A multicancer blood test did not cut late-stage diagnoses

Design

Randomised controlled trial, three annual screening rounds

Population

142,250 adults aged 50 to 77 in England

Primary outcome

Incidence of stage III or IV cancer across 12 cancer types

Effect

Incidence rate ratio 1.03 (95% CI 0.92-1.14); stage IV 0.86 (0.74-1.00)

Blood-based multicancer early-detection tests have generated enormous interest, and the NHS-Galleri trial is the first large randomised test of whether they help. It randomised 142,250 people aged 50 to 77 in England to have blood tested or stored, across three annual rounds.

The primary endpoint was negative. The incidence of stage III or IV cancer across 12 prespecified types did not differ between groups (incidence rate ratio 1.03, 95% CI 0.92 to 1.14). A key secondary endpoint, stage IV cancer, showed a borderline reduction (incidence rate ratio 0.86, 95% CI 0.74 to 1.00). Trial-related adverse events were rare and none serious.

The practical message is that a multicancer blood test should not yet be offered as screening to shift cancers to earlier stages, let alone to reduce mortality, which this trial did not measure. The borderline stage IV signal and planned longer follow-up mean the story is not over, but the headline primary result does not support adoption now.

• Randomised trial of 142,250 adults aged 50 to 77 in England, three annual screening rounds.
• Stage III or IV cancer incidence did not differ (incidence rate ratio 1.03, 95% CI 0.92-1.14).
• A key secondary endpoint, stage IV cancer, was borderline (incidence rate ratio 0.86, 95% CI 0.74-1.00).
• Trial-related adverse events were rare and none were serious.
• The trial did not measure cancer mortality.

Why it matters

It directly challenges the assumption that a multicancer blood test will catch cancers earlier at a population level.

Don't overread it

The borderline stage IV signal and absence of mortality data mean this neither proves nor disproves long-term benefit; it shows the primary stage-shift endpoint was not met.

The statistics, in plain English

An incidence rate ratio of 1.03 with a confidence interval spanning 1.0 means no effect on stage III or IV cancer was detected. The stage IV ratio of 0.86 just touches 1.0, a hint rather than proof, which is why longer follow-up is planned.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.