Digestive Diseases

Jan. 08, 2025

Over the past 40 years, the incidence of esophageal adenocarcinoma (EAC) has increased significantly. Because EAC is often diagnosed after the onset of obstructive symptoms, the mortality rate associated with this form of cancer is high.

Barrett esophagus (BE) is the precursor of most EACs. Most gastroenterology societies suggest endoscopic screening for BE in individuals with multiple risk factors, followed by endoscopic surveillance and endoscopic treatment of dysplasia. Despite these recommendations, however, BE screening rates are quite low.

Endoscopic detection of BE and EAC is invasive, expensive and difficult to access. Researchers and clinicians agree that there is a critical need to develop accurate and minimally invasive BE and EAC risk assessment tools. Prasad G. Iyer, M.D., M.S., a gastroenterologist and researcher at Mayo Clinic in Rochester, Minnesota, is working with colleagues to address this need.

Several case control studies conducted by Dr. Iyer, colleagues and others have demonstrated that a swallowable encapsulated collection sponge is a safe, less expensive and well-tolerated BE screening tool. Additional research has shown that this nonendoscopic screening tool can increase BE detection and potentially increase the detection of dysplastic BE and early-stage, endoscopically treatable EAC.

In a 2020 case control study published in the American Journal of Gastroenterology, Dr. Iyer and co-authors demonstrated that methylated DNA markers (MDMs) assayed on samples obtained with the sponge collection device could detect BE. In a 2021 publication in Gastrointestinal Endoscopy, Dr. Iyer and co-authors reported on the validation of a five-marker MDM panel with excellent performance characteristics.

Building on this body of work, Dr. Iyer and co-investigators sought to train and test an algorithm using a three-MDM panel and test its performance in prospective multicenter cohorts. The results of this study were published in Clinical Gastroenterology and Hepatology in 2024.

Methods

The researchers recruited 352 patients for the training set (154 with BE, 198 controls) from six medical centers, and 125 patients for the test set (eight with BE, 44 controls) from three medical centers. Participants were classified as cases, control subjects or indeterminate based on predefined criteria.

Participants in the BE-EAC case group were patients with a BE segment ≥ 1 cm and histology demonstrating intestinal metaplasia, excluding patients with a history of ablation and a history of esophageal or gastric neoplasia. Participants in the control group were patients with no endoscopic evidence of esophageal columnar metaplasia and patients with gastroesophageal reflux disease and one or more BE risk factors. Participants in the indeterminate group were patients who were likely to be encountered in a population at risk for BE but who did not meet the diagnostic criteria for the case or control group.

All patients underwent the sponge collection device procedure followed by endoscopy within 24 hours. The researchers extracted DNA from collection samples and blindly assayed the MDMs. They used cross-validated logistic regression to set and lock the algorithm (training set) and assessed its performance in an independent test set. Patients in the training set also completed a tolerability assessment survey immediately after sponge collection device retrieval.

The researchers determined whether BE was present or absent using biopsy samples taken for histological evaluation at endoscopy. They assigned a mucosal injury score ranging from 1 (no trauma) to 5 (bleeding requiring endoscopic therapy) to all participants in the training set. An expert gastrointestinal pathologist reviewed all pathology. The researchers also obtained Seattle protocol biopsies in participants in the BE-EAC case group who were undergoing clinical endoscopy for surveillance.

Results

The clinical characteristics of the training and test sets were comparable. The final marker panel included three MDMs — NDRG4, VAV3 and ZNF682.

• Overall sensitivity of the three-MDM panel was 82% in the training set and 88% in the test set. Overall specificity was 90% in the training set and 84% in the test set.
• Sensitivity was 90% for all long-segment BE and 68% for all short-segment BE.
• Sensitivity for BE with high-grade dysplasia and EAC was 100% in both the training and test sets.
• Overall sensitivity for nondysplastic BE was 82%.

"Our findings demonstrated that this locked three-MDM panel algorithm for BE-EAC detection using a nonendoscopic, swallowable sponge collection device demonstrated excellent sensitivity for high-risk BE cases in independent validation samples," explains Dr. Iyer. "These results further establish that nonendoscopic detection of BE is a safe and well-tolerated alternative to sedated EGD, with excellent sensitivity across all dysplasia grades where intervention is recommended."

Next steps

"We are conducting a large prospective trial in a screening population to assess the positive and negative predictive value of the sponge collection device test for detecting BE," explains Dr. Iyer. "We recognize that implementation pathways in primary care practices still need to be delineated."

For more information

Iyer PG, et al. Accurate nonendoscopic detection of Barrett's esophagus by methylated DNA markers: A multisite case control study. The American Journal of Gastroenterology. 2020;115:1201.

Iyer PG, et al. Validation of a methylated DNA marker panel for the nonendoscopic detection of Barrett's esophagus in a multisite case-control study. Gastrointestinal Endoscopy. 2021;94:498.

Iyer PG, et al. Algorithm training and testing for a nonendoscopic Barrett's esophagus detection test in prospective multicenter cohorts. Clinical Gastroenterology and Hepatology. 2024;22:1596.

Refer a patient to Mayo Clinic.