AI-Based ‘Liquid Biopsies’ Use Cell-Free DNA and Protein Biomarkers for Early Detection of Ovarian Cancer
Posted on 02 Oct 2024
Ovarian cancer ranks as the fifth leading cause of cancer deaths among women in the United States, with a five-year survival rate of roughly 50%. Early detection is crucial for improving survival rates, but many women are diagnosed at later stages when the prognosis is much worse. Early detection challenges stem from a lack of specific early symptoms and the absence of effective biomarkers. Now, a blood test using artificial intelligence (AI) to identify genetic changes and protein biomarkers related to cancer may help screen women for early signs of ovarian cancer.
In a new study, researchers at the Johns Hopkins Kimmel Cancer Center (Baltimore, MD, USA), in collaboration with various other institutions in the United States and Europe, utilized AI-driven analysis of DNA fragments and two known protein biomarkers to detect ovarian cancer. The two biomarkers, cancer antigen 125 (CA-125) and human epididymis protein 4 (HE4), were previously linked to ovarian cancer but lacked the reliability to detect the disease on their own. However, combining these biomarkers with AI-based analysis of cancer-related DNA fragment patterns in the blood significantly improved screening accuracy, enabling better differentiation between malignant and benign growths.
Previously, the team demonstrated that their AI-powered DELFI (DNA Evaluation of Fragments for early Interception) method, which uses a new liquid biopsy approach called fragmentomics, enhanced the detection of DNA fragments in the blood, successfully identifying lung cancer. The technology capitalizes on the fact that DNA from healthy cells is organized, while DNA from cancer cells is chaotic. As healthy cells die, they release orderly DNA fragments into the bloodstream. In contrast, cancer cells produce irregular DNA fragments when they die. In the latest study, published in Cancer Discovery, the researchers analyzed blood samples from 94 women with ovarian cancer, 203 with benign ovarian tumors, and 182 without known ovarian growths. These samples were drawn from patients treated at hospitals in the Netherlands and Denmark.
The researchers used the DELFI-Pro test, which combines AI-driven cell-free DNA analysis with CA-125 and HE4 biomarker tests, to screen for ovarian cancer. The DELFI-Pro test outperformed tests for either biomarker alone, detecting significantly more cases of ovarian cancer while minimizing false positives. It identified 72%, 69%, 87%, and 100% of ovarian cancer cases at stages I–IV, respectively. In comparison, CA-125 alone detected 34%, 62%, 63%, and 100% of cases at these stages. To further confirm the results, the researchers tested the method in a second sample group from the U.S., which included 40 patients with ovarian cancer, 50 with benign ovarian growths, and 22 without ovarian lesions. Despite the smaller sample size, the test delivered similar results, detecting 73% of all cancers and 81% of high-grade serous ovarian carcinoma, the most aggressive form of ovarian cancer, with nearly no false positives among cancer-free women.
The DELFI-Pro test also effectively distinguished benign from malignant ovarian tumors, something that standard ultrasound exams struggle to do. This distinction is critical because, after ovarian growths are detected via ultrasound, women often undergo exploratory surgery to confirm cancer. By using the DELFI-Pro liquid biopsy test, many women with benign growths could avoid unnecessary surgery. The researchers now plan to validate the test in larger randomized clinical trials, but the current results are promising.
“The combination of artificial intelligence, cell-free DNA fragmentomes and a pair of protein biomarkers in a simple blood test improved detection of ovarian cancer even in patients with early-stage disease,” said Victor E. Velculescu, M.D., Ph.D., senior author of the study, professor of oncology, and co-director of the Cancer Genetics and Epigenetics Program at the Johns Hopkins Kimmel Cancer Center. “This AI-enabled approach has the potential to be an affordable, accessible method for widespread screening for ovarian cancer.”
