We just had the Ad Comm for GRAIL - whose stock nearly doubled from 9/17 to 9/24. Market cap rising from $2.5B to $5.5B.
You'll be excused for not keeping this week's six publications straight in your head. See an article and op ed in NEJM, plus two articles and op ed in Nature Medicine, and finally a concensus publication by Etzioni in Cancer on late-stage-cancer as a screening endpoint.
##
Article by Chat GPT 6.
##
GRAIL’s Publication Wave: A Guide to Six Papers
Around GRAIL’s FDA review, readers encountered six related publications addressing different questions: Can Galleri find cancers? Does screening reduce advanced disease? What evidence should justify adoption? The essential distinