Incidence of esophageal adenocarcinoma (EA), a rapidly fatal cancer, has increased sharply over recent decades. Although several risk factors have been defined for EA and its precursor, Barrett's esophagus (BE), including reflux, Caucasian race, male gender, obesity, and smoking, a key challenge remains the identification of individuals at highest risk, since most with GERD do not develop BE, and most with BE do not progress to cancer. Cellular energy metabolism is fundamentally altered in a broad spectrum of cancers, and cancer-associated metabolic changes are detectable in blood and urine. To determine whether serum metabolites can discriminate between intermediate disease states preceding development of EA, we used high performance liquid chromatography-mass spectrometry (LC-MS) to profile 106 metabolites in three case groups: GERD (n=50), BE (n=50), and high-grade dysplasia (HGD)/early-stage EA (n=50). Multiple metabolites differed significantly (P<0.05) in cases with BE versus GERD (n=8), HGD/EA versus BE (n=9), or HGD/EA versus GERD (n=30). Four candidates – linolenic acid, fructose, malonic acid, D-leucic acid – remained significant after Bonferroni correction (P<4.7x10-4) and exhibited linear trends in signal across the GERD-BE-HGD/EA spectrum. Multivariate modeling procedures based on partial least squares-discriminant analysis (PLS-DA) or “elastic net” regression, coupled with Monte Carlo cross validation, achieved robust discrimination between HGD/EA and GERD cases (AUC=0.93), with less pronounced separation observed between intermediate case groups (AUC=0.63-0.66). Metabolite set enrichment analyses identified two pathways significantly altered in cases with HGD/EA versus GERD – bile acid biosynthesis (P=0.01, q=0.28) and fatty acid biosynthesis (P=0.03, q=0.40) – and several amino acid metabolism pathways altered in cases with BE (P≤0.2, q<0.02). These findings provide further insight into metabolic perturbations potentially linked to BE/EA pathophysiology, and suggest that small panels of serum metabolites may help differentiate individuals at different stages of progression to EA. Additional studies are needed to validate these marker candidates and assess their utility for non-invasive clinical assessment and stratification of persons at risk of EA.

Citation Format: Matthew F. Buas, Haiwei Gu, Danijel Djukovic, Jiangjiang Zhu, Brian J. Reid, Daniel Raftery, Thomas L. Vaughan. Identification of novel candidate serum metabolite biomarkers for distinguishing between gastro-esophageal reflux disease, Barrett's esophagus, and high-grade dysplasia/esophageal adenocarcinoma. [abstract]. In: Proceedings of the Thirteenth Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2014 Sep 27-Oct 1; New Orleans, LA. Philadelphia (PA): AACR; Can Prev Res 2015;8(10 Suppl): Abstract nr PR10.

©2015 American Association for Cancer Research.
2015