Abstract
EsopredictTM is a prognostic assay that risk-stratifies Barrett's esophagus patients to predict future progression to high-grade dysplasia (HGD) or esophageal adenocarcinoma (EAC). Established based on foundational studies at Johns Hopkins University, a risk algorithm was developed and clinically validated in two independent studies (n = 320). EsopredictTM is currently offered as a clinical test under the Clinical Laboratory Improvement Amendments (CLIA) guidelines. Here we present the analytical validation by repeated testing of FFPE tissues (n = 26 patients), cell lines, and contrived DNA controls to determine assay performance regarding analytical sensitivity (as defined by the limit of detection (LOD)), analytical specificity (as defined by the limit of blank (LOB)), accuracy as determined from the average positive and negative agreement, repeatability, and reproducibility. The LOD for the assay at 1.5% DNA methylation was significantly higher than the LOB, as determined by an unmethylated DNA control (0% methylated DNA). Inter- and intra-assay average positive agreement (APA) were 88% and 94%, respectively, while average negative agreement (ANA) values were 90% and 94%, respectively. Average inter- and intra-assay precision were <9% and <5% coefficient of variation (CV), respectively. These results confirm that EsopredictTM is a highly reproducible, sensitive, and specific risk categorization assay for the prediction of progression to HGD or EAC within 5 years.