The NHS-Galleri trial: What to expect in 2026 from the world’s first randomized controlled trial of a Multi-Cancer Early Detection test, Galleri®

As the multi-cancer early detection (MCED) field evolves, we and others have asserted that it is essential - in order for individuals to have confidence in a test - that it must have demonstrated performance in a prospective trial in the intended use population (Ofman 2025). In addition, payor and large health system adoption of a test across their populations requires evidence that the use of the test improves outcomes across a population, and the benefits outweigh potential harms. To understand whether Galleri provides a meaningful benefit at population scale, we needed a randomized controlled trial (RCT) to evaluate whether screening with Galleri improves outcomes in real-world practice.

That is why we launched the NHS-Galleri trial in England in 2021 (NCT05611632; ISRCTN91431511). The NHS-Galleri trial - the only population-scale RCT of an MCED - is evaluating annual screening with Galleri alongside standard NHS cancer care (Neal 2022). Reflecting the huge public appetite for a test like this, we enrolled more than 140,000 people in just over 10 months during 2021-2022 (Swanton 2025) - a remarkable achievement, particularly in the shadow of the COVID-19 pandemic.

As we look ahead to the main trial results in 2026, here's what to expect - and why the trial was designed the way it was.

A Large, Pragmatic RCT Embedded In The NHS
We designed the trial within the NHS to leverage high quality, nationally collected, consistently curated and centralized cancer datasets, enabling prospective follow-up of every participant's cancer and health status. This infrastructure supports robust evaluation of outcomes, diagnostic pathways, treatments, complications and healthcare resource use (Neal 2022).

Leveraging the NHS DigiTrials invitation system enabled us to enroll more than 140,000 individuals aged 50 to 77, who had no signs or symptoms raising clinical suspicion of cancer and who had not been diagnosed or treated for cancer in the prior three years. Critically, we designed the invitation system and trial operations to mitigate the "healthy volunteer" effect seen in almost all large cancer screening trials conducted to date and successfully recruited a demographically representative cohort, including from the most socio-economically deprived communities, where the cancer burden is highest (Swanton 2025).

Participants were randomized after their first blood draw to either:

• Intervention: blood tested with Galleri annually for three years, or
• Control: blood collected and banked for potential future analysis.

Crucially, Galleri's "Cancer Signal Origin" (CSO) feature - which indicates the tissue or organ most likely to be the source of a detected cancer signal - enabled diagnostic investigation through existing NHS pathways. This helped focus diagnostic workups more efficiently and consistently across sites, while ensuring parity between intervention and control arms through standard-of-care pathways (Lowenhoff 2025; Neal 2022).

Late-stage Reduction as the Primary Objective
The primary objective of the NHS-Galleri trial is to demonstrate a statistically significant reduction in the incidence of late-stage cancers in the intervention arm compared with the control arm. This will be assessed after the follow-up of the three annual screens and should be available in mid 2026 (Neal 2022).

To maximize clinical relevance and statistical rigor, the analysis will be conducted sequentially across three prespecified cancer groupings. First, we will look for a statistically significant reduction in stage III and IV cancer in a pre-specified group of 12 cancer types that together represent approximately two-thirds of cancer deaths in the UK and US, most of which do not have any screening available today (Note 1). If a statistically significant reduction is found, the primary objective is met. A subsequent analysis will look for a difference in all routinely staged cancer types other than prostate cancer (given that prostate cancer is known to be subject to overdiagnosis, which impacts the perceived benefit of late-stage reduction). If the second evaluation also shows a significant reduction, then all stageable cancers, including prostate, will be analyzed. This stepwise approach concentrates first on cancers where there is the greatest unmet need and where earlier detection could have the greatest impact on outcomes.

The primary objective of the NHS-Galleri trial is grounded in a growing body of evidence showing that, across multiple cancers, reductions in late-stage disease are strongly associated with reductions in cancer mortality (Dai 2024; Sasieni 2025a). In other words, if a screening program reliably shifts diagnoses from late to earlier stages, we should expect to see mortality benefits emerge over time. Measuring late-stage incidence after three annual screens, rather than waiting solely for mortality data, allows earlier, evidence-based decisions on next steps, enabling us to start improving outcomes much earlier than would otherwise be the case.

