A blood test finds esophageal cancer that is often missed by physical tissue biopsies.
April 23, 2026
Original Paper
Noninvasive Detection And Subtyping of Esophageal Cancer by Circulating Cell-Free mtDNA Fragmentomics: A Multi-Center Validation Study
SSRN · 6620166
The Takeaway
Doctors usually rely on a biopsy to diagnose cancer, but this method often fails in the early stages. This new test analyzes fragments of mitochondrial DNA floating in the plasma with over 92% sensitivity. The machine learning model identifies specific fragmentomics patterns that signal the presence of a tumor. It can even tell the difference between different subtypes of cancer with high accuracy, leading to much earlier and more successful treatment.
From the abstract
Esophageal cancer (EC) is lethal due to late diagnosis and lacks accurate non-invasive tools for early detection and subtyping. Circulating cell-free mitochondrial DNA (ccf-mtDNA) fragmentomics offers a promising biomarker source. This multi-center study analyzed plasma ccf-mtDNA fragmentomics in 1,180 participants, including healthy control (HC) and patients with benign esophageal diseases (BED), precancerous lesions, or EC. And we developed an EC detection (ED) model and a pathological subtypi