NHS-Galleri: what the multi-cancer blood test trial showed

NHS-Galleri, a randomised trial of 142,942 adults aged 50 to 77 in the UK, did not meet its primary endpoint. Even so, stage IV diagnoses were about 14% lower in the group offered the test, and screening found four times as many cancers as in the comparison group. Whether that leads to fewer deaths is still unknown.

What was the NHS-Galleri trial testing?

The trial compared an annual Galleri blood test added to standard screening against standard screening alone, over three years, in adults with no cancer symptoms. The primary endpoint was a reduction in stage III and IV diagnoses across 12 prespecified cancers. According to ASCO, that endpoint was not statistically significant.

Galleri is a multi-cancer early detection (MCED) test: a single blood draw designed to look for signals from many cancer types at once. The trial’s team, from Queen Mary University of London, reported first results at an international cancer conference in 2026.

What did the results show, and what is still unknown?

Several numbers point in the same direction, with caveats:

• Stage IV diagnoses were 14% lower in the test group during the testing period (ASCO). The Queen Mary team reported a reduction of more than 20% in stage 4 cancers in the second and third screening rounds.
• Screening-detected cancers were 1,173 in the test group versus 290 in the control group. Total cancers diagnosed were 3,637 versus 3,400.
• Specificity was 99.55%, and the positive predictive value was 52%, meaning about one in two positive results identified a cancer.

The limits matter as much. ASCO’s summary states it is unknown whether all the cancers found with this test would ever have become symptomatic, or whether earlier diagnosis improves survival. A mortality benefit was not demonstrated. The investigators say they will keep analysing the data, including psychological effects, acceptability of screening and the effect of annual testing on cancer deaths.

GRAIL, the company that makes the test, has published its own summary of the results, and it frames them more positively than the primary-endpoint outcome alone. Reading the trial team’s and ASCO’s wording alongside a company’s is a useful habit with any screening result.

The bottom line

• In 142,942 adults aged 50 to 77, an annual multi-cancer blood test did not significantly reduce stage III and IV diagnoses across 12 prespecified cancers (primary endpoint not met).
• Stage IV diagnoses were about 14% lower, screen-detected cancers were fourfold higher (1,173 versus 290), and positive predictive value was 52%.
• The effect on survival has not been shown; follow-up on cancer deaths is ongoing.