A multibiomarker risk score helps predict risk for Barrett's esophagus

Area of Science:

• Gastroenterology
• Biomarker Discovery
• Medical Diagnostics

Background:

• Barrett's esophagus (BE) risk prediction models using demographic and clinical data have shown modest performance.
• There is a need for improved discriminatory ability in BE risk assessment.

Purpose of the Study:

• To evaluate if adding a multibiomarker risk score to existing models improves the prediction of Barrett's esophagus.
• To compare the predictive accuracy of models with and without the multibiomarker score.

Main Methods:

• A case-control study involving 141 BE patients and 138 controls.
• Three risk prediction models were developed: 1) GERD frequency/duration, 2) Model 1 + demographics/H. pylori, 3) Model 2 + multibiomarker score (IL12p70, IL6, IL8, IL10, leptin).
• Predictive accuracy was assessed using area under the receiver operating characteristic curve (AUC) and calibration.

Main Results:

• The multibiomarker risk score was significantly associated with BE risk.
• Individuals with a biomarker score of 3 or higher had over a 10-fold increased risk for BE (OR, 11.9).
• The model incorporating the multibiomarker score demonstrated improved discrimination (AUC, 0.85) compared to models without it (AUC, 0.74).

Conclusions:

• A risk prediction model including a multibiomarker score, demographic, and clinical factors accurately identifies individuals with BE.
• This multibiomarker approach, using serum cytokines and leptin, offers improved diagnostic accuracy over previous methods for BE detection.