Researchers identified coordinated lipid, metabolite, and protein signatures in symptomatic patients, supporting a multi-omic strategy for earlier ovarian cancer detection.
AOA Dx announced the publication of a peer-reviewed study in the journal Diagnostics that validates the biology behind its multi-omic approach to detecting ovarian cancer. The study analyzed ovarian cancer serum in a symptomatic population to provide a biological foundation for the company’s AKRIVIS GD diagnostic test and GlycoLocate discovery platform.
The research enrolled 503 participants across six clinical groups, including early- and late-stage ovarian cancer, borderline tumors, benign gynecological conditions, gastrointestinal disorders, and healthy controls. Researchers performed lipidomic, metabolomic, and protein profiling within the cohort using liquid chromatography-mass spectrometry and immunoassay.
According to the study, cancer exhibits distinct molecular signatures across lipids, metabolites, and proteins. While each molecular layer showed specific profiles compared with benign serum, integrating all three revealed biological connections linking membrane remodeling, redox balance, energy metabolism, and immune signaling.
The findings indicated that no single biomarker class is sufficient for early detection. Standard-of-care markers such as CA125 and HE4 showed significant overlap between early-stage ovarian cancer and benign conditions in symptomatic women. The multi-omic framework identified biological connections between molecular classes that are not visible to individual tests.
“This study reflects years of methodical work to understand not just whether molecular signatures exist in ovarian cancer serum, but how they are organized across biological systems,” says Oriana Papin-Zoghbi, CEO and co-founder of AOA Dx, in a release. “The cross-omic network we’ve characterized here demonstrates that ovarian cancer rewires multiple molecular pathways simultaneously and that the most diagnostically informative signals often live in the interactions between those pathways, not within any single one.”
The study also suggests the platform can detect a biological continuum of disease rather than a binary signal, as borderline tumors and benign conditions showed molecular profiles that fall between healthy and malignant specimens.
Industry data from the TD Cowen 5th Annual Tools/Dx Revolution Conference in June 2026 indicates that 81% of surveyed investors believe multi-omics will become the standard of care or see broad adoption within five years. This trend is driven by improving technology, decreasing costs, and evidence that integrating multiple biological layers delivers superior clinical performance over single-modality tests.
The GlycoLocate platform is designed to identify coordinated cross-omic biomarker signatures using lipidomics, metabolomics, and artificial intelligence-driven discovery. The AKRIVIS GD test combines lipidomic profiling, protein immunoassay, and machine learning to provide a blood-based signal for physicians when a patient presents with symptoms.
