Multi-Cancer Early Detection Tests Have Insufficient Sensitivity for Population Screening

GRAIL's Galleri test can detect 50+ cancer types from a blood draw, but sensitivity varies dramatically by cancer stage and type — high for late-stage cancers but potentially missing many early-stage ones. GRAIL submitted FDA PMA in January 2026 with data from 25,490 PATHFINDER 2 participants and 140,000+ NHS-Galleri trial participants. Over 475,000 commercial tests sold.

Galleri uses targeted methylation-based cell-free DNA analysis. In PATHFINDER 2, the test showed improved cancer detection vs standard of care. The NHS-Galleri trial is the first prospective RCT of an MCED test in 140,000+ participants. Key limitation: sensitivity for Stage I cancers (particularly pancreatic, ovarian, and brain) remains lower than for Stage III/IV. Specificity is high (~99.5%) but at population scale, even 0.5% false positive rate means thousands of unnecessary follow-up procedures.

Why It Matters

Cancer kills 600,000 Americans/year. 5-year survival for Stage I vs Stage IV differs dramatically: breast (99% vs 29%), lung (63% vs 7%), colorectal (91% vs 14%). Earlier detection could prevent hundreds of thousands of deaths annually.

Startup Approach

Develop complementary biomarker panels that improve sensitivity for the cancers MCED tests miss most (pancreatic, ovarian, brain). Or build AI integration platform that combines MCED results with imaging, family history, and wearable health data for risk-stratified screening protocols.

NIH Funding

NCI Early Detection Research Network (EDRN) funds biomarker validation. NCI budget: $7.3B (FY2024). Cancer Moonshot initiative supports early detection research.

Who's Working On It

GRAIL/Illumina (Galleri, 475K+ tests sold, FDA PMA submitted Jan 2026), Exact Sciences (multi-cancer blood test), Guardant Health (Shield for CRC), Freenome (AI + multi-omic), Thrive Earlier Detection (CancerSEEK)