An Extracellular Vesicle Protein-Based Machine Learning Framework for Early Detection of Oesophageal Squamous Cell Carcinoma: A Multicentre, Prospective Study - PubMed
5 hours ago
- #biomarkers
- #liquid biopsy
- #early detection
- Developed a blood test using extracellular vesicle (EV) proteins for early detection of oesophageal squamous cell carcinoma (ESCC).
- Created BarFlare, a high-sensitivity platform for serum EV protein analysis, identifying novel biomarkers SCC and MMP13.
- Integrated biomarkers with clinical factors into a machine-learning framework (MCF) for accurate ESCC detection.
- Validated the MCF model in multicentre diagnostic cohorts (n=1018), achieving high AUC scores (0.901-0.987).
- Detected ESCC in preclinical stages with a median lead time of 34.9 months before clinical diagnosis.
- Demonstrated superior performance over traditional serum SCC for risk stratification.
- Provides a non-invasive tool for risk-stratified screening and early intervention.