The use of endoscopic mucosal resection (EMR) before radiofrequency ablation significantly reduced the risk for treatment failure among patients with Barrett’s esophagus–associated intramucosal adenocarcinoma (IMC) and dysplasia, according to new data published in the American Journal of Surgical Pathology.
Complete eradication of IMC/dysplasia on the first follow-up endoscopy after treatment was achieved in 86% of patients, while durable eradication, defined as a complete recurrence that persisted until the last follow-up, was achieved in 78% of patients. However, there was significant variation between the different study sites (P = .03) and outcomes were significantly impacted by the baseline extent of IMC and the use of EMR prior to radiofrequency ablation (RFA) therapy.
In addition, almost a quarter of all patients developed treatment-related strictures, usually in the setting of multiple EMRs, and recurrence occurred as a malignant stricture in one patient.
Radiofrequency ablation used with or without EMR, is a safe, effective, and durable treatment option for the treatment of dysplasia associated with Barrett’s esophagus. However, studies that have assessed the predictors of treatment failure in Barrett’s esophagus-associated intramucosal adenocarcinoma (IMC) are limited.
In this study, Dr. Agoston T. Agoston of the department of pathology at Brigham and Women’s Hospital, Boston, and his colleagues investigated the rate of Barrett’s esophagus–associated IMC eradication when using RFA, with or without EMR, in a multicenter setting. In addition, they attempted to identify clinical and pathologic predictors of treatment failure.
“We anticipate that these data will have significant implications for a personalized treatment approach to patients with BE [Barrett’s esophagus]–associated IMC,” wrote the authors.
They conducted a retrospective review of medical records from four tertiary care academic medical centers, and identified 78 patients who underwent RFA with or without EMR as the primary treatment for biopsy-proven IMC.
Some notable baseline differences were observed in patient characteristics at the different study sites, including baseline Barrett’s esophagus segment length (P = .06), baseline nodularity (P = .08), and percentage of tissue involved by IMC at pretreatment endoscopy and biopsy (P = .01).
Over a mean follow-up time of 26.4 months (range, 2-116 months), 86% of patients achieved complete eradication and 78% durable eradication of IMC/dysplasia.
Within the cohort, 11 patients failed to achieve complete eradication, and of the 67 patients who initially did, 6 patients (9.0%) had a subsequent recurrence of neoplasia (3.91 recurrences per 100 patient-years). This extrapolated to an overall rate of 22% for treatment failure (17/78 patients). Of the 17 patients who failed the treatment, 3 subsequently underwent esophagectomy, 1 received palliative measures in the setting of advanced neurological disease, and 1 patient is currently undergoing a repeat ablation procedure with curative intent.
Dr. Agoston and his team also identified 2 clinicopathologic factors that were significantly associated with treatment failure, and both remained significant on univariate and multivariate analysis. The first was that the use of EMR prior to RFA was associated with a significantly reduced risk for treatment failure (hazard ratio, 0.15; 95% confidence interval, 0.05-0.48; P = .001), and the second was that the extent of IMC involving at least 50% of the columnar metaplastic area was associated with a significantly increased risk for treatment failure (HR, 4.24; 95% CI, 1.53-11.7; P = .005).
They also observed similar results when the analysis of extent of IMC as a predictor was restricted to a subset of 43 cases in which the diagnosis of IMC was made on EMR specimens only (HR, 10.8; 95% CI, 2.30-50.8; P = .003).
“In conclusion, we have identified endoscopic and pathologic factors associated with treatment success in patients with BE-associated IMC treated with RFA with or without EMR,” they wrote. “Utilization of these predictors can help in identifying patients with a high probability of success and also those patients with a higher risk for treatment failure for whom a more aggressive initial approach may be justified” (Am J Surg Pathol. 2015 Dec 5. doi: 10.1097/PAS.0000000000000566).
Dr. Rothstein and Dr. Abrams have received research support previously from Barrx/ Covidien and C2 Therapeutics/Covidien, respectively. None of the other authors reported significant conflicts of interest.