A Blood Test That Reads DNA Wreckage Could Catch Liver Scarring Before It Turns Lethal

About 100 million Americans are living with some form of liver disease. Most of them have no idea.

The liver is a quiet organ. Fibrosis - the accumulation of scar tissue that builds up when the liver is chronically inflamed - produces no reliable symptoms in its early stages. By the time most patients are diagnosed, the scarring has advanced to cirrhosis, a condition that dramatically raises the risk of liver cancer and liver failure. Current blood tests detect cirrhosis only about half the time. Imaging requires specialized ultrasound or MRI equipment that many patients cannot easily access.

Reading the genome's wreckage for signs of disease

A team at the Johns Hopkins Kimmel Cancer Center has been working on a different approach. Their tool does not search for a single biomarker or a specific gene mutation. Instead, it analyzes the physical state of DNA circulating in the blood - how it has been cut, packaged, and scattered across the genome - to infer what is happening inside the body's tissues.

This approach, called fragmentome analysis, was originally developed to detect cancer. The idea is that when cells die, they release fragments of DNA into the bloodstream. The pattern of those fragments - their size, their location across the genome, how they cluster in repetitive regions - carries information about the cells that made them. In cancer patients, fragmentation patterns are distinctly abnormal. The question the Hopkins team asked: do other diseases leave their own readable signatures?

The answer, according to a study published March 4 in Science Translational Medicine, appears to be yes - at least for liver disease.

40 million fragments per sample

The researchers analyzed cell-free DNA from 1,576 people with liver disease and other comorbidities, using whole-genome sequencing. For each sample, roughly 40 million DNA fragments spanning thousands of genomic regions were examined - including repetitive sequences that most liquid biopsy tests ignore. Machine-learning algorithms then sorted through that data to find patterns distinguishing healthy tissue from diseased.

The result was a classifier that detected early liver fibrosis and advanced cirrhosis with high sensitivity. It represents the first systematic use of fragmentome technology outside of cancer detection.

"This builds directly on our earlier fragmentome work in cancer, but now using AI and genome-wide fragmentation profiles of cell-free DNA to focus on chronic diseases," said Victor Velculescu, M.D., Ph.D., co-director of the cancer genetics and epigenetics program at Hopkins and co-senior author of the study. "Liver fibrosis is reversible in its early stages, but if left undetected, it can progress to cirrhosis and ultimately increase the risk of liver cancer."

Not mutations - patterns

The distinction between this test and most liquid biopsies is worth pausing on. Conventional blood-based cancer tests look for specific mutations in known oncogenes or tumor suppressors. The fragmentome approach does something broader. It treats the genome's physical fragmentation state as a readout of cellular biology.

"The fact that we are not looking for individual mutations is what makes this study so powerful," said first author Akshaya Annapragada, an M.D./Ph.D. student working in the Velculescu lab. "We are analyzing the entire fragmentome, which contains a tremendous amount of information about a person's physiologic state."

Crucially, the liver fibrosis classifier is disease-specific and does not cross-react with cancer classifiers built from the same platform. The same underlying technology can generate distinct tests for distinct conditions.

A comorbidity index built from blood

The team also explored whether fragmentome data could predict overall health burden. In a cohort of 570 individuals presenting with suspected serious illness, they developed a "fragmentation comorbidity index" - a score derived from DNA patterns that correlated with the Charlson Comorbidity Index, a standard tool clinicians use to estimate mortality risk from co-existing conditions.

The fragmentome index independently predicted overall survival and, in some analyses, outperformed traditional inflammatory markers. Certain fragmentation signatures correlated with worse clinical outcomes across cardiovascular, inflammatory, and neurodegenerative conditions - though the sample sizes for those conditions were too small to build disease-specific classifiers. That work is continuing.

What this is - and what it isn't yet

The liver fibrosis assay is a research prototype, not a clinical test. The study enrolled patients who already had known liver disease or comorbidities, which could make the classifier's task easier than it would be in a general screening population. Real-world performance will depend on validation in larger, more diverse cohorts - studies the Hopkins team identifies as the next priority.

There are also competing interests to note: several authors, including Velculescu, are founders of or investors in companies - Delfi Diagnostics and Artemyx - commercializing related cell-free DNA technologies. Johns Hopkins University holds equity in Delfi Diagnostics. These arrangements have been reviewed under the university's conflict-of-interest policies.

The research originated from a 2023 study of liver cancer patients, where Velculescu's team noticed that people with fibrosis or cirrhosis showed subtle but detectable fragmentome signals even before cancer developed. That observation seeded the current investigation.

"If we can intervene earlier - before fibrosis progresses to cirrhosis or cancer - the impact could be substantial," Velculescu said. For a condition affecting a hundred million Americans with no adequate early warning system, that is not a small claim. But the science, at this stage, appears to back it up.