Two biotechnology companies are in the process of developing liquid biopsy tests that could detect ovarian cancer at early stages.

While these tools by Mercy BioAnalytics and DELFI Diagnostics are not yet FDA-approved, recent trials showed that they are much more sensitive than current screening methods for identifying stages I and II ovarian cancer.

“Early detection has been the holy grail in the field for decades,” said Ron Drapkin, MD, PhD, director of the Penn Ovarian Cancer Research Center in Philadelphia. This is because the current methods used to detect ovarian cancer often miss the disease at earlier stages, delaying diagnosis until after the cancer advances and becomes much harder to treat.

When a patient is diagnosed with stage I ovarian cancer, for example, the disease has a 5-year survival rate of nearly 90%. But when the cancer isn’t detected until stage III, after it spreads beyond the pelvis, this rate plummets to 40% — and by stage IV, as the cancer expands to the lungs, liver, or outer lymph nodes, the 5-year survival rate is only about 20%.

What Are the Shortcomings of Current Tests Used to Diagnose Ovarian Cancer?

Currently, more than 80% of patients are not diagnosed until these later stages. This is not only because early-stage ovarian cancer causes no symptoms but also because CA-125 and HE4, proteins used as the primary blood biomarkers of ovarian cancer, are usually not released at high enough levels for detection during early stages. In addition, these biomarkers can also be elevated in patients with benign conditions like endometriosis and fibroids. “[These tests] work better if there is some sort of mass that’s already seen,” said Andrew Wagner, MD, a clinical geneticist at Northwestern Medicine in Chicago.

The shortcomings of these techniques have manifested as failing to significantly reduce mortality, even when utilized for routine screening.

Meanwhile, while preventive interventions like bilateral salpingo-oophorectomy can reduce the risk for ovarian cancer by more than 90%, it is often difficult to determine which patients need this treatment. Currently, the only factors used for risk stratification are family history and mutations in genes like BRCA1 and BRCA2, which, while linked with significantly higher risk for ovarian cancer, are not present in the majority of patients diagnosed with the disease.

Researchers have identified a number of other genes, including PALB2 and RAD51, with smaller, but significant, links to the risk for ovarian cancer. However, a polygenic risk score, which could distill patients’ genetic risk into a single number, has not yet been developed for ovarian cancer.

“That’s been something that people have been trying to get at for a long time,” said Drapkin, “because outside of the BRCA1 and 2 genes and very strong family histories, we don’t yet have a clinically useful scoring system to define someone’s risk.”

What Tests Are Being Investigated That Show Potential for Early Detection?

While there is currently no reliable tool for detecting ovarian cancer before it advances, this could change in the near future with the advent of liquid biopsy tools that researchers are working hard to validate for clinical use.

For example, Mercy BioAnalytics developed the Evident test, which uses blood-based liquid biopsy to identify cancer-associated signatures carried by extracellular vesicles and particles (EVPs). This approach is similar to circulating tumor DNA (ctDNA) testing, which looks for ctDNA shed by tumor cells as they die and is often used for ovarian cancer treatment monitoring. Because most early-stage tumors shed very little ctDNA, ctDNA testing cannot reliably detect ovarian cancer at early stages. EVPs, however, are secreted by living tumor cells beginning in early stages of tumor development, meaning Mercy BioAnalytics’ test can detect the disease much earlier.

The Evident test uses biomarker colocalization, where multiple biomarkers are detected in combinations of three.

“If you would look at these proteins by themselves, just one by one, they wouldn’t be very specific,” said Toumy Guettouche, PhD, chief scientific officer at Mercy BioAnalytics. But with the Evident test, he said, “[y]ou only get a signal if you have all three biomarkers on the same [extracellular vesicle],” making it easier to determine which EVPs came from cancer cells. “Our test is very good at distinguishing the healthy background from the tumor signal,” Guettouche said.

The test uses an algorithm to generate a probability score between 0 and 1, and patients with a higher score receive a positive result indicating suspected ovarian cancer. To diagnose a patient, physicians would need to follow this result with further testing, including a transvaginal ultrasound and CT/MRI.

Mercy BioAnalytics analyzed the test using a retrospective sample from the UKCTOCS trial, a longitudinal study on postmenopausal patients without increased risk for ovarian cancer. Using preclinical samples of patients eventually diagnosed with ovarian cancer, Mercy BioAnalytics’ test performed with 89% sensitivity and 98% specificity for early-stage detection.

While this study was retrospective, it was designed to minimize bias by using a preclinical sample rather than patients who had already been diagnosed, which is common in prospective studies.

“We are the only test that, in the intended use population, showed statistically significantly better performance than CA-125,” Guettouche said.

Mercy BioAnalytics’ test was given FDA Breakthrough Device Designation in 2024, and the company also submitted an application for full FDA approval, with the goal of receiving this in 2028. Currently, the test is available as a laboratory-developed test (LDT) in Georgia, Florida, and Texas.

Another test being studied is agnostics’ DELFI-Pro liquid biopsy test, which uses a machine learning model to analyze fragmentation patterns in cell-free DNA (cfDNA) in order to detect tumor DNA, combining this with measures of CA-125 and HE4. Unlike these protein biomarkers, tumor DNA patterns are not influenced by benign conditions. Therefore, Drapkin said, “the DNA coming from the tumor specifically adds a layer of stringency that we don’t have at the protein level right now.”

A 2025 Cancer Discovery paper, co-authored by Drapkin, showed that DELFI-Pro had a sensitivity of about 72% for stage I ovarian cancer, with a specificity of > 99%. At the same specificity, CA-125 alone only had a sensitivity of 34% for stage I, whereas HE4 alone had a sensitivity of 28%. Prior research has estimated the sensitivity of combined CA-125 and HE4 to be about 58.1% with a specificity of 97.7% when testing for stages I and II ovarian cancer, meaning DELFI-Pro outperforms the combined proteins as well.

Altogether, “it looks like there’s a lot of potential for [cell-free DNA], especially if people start combining it with other more traditional approaches,” Drapkin said.

A test using DELFI Diagnostics’ cfDNA fragmentation technology from DELFI-Pro is currently available as an LDT for lung cancer, and the company plans to follow this by making its ovarian cancer test available too.

“The goal is to do it in a very common tumor type where you have a lot of patients and then get into the less common subtypes like ovarian,” said Drapkin.

Drapkin’s own team is working on a precancer atlas, which would map ovarian cancer’s development beginning with precancer changes and help researchers uncover new biomarkers for early detection. While the new markers identified so far are still at an experimental stage, the researchers have started collaborating with DELFI Diagnostics to see if any of them can improve DELFI-Pro’s sensitivity.

Overall, researchers are hopeful that the day when clinicians can routinely diagnose ovarian cancer at early stages is not far away.

Drapkin reported serving as co-chair in the Scientific Advisory Board for the Honorable Tina Brozman Foundation, a foundation focused on early detection of ovarian cancer. He also reported serving on the Board of the Powell-Drescher Ovarian Cancer Research Foundation, the scientific advisory board of the Pershing Square Philanthropies Lotus Award committee, and the advisory board of the Ovarcome Foundation.

Guettouche reported being an employee of Mercy BioAnalytics, holding stock options, and receiving a salary from the company.