Failure to Consider Atopic Status Limits Existing Minimally-Invasive Biomarker Studies in Eosinophilic Esophagitis: A Systematic Review
Sunday, March 4, 2018
South Hall A2 (Convention Center)
Brittany Hines, MD, Matthew A. Rank, MD FAAAAI, Benjamin L. Wright, MD, Lisa Marks, Matthew J. Greenhawt, MD MBA MSc, Evan S. Dellon, MD MPH

RATIONALE: A systematic review may help identify promising minimally-invasive EoE biomarkers and remaining gaps in biomarker validation.

METHODS: EMBASE, Ovid Medline, PubMed, and Web of Science were searched from inception to June 6, 2017. Studies were included if individuals in the study met the 2007 consensus definition for EoE diagnosis, used a minimally-invasive test of a biomarker defined as an objective measure of biological processes, pathogenic processes, or responses to therapeutic interventions, and included at least 1 control comparison.

RESULTS: The search yielded 1,507 studies, 236 of which were reviewed at full text level, and 49 of which were included in the full text review (20 pediatric, 18 adult, 7 pediatric and adult, and 4 not stated). The majority (26 of 49) were reported since 2014. Thirty-six studies included normal controls, 8 analyzed separate atopic control groups, and 29 compared active and controlled EoE samples. Minimally invasive biomarkers were from blood (n=41 studies), sponge/string (3), oral/throat swabs (2), breath (2), stool (2), and urine (2). The most commonly reported minimally-invasive biomarkers were eosinophils (n=17), cytokines/chemokines (13), eosinophil surface or intracellular markers (11), and eosinophil granule proteins (9). There were 23 studies that reported a significant difference for a biomarker comparing EoE to normal controls, 2 studies for comparing EoE to atopic controls, and 18 studies for comparing active and controlled EoE.

CONCLUSIONS: Many promising minimally-invasive biomarkers for EoE have emerged for diagnosis and monitoring but only 8 performed separate atopic control analyses and only 2 of these found significant differences.