AI-Powered Liquid Biopsy Shows Promise for Early Detection of Chronic Diseases

A new artificial intelligence (AI)-based liquid biopsy test analyzing cell-free DNA (cfDNA) fragmentation patterns is demonstrating potential for the early detection of liver fibrosis, cirrhosis, and signals of broader chronic disease burden, according to research published in Science Translational Medicine.¹ This advancement builds upon the established utility of liquid biopsies in cancer detection, expanding its application to a wider range of health conditions.

How the Test Works: Analyzing cfDNA Fragmentomes

Unlike traditional liquid biopsies that focus on identifying specific genetic mutations, this new approach examines the entire “fragmentome” – the size and distribution of cfDNA fragments within the genome. Researchers used whole-genome sequencing to analyze cfDNA fragmentomes from 1,576 individuals, including those with liver disease, vascular conditions, autoimmune diseases, and neurodegenerative disorders.¹

By analyzing these fragmentation patterns, a machine learning classifier was developed to detect signatures of early liver disease, advanced fibrosis, and cirrhosis. The classifier demonstrated high sensitivity and limited cross-reactivity with other diseases when tested on discovery (n=423) and validation (n=221) cohorts.¹

Beyond Liver Disease: Identifying Disease-Specific Signatures

Analysis of both fragmentomes and methylomes – patterns of chemical modifications to DNA – revealed liver-derived and immune-mediated changes in cfDNA from patients with liver disease. Importantly, disease-specific changes in fragmentomes were also observed in individuals with other conditions, suggesting the potential for developing diagnostic classifiers for a variety of illnesses.¹

A machine learning model utilizing cfDNA fragmentomes was then created to predict survival rates in patients with several of the identified diseases. This model was tested on separate discovery (n=571) and validation (n=231) cohorts.¹

The Power of a Genome-Wide Approach

“The fact that we are not looking for individual mutations is what makes this study so powerful,” explains Akshaya Annapragada, the study’s first author and an MD/PhD student at Johns Hopkins Kimmel Cancer Center. “We are analyzing the entire fragmentome, which contains a tremendous amount of information about a person’s physiologic state. The scale of these data, coupled with machine learning, enables development of specific classifiers for many different health conditions.”¹

The researchers emphasize that the fragmentome can serve as a foundation for building disease-specific classifiers. A classifier designed to detect liver fibrosis, for example, is distinct from one designed to detect cancer, highlighting the test’s specificity.¹

Implications for Early Detection and Treatment

Early detection is crucial for many chronic diseases, particularly liver fibrosis, which is reversible in its early stages but can progress to cirrhosis and increase the risk of liver cancer if left untreated.¹ This AI-powered liquid biopsy offers a minimally invasive approach to identify these conditions at an earlier stage, potentially improving treatment outcomes.

Future Directions

While promising, further research is needed to validate these findings in larger and more diverse populations. The development of additional disease-specific classifiers and the integration of this technology into clinical practice hold significant potential for transforming the diagnosis and management of chronic diseases.

References

1. Annapragada, A., et al. (2026). Genome-wide cell-free DNA fragmentation patterns reveal disease-specific signatures for early detection of chronic conditions. Science Translational Medicine. https://www.science.org/doi/10.1126/scitranslmed.abp8704