Artificial intelligence-based liquid biopsies, utilizing free circulating DNA and protein biomarkers, could enable the early detection of ovarian cancer, according to a study by Prof. Dr. Victor E. Velculescu and his collaborators from multiple institutions in the United States and Europe.
Ovarian cancer is the fifth leading cause of cancer-related death among women in the United States, with a five-year survival rate of approximately 50%, according to the U.S. Centers for Disease Control and Prevention.
The study developed a blood test to detect early signs of ovarian cancer by using artificial intelligence (AI) and protein biomarkers to identify genetic changes associated with the disease.
The study, published in "Cancer Discovery", a journal of the American Association for Cancer Research, utilized AI-based analysis of circulating free DNA fragments and two protein biomarkers to identify women with ovarian cancer. The two protein biomarkers, cancer-associated antigen CA-125 and human epididymal epidididymal protein 4 (HE4), have previously been recognized as biomarkers of ovarian cancer but, on their own, cannot accurately detect the disease. However, combining these biomarkers with AI-assisted detection of circulating free DNA fragments has improved screening accuracy and enhanced the ability to distinguish malignant from benign tumors.
"The combination of artificial intelligence, free circulating DNA fragments, and a set of protein biomarkers in a simple blood test has improved the detection of ovarian cancer, even in patients with early-stage disease," reports Prof. Victor E. Velculescu, M.D., Ph.D., lead study author, professor of oncology, and co-director of the Genetics and Epigenetics in Cancer Program at Johns Hopkins Kimmel Cancer Center. "This AI-based approach has the potential to be an affordable and effective method for large-scale ovarian cancer screening."
"Early detection of ovarian cancer can save lives, but most women are diagnosed at a late stage of the disease, when survival rates are significantly lower," explains co-senior author Jamie Medina, Ph.D., a postdoctoral researcher at Johns Hopkins Kimmel Cancer Center. "The absence of specific symptoms in the early stages of the disease, along with the lack of effective biomarkers, has hindered early detection of ovarian cancer."
Researchers have previously demonstrated that the DELFI (DNA Evaluation of Fragments for Early Interception) test method, which is based on artificial intelligence and fragmentomics, enhances the detection of DNA fragments in the blood and effectively identifies lung cancer. The technology leverages the fact that while DNA is organized in healthy cells, it becomes disorganized in cancer cells. When healthy cells die and lyse, they leave behind a predictable and orderly set of DNA fragments in the blood. In contrast, when cancer cells die and break down, the DNA fragments left behind are irregular and chaotic.
"Ovarian cancers have a unique pattern of DNA fragmentation that is not present in benign tumors," says Akshaya Annapragada, co-senior author and a student at Johns Hopkins University School of Medicine. The ability to distinguish between benign and cancerous tumors is important because the next step in screening women with ultrasound-detected ovarian tumors is exploratory surgery. The use of liquid biopsy-based tests could spare women with benign tumors from unnecessary surgery.
Professor Velculescu and colleagues plan to validate the usefulness of the test in larger samples from randomized clinical trials, but he considers the current results highly encouraging: "This study provides further evidence demonstrating the benefits of free circulating DNA fragmentation and artificial intelligence in detecting cancers with high accuracy. Our results show that this combined approach outperforms existing biomarkers for screening."
Professor Victor Velculescu, in his role as Chairman of the Advisory Board, actively supports the development of the Romanian Genomics Research and Development Institute. In collaboration with the "Carol Davila" University of Medicine and Pharmacy in Bucharest and the Fundeni Clinical Institute, these three institutions are currently working to validate this new early detection technology, based on artificial intelligence and fragmentomics, for other types of cancer, such as liver cancer.
Source: Ro Health Review
