Whole Genome Methylation Analysis of Nondysplastic Barrett Esophagus that Progresses to Invasive Cancer

Ann Surg. 2019 Mar;269(3):479-485. doi: 10.1097/SLA.0000000000002658.

Abstract

Objective: To investigate differences in methylation between patients with nondysplastic Barrett esophagus who progress to invasive adenocarcinoma and those who do not.

Background: Identifying patients with nondysplastic Barrett esophagus who progress to invasive adenocarcinoma remains a challenge. Previous studies have demonstrated the potential utility of epigenetic markers for identifying this group.

Methods: A whole genome methylation interrogation using the Illumina HumanMethylation 450 array of patients with nondysplastic Barrett esophagus who either develop adenocarcinoma or remain static, with validation of findings by bisulfite pyrosequencing.

Results: In all, 12 patients with "progressive" versus 12 with "nonprogressive" nondysplastic Barrett esophagus were analyzed via methylation array. Forty-four methylation markers were identified that may be able to discriminate between nondysplastic Barrett esophagus that either progress to adenocarcinoma or remain static. Hypomethylation of the recently identified tumor suppressor OR3A4 (probe cg09890332) validated in a separate cohort of samples (median methylation in progressors 67.8% vs 96.7% in nonprogressors; P = 0.0001, z = 3.85, Wilcoxon rank-sum test) and was associated with the progression to adenocarcinoma. There were no differences in copy number between the 2 groups, but a global trend towards hypomethylation in the progressor group was observed.

Conclusion: Hypomethylation of OR3A4 has the ability to risk stratify the patient with nondysplastic Barrett esophagus and may form the basis of a future surveillance program.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adenocarcinoma / genetics*
  • Adenocarcinoma / pathology
  • Adult
  • Barrett Esophagus / genetics*
  • Barrett Esophagus / pathology
  • Biomarkers, Tumor / genetics*
  • Case-Control Studies
  • Computational Biology
  • DNA Methylation*
  • Databases, Factual
  • Disease Progression
  • Esophageal Neoplasms / genetics*
  • Esophageal Neoplasms / pathology
  • Female
  • High-Throughput Nucleotide Sequencing
  • Humans
  • Immunohistochemistry
  • Male
  • Middle Aged
  • Neoplasm Invasiveness
  • Precancerous Conditions / genetics*
  • Precancerous Conditions / pathology
  • Risk Assessment

Substances

  • Biomarkers, Tumor