Oncological impact of universal endoscopic submucosal dissection for large Barrett’s cancers

BACKGROUND: Oncological principles favour en bloc R0 excision for curative endoscopic resection. In Barrett's neoplasia, endoscopically curable cancers include T1a and selected early T1b disease. Although endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) are established treatments, optimal lesion selection remains debated. OBJECTIVE: To evaluate the oncological impact of two selective resection strategies: (1) prioritising ESD for suspected Barrett's cancers >15 mm and (2) a historical approach reserving ESD mainly for advanced cancers. DESIGN: Multicentre retrospective observational study comparing an ESD-first strategy (period 2, 2017-2024) with a historical selective ESD approach (period 1, 2004-2016). Lesion allocation was based on endoscopic assessment of invasion in both periods. Outcomes included basal R0 resection, curative resection, recurrence and adverse events. RESULTS: A total of 581 resections were performed in 542 patients (median lesion size 20 mm). Cancer was present in 271 cases (178 T1a and 93 T1b). Period 2 had a higher cancer burden (52.3% vs 34.9%) and greater ESD use (77.1% vs 21.2%). Basal R0 resection improved from 69.7% to 91.2% (p<0.001), with the greatest benefit in T1b lesions (33.3% to 81.9%, p<0.001). In T1b cancers, curative resection increased (9.5% to 30.5%, p=0.043) and recurrence decreased (55.6% to 23.6%, p=0.043). ESD achieved higher 2-year cancer-free survival than EMR (87.4% vs 50%, p=0.021). Adverse events were infrequent (2.2%) and similar between techniques. CONCLUSION: Prioritising ESD for Barrett's cancers >15 mm improves basal R0 resection, reduces recurrence and improves short-term survival for T1b disease, supporting routine ESD for all larger Barrett's cancers.