Multi-cancer early detection: companies, research and trends

GRAIL, Exact Sciences and Freenome Inc. are among the companies developing multi-cancer early detection tests. GRAIL’s Galleri uses blood-based signals and predicts a likely tissue of origin; Exact Sciences is developing CancerGuard; Freenome combines molecular signals from tumor and non-tumor sources with machine learning. SeekIn Inc. is also developing blood-based tests and has patent applications covering multi-omics and protein markers.

The National Cancer Institute, Dana-Farber Cancer Institute and Johns Hopkins University are among the research organizations active in MCED. The National Cancer Institute studies randomized screening-trial design, while Dana-Farber examines follow-up after positive results and Johns Hopkins studies clinical outcomes and false positives. Mayo Clinic, MD Anderson and Fred Hutchinson Cancer Center also contribute research on evaluation, implementation and screening performance.

The dataset records $8,369,024 in grants across 14 awards since 2020. It also lists a September 2026 grant, Enlighten, for Proteotype Diagnostics Ltd’s work on affordable, immune-based multi-cancer early detection, but gives no award amount. These grant figures are distinct from funding-round records, which may include activity not specific to MCED.

SeekIn, Inc., Shenzhen, China, has 5 listed patents, while GRAIL and SeekIn Inc. each have 4. Geneseeq Technology Inc. has 2. The listed patent holders also include the California Institute of Technology, the University of Southampton, SEEK Ltd. and Tencent Technology Shenzhen. SeekIn’s patent applications include multi-omics marker panels and protein tumor markers.

No: the dataset lists 21 patents in the 24-month window and reports patent activity down 25% against the previous 12 months. Activity varies by approach: cfDNA methylation profiling accounts for 8 patents in the window, genomic alteration detection for 5, and multi-analyte and multi-omic models for 4. Patent counts alone do not establish clinical maturity.

Recent developments include GRAIL’s Galleri receiving a favorable FDA advisory panel recommendation in September 2026 and a proteomics review published in October 2026. The review considered Olink, SomaScan and mass spectrometry evidence. Clinical research is also examining hereditary-risk screening, trial feasibility, cancer natural history and how patients receive confirmatory diagnostics after a positive result.

The dataset identifies 5 approaches: cfDNA methylation profiling, genomic alteration detection, protein biomarker panels, cfDNA fragmentomics, and multi-analyte and multi-omic models. They differ in the biological signals measured and the way those signals are interpreted. DNA approaches can seek tumor-derived changes, protein panels can capture tumor and host-response signals, and multi-omic models combine evidence, with added complexity in validation.

The United States leads the dataset with 3,376 documents, followed by the United Kingdom with 1,191 and China with 1,125. Germany and India are also represented, with 528 and 509 documents respectively. These counts indicate where activity appears in the dataset; they do not by themselves measure clinical adoption, research quality or the location of all test development.

The listed applications include average-risk population screening, high-risk screening, occupational health screening, and primary care or direct-to-consumer testing. The dataset associates 120 documents with average-risk screening, 98 with high-risk screening, 10 with occupational screening and 118 with primary care or direct-to-consumer testing. Recent work also examines hereditary cancer risk and firefighter screening initiatives.