Cryotherapy for Persistent Barrettʼs Dysplasia after Radiofrequency Ablation: If You Canʼt Burn It, Freeze It!

Abstract

Purpose: Cryospray ablation (CRYO) uses low-pressure liquid nitrogen for the treatment of dysplasia in Barrett's esophagus (BE). Current endoscopic therapy for Barrett's dysplasia is often performed with radiofrequency ablation (RFA) and endoscopic mucosal resection (EMR) for nodular lesions. However, there are patients who continue to have persistent dysplasia despite treatment with RFA and may benefit from CRYO. We present a case series of 5 patients who underwent CRYO for persistent high grade dysplasia (HGD) after RFA. Methods: We reviewed our prospectively-maintained database from 2007 to June 2012, to identify all patients who had CRYO as rescue therapy for persistent HGD in spite of serial RFA treatments at our tertiary referral center. Failure of RFA was defined as presence of histologically-confirmed dysplasia after at least two RFA treatments in the last 12 months. We retrieved baseline patient characteristics, endoscopic findings and histology. We assessed for the improvement in histology and regression in BE length before and after CRYO. Results: There were 5 patients with biopsy-confirmed HGD who received CRYO after failing RFA treatments. They were all males with a median age of 68 yrs. (range: 63-69 y.o.). The median size of their hiatal hernia was 5 cm (range: 3-6 cm). All patients underwent EMR for nodular lesions prior to RFA. Per standard clinical practice, RFA was initiated with HALO360 followed by HALO90, device with surveillance endoscopy and biopsies every 3 months. The median RFA session was 3 (range 3-4). After RFA, 3 patients had persistent HGD and 2 with low grade dysplasia (LGD). The median length of remaining BE segment post-RFA was 9 cm (range 7-10 cm). The patients were then treated with CRYO for a median of 3 sessions (median: 2-4) per standard dosimetry of 2 cycles for 20 seconds per application. All patients had follow-up biopsies after CRYO which showed 2 patients with eradication of dysplasia, 2 were downgraded to LGD, and 1 with persistent HGD. Overall improvement in histology occurred in 4 out of 5 patients. There was no progression to esophageal adenocarcinoma. The median post-CRYO BE length was 5 cm (range 0-8 cm). There was significant improvement in BE length after CRYO using a one-tailed Wilcoxon signed rank test (p= 0.03). There were no adverse events reported. Conclusion: In our series of 5 patients with HGD who failed to eradicate dysplasia after RFA, CRYO was a safe and effective ablative technique that improved their histologic grade and BE segment length. Prospective studies with adequate sample population and long-term follow-up will be needed to further assess the efficacy of CRYO in patients with dysplasia refractory to RFA. Disclosure: Dr. Wang - Consultancy and research support from Covidien, NinePoint Medical, CDX diagnostics, Fujinon, and Pinnacle. Dr. Iyer - Research support from Takeda. Ms. Lutzke - Research support from Covidien, Fujinon, and NinePoint. All rest of the authors have no disclosures.