG group showed significantly better overall survival compared to the 3RG group (p=0.02). Moreover, whereas chemotherapy produced a significant increase in survival for the non-3RG group patients, those in the 3RG group obtained no survival benefit from chemotherapy. When patients were classified by low or high TS mRNA expression levels, low TS expressers obtained survival benefit from chemotherapy while high TS expressers did not, although there was no difference of median TS mRNA levels between 3RG and non-3RG group. The 3'-UTR polymorphism was not associated with overall survival. These results suggest that the status of the TS 5'-UTR polymorphism and TS mRNA expression are independent predictive markers for survival benefit from 5-FU-based therapy.