Background
Randomized, controlled trial (NHS-Galleri) in England. N=142,250 (71,122 intervention vs 71,128 control), ages 50–77, blood drawn at up to 3 annual visits. Intervention arm: blood tested with GRAIL multicancer early-detection (MCED) test in addition to usual care; control arm: blood stored, not tested. Participants and staff initially unaware of allocation. Question: does MCED screening shift diagnosis away from late-stage cancer?
Results
Interventions and follow up: Arm A: MCED test (results returned) + usual care, up to 3 annual rounds
Arm B: Usual care; blood stored (not tested)
Primary endpoint: Incidence rate of stage III–IV cancer among 12 prespecified cancer types
Key secondary: Incidence rate of stage IV cancer (12 prespecified types)
mFollow up: NR in abstract (analysis after 3 screening rounds + ≥12 months after last participant's third visit)
Results: Stage III–IV incidence (primary): IRR 1.03 (95% CI 0.92–1.14); P=.63 ➡️ not met
Stage IV incidence (key secondary): IRR 0.86 (95% CI 0.74–1.00)
Cancer-specific mortality / OS: NR
Arm B: Usual care; blood stored (not tested)
Primary endpoint: Incidence rate of stage III–IV cancer among 12 prespecified cancer types
Key secondary: Incidence rate of stage IV cancer (12 prespecified types)
mFollow up: NR in abstract (analysis after 3 screening rounds + ≥12 months after last participant's third visit)
Results: Stage III–IV incidence (primary): IRR 1.03 (95% CI 0.92–1.14); P=.63 ➡️ not met
Stage IV incidence (key secondary): IRR 0.86 (95% CI 0.74–1.00)
Cancer-specific mortality / OS: NR
Adverse events
Trial-related AEs: <1% of participants
Serious AEs: none
Serious AEs: none
Conclusions
Three annual rounds of MCED testing added to usual care did not reduce the incidence of stage III–IV cancer across 12 prespecified cancers. A possible signal for fewer stage IV cancers (IRR 0.86, upper CI 1.00) needs longer follow-up before any clinical-utility claim.
Key Limitations
Primary endpoint (stage shift) is a surrogate for mortality benefit, which is not reported. Stage IV result is a secondary endpoint with CI touching 1.00. Early analysis after 3 rounds with further follow-up planned. Test performance (sensitivity, PPV, false positives) not in abstract. Industry-funded (Grail). Single-country, age 50–77.
Clinical Context
First large RCT of an MCED blood test with a clinical endpoint. Does not support population MCED screening on current data; remains investigational, and screening guidelines continue to rely on organ-specific programs. Longer follow-up (stage IV and mortality) will determine whether this changes practice.
