Initial Treatment With Endoscopic Mucosal Resection Decreases the Incidence of Esophageal Adenocarcinoma in Patients With Barrettʼs, With High-Grade Dysplasia

Abstract

Introduction: Endoscopic mucosal resection (EMR) is effective for the removal of nodular neoplastic mucosa in Barrett's esophagus (BE) with high grade dysplasia (HGD). While EMR is commonly performed prior to ablation, the impact of early EMR on the subsequent risk of esophageal adenocarcinoma (EAC) after radiofrequency ablation (RFA) is unknown. We compared outcomes of patients with HGD who had EMR then RFA to patients with HGD who had RFA alone. Methods: The U.S. RFA Registry is a study of RFA for BE at 148 centers. Patients were enrolled from 2007-2011 and followed prospectively until 2014. For inclusion in the current investigation, patients had HGD and were free from EAC at baseline. We compared HGD patients receiving EMR prior to RFA to HGD patients receiving RFA alone using Wilcoxon rank sum test for continuous variables and chi-square testing for categorical variables. Logistic regression models were constructed to assess odds ratios (OR) for EAC. Results: Among 5521 patients in the registry, 974 (18%) had HGD. Of these 974, 237 (24%) underwent EMR prior to RFA. Patients who underwent EMR prior to RFA did not differ from those who got RFA alone on age, race, gender, frequency of complete eradication of intestinal metaplasia (CEIM), or time in study, but did have shorter BE at baseline (4.7 vs 5.3 cm, p = 0.008) (Table). Amongst all patients with HGD, 83 (9%) developed EAC. Patients undergoing EMR prior to RFA were less likely to develop EAC (5% vs 10%, p = 0.03) or invasive EAC (2% vs 5%, p = 0.04). On logistic regression, after controlling for age, race, gender, BE length, and time in study, patients who did not undergo initial EMR had doubled odds of developing EAC (OR 1.9, 95% CI [1.01-3.6]). The impact of EMR on invasive EAC was similar in magnitude but did not reach significance after adjustment for the same factors (OR 2.7, 95% CI [0.94-7.7]). Conclusion: After controlling for multiple potential confounders, patients with HGD who received EMR prior to RFA were less likely to develop EAC than HGD patients treated with RFA alone. Since nodularity within BE is a risk factor for EAC, this finding is counter-intuitive. The performance of EMR may be a marker for closer mucosal inspection, leading to superior outcomes. Alternatively, facilities able to perform EMR may also perform more effective RFA. Facilities not able to perform EMR may be more likely to perform RFA over areas of nodularity. Early EMR appears to have a role in protecting against EAC.Table 1: Patient Characteristics by EMR performed prior to RFA