Development and Validation of a Model to Determine Risk of Progression of Barrett's Esophagus to Neoplasia

Gastroenterology. 2018 Apr;154(5):1282-1289.e2. doi: 10.1053/j.gastro.2017.12.009. Epub 2017 Dec 19.

Abstract

Background & aims: A system is needed to determine the risk of patients with Barrett's esophagus for progression to high-grade dysplasia (HGD) and esophageal adenocarcinoma (EAC). We developed and validated a model to determine of progression to HGD or EAC in patients with BE, based on demographic data and endoscopic and histologic findings at the time of index endoscopy.

Methods: We performed a longitudinal study of patients with BE at 5 centers in United States and 1 center in Netherlands enrolled in the Barrett's Esophagus Study database from 1985 through 2014. Patients were excluded from the analysis if they had less than 1 year of follow-up, were diagnosed with HGD or EAC within the past year, were missing baseline histologic data, or had no intestinal metaplasia. Seventy percent of the patients were used to derive the model and 30% were used for the validation study. The primary outcome was development of HGD or EAC during the follow-up period (median, 5.9 years). Survival analysis was performed using the Kaplan-Meier method. We assigned a specific number of points to each BE risk factor, and point totals (scores) were used to create categories of low, intermediate, and high risk. We used Cox regression to compute hazard ratios and 95% confidence intervals to determine associations between risk of progression and scores.

Results: Of 4584 patients in the database, 2697 were included in our analysis (84.1% men; 87.6% Caucasian; mean age, 55.4 ± 20.1 years; mean body mass index, 27.9 ± 5.5 kg/m2; mean length of BE, 3.7 ± 3.2 cm). During the follow-up period, 154 patients (5.7%) developed HGD or EAC, with an annual rate of progression of 0.95%. Male sex, smoking, length of BE, and baseline-confirmed low-grade dysplasia were significantly associated with progression. Scores assigned identified patients with BE that progressed to HGD or EAC with a c-statistic of 0.76 (95% confidence interval, 0.72-0.80; P < .001). The calibration slope was 0.9966 (P = .99), determined from the validation cohort.

Conclusions: We developed a scoring system (Progression in Barrett's Esophagus score) based on male sex, smoking, length of BE, and baseline low-grade dysplasia that identified patients with BE at low, intermediate, and high risk for HGD or EAC. This scoring system might be used in management of patients.

Keywords: Esophageal Cancer; Multi-Center Trial; PIB; Progression in Barrett’s Esophagus score.

Publication types

  • Multicenter Study
  • Research Support, Non-U.S. Gov't
  • Validation Study
  • Video-Audio Media

MeSH terms

  • Adenocarcinoma / diagnosis
  • Adenocarcinoma / epidemiology*
  • Adenocarcinoma / mortality
  • Adult
  • Aged
  • Barrett Esophagus / diagnosis
  • Barrett Esophagus / epidemiology*
  • Barrett Esophagus / mortality
  • Biopsy
  • Cigarette Smoking / adverse effects
  • Databases, Factual
  • Decision Support Techniques*
  • Disease Progression
  • Esophageal Neoplasms / diagnosis
  • Esophageal Neoplasms / epidemiology*
  • Esophageal Neoplasms / mortality
  • Esophagoscopy
  • Esophagus / pathology*
  • Female
  • Humans
  • Incidence
  • Kaplan-Meier Estimate
  • Longitudinal Studies
  • Male
  • Middle Aged
  • Neoplasm Grading
  • Netherlands / epidemiology
  • Predictive Value of Tests
  • Proportional Hazards Models
  • Reproducibility of Results
  • Risk Assessment
  • Risk Factors
  • Sex Factors
  • Time Factors
  • United States / epidemiology

Supplementary concepts

  • Adenocarcinoma Of Esophagus