Cellular and Molecular Medicine Research, ISSN 2817-6359 online, Open Access
Article copyright, the authors; Journal compilation copyright, Cell Mol Med Res and Elmer Press Inc
Journal website https://cmmr.elmerpub.com

Review

Volume 4, Number 1, September 2026, pages 1-9

Molecular Diagnostics in Barrett’s Esophagus: Bridging Endoscopy and Cellular Pathobiology to Detect Early Esophageal Adenocarcinoma

Figure

↓  Figure 1. The molecular natural history of the Barrett’s esophagus-to-adenocarcinoma sequence (top: metaplasia of gastric-cardia origin, with TP53 loss and clonal expansion in a subset of progressing lineages, often before whole-genome doubling, followed by late oncogene amplification) mapped to the diagnostic modalities by stage (bottom: non-endoscopic cell collection for screening; AI-assisted endoscopy and WATS3D for detection; p53 immunohistochemistry and tissue systems pathology for risk stratification; and cell-free DNA for monitoring).
Figure 1.

Table

↓  Table 1. Comparative Performance, Strengths, Limitations, and Current Clinical Role of Diagnostic Modalities in Barrett’s Esophagus and Early Esophageal Adenocarcinoma
 
Modality/sampleStrengthMain limitationPerformance estimateCurrent role/accessEvidence
Performance values are study- or meta-analysis-specific and should not be compared as if obtained in a single head-to-head population. AGA: American Gastroenterological Association; BE: Barrett’s esophagus; CAD: computer-aided detection; cfDNA: cell-free DNA; EAC: esophageal adenocarcinoma; FFPE: formalin-fixed paraffin-embedded; HGD: high-grade dysplasia; LGD: low-grade dysplasia; NPV: negative predictive value; RCT: randomized controlled trial; TFF3: trefoil factor 3; WATS3D: wide-area transepithelial sampling with three-dimensional computer-assisted analysis.
p53 IHC/biopsyLow-cost, widely available adjunct to dysplasia interpretationSingle pathway; retains biopsy-sampling and scoring variabilityProgression association RR 14.25; standalone sensitivity/specificity not establishedAdjunct to expert pathology; routine biomarker-directed surveillance is not recommended[14, 32]
TSP-9/TissueCypher/FFPE biopsyObjective nine-biomarker risk class with high NPVProprietary centralized assay; moderate sensitivity; mainly retrospective or industry-associated evidencePooled sensitivity 61%, specificity 81%, NPV 97%Selective risk-stratification adjunct where available; no current AGA recommendation[17–19, 32, 33]
WATS3D/brushBroad spatial sampling and incremental dysplasia detectionAdjunct only; heterogeneous studies and uncertain natural history of WATS3D-only dysplasiaIncremental dysplasia yield 7.2%; standalone sensitivity/specificity not establishedAdjunct to Seattle-protocol biopsy where expertise exists; no current AGA recommendation[20, 32]
Cytosponge-TFF3/spongeScalable non-endoscopic screening/triage with randomized and economic evidencePositive tests require endoscopy; uptake, laboratory pathways, and availability varyPooled sensitivity 81% and specificity 89% for BE of any circumferential lengthScreening/triage in selected systems; not a replacement for diagnostic endoscopy[22, 23, 34]
EsoCheck/EsoGuard/balloonRapid sampling with methylated VIM and CCNA1 detectionCommercial assay; confirmatory endoscopy; predominantly White cohort; industry involvementSensitivity 85%, specificity 85%; 18/18 cancers detected in 243 evaluable participantsSelected non-endoscopic screening/triage where available; outcome validation is needed[25]
Liquid biopsy/cfDNAMinimally invasive and suitable for serial monitoringInsufficient sensitivity for nondysplastic BE or early EAC; specialized analysisSignal mainly detected in metastatic disease; early-disease sensitivity/specificity not establishedResearch and response/recurrence monitoring; not primary BE screening[26]
AI CAD/endoscopic image or videoReal-time lesion localization and targeted-biopsy supportImage-quality and external-validation dependence; workflow, regulatory, and training requirementsValidation dataset: sensitivity 90%, specificity 88%, accuracy 89%Adjunct to high-quality endoscopy; not a molecular test or replacement for structured biopsy[27]