S858 Minimally Invasive Cell Sampling Device for Screening of Barrett’s Esophagus in Patients With GERD: A Meta-Analysis of Diagnostic Accuracy

Introduction: International guidelines now recommend non-endoscopic, swallowable sponge devices with biomarkers as valid alternatives to endoscopy for detecting Barrett’s esophagus (BE) in at-risk patients. Understanding their screening performance is essential to ensure cases are not missed due to suboptimal test accuracy. In this study, we analyzed the screening performance of the Cytosponge with Trefoil Factor 3 (TFF3) in patients with symptomatic acid reflux and no prior history of BE. We also assessed test performance in patients with and without additional BE risk factors. Methods: Eligibility criteria included adults with reflux symptoms with no prior diagnosis of BE. Included studies reported results for both Cytosponge with TFF3 and the gold standard of endoscopic assessment with biopsy for BE diagnosis. Scopus, Embase, PubMed, CINAHL, Web of Science, Cochrane Central, and ClinicalTrials.gov were searched from January 2009, through March 29, 2025. Two reviewers independently screened and extracted data using the COVIDENCE tool; risk of bias was assessed via QUADAS-2. Sensitivity and specificity were estimated in Statistical Analysis Software v 9.4 using bivariate mixed models, which were used to estimate likelihood ratios, diagnostic odds ratios, and predictive values; meta-regression models were estimated for individual risk factors. Forest plots were created in Stata v. 19.5. Results: We identified 7 studies involving 1,735 participants from the general population in our meta-analysis. The pooled performance for Cytosponge with TFF3 for detecting BE included sensitivity 0.87 (95% confidence interval [CI] 0.81-0.93), specificity 0.84 (95% CI 0.82-0.86), and diagnostic odds ratio of 35.84 (95% CI 15.57-56.10). There were no statistically significant differences in sensitivity when comparing groups based on age, sex, race, smoking status, body mass index (BMI), or acid suppressant use; all these risk factors, except for BMI, were associated with increased specificity. Conclusion: Overall, our findings show that Cytosponge is an effective minimally invasive screening tool for patients at risk of BE. All studies were conducted in the U.K., which may limit the generalizability of results. Still, sensitivity remained consistent across risk factors, demonstrating reliable detection of BE in patients with gastroesophageal reflux disease. Increased specificity in individuals over the age of 50 years and higher than 95% negative predictive value in populations with 20% prevalence for BE may help reduce unnecessary endoscopies in these higher-risk groups.