July 13, 2020

Family history of GI cancer and risk of Barrett's

Joel Rubenstein and colleagues report increased risk of Barrett's associated with a family history of esophageal or colorectal cancer, supporting previous research.

Family history of GI cancer and risk of Barrett's

Clin Transl Gastroenterol. 2020 Apr;11(4):e00151.
doi: 10.14309/ctg.0000000000000151.

Family History of Colorectal or Esophageal Cancer in Barrett's Esophagus and Potentially Explanatory Genetic Variants

Joel H Rubenstein  1   2   3 , Anna Tavakkoli  4 , Erika Koeppe  3 , Peter Ulintz  3 , John M Inadomi  5 , Hal Morgenstern  6   7 , Henry Appelman  8 , James M Scheiman  9 , Philip Schoenfeld  10 , Val Metko  2 , Elena M Stoffel

PMID: 32251017 PMCID: PMC7263651 DOI: 10.14309/ctg.0000000000000151 


Introduction: We aimed to estimate the effects of a family history of colorectal cancer (CRC) or esophageal cancer on the risk of Barrett's esophagus (BE) and identify variants in cancer genes that may explain the association.

Methods: Men scheduled for screening colonoscopy were recruited to undergo upper endoscopy. Cases and noncases were screenees with and without BE, respectively. The effects of family histories on BE were estimated with logistic regression, adjusting for the potential confounders. We additionally recruited men recently diagnosed with BE by clinically indicated endoscopies. Banked germline DNA from cases of BE with =2 first-degree relatives (FDRs) with CRC and/or an FDR with esophageal cancer underwent next-generation sequencing using a panel of 275 cancer genes.

Results: Of the 822 men screened for CRC who underwent upper endoscopy, 70 were newly diagnosed with BE (8.5%). BE was associated with family histories of esophageal cancer (odds ratio = 2.63; 95% confidence interval = 1.07-6.47) and CRC in =2 vs 0 FDRs (odds ratio = 3.73; 95% confidence interval = 0.898-15.4). DNA analysis of subjects with both BE and a family history of cancer identified one or more germline variants of interest in genes associated with cancer predisposition in 10 of 14 subjects, including the same novel variant in EPHA5 in 2 unrelated individuals.

Discussion: We found an increased risk for BE associated with a family history of esophageal cancer or CRC. Although analysis of germline DNA yielded no clinically actionable findings, discovery of the same EPHA5 variant of uncertain significance in 2 of 14 cases merits additional investigation.