AI Blood Assay Detects Liver Cancer Across Diverse Populations

Researchers at the Johns Hopkins Kimmel Cancer Center have validated an artificial intelligence (AI)-powered blood test that accurately detects liver cancer in people from two geographically and biologically distinct populations. The team also uncovered underlying biological signals that make the test effective.

The findings, published in Cell Press Blue, build on the team’s earlier development of the DELFI (DNA Evaluation of Fragments for Early Interception) liquid biopsy platform, which analyzes millions of fragments of cell-free DNA circulating in the bloodstream. The new research validates a previously developed liver cancer classifier in independent high-risk populations and provides new insights into the biology behind its performance. This study also advances research from March 2026, in which the researchers demonstrated that a similar genome-wide fragmentome technology could detect liver fibrosis and cirrhosis, conditions that often precede liver cancer.

“Our earlier studies showed that fragmentome analyses could detect liver cancer and, more recently, chronic liver diseases that increase cancer risk,” said Victor Velculescu, M.D., Ph.D., the Cancer Genetics and Epigenetics professor, co-director of the cancer genetics and epigenetics program, and co-senior study author. “This study demonstrates the approach works with high performance across different patient populations while revealing the biological signals in the bloodstream that make this type of detection possible.”

Liver cancer is one of the leading causes of cancer deaths worldwide, and its incidence continues to rise due to increasing rates of metabolic liver disease and other risk factors. Detecting liver cancer early dramatically improves treatment options, but current screening methods—which rely primarily on ultrasound imaging and the blood protein alpha-fetoprotein (AFP)—can miss many early cancers, the researchers explain.

To determine whether the DELFI approach could reliably detect liver cancer regardless of its underlying cause, investigators analyzed blood samples from 377 people from Guatemala and Romania, with and without hepatocellular carcinoma, the most common form of liver cancer. The two populations represent markedly different causes of liver cancer. Most participants in Romania developed liver disease related to viral hepatitis or alcohol use, while participants in Guatemala primarily had metabolic liver disease, obesity, and diabetes, with many also exposed to aflatoxin, a naturally occurring toxin linked to liver cancer. Study co-author John Groopman, Ph.D., Anna M. Baetjer Professor at the Johns Hopkins Bloomberg School of Public Health and associate director of the population sciences at the Johns Hopkins Kimmel Cancer Center, is a leading expert on the role of aflatoxins in initiating liver cancer.

Despite these differences, the blood test consistently detected liver cancer across both populations. When combined with AFP and simple clinical risk factors, such as age and sex, the approach identified early- and late-stage cancers with greater sensitivity than existing blood testing alone.

Using a newly developed method for tracing where DNA fragments originate, called MethID, the team showed that the DELFI test works because it captures more than just signals from tumor cells. It also captures signals from liver cells, blood vessels, and immune cells responding to the cancer.

“As a result, these DNA fragments contain much more information than whether cancer is present,” stated Zachariah Foda, M.D., Ph.D., assistant professor of medicine at the Johns Hopkins University School of Medicine and co-senior study author. “It tells us where these fragments originate and how they change during cancer development, allowing us to better understand the biology of the disease and improve our ability to detect it.”

The researchers also identified molecular signatures that differed between populations. For example, they detected a distinctive mutation pattern across the genome associated with aflatoxin exposure in participants from Guatemala, but they also showed that the overall fragmentome classifier remained effective regardless of the underlying cause of the liver cancer.

The study suggests that genome-wide fragmentome analysis captures both universal biological features of liver cancer and region-specific molecular changes, making it adaptable to diverse patient populations globally.

The findings also extend the researchers’ broader vision for fragmentome technology, suggesting that genome-wide analysis of cell-free DNA could ultimately provide a foundation for noninvasive blood tests capable of detecting multiple diseases using a common technology platform.  As an example, Velculescu and colleagues working with DELFI Diagnostics recently reported the clinical validation of a blood test for lung cancer screening, called FirstLook Lung, which is already available in certain health systems nationwide.  

The investigators claim future studies will focus on prospective clinical validation and further refining multimodal liquid biopsy approaches that combine fragmentome analysis with protein biomarkers and clinical risk factors to improve early detection of liver cancer.

In addition to Foda and Velculescu, other researchers were Hope Orjuela, Carter Norton, Shashikant Koul, Daniel C. Bruhm, Akshaya Annapragada, Jason Zavras, Sarah Short, Keerti Boyapati, Adrianna Bartolomucci, Vilmos Adleff, Nicholas Vulpescu, Kaui Patricio Lebarbenchon, Jacob Carey, Carter Portwood, Andrei Sorop, Razvan Iacob, Speranta Iacob, Liana Gheorghe, Simona Dima, Manuel Ramírez-Zea, Katherine McGlynn, Jillian Phallen, Robert Scharpf, and John Groopman.

This research was supported by the Dr. Miriam and Sheldon G. Adelson Medical Research Foundation, the Commonwealth Foundation, the Cole Foundation, a research grant from Delfi Diagnostics, National Institutes of Health grants CA121113, CA006973, CA233259, CA062924, CA271896, T32GM136577, and T32GM007814-40, and Department of Defense grant HT94252510480.

The researchers disclose the following competing interests. Phallen, Adleff and Scharpf are co-founders of Delfi Diagnostics, and Adleff, Scharpf, and Foda are consultants. Phallen, Scharpf, and Foda are inventors on patent applications submitted by Johns Hopkins University related to cfDNA and cancer detection that have been licensed to Delfi Diagnostics.  Velculescu is a founder of Delfi Diagnostics and Artemyx, serves on the board of directors and owns stock of both organizations. The Johns Hopkins University owns equity in Delfi Diagnostics. Velculescu divested his equity in Personal Genome Diagnostics (PGDx) to LabCorp in February 2022. He is an inventor on patent applications submitted by Johns Hopkins University related to cancer genomic and cell-free DNA analyses that have been licensed to one or more entities, including Delfi Diagnostics, Artemyx, LabCorp, Qiagen, Sysmex, Agios, Genzyme, Esoterix, Ventana and ManaT Bio. Under the terms of these license agreements, the university and inventors are entitled to fees and royalty distributions. Velculescu is an adviser to Viron Therapeutics and Epitope. Annapragada, Phallen, Scharpf and Foda are co-founders of Artemyx and hold equity in Artemyx. All arrangements have been reviewed and approved by The Johns Hopkins University in accordance with its conflict-of-interest policies.