The study will compare EMR versus ESD technique (both combined with subsequent ablative therapy) of mucosal resection in Barrett's esophagus with regard to efficacy and risk in a long term setting.

For Barrett's Esophagus neoplasia of at least LGIN up to early adenocarcinoma, the aim is to debulk or completely treat polypoid dysplastic or malignant lesions in Barrett's esophagus. The Endoscopic Mucosal Resection EMR has been established to be a less invasive, safe, and effective nonsurgical therapy. The most commonly employed modalities of EMR include snare resection with and without prior submucosal injection of fluid, and resection using a cap. Since resection of larger areas can only be done piece - by- piece this kind or resection is also called piecemeal resection or piecemeal EMR. Meanwhile, another endoscopic resection has been developed called Endoscopic Submucosal Dissection ESD.It enables complete resection of neoplasms that were impossible to resect en bloc by EMR. After circumferential cutting of the surrounding mucosa of the lesion, fluid is injected into the submucosa to elevate the lesion from the muscle layer, and subsequently the connective tissue beneath the lesion is dissected. As a basic principle on histopathological and oncological terms, the en bloc resection is to be preferred since resection integrity can be evaluated much more securely. However, complexity of this kind of resection technique as well as complication rates can be different and sometimes higher than with EMR. Current approach treating Barrett's esophagus is to eradicate both neoplastic as well as pre neoplastic or non neoplastic Barrett mucosa in order to lower the relapse risk. Current treatment standard is to combine resection of visible neoplastic areas with subsequent thermo-ablation such as RFA or APC, so this approach will also be the basis of the present study. Since RFA has the largest volume of data screened it shall be the preferred method of ablation in this study.In total, data situation ist inconsistent. Short- and Long term of EMR is excellent in centres(Pech et al, Gastroenterology 2014) whereas ESD achieved only suboptimal outcomes in tree minor western studies (Neuhaus et al. Endoscopy 2012, Höbel et al., Surg Endosc 2015, Chevaux et al. Endoscopy 2015). One randomised study published in 2016 (Terheggen et al. Gut 2016) had a higher rate of R0 resections with ESD on 40 patients but no difference in complete remissions in combination with RFA. Although, this study was not empowered sufficientliy, and also showed a higher complication rate on ESD . At present no randomised study data are availale to allow statements about long term developments, so we will set up this current randomised study. We will compare data with regard to efficacy (histological completeness and relapse rates), as well as risks, e.g. perforations and strictures or stenosis by scarring.

Inclusion Criteria:

• patients to be treated for Barrett's esophagus by mucosal resection and following ablative therapy
• Barrett's mucosal extension up to 10 cm maximum.
• patient's ability for compliance to therapy
• signed Informed Consent

Exclusion Criteria:

• any lesion questionable to be resectable by mucosectomy, e.g. bulky lesions ≥10 mm in endoscopy und endosonography, suspected deep submucosal infiltration, ulcers, suspected or by FNA confirmed lymph node infiltration
• Barrett's esophagus > 10 cm
• lesions that would afford resection of more than 2/3rd of esophagal circumference
• two or more single Barrett's lesions with bulky HGIN or early cancer histology, not to be resectable in one half of esophageal circumference
• planned circumferencial resections
• very serious general illness and metastatic carcinoma
• coagulation disorder or anticoagulants that make biopsies and resections impossible
• American Society of Anesthesiologists (ASA) status > III
• pregnancy and lactation
• remainders or recurrences after therapeutic history of Barrett's espohagus