TBA (25W107)

Risk Stratification in Barrett’s Oesophagus Using the Oesophageal Capsule Sponge: Real-World Data from Northern Ireland

Author(s)

D.N. Johnston (1), D. Concannon (2), H. Graham (3), F. Gregg (3), K. Diong (4), I. Mainie (3), J. McGoran (2), R.C. Turkington (1)

Department(s)/Institutions

1. Centre for Public Health, Queen’s University Belfast, Belfast, United Kingdom 2. Department of Gastroenterology, Western Health and Social Care Trust, Londonderry, United Kingdom 3. Department of Gastroenterology, Belfast Health and Social Care Trust, Belfast, United Kingdom 4. Department of Gastroenterology, Northern Health and Social Care Trust, Antrim, United Kingdom

Introduction

Patients with Barrett’s oesophagus (BO) currently undergo endoscopic surveillance. This is invasive and expensive, and recent evidence suggests it does not reduce mortality. There is an urgent need for novel risk-stratification approaches to identify which patients would benefit from endoscopy.

Aims/Background

The capsule sponge is a non-endoscopic oesophageal cell collection device, enabling immunohistochemical biomarker testing of the collected samples. We aimed to evaluate whether this could risk-stratify patients with BO.

Method

Patients under surveillance for BO were recruited from 4 Northern Irish sites, undergoing capsule sponge testing between June 2022 and May 2023. Samples were analysed for p53 and atypia (high-risk findings associated with dysplasia). Follow-up surveillance endoscopy and histology data were collected until July 2025. Patient acceptability data were recorded.

Results

188 patients had capsule sponge and follow-up data available. 5 (2.7%) were positive for p53/atypia on capsule sponge, and considered high-risk. The remaining 183 (97.3%) were considered non-high-risk. The positive predictive value of a high-risk capsule sponge for any dysplasia was 20.0%. The prevalence of dysplasia in the overall cohort was 3.7%, therefore a high-risk capsule sponge enriched over 5-fold for dysplasia. The negative predictive value of a non-high-risk capsule sponge for any dysplasia was 96.7%, and for high-grade dysplasia was 98.4%. No cases of adenocarcinoma were observed in the non-high-risk group. 97.8% of patients indicated they would be willing to undergo the capsule sponge again.

Conclusions

Capsule sponge testing in patients with BO could potentially identify the high-risk minority who require endoscopy, whilst the majority could undergo non-endoscopic capsule sponge surveillance.