NHS-Galleri: Does a multicancer blood test reduce late-stage diagnoses?

NHS-Galleri did not show fewer late-stage cancers on its primary endpoint; stage IV disease was less frequent, but uncertainty around this secondary finding and the need for longer follow-up remain (NEJM, 2026). Reading these findings together helps distinguish a test detecting cancer from screening benefiting the population.

How was NHS-Galleri conducted?

In England, 142,250 people aged 50–77 were randomized: 71,122 to testing and 71,128 to control (NEJM, 2026). Blood was collected at up to three annual visits. Samples from the testing group were analyzed; control samples were stored without testing. Participants and trial personnel were initially unaware of assignments. Testing added to usual care was thus compared with usual care alone.

According to the NEJM 2026 methods abstract, assessment followed three screening rounds and at least 12 months after the last participant’s third visit. Randomization enables comparison beyond counting cancers found among people tested: it examines whether adding screening changes late-stage diagnosis rates in the population.

What were the primary and key secondary findings?

The primary endpoint covered stage III–IV disease in 12 prespecified cancers: the incidence rate ratio was 1.03 (95% confidence interval 0.92–1.14; P=0.63). Stage IV alone in the same cancers was a key secondary endpoint; after three rounds, its rate ratio was 0.86 (95% confidence interval 0.74–1.00; NEJM, 2026). The primary rate ratio of 1.03 means that late-stage cancer occurred at almost the same rate in the testing and control groups.

The stage IV finding suggests a lower rate, but the interval’s upper bound corresponds to no difference. The nonsignificant primary finding does not justify ignoring the secondary result; nor does that result establish definitive screening benefit. The authors call for longer follow-up to assess how findings evolve.

How do the design and endpoints differ from PATHFINDER 2?

PATHFINDER 2 was a single-arm study without a comparison group, descriptively assessing test performance and diagnostic procedure safety (Nature Medicine, 2026). NHS-Galleri compared late-stage diagnoses in randomized groups. Our earlier PATHFINDER 2 report therefore addresses a different question. Detecting a cancer signal does not by itself establish screening’s clinical benefit.

What are the safety findings and other limitations?

Trial-related adverse events affected fewer than 1% of participants; none were serious (NEJM, 2026). This does not guarantee risk-free future use. Late-stage diagnosis rates and cancer deaths are different outcomes. Review was limited to the accessible author abstract; subgroup and diagnostic-workup details in the full text and supplements were not verified. Applying findings from England to Türkiye requires separate assessment.

What are the implications for clinical practice?

Screening does not establish diagnosis: positive signals require investigation, negative results do not exclude cancer, and established screening continues (NCI; accessed 2026). Conventional circulating tumor markers also generally perform inadequately in general-population screening (NCI, 2023). Our article on elevated CA 15-3 (Turkish) explains the marker–diagnosis distinction. This report does not recommend changing personal screening plans; these are assessed with a physician.

What are the answers to frequently asked questions?

Does a negative result definitively exclude cancer? A negative multicancer test does not definitively exclude cancer (NCI; accessed 2026).

Which national screenings continue in Türkiye? The Ministry of Health (2024) describes mammography every two years for women aged 40–69, HPV-DNA testing every five years for women aged 30–65, and fecal occult blood testing every two years plus a colonoscopy recommendation every ten years for women and men aged 50–70. Multicancer testing does not replace these (NCI; accessed 2026).

Does the stage IV finding prove longer survival? No; stage at diagnosis and mortality are assessed separately. Uncertainty surrounding the secondary finding requires longer follow-up.

Who funded the research?

NEJM’s 2026 funding statement reports that Grail supported NHS-Galleri.

Which sources were used?

• NEJM / NHS-Galleri, 2026. DOI: 10.1056/NEJMoa2505723. Accessed: 15 October 2026.
• Nature Medicine / PATHFINDER 2, 2026. DOI: 10.1038/s41591-026-04618-w. Accessed: 15 October 2026.
• NCI / Multicancer tests, accessed 2026. Accessed: 15 October 2026.
• NCI / Tumor markers, 2023. Accessed: 15 October 2026.
• Ministry of Health / National cancer screenings, 2024. Accessed: 15 October 2026.

Last updated: 15 October 2026 · Doktorclub Sağlık Editörleri

This English translation has not undergone physician review.