Plasma cfDNA methylation profile supports non-invasive gastric cancer detection in small cohort study
Researchers publishing in PLOS Genetics report that whole-genome bisulfite sequencing of circulating free DNA, combined with a machine learning classifier, distinguished gastric cancer patients from healthy controls in a 90-person dataset.
A peer-reviewed study published in PLOS Genetics by Hailing Ou, Jinli Miao, and colleagues describes the development of a machine learning model for non-invasive detection of gastric cancer (GC) using whole-genome methylation profiling of circulating free DNA (cfDNA) in plasma.
The researchers performed whole-genome bisulfite sequencing (WGBS) on paired tumour tissue and plasma samples from 61 patients with GC and 29 healthy controls. By comparing tissue and plasma methylation landscapes, they identified novel differentially methylated regions that were detectable in cfDNA. A machine learning classifier trained on these cfDNA methylation features is reported to distinguish GC cases from controls, with the authors arguing the approach shows superior sensitivity compared with existing non-invasive methods.
Gastric cancer carries a poor prognosis when diagnosed at advanced stage; non-invasive early-detection strategies based on liquid biopsy remain an active area of research across several tumour types. The study's cohort is small, and external validation in independent populations will be required before the performance characteristics described can be assessed in broader contexts. The work represents a research advance in cfDNA biomarker discovery rather than a validated diagnostic tool.
Plain-language version
For patients, families, and general readers. Educational only — not medical advice.
Gastric cancer (cancer of the stomach) is often found late because early-stage disease causes few symptoms. Researchers are working on blood tests that could detect cancer earlier by looking for tiny fragments of DNA shed by tumour cells into the bloodstream — this is called a "liquid biopsy".
A study published in the scientific journal PLOS Genetics reports that scientists were able to use a type of blood test, combined with a computer programme, to tell apart 61 patients with gastric cancer from 29 healthy people. The test looks at chemical tags on DNA fragments in the blood called methylation marks. The study involved a small number of people, so much more research is needed before anything like this could be used routinely.
This is an educational summary, not medical advice. If anything here raises questions for you, please speak with your GP or a clinical professional.
Sources
Read the original reporting — these are the public sources this summary draws from.
-
Primary source Public Library of Science · 2026-10-15Genome-Wide DNA methylation profiling of tissue-plasma pairs identifies Novel cfDNA biomarkers for the diagnosis of gastric cancer