NHS-Galleri: no fall in stage III–IV cancer after three rounds

Design

Randomised controlled trial, three annual screening rounds

Population

142,250 people aged 50–77 in England

Primary outcome

Stage III or IV cancer among 12 prespecified types

Effect

IRR 1.03 (95% CI 0.92 to 1.14); stage IV IRR 0.86 (0.74 to 1.00)

NHS-Galleri randomised 142,250 people aged 50 to 77 in England. All gave blood at up to three annual visits; in the intervention group it was tested with a multicancer early-detection test, in the control group it was stored. The primary end point was stage III or IV cancer among 12 prespecified types. The trial was funded by the test's manufacturer and published in NEJM on 22 September.

The incidence of stage III or IV cancer did not differ (incidence rate ratio 1.03, 95% CI 0.92 to 1.14). Stage IV cancer, a key secondary end point, was lower with testing (0.86, 0.74 to 1.00). Fewer than 1% had trial-related adverse events, none serious.

The test did not deliver its main promise: moving cancers from late to early stage. The stage IV signal is interesting but secondary and borderline, and stage shift is itself a surrogate for mortality. Patients asking about commercially available multicancer tests should hear that the largest randomised trial so far did not show a benefit on its primary end point.

• Do not offer multicancer blood tests as a substitute for organised screening.
• Keep cervical, breast, colorectal and, where indicated, lung screening on schedule.
• Tell patients asking about these tests that the main trial result was negative.
• Watch for longer follow-up and any mortality data.

Why it matters

It is the first large randomised test of the idea that a blood test can shift cancers to earlier stages, and it did not.

Don't overread it

The lower stage IV rate is a borderline secondary finding, not evidence that the test saves lives.

The statistics, in plain English

An incidence rate ratio of 1.03 with an interval from 0.92 to 1.14 means late-stage cancers were essentially as common with testing as without. The stage IV ratio of 0.86 has an upper limit of exactly 1.00, so the result is borderline, and it is a secondary end point, which carries less weight than the primary.

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.