According to a global analysis of the burden of liver disease, 400 million people worldwide are thought to be at an increased risk of hepatocellular cancer (HCC), a type of liver cancer, as a result of cirrhosis from chronic liver diseases like chronic viral hepatitis or non-alcoholic fatty liver disease. Researchers at the Johns Hopkins Kimmel Cancer Center have now reported that a recent trial using their cutting-edge artificial intelligence blood testing equipment has successfully diagnosed more than 80% of liver malignancies.
Their research is presented in an article titled “Detecting liver cancer using cell-free DNA fragmentomes” that appears in the journal Cancer Discovery.
The leading cause of cancer-related deaths worldwide is liver cancer. Although current screening techniques are insufficient, screening high-risk individuals, such as those with cirrhosis and viral hepatitis, offers a chance for increased survival. In this study, 724 people from the United States, the European Union, or Hong Kong who had HCC or were at average or high risk for developing it were assessed using whole-genome cell-free DNA fragmentome analysis.
The study was co-led by Victor Velculescu, MD, PhD, professor of oncology and co-director of the Cancer Genetics and Epigenetics Program at the Johns Hopkins Kimmel Cancer Center, Zachariah Foda, MD, PhD, gastroenterology fellow, Akshaya Annapragada, MD/PhD student, and Amy Kim, MD, assistant professor of medicine at the Joss-Hopkins Kimmel Cancer Center.
The machine learning model, a sort of artificial intelligence that uses data and algorithms to increase accuracy, was trained and validated using 501 of the 724 plasma samples examined, including samples from 75 persons with HCC.
The DELFI technology analyses the size and quantity of cell-free DNA present in the circulation from various areas of the genome to measure the way DNA is packaged inside the nucleus of a cell using a blood test. DNA is packaged by healthy cells like a well-organized suitcase, with distinct parts of the genome packed with care in discrete compartments. In contrast, the nuclei of cancer cells are like more jumbled luggage, with pieces of the DNA thrown in randomly. Cancer cells expel bits of DNA into the circulation as they decompose.
In the most recent investigation, cfDNA fragments extracted from plasma samples were subjected to the test. To create a DELFI score, they examined the fragmentation patterns present in each sample.
The DELFI technique had an overall sensitivity of 88% and a specificity of 98% for detecting liver cancer in its earliest stages. The test had an 85% sensitivity and an 80% specificity in samples taken from individuals with a high risk of developing HCC.
Due to accessibility issues and subpar test results, less than 20% of those at high risk for liver cancer are now checked. According to Kim, a co-senior author of the study, “This novel blood test can boost early cancer detection and potentially double the number of liver cancer cases diagnosed.”
The researchers note that one of their upcoming tasks will be to validate this strategy in larger studies for therapeutic application.
