Multicenter Prospective Cohort on Esophageal Endoscopic Submucosal Dissection

In France in 2018, there were an estimated 2,074 new cases of esophageal adenocarcinoma and 3,224 cases of squamous cell carcinoma. The estimated deaths from esophageal cancer were 3,725, with a standardized 5-year net survival rate of 20% for cases diagnosed between 2010 and 2015, mainly due to late diagnosis. Surgery was historically the standard treatment for localized disease but carries significant morbidity. Over the past decade, endoscopic treatments, particularly endoscopic submucosal dissection (ESD), have become the reference approach for superficial esophageal cancers. After endoscopic resection, histological analysis allows classification of recurrence risk into very low, low, and high categories. Predicting lymph node or distant recurrence is complex, depending on factors such as depth of wall infiltration, lymphovascular invasion, and tumor differentiation. The frequent combination of unfavorable histological features may have led to an overestimation of lymph node involvement risk in T1b cancers. ESD is widely performed in France, with over 1,600 procedures reported in 2023 for esophageal and gastric lesions, demonstrating the feasibility of a large observational study. This multicenter French cohort will evaluate technical, oncological, and organizational outcomes of esophageal ESD, including overall survival, recurrence-free survival, and management of residual Barrett's esophagus. It will also identify predictive factors for treatment success, recurrence, and complications, providing real-world evidence to guide patient management.

Inclusion Criteria: * Adults (≥ 18 years old) * Patients referred for endoscopic resection of an esophageal lesion and treated by submucosal dissection * Patients managed in one of the participating centers * Patient who has received oral and written information about the study and has not objected to participation * Patients covered by a social security or health insurance scheme. Exclusion Criteria: * Patient refusal after reading the information sheet * Patient under legal protection (guardianship, curatorship, court-ordered protection) * Inability to inform the patient due to cognitive impairment, language barrier, or medical emergency * Patients previously included for another lesion in the same study * Pregnant or breastfeeding women. * Patient deprived of liberty. * Patient currently enrolled in another clinical trial.