Nonendoscopic Screening for Barrett’s Esophagus in Veterans Without Chronic Reflux

Description

This research study addresses early detection of esophageal adenocarcinoma (EAC). Incidence of EAC has increased more than 6-fold in the past four decades yet prognosis for patients with EAC remains poor with less than 20% of patients surviving beyond 5 years. The veteran population is at increased risk for EAC and its precursor lesion, Barrett’s esophagus (BE), due to increased prevalence of disease risk factors (age, male, obesity, smoking, family history of BE or EAC), compared to the general population. Current society guidelines recommend one-time upper endoscopy for persons with gastro-esophageal reflux (GERD) plus additional risk factors for BE and EAC. Still, up to 40% of patients who develop esophageal cancer do not report history of GERD preceding cancer diagnosis. Screening benefit needs to be explored in these patients without history of GERD.

Hypothesis: The incorporation of a non-endoscopic detection method in clinical practice will increase the positive predictive value of esophagogastroduodenoscopy (EGD) and increase the detection of BE among veterans who do not report GERD but have multiple risk factors for EAC.

Objectives: Unsedated non-endoscopic BE screening method will be offered to veterans at risk of EAC but without GERD history.

The Specific Aims are: 1. To detect BE in high-risk subjects without GERD, using a non-endoscopic biomarker-based screening tool. 2. Measure screening acceptance rate, reasons for refusal and screening yield.

Preliminary Data: Prospectively collected data from Cleveland VA showed that the sensitivity of this non-endoscopic screening tool in patients with chronic GERD and/or PPI use was 92.9% (95% CI 66.1-99.8), and specificity of 72.2% (95% CI 62.1-80.8). A retrospective study of 113 patients with EAC at Louis Stokes Cleveland VA Medical Center found that 52 (46%) presented with Stage IV disease. Of these patients with metastatic cancer who were eligible for screening, 20 (39%) had been seen in clinic at least one year prior to cancer diagnosis and had not been screened with EGD. In the overall group, 42 of the 117 patients had been seen in clinic with GERD previously and 26 (62%) had not been screened.

Research Design: Patients without history of GERD will be approached regarding BE screening. Reasons for refusal of screening will be recorded. Adult subjects without history of gastroesophageal reflux with three or more risk factors- white race, male gender, obesity (BMI > 30), smoking, or family history will be included. Only those patients who have not had prior EGD will be candidates for Esocheck screening. Distal esophageal samples will be assayed for mVIM + mCCNA1 (Esoguard assay). All patients will be asked to complete upper endoscopy.

Sample Size and Analysis: The prevalence of BE found on EGD performed in Caucasian men without chronic GERD is around 8.5%. Assuming prevalence of BE of 9%, and sensitivity/specificity of Esocheck/Esoguard equal to 90/90%, respectively, this study plans to recruit 400 patients. Sensitivity, specificity, diagnostic accuracy of Esocheck/Esoguard in patients without GERD will be calculated. Acceptance rates of screening will be reported. Latent class analysis will be performed to examine common patterns among patients who decline EAC screening.