Lifesaver! An AI-Based Blood Test that can Detect Liver Cancer

DELFI an AI-based blood test developed by Johns Hopkins uses DNA referencing to detect cancer

DELFI functions by identifying alterations in the DNA fragmentation of cancer cells that are lost in circulation. Cell-free DNA is the name for this kind of DNA (cfDNA). Researchers can learn more about how DNA is organized inside a cell's nucleus by analyzing the quantity and size of cfDNA found in various regions of the genome discovered in a blood sample. These discoveries can help differentiate between cancer cells and healthy cells because when cancer cells die, DNA fragments are released into the bloodstream. DELFI can detect the presence of cancer by scanning these cfDNA fragments for aberrant patterns using AI and machine learning (ML). Based on recent research of 724 individuals, scientists at the Johns Hopkins Kimmel Cancer Center have created an AI-based blood test that can identify liver cancer. In a 2021 study, the DNA assessment of fragments for early interception (DELFI) blood test technique helped researchers successfully categorize lung cancer, according to a press release.

The research team wanted to use DELFI to detect liver cancer because it was successful at finding lung cancer. The researchers used DELFI on 724 blood plasma samples from the US, the European Union (EU), and Hong Kong to assess its usefulness in detecting liver cancer. According to the Mayo Clinic, they asked it to look for hepatocellular carcinoma (HCC), the most typical form of primary liver cancer. 501 of the 724 samples were taken in the US and the EU. The ML model was trained using samples from 75 individuals who had HCC. An additional 223 plasma samples from people in Hong Kong were examined to confirm the model. These comprised samples from 90 HCC patients, 66 HBV patients, 35 HBV-related liver cirrhosis patients, and 32 individuals without any known risk factors. These samples allowed the researchers to examine cfDNA fragmentation patterns and create a DELFI score. Higher scores indicate a higher probability of cancer, and these values correspond with cancer presence.

In patients with an average risk of developing HCC, DELFI had an 88 percent sensitivity and a 98 percent specificity. It demonstrated 85% sensitivity and 80% specificity in patients who were at high risk for the illness. Individuals without cancer who had viral hepatitis or cirrhosis generally scored poorly, with median DELFI scores of 0.078 and 0.080, respectively. The 75 HCC patients in the US and EU samples had scores that were, on average, five to ten times higher, and these high scores were seen in cancers of all stages, including those in the early stages.

According to the press release, Victor Velculescu, MD, PhD, professor of oncology and co-director of the Cancer Genetics and Epigenetics Program at the Johns Hopkins Kimmel Cancer Center, "Increased early detection of liver cancer could save lives, but currently, available screening tests are underutilized and miss many cancers." Due to access issues and subpar test results, less than 20% of people at high risk for liver cancer are now screened. According to a press release from Amy Kim, MD, assistant professor of medicine at the Johns Hopkins University School of Medicine and co-senior author of the study, "This new blood test can double the number of liver cancer cases detected, compared to the standard blood test available, and increase early cancer detection." The researchers came to the conclusion that DELFI could enhance early detection but that it needs to first undergo additional validation in larger trials before being applied in a clinical context.