Multi-cancer early detection (MCED)

Multi-cancer early detection (MCED) screens asymptomatic, average-risk individuals from a single blood draw for molecular signals shared across dozens of cancer types simultaneously, aiming to catch cancer at an earlier, more treatable stage before symptoms or standard single-cancer screening would detect it — a population-screening use case distinct from the treatment-matched companion diagnostics covered elsewhere in this catalog. GRAIL’s Galleri test reported full results from the NHS-Galleri trial at the 2026 ASCO Annual Meeting, demonstrating a substantial reduction in stage IV cancer diagnoses, increased stage I/II detection of deadly cancers, and a four-fold higher cancer detection rate; GRAIL also presented Pathfinder 2 results from more than 35,000 participants and has submitted an FDA premarket approval (PMA) application for Galleri. Exact Sciences launched Cancerguard, which it calls the first-of-its-kind multi-cancer early detection blood test. Freenome’s SimpleScreen colorectal-cancer blood test met all primary and secondary endpoints with 85% sensitivity for CRC and 22% for advanced precancerous lesions, alongside separate multiomic blood-screening data for lung cancer. In Switzerland, Novigenix runs an AI-powered blood-transcriptomic platform with published results on early colorectal-cancer detection, recently securing funding to scale its precision-oncology platform. In China, Singlera Genomics develops multi-cancer early-screening panels including its ColonAiQ product.

The key directions of multi-cancer early detection are:

1. Cell-free DNA methylation screening (Cell-Free DNA Methylation): methylation-pattern analysis of cell-free DNA to detect and localize a cancer signal across dozens of cancer types from one blood draw — GRAIL’s Galleri.
2. Multiomic blood screening (Multiomic Blood Screening): combining genomic, proteomic and other molecular layers from blood to improve single- and multi-cancer detection sensitivity — Freenome, Exact Sciences’ Cancerguard.
3. Blood transcriptomic AI screening (Blood Transcriptomic Screening): RNA expression signatures in blood analyzed with machine learning to flag early cancer signals — Novigenix.
4. Regional MCED panels (Regional MCED Panels): multi-cancer screening products developed and validated for a domestic population and regulatory pathway — Singlera Genomics.

Sectoral value chain#

[blood draw] ──> [cfDNA/multiomic/transcriptomic signal extraction] ──> [ML classifier & cancer-signal-origin prediction]
                                                    │
                                       (screening result & confirmatory workup)
                                                    │
                                                    ▼
[population screening program adoption] <─── [regulatory approval & clinical validation] <─────┘

Value chain levels#

Cross-cutting technologies of the sector:

• Cell-free DNA methylation analysis (Cell-Free DNA Methylation): cancer-specific methylation patterns in cell-free DNA distinguish tumor-derived signal from background and predict the cancer’s tissue of origin, GRAIL’s core Galleri technology.
• Multiomic integration (Multiomic Blood Screening): combining multiple molecular data types (genomic, proteomic) from a single blood sample improves detection sensitivity beyond any single biomarker class, Freenome’s core approach.
• AI-based transcriptomic classification (Blood Transcriptomic Screening): machine-learning models trained on blood RNA expression signatures flag early cancer-associated immune and tissue response patterns, Novigenix’s core technology.

The stack converts a single blood draw into a multi-cancer signal by extracting a molecular layer (cfDNA methylation, multiomic markers, or transcriptomic RNA signatures) and classifying it with machine learning trained on large prospective screening cohorts.

1. cfDNA methylation-based detection (Cell-Free DNA Methylation):
   • GRAIL’s Galleri analyzes methylation patterns across cell-free DNA fragments to detect a shared cancer signal spanning dozens of cancer types and predict the likely tissue of origin, validated in the NHS-Galleri and Pathfinder 2 trials.
2. Multiomic blood-based screening (Multiomic Blood Screening):
   • Freenome combines genomic and proteomic signal layers from a single blood draw, achieving 85% sensitivity for colorectal cancer in its SimpleScreen test while extending the same multiomic approach to lung-cancer screening.
3. AI-powered blood transcriptomics (Blood Transcriptomic Screening):
   • Novigenix applies machine learning to blood RNA expression signatures, identifying immune and tissue-response patterns associated with early-stage colorectal cancer.
4. Domestic MCED panel development (Regional MCED Panels):
   • Singlera Genomics builds multi-cancer screening panels (including ColonAiQ) validated against China’s domestic population and regulatory requirements.