Breaking News:cfDNA Fragmentome Test Shows Promise for Liver Disease– What Just Happened

A NEW study suggests that analysing the cell free DNA (cfDNA) fragmentome could help detect liver disease earlier and even predict survival across a range of illnesses. Liver disease, which includes conditions such as fibrosis and cirrhosis, occurs when damage to liver tissue gradually impairs function. Early stages can be asymptomatic, making timely intervention difficult. Researchers investigated whether cfDNA fragmentome analysis, a minimally invasive liquid biopsy, could reveal disease-related changes in the blood.

Liquid Biopsy Reveals Signals of Liver Disease

cfDNA consists of small DNA fragments released into the bloodstream when cells break down through biological processes such as apoptosis. By studying genome-wide fragmentation patterns, scientists can identify which tissues are being affected and how the body responds to disease.

In the study, whole-genome sequencing to analyse cfDNA fragmentomes from 1,576 participants, including those with liver disease and other conditions, such as vascular, autoimmune and neurodegenerative disorders.

A machine learning classifier was developed to detect liver disease using cfDNA fragmentome data. The tool identified early liver disease, advanced fibrosis and cirrhosis with high sensitivity.  The algorithm was trained in one group of participants (n = 423) and then tested in a second, independent group (n = 221) to confirm reproducibility. The model also showed limited cross-reactivity with other diseases, suggesting disease-specific patterns.

Molecular Clues from Immune and Liver Cells

Further analyses combining fragmentome and methylome profiling indicated that the circulating DNA changes reflected both liver-derived signals and immune-mediated processes in people with liver disease. cFDNA fragmentation patterns also differed in participants with other conditions, highlighting their potential as broad biomarkers of physiological health.

Potential to Predict Survival

Beyond disease detection, the researchers trained another machine learning model using cfDNA fragmentome patterns to estimate overall survival. This approach was tested in separate morbidity cohorts, including 571 individuals in a discovery set and 231 in a validation group.

Implications for Future Liquid Biopsies

These findings suggest that cfDNA fragmentomes may act as biomarkers of an individual’s physiological state, offering a minimally invasive way to detect liver disease and monitor other conditions.

It is important to note that machine learning models can inadvertently learn gender or sex biases if the training data is unbalanced, meaning that predictive tools might perform differently for men and women if sex-specific biological differences are not considered. Open access to sex-disaggregated data is therefore crucial, allowing researchers to detect and correct biases, improve model accuracy, and ensure that innovations like cfDNA fragmentome tests are equitable for all patients.

Reference

Annapragada AV et al. Cell-free DNA fragmentomes for noninvasive detection of liver cirrhosis and other diseases. Sci Transl Med. 2026;18(839):eadw2603

Featured image: Ultramansk on Adobe Stock