Radiofrequency Ablation Combined With Endoscopic Resection Is Highly Effective for Eradication of Early Barrett’s Neoplasia: Final Results of a Large Prospective European Multicenter Study (EURO-II) Nadine Phoa*, Roos E. Pouw, Raf Bisschops, Oliver Pech, Krish Ragunath, Bas L. Weusten, Brigitte Schumacher, Bjorn Rembacken, Alexander Meining, Helmut Messmann, Erik J. Schoon, Liebwin Gossner, Jayan Mannath, Cees a. Seldenrijk, Mike Visser, Antoon E. Lerut, Thomas RoSch, Stefan Seewald, Fiebo J. Ten Kate, Christian Ell, Horst Neuhaus, Jacques J. Bergman Gastroenterology and Hepatology, Academic Medical Center, Amsterdam, Netherlands; University Hospital Leuven, Leuven, Belgium; Dr. Horst-Schmidt-Kliniken, Wiesbaden, Germany; Nottingham Digestive Diseases Centre, Nottingham University Hospital, Nottingham, United Kingdom; Gastroenterology and Hepatology, St Antonius Hospital, Nieuwegein, Netherlands; Evangelisches Krankenhaus Dusseldorf, Dusseldorf, Germany; The General Infirmary at Leeds, Leeds, United Kingdom; Technical University of Munich, Munich, Germany; Klinikum Augsburg, Augsburg, Germany; Catharina Hospital, Eindhoven, Netherlands; Klinikum Karlsruhe, Karlsruhe, Germany; Pathology, St Antonius Hospital, Nieuwegein, Netherlands; Pathology, Academic Medical Center, Amsterdam, Netherlands; Universitatsklinikum HamburgEppendorf, Hamburg, Germany Background: Radiofrequency ablation (RFA) after prior endoscopic resection (ER) of focal lesions, has proven highly effective for Barrett’s esophagus (BE) containing high-grade dysplasia (HGD) or early cancer (EC), in a number of small-sized single center studies. Aim: To prospectively evaluate the efficacy of RFA combined with ER in case of focal lesions, for BE with HGD/EC in 13 European centers with expertise in BE neoplasia. Methods: Patients with BE 12 cm and HGD/EC on 2 separate endoscopies were included. Visible lesions ( 2cm length; 50% circumference) were removed with ER, residual EC was excluded on 2 mapping endoscopies post-ER. Subsequent RFA was scheduled every 3 months until clearance of BE was achieved, with max 5 RFA sessions allowed. Escape treatment was permitted for residual BE after RFA (max 2 APC sessions for islands 5mm, ER for islands 5mm or suspicious lesions). Followup (FU) endoscopy was scheduled at 3 9 mo after last treatment and annually thereafter, with 4Q/2cm biopsies from neosquamous epithelium and gastric cardia. Endpoints: complete eradication of neoplasia (CE-neo) and intestinal metaplasia (CE-IM); durability of CE-neo/CE-IM. To ensure uniformity and protocol compliance, investigators were trained at the coordinating site and a study monitor attended all treatments and first FU on-site. Central pathology review of all ER/biopsies was performed at the coordinating site. Results: 132 patients (107M, mean 65yrs, median BE C3M6) underwent en-bloc (n 63) or piecemeal ER (n 56); or no-ER (n 13). Worst ER histology: EC (n 78), HGD (n 31), LGD (n 7), no dysplasia (n 3). Worst grade post-ER/pre-RFA: HGD (n 36), LGD (n 45), no dysplasia (n 51). 124 pts completed the treatment phase, 8 discontinued due to unrelated causes. After a median of 3 (IQR 3-4) treatments, including ER (n 14) or APC (n 14), per intention-to-treat analysis (counting drop-outs as failures) CE-neo was reached in 122/132 (92%) and CEIM in 115/132 (87%) pts. In a per-protocol analysis (censoring for drop-outs) CEneo/CE-IM were achieved in 98% and 93%, respectively. Of 2 CE-neo failures, 1 was referred for surgery (T1bN0M0), 1 patient was treated endoscopically (offprotocol). Of 115 pts who reached CE-neo/CE-IM per-protocol, CE-neo was maintained in 112/115 (97%) pts during median 25 (IQR 18-34) mo FU since last treatment, with median 4 (2-5) FU endoscopies and 41 biopsies per patient. 3 pts with recurrent neoplasia (EC n 1; HGD n 2) were effectively re-treated by ER or APC. Conclusion: This is the largest prospective multicenter trial on RFA combined with ER for treatment of BE with HGD/EC. Our outcomes suggest that if performed by trained, expert endoscopists in carefully selected patients, this combined approach is highly effective, and appears durable in the majority (97%) of patients once complete eradication of neoplasia and IM is established. Fig 1. Inflammation score (range 0-13) of inflamed pediatric Crohn’s disease ileum increased compared to both non-inflammed Crohn’s diseae and control. Results expressed as mean SEM (*P 0.001).
Read full abstract