U.S. scientists and biotech company AOA Dx have developed a new blood test that can detect early-stage ovarian cancer with high accuracy, offering a potential breakthrough in diagnosing one of the deadliest gynecologic cancers. The findings were presented at the 2025 American Association for Cancer Research Annual Meeting and are part of a growing body of research aimed at improving early detection, when treatment outcomes are significantly more favorable.
The test, developed in collaboration with the University of Colorado Cancer Center and the University of Manchester, combines lipidomics, ganglioside profiling, and protein biomarkers to analyze blood samples. It uses liquid chromatography mass spectrometry and artificial intelligence to interpret the molecular data. In an independent testing group, the test achieved an area under the curve of 92 percent for stage I and II ovarian cancer, surpassing the performance of conventional biomarkers such as CA125, which has been the standard in clinical use but lacks the specificity and sensitivity needed for effective screening.
Researchers at the University of Colorado are currently validating the test using samples from their Gynecologic Tissue and Fluid Bank. The platform uses machine learning algorithms to distinguish between malignant and benign conditions, reducing false positives that often lead to unnecessary surgical procedures. The high accuracy in early-stage detection could make it suitable for screening high-risk populations, including postmenopausal women and those with a family history of ovarian or related cancers.
Clinical validation underway for new AI-driven blood tests
Ovarian cancer is the fifth most common cause of cancer death among women and is frequently diagnosed at advanced stages due to the absence of reliable early detection tools. Symptoms are often vague and include abdominal bloating, pelvic pain, and changes in urinary habits, making it difficult to diagnose without invasive procedures. Most cases are only identified once the cancer has spread beyond the ovaries, significantly lowering survival rates.
If detected at stage I, the five-year survival rate exceeds 90 percent, but it drops to below 30 percent for stage III or IV diagnoses. In Australia, scientists at the University of Queensland, in partnership with INOVIQ Ltd, have developed a blood test that detects cancer by identifying extracellular vesicles released by tumor cells. Preliminary testing showed 94 percent accuracy and a 4 percent false positive rate.
Researchers explore applications beyond diagnosis
A clinical trial involving 1,500 postmenopausal, asymptomatic women is now underway to assess its performance in real-world conditions. At Johns Hopkins Kimmel Cancer Center, researchers are developing a diagnostic approach that combines tumor DNA fragmentation analysis with established biomarkers CA125 and HE4. Using artificial intelligence, the test analyzes genetic patterns in blood to identify ovarian cancer earlier than current diagnostic tools.
This method aims to address the limitations of existing tests by improving specificity and reducing missed cases. In a separate effort, the University of New South Wales has identified three DNA biomarkers capable of detecting all major ovarian cancer subtypes in blood samples. Clinical trials are expected to begin by 2026. These biomarkers have the potential to form the basis of a universal screening tool applicable across different forms of the disease.
Collectively, these advancements reflect a global initiative to improve ovarian cancer outcomes by introducing accurate, non-invasive screening methods. If confirmed in large-scale trials, these blood tests could represent a major shift in how ovarian cancer is detected and treated, enabling intervention before the disease progresses to advanced stages. – By Content Syndication Services.