DESCRIPTION (provided by applicant): The incidence of esophageal adenocarcinoma in the United States has risen 6-fold over the last 4 decades. Barrett's esophagus is a change in the lining of the esophagus that is a precursor to the cancer. Major factors associated with both Barrett's Esophagus and esophageal adenocarcinoma are heartburn and regurgitation, which are symptoms of gastroesophageal reflux disease (GERD). Roughly 20% of Americans have GERD symptoms on a weekly basis, and over 2 million upper endoscopies are performed annually in the U.S. in adults with GERD symptoms. Nonetheless, fewer than 15% of patients with esophageal adenocarcinoma have had an upper endoscopy prior to their diagnosis of the cancer. Patients with GERD symptoms are more likely to be referred to upper endoscopy if their symptoms have responded inadequately to proton pump inhibitors (PPIs). Approximately 50% of such patients do not have abnormal amounts of gastroesophageal reflux accounting for their symptoms, and often have psychological distress characterized by features of somatization, anxiety and depression. Clearly, there is a need for improved selection of patients for endoscopic screening. A quick, personalized clinical tool for identifying men at risk for Barrett's esophagus has been developed, the Michigan Barrett's Esophagus pREdiction Tool (M-BERET). The proposed study aims to validate that tool in a clinical population, extend it to women, and to determine if inclusion of specific circulating biomarkers and/or features of psychological distress can improve its accuracy. The study will enroll consecutive patients undergoing their first endoscopy and compare the accuracy of the tool to GERD symptoms alone for predicting the presence of Barrett's esophagus. Patients referred for their first endoscopic treatment of Barrett's esophagus with high grade dysplasia or intramucosal cancer will also be enrolled, and the accuracy of the tool will be assessed for discriminating these patients from patients with non-dysplastic Barrett's esophagus and no Barrett's esophagus. The study will also compare the accuracy of the tool to GERD symptoms for predicting the incidence of esophageal adenocarcinoma in a large retrospective longitudinal cohort. If the personalized tool is validated in the proposed study and widely used, it is expected to improve the allocation of endoscopy, decreasing over-utilization in low-risk patients, and increasing utilization in high-risk patients, ultimately decreasing the burden of esophageal adenocarcinoma in an efficient manner.

PUBLIC HEALTH RELEVANCE: The number of new cases of esophageal adenocarcinoma each year has risen 6-fold over the last decade. Its risk factors are common among veterans, including male sex, obesity, tobacco, and symptoms of gastroesophageal reflux disease (GERD). Endoscopic screening among individuals with GERD for the precancerous lesion of Barrett's esophagus is widely practiced. But screening based on GERD alone is very inefficient because GERD symptoms are very common and the cancer is uncommon: approximately 150,000 veterans receive a new diagnosis of GERD annually, and approximately 16% undergo upper endoscopy, but fewer than 1% of them are found to have esophageal cancer. And yet the vast majority of patients who develop the cancer never had undergone upper endoscopy prior to their diagnosis of the cancer. The proposed study aims to validate a quick personalized tool for predicting the presence of Barrett's esophagus and future esophageal adenocarcinoma that could greatly improve the efficiency of endoscopic screening among veterans.