The goal of this study was to identify a set of fluorescence in situ hybridization probes for the detection of dysplasia and adenocarcinoma in patients with Barrett's esophagus. We examined 170 brushing specimens from 138 patients with Barrett's esophagus or a history of Barrett's esophagus using fluorescence in situ hybridization with probes to 5p15, 5q21-22, centromere 7, 7p12, 8q24.12-13, centromere 9, 9p21, centromere 17, 17p13.1, 17q11.2-12, 20q13.2, and centromere Y. Receiver-operator curves were used to determine the sensitivity and specificity of various four-probe combinations for detecting low-grade dysplasia, high-grade dysplasia, and esophageal adenocarcinoma. Endoscopic biopsy results were used as the gold standard. Numerous four-probe combinations provided a similarly high sensitivity and specificity. Of these, a set consisting of probes to 8q24, 9p21, 17q11.2, and 20q13.2 was found to have a sensitivity and specificity, respectively, of 70% and 89% for low-grade dysplasia, 84% and 93% for high-grade dysplasia, and 94% and 93% for esophageal adenocarcinoma. This probe set was chosen for future prospective clinical evaluations based on its high sensitivity and specificity, its ability to distinguish adenocarcinoma and high-grade or low-grade dysplasia from lesser diagnostic categories, and the favorable signal quality for each of the probes. The majority of esophageal adenocarcinomas is thought to arise in patients with Barrett's esophagus. Barrett's esophagus is a preneoplastic condition caused by metaplasia of the normal squamous mucosa of the distal esophagus into specialized intestinal mucosa containing goblet cells.1 Individuals with Barrett's esophagus have a 30-to 60-fold elevated risk of developing adenocarcinoma and develop adenocarcinoma at a rate of Ïł0.5 to 1.0% per year of follow-up.
2-5The American College of Gastroenterology has recommended that Barrett's esophagus patients be monitored for the development of dysplasia and adenocarcinoma by undergoing regular endoscopic examinations of the esophagus and obtaining four-quadrant biopsies for every 1 to 2 cm of affected esophagus.6,7 However, there are problems associated with the use of biopsies for monitoring Barrett's esophagus patients for the development of neoplasia including: 1) limited sampling of the affected mucosa leading to false-negative biopsy results, 8 2) lengthy procedure time required to take fourquadrant biopsies every 1 to 2 cm, 7 3) poor interobserver reproducibility of pathologists for the diagnosis of dysplasia, 9 and 4) relatively poor ability of histological findings to predict which patients are likely to progress to adenocarcinoma.10 Assays that improve the ability to accurately detect dysplasia and esophageal adenocarcinoma as well as predict which patients will progress to adenocarcinoma could markedly improve the clinical management of these patients. Conventional cytology is not routinely performed on endoscopic brushings obtained from patients with Barrett's esophagus in most institutions, and there are rel...