To improve ovarian cancer screening, researchers led by Johns Hopkins University’s Victor E. Velculescu, MD, PhD, developed a combination liquid biopsy approach. It has three components: measurements of two protein biomarkers, measurements of cell-free DNA (cfDNA) fragments, and artificial intelligence (AI)-driven analysis of the protein measurements and the cfDNA “fragmentomes.” The researchers found that their approach could be used to differentiate benign masses from ovarian cancers with high accuracy.
The two protein biomarkers, cancer antigen 125 (CA-125) and human epididymis protein 4 (HE4), were previously identified as ovarian cancer biomarkers. However, on their own, CA-125 and HE4 do not enable the reliable detection of ovarian cancer. Better results could be achieved if assays were to use a larger number of proteins. Indeed, such assays have shown some promise, but they are not yet broadly available for research or clinical use.
Rather than add proteins to an ovarian cancer screen, the researchers chose to test whether protein measurements and cfDNA measurements could provide complementary information and, upon analysis by artificial intelligence, enable more accurate—but still accessible—results.
The researchers reported their findings in Cancer Discovery, in an article titled, “Early detection of ovarian cancer using cell-free DNA fragmentomes and protein biomarkers.”
“[We evaluated] 591 women with ovarian cancer, benign adnexal masses, or without ovarian lesions,” the article’s authors wrote. “Using a machine learning model with the combined features, we detected ovarian cancer with specificity >99% and sensitivity of 72%, 69%, 87%, and 100% for stages I–IV, respectively. At the same specificity, CA-125 alone detected 34%, 62%, 63%, and 100% of ovarian cancers for stages I–IV.”
The researchers added that their approach differentiated benign masses from ovarian cancers with high accuracy. They also noted that they obtained their initial results by evaluating samples from 479 women, who were treated at hospitals in the Netherlands and Denmark. These results were confirmed in a second population of 112 women, who were treated in the United States.
Even in this smaller sample, the test achieved similar success rates, with 73% of all cancers detected and 81% of the high-grade serous ovarian carcinoma, the most aggressive form of the disease, with almost no false positives in women without cancer. The combination approach was also able to effectively distinguish between benign growths and cancerous tumors—something ultrasound exams cannot.
“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 Velculescu, who is 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.”
In their study, the investigators incorporated the AI-powered DELFI (DNA Evaluation of Fragments for early Interception) test method, which utilizes a new approach for liquid biopsies, called fragmentomics, that improves detection of DNA fragments in the blood and effectively detects lung cancer. The technology takes advantage of the fact that DNA, neatly packaged in healthy cells, becomes disorganized in cancer cells. When healthy cells die and break apart, they leave behind a predictable, orderly set of DNA fragments in the blood. However, when cancer cells die and break apart, the DNA fragments left behind are irregular and chaotic.
“Ovarian cancers have a unique DNA fragmentation signature that is not present in benign lesions,” said Akshaya Annapragada, co-first author and an MD/PhD student at the Johns Hopkins University School of Medicine. Being able to distinguish benign from cancerous ovarian growth is important because the next step in cancer screening for women with ovarian growths detected via ultrasound is exploratory surgery. Using the liquid biopsy tests could spare women with benign growths having to undergo unnecessary surgery.
“This study provides further evidence demonstrating the benefit of genome-wide, cell-free DNA fragmentation, and artificial intelligence to detect cancers with high accuracy,” Velculescu stated. “Our results show that this combined approach has higher performance for screening than existing biomarkers.” Velculescu and his colleagues intend to validate the test’s utility in larger samples from randomized clinical trials.
