S980 Impact of Radiofrequency Ablation Energy Settings on Risk of Recurrence After Complete Remission of Intestinal Metaplasia in Barrett’s Esophagus

Introduction: The impact of focal radiofrequency ablation (RFA) energy settings on recurrence risk after complete remission of intestinal metaplasia (CRIM) is unclear following endoscopic eradication therapy (EET) for Barrett’s esophagus (BE). Methods: We retrospectively analyzed 580 patients who achieved CRIM following RFA for dysplastic BE or intramucosal adenocarcinoma (IMCa) at a high-volume center between 2003 and 2024. CRIM was defined as the first histologic confirmation of absence of intestinal metaplasia (IM) following EET. Recurrence was defined as any detection of IM, dysplasia, or IMCa on surveillance biopsies after CRIM. Patients were stratified based on the energy used: 12 J (12 Jx2 per treatment site covering entire BE segment, coagulum cleaned, then repeat 12 Jx2; N = 284), 15 Jx3 per treatment site covering entire BE segment; N = 95), or mixed/other (N = 201). Outcomes assessed included incidence of first and second recurrence and recurrence histology. A Prentice-Williams-Peterson conditional Cox model evaluated predictors of multiple recurrences. A secondary Cox model assessed predictors of second recurrence after second CRIM. Results: Two hundred twenty-five (38.8%) CRIM patients developed recurrence during a median follow up of 3.7 years (interquartile range: 1.3-6.7). The recurrence rates per 100 person-years were: 12 J (12.8), 15 J (15.6), and other (13.5). In the PWP model, increasing BE segment length (HR 1.06; 95% CI 1.03-1.10; P < 0.001) and smoking history (HR 1.46; 95% CI 1.10-1.93; P = 0.009) were significantly associated with recurrence, but not variability in RFA energy delivery. Of those with recurrence, 52 underwent repeat RFA and achieved second CRIM. Nineteen patients among these developed a second recurrence. Second recurrence rates per 100 person-years were: no RFA (16.5), 12 J (13.9), 15 J/other (21.2). In the Cox model for second recurrence, energy delivery was not associated with recurrence risk. However, a dysplastic first recurrence was associated with a higher likelihood of dysplastic second recurrence (16.7% vs 5.4%; P < 0.01). Conclusion: While RFA energy was not associated with recurrence, patients treated with 15 J had numerically higher recurrence rates. Smoking and longer BE segment length remained significant recurrence predictors. Importantly, dysplastic first recurrence substantially increased the risk of a dysplastic second recurrence, emphasizing the need for closer surveillance in this subgroup. Future prospective evaluation of energy dosing and recurrence risk is warranted.