Barrett's Esophagus: Endoluminal Therapy Safe, Effective

Troy Brown

February 15, 2013

Endoluminal therapy combining endoscopic mucosal resection and ablation is successful long-term in patients with Barrett's esophagus (BE) and high-grade dysplasia (HGD) or intramucosal carcinoma, according to a retrospective cohort study of 166 patients with dysplastic BE.

Carlos Guarner-Argente, MD, PhD, from the Department of Medicine and the Division of Gastroenterology at the University of Pennsylvania Perelman School of Medicine in Philadelphia, and colleagues report their findings in the February issue of Gastrointestinal Endoscopy.

Standard treatment for BE and HGD or intramucosal carcinoma is esophagectomy, but endoluminal therapy is less invasive than surgical resection. Patients included in the study received endoluminal therapy consisting of resection and photodynamic therapy, radiofrequency ablation, and/or argon plasma coagulation treatment.

Dysplasia and neoplasia were completely eradicated in 95% (157/166) of patients, and intestinal metaplasia was completely eradicated in 83% (137/166) of patients in an intention-to-treat analysis.

The 156 patients who completed treatment were followed-up for an additional 33 months (interquartile range [IQR], 18 - 58 months) and underwent final endoscopy at 26 months (IQR, 15 - 21 months).

Intestinal metaplasia, dysplasia, or adenocarcinoma recurred in 66 cases (42%). In 13 of those 66 patients (20%), repeat recurrence occurred after interval re-eradication.

Among those in whom intestinal metaplasia was completely eradicated, intestinal metaplasia recurred in 35% (48/137), and dysplasia recurred in 9% (12/137). Dysplasia recurred in 32% (6/19) of those with complete eradication of dysplasia and neoplasia only.

Recurrence occurred 16 months after the end of treatment (IQR, 11 - 29), but during the third year or later in 20% and after 5 years in 5%. Intestinal metaplasia recurred in 35% of patients, and dysplasia or carcinoma recurred in 12%. Recurrence occurred more often in those without complete elimination of intestinal metaplasia.

The researchers conducted univariable and multivariable analyses of predictive factors for recurrence. They controlled for age, multifocal dysplasia, and complete eradication of intestinal dysplasia. Multifocal dysplasia and patient age were risk factors for dysplasia and/or carcinoma recurrence in the multivariable and univariable analyses. Complete eradication of intestinal metaplasia was protective for dysplasia and/or carcinoma recurrence in the univariable analysis only.

Treatment-related adverse events occurred in 23.9%; stricture occurred in 11.9%. Most events were minor and were resolved with endoscopic or supportive therapy. One death (0.6%) occurred that was related to treatment.

Thomas M. Deas Jr, MD, MMM, president of the American Society for Gastrointestinal Endoscopy, commented on the study in a telephone interview with Medscape Medical News.

Clinicians "should be more aware of these minimally invasive techniques and then consider them in lieu of the higher-risk surgery," Dr. Deas said.

Patients also need close follow-up. "Once the endoscopic therapy is completed and these patients return to their referring physicians, they should be monitored at intervals to detect recurrence, which may develop in up to a third of these patients," Dr. Deas explained.

Dr. Deas also urged caution when referring patients for evaluation for endoluminal therapy. "When we refer our patients, we should refer them to endoscopists who have experience with these procedures.... These techniques are not common to all gastrointestinal endoscopists, so you should select an endoscopist who has special training and experience," Dr. Deas concluded.

Dr. Guarner-Argente received financial support from the Instituto de Salud Carlos III, Government of Spain, and from the Societat Catalana de Digestologia. One coauthor is a consultant for CDX and C2-Theraptutics. The other authors and Dr. Deas have disclosed no relevant financial relationships.

Gastrointest Endosc. 2013;77:190-199. Full text