Importantly, although the trial is not designed or statistically powered to detect a reduction in cancer-specific mortality, this will be assessed in several complementary ways. First, in the near-term, we will model predicted cancer-specific mortality based on observed differences in stage distribution between the trial arms, benchmarking against outcomes for similar NHS cancer patients. Second, we will conduct a novel "nested mortality" analysis that focuses on where screening would confer benefit - among test-positive individuals - by retrospectively testing stored samples from control participants who are later diagnosed with cancer (Hackshaw 2021; Sasieni 2025b). Finally, observed cancer-specific mortality will be reported at three and six years after the final screen; most screening studies have taken a decade or two to read out their mortality findings.

What Else Will the Trial Tell Us?
The NHS-Galleri trial will also consider other very important objectives, including: reduction in stage IV cancer; performance of the Galleri test, including PPV and false positive rate; increase in overall cancer detection rate (see our recent performance primer for more detail on these measures); safety; and healthcare resource utilization (see Neal 2022 for more detail on key secondary and exploratory objectives).

Taken together, these data will provide a comprehensive view of the benefits and risks of introducing annual MCED screening alongside existing NHS programs.

In particular, a reduction in stage IV cancer diagnoses will be a critically important outcome. With a substantial and growing arsenal of effective treatments for many stage III cancers (Cohen 2025, Gondal 2023, Forde 2025, Uppaluri 2025), it is increasingly the case that a shift from stage IV to stage III will provide the biggest contribution to improving outcomes across many cancer types.

Furthermore, trial results will tell us whether adoption could shift diagnoses away from emergency presentations - currently around 20% of cases - which have the worst outcomes (NDRS), and increase the proportion of all cancer cases found through screening, currently around 6% in the UK and 14% in the US (NDRS, NORC).

Conclusion: What To Expect In 2026 And Beyond
In 2026, the NHS-Galleri trial will report on whether annual screening with Galleri reduces the incidence of late-stage cancer relative to standard care. These results will be of global significance. In the UK, the results will inform recommendations from the National Screening Committee and potential next steps, such as an expanded in-service evaluation within the NHS. Longer term read-outs will add important context over time.

The opportunity is profound. Positive results, including a reduction in late-stage cancers, would mark a long-overdue shift in how we find cancer - one that could be scaled equitably through public health systems worldwide to potentially benefit millions.

Note 1: The 12 pre-specified cancer types are: anus, bladder, colon/rectum, esophagus, head and neck, liver/bile duct, lung, lymphoma, myeloma/plasma cell neoplasm, ovary, pancreas, stomach.

References

Clarke CA, et al. 2020. Cancer Epidemiol Biomarkers Prev. 29(5):895-902. doi: 10.1158/1055-9965.EPI-19-1366.

Cohen R et al. 2025. ESMO Open. 10(3):104481. doi: 10.1016/j.esmoop.2025.104481.

Dai JY, et al. 2024. J Med Screen. 31(4):211-222. doi: 10.1177/09691413241256744.

Forde PM, et al. 2025. 393(8):741-752. doi: 10.1056/NEJMoa2502931.

Gondal TA, et al. 2023. Curr Oncol. 30(3):3232-3250. doi: 10.3390/curroncol30030246.

Hackshaw A, Berg CD. 2021. Lancet Oncol. 22(10):1360-1362. doi: 10.1016/S1470-2045(21)00204-7.

Lowenhoff I, et al. 2025. Front Oncol. 15:1511816. doi: 10.3389/fonc.2025.1511816.

National Disease Registration Service. Routes to Diagnosis 2006-2020. Available at: https://nhsd-ndrs.shinyapps.io/routes_to_diagnosis/.

Neal RD, et al. 2022. Cancers (Basel). 14(19):4818. doi: 10.3390/cancers14194818.

NORC at the University of Chicago. (2022). Percent of Cancers Detected by Screening in the U.S. Available at: https://cancerdetection.norc.org/

Ofman J et a. 2025. Available at: https://grail.com/stories/not-all-mced-tests-are-created-equal-the-realities-of-mced-test-development-and-validation/

Sasieni P, et al. 2025a. Ann Oncol. 36(6): 706-708. doi: 10.1016/j.annonc.2025.03.001.

Sasieni P, Brentnall AR. 2025b. JNCI. 117(3):450-455. doi: 10.1093/jnci/djae251.

Swanton C, et al. 2025. Clin Trials. 22(2):227-238. doi: 10.1177/17407745241302477.

Uppaluri R, et al. 2025. New Engl J Med. 393(1):37-50. doi: 10.1056/NEJMoa2415434.