Endoscopic resection in early Barrett’s esophagus-associated adenocarcinoma: does a universal endoscopic submucosal dissection approach make the cut?

10.1055/a-2767-6165

Endoscopic resection, either endoscopic mucosal resection (EMR) or endoscopic submucosal dissection (ESD), remains the standard of care for Barrett’s esophagus (BE)-associated early esophageal adenocarcinoma (EAC) [1] [2]. However, the preferred resection technique is often debated in educational forums. To expand our knowledge on this topic, Younis et al. performed a retrospective study from two academic centers, examining 311 early cancers that underwent endoscopic resection [3]. The technique chosen was at the discretion of the endoscopist, but per institutional practice, EMR was performed for suspected T1a lesions and ESD for suspected T1b lesions. Lesions >20 mm in size and with features of submucosal invasion were allocated to the ESD technique. The primary outcome was complete resection (R0) defined by a negative basal (deep) margin.

“Until there is clear prospective evidence that ESD is superior to EMR with regard to EAC recurrence or survival, it will not be the universal approach”

During 2009–2023, 178 EMRs and 133 ESDs were performed. Of the EMR specimens 148 (83%) were T1a lesions compared with 87 (65%) in the ESD group. High-risk lesions were defined as deep submucosal infiltration, advanced tumor grading, and lymphovascular invasion; which were equally present among both groups (33% vs. 39%; P = 0.23). A complete R0 resection was achieved for 80% of T1a and 27% of T1b lesions in the EMR group vs. 92% of T1a and 63% of T1b lesions in the ESD group. In multivariable logistic regression analysis, ESD was the only factor associated with an R0 resection. The authors concluded that given the challenges of correctly identifying T1a vs. T1b lesions, and that ESD is more likely to achieve an R0 resection, a universal approach using ESD should be employed.

We applaud the authors for adding to the literature on this controversial topic. The number of early EAC cases undergoing endoscopic therapy is certainly impressive in the expert centers. However, we want to express caution in endorsing a paradigm shift of a universal ESD approach in early EAC. When considering adopting a universal change in how early EAC is managed, it is important to evaluate the most essential or critical outcomes. In this study, R0 is the main measured outcome. Although this is an important outcome, the critical outcome that should be examined is the presence of EAC at 1–2 years after endoscopic eradication therapy, which is the outcome used in society guidelines for BE [2] [4]. The therapeutic end point in Barrett’s endotherapy is complete eradication of intestinal metaplasia with ablation therapy after resection of visible neoplasia. Thus, while ESD indisputably allows higher rates of R0 en bloc resection, subsequent ablation therapy likely diminishes this clinical benefit. The limited data that do exist with this critical outcome (one small randomized controlled trial favoring EMR and six observational studies) demonstrate no difference in EAC outcomes during follow-up between EMR and ESD groups [4] [5]. Results of an ongoing large randomized controlled trial comparing EMR with ESD in early Barrett’s EAC with this main outcome will help answer this question (NCT05276791).

Other factors also limit this study. Of patients allocated to the EMR cohort, 40% did not meet the EMR criteria as outlined by the authors. EMR was supposed to be performed on small superficial lesions that measured <20 mm in size. However, in the EMR cohort, 23% of the lesions were >20 mm, and thus met the criteria for ESD. EMR of a bulky lesion will fill the EMR cap, and the likelihood of obtaining a negative deep resection margin is lower. In addition, 17% of the EMR cohort had invisible cancer (no discrete lesion) compared with 4% in the ESD (P < 0.001). It is unclear how an endoscopist can adequately resect an invisible lesion. Ideally, these patients should be excluded from the analysis. Finally, the time period from 2009 to 2023 is wide. Endoscopic technology and techniques improved throughout this period. In fact, ESD became increasingly utilized toward the end of this period, together with the propagation of classification schemes to help identify neoplasia that could favor its outcomes. A sensitivity analysis with regard to year of procedure performed would have been a welcomed addition.

Although the evidence presented in this study may not support a universal approach for ESD in all early EAC, there is no denying a role for ESD in the management of dysplastic BE. As the 2023 European Society of Gastrointestinal Endoscopy and 2024 American Gastroenterological Association guidelines suggest, ESD may be used when lesions are bulky, greater than 2 cm, and for suspected submucosal involvement [1] [2]. Characteristics of submucosal involvement include depressed lesions (Paris 0-llc), lesions that do not lift, or poor mobility. EMR of these lesions would require a piecemeal approach whereas ESD provides an en bloc specimen that is better for staging purposes. However, as stated above, clear evidence for long-term recurrence and survival data is not yet available. ESD is also a great tool for recurrent, residual, or fibrotic lesions in which cap EMR would be technically challenging. Finally, ESD may be useful in patients who are poor candidates for subsequent ablation or may have adherence issues related to subsequent endoscopy. Fujiyoshi et al. found lower residual rates of neoplasia in patients undergoing ESD vs. EMR on biopsies performed after remission of neoplasia [6].

Currently most lesions in need of endoscopic resection can be handled by the EMR technique [2]. The technique is straightforward, less resource intense, and more accessible, and the procedure time is short and the learning curve quick [1] [2] [4]. ESD is technically demanding, and patients should be selectively referred to specialist centers for good outcomes. In addition, although EMR and ESD are both considered safe procedures, stricture formation may be higher following the ESD technique, especially for large lesions [2].

In summary, the debate regarding ESD vs. EMR for endoscopic resection of early EAC will continue to live on for now! While this study has added to our understanding, it has not definitively settled the question. Until there is clear prospective evidence that ESD is superior to EMR with regard to EAC recurrence or overall survival data, it will not be the universal approach, and hence does not make the cut today. Certainly, carefully selected lesions will benefit from ESD in expert centers.

Publication History

Article published online:
13 January 2026

• References

• 1 Weusten B, Bisschops R, Dinis-Ribeiro. et al. Diagnosis and management of Barrett’s esophagus: European Society of Gastrointestinal Endoscopy (ESGE) guideline. Endoscopy 2023; 55: 1124-1146
• 2 Rubenstein JH, Sawas T, Wani S. et al. AGA clinical guideline on endoscopic eradication therapy of Barrett’s esophagus and related neoplasia. Gastroenterology 2024; 166: 1020-1055
• 3 Younis F, Rösch T, Anders M. et al. Universal submucosal dissection outperforms selective mucosal resection in early esophageal adenocarcinoma. Endoscopy 2026; 58
• 4 Bos V, Nuehaus H, Pouw RE. Endoscopic mucosal resection and endoscopic submucosal dissection in Barrett’s esophagus. Gastrointest Endosc Clin of N Am 2026; 1: 211-223
• 5 Radadiya D, Desai M, Patel H. et al. Endoscopic submucosal dissection and endoscopic mucosal resection for Barrett’s-associated neoplasia: a systematic review and meta-analysis of the published literature. Endoscopy 2024; 56: 940-954
• 6 Fujiyoshi Y, Khalaf K, Tham D. et al. Recurrence following successful eradication of neoplasia with endoscopic mucosal resection versus endoscopic submucosal dissection in Barrett’s esophagus: a retrospective study. Endoscopy 2025; 57: 5-13