The purpose of this study is to determine whether High Resolution Magnification Endoscopy (HRME) and Computed Virtual Chromoendoscopy (CVC) with targeted biopsies is superior to conventional white light endoscopy (WLE) with 4 quadrant biopsies of the metaplastic epithelium every 1-2 cm (Seattle Protocol) for detection of pre-malignant lesions in patients with Barrett's Esophagus (BE).

BE is a metaplastic mucosal transformation adjacent to the esophagogastric junction, due to chronic reflux of gastric juices, Gastro Esophageal Reflux Disease (GERD). The continuous esophageal exposure of acid- and/or bile- containing fluids leads, untreated, to chronic esophagitis. In certain patients a mucosal transformation takes place. The epithelium in the distal part of esophagus is transferred from squamous into a more intestinal-like mucosa, called Specialized Intestinal Metaplasia (SIM).Patients with BE is believed to run a higher risk of developing esophageal adenocarcinoma (EAC).

EAC is a rare condition in the western society, but the prevalence is rising compared with other malignancies, and a substantial increase has been seen during the last four decades. The pathogenesis of cancer development is believed to be that SIM in some patients can undergo dysplastic transformation, from low to high grade, and from high grade dysplasia (HGD) develop into AC. Advanced EAC is associated to a poor prognosis whereas HGD or carcinoma in situ may be treated endoscopically with a favorable outcome.

The need for surveillance endoscopy in order to discover early cancer lesions available to curable treatment is up against cost effectiveness and evidence level regarding screening. The conventional endoscopic (CE) surveillance algorithm for BE is standard WLE and 4 quadrant biopsies of the metaplastic epithelium every 1-2 cm above the esophagogastric junction. The development of advanced endoscopic techniques have made it possible to distinguish minimal polypoid lesions but also the microvasculature and pit-pattern structures that in certain grading systems have been associated to presence of dysplasia. Attempts have been made in exploring the benefits of advanced endoscopic technologies against standard WLE. Feasibility-studies suggests that the new techniques improves the biopsy-yield for dysplasia, however only a limited number of prospective studies exist.

Study aim: To determine whether HRME and CVC with targeted biopsies is superior to conventional WLE with 4 quadrant biopsies of the metaplastic epithelium every 1-2 cm (Seattle Protocol) for detection of pre-malignant lesions in patients with BE.

Primary endpoint: Incidence of detected dysplasia by each endoscopic technique. Secondary endpoints: 1.The yield of low- and/or high-grade dysplasia by each endoscopic technique. 2.The number of biopsies taken and the duration of the different endoscopic techniques. 3.The endoscopic prediction capability of present dysplasia compared to histopathology for HRME.

Statistical power: Based on the primary endpoint, the amount of dysplasia in a Barrett-population is approximately 10%. We calculated a raise in positive yield with using advanced endoscopy to 8%. At p<0,05 and a power of 80% the need for 105 patients.

Setting: Tertiary referral high volume endoscopy center at Sahlgrenska University Hospital, Sweden.

Inclusion Criteria:

• Presence of specialized intestinal metaplasia in biopsies from the esophagus

Exclusion Criteria:

• Dysplasia or cancer