BALTIMORE: Researchers at Johns Hopkins Kimmel Cancer Center have validated an artificial intelligence powered blood test that can detect liver cancer in people from different regions and backgrounds, offering new hope for earlier diagnosis and improved treatment options worldwide.
The research builds on the team’s earlier development of the DELFI (DNA Evaluation of Fragments for Early Interception) liquid biopsy platform, which examines millions of pieces of cell free DNA found in the bloodstream. The latest study tested a previously developed liver cancer detection model in independent high risk populations from Guatemala and Romania.
Liver cancer remains one of the world’s leading causes of cancer related deaths, with cases increasing due to rising rates of metabolic liver disease and other risk factors. Detecting the disease at an early stage can improve treatment success, but current screening methods, including ultrasound scans and alpha fetoprotein blood tests, can fail to identify many early cancers.
Researchers analyzed blood samples from 377 participants, including people with and without hepatocellular carcinoma, the most common type of liver cancer. The participants came from Guatemala and Romania, where the causes of liver cancer were different.
Despite these differences, the AI powered blood test successfully detected liver cancer in both groups. Researchers found that combining the test results with alpha fetoprotein levels and basic health information, including age and sex, improved detection accuracy compared with traditional blood tests alone.
Scientists also used a new method called MethID to understand how the test works. The technology tracks where DNA fragments in the blood come from and showed that the test detects signals not only from cancer cells but also from liver cells, blood vessels and immune cells responding to the disease.
Researchers said the DNA fragments provide information about the presence of cancer as well as changes happening inside the body during cancer development.
The study also identified differences in molecular patterns between populations, including genetic changes linked to aflatoxin exposure among participants. However, researchers found that the overall AI based detection system remained effective regardless of the cause of liver cancer.
Scientists said the findings support the possibility of using genome wide DNA fragment analysis as a foundation for future blood tests that could detect multiple diseases through a single technology platform.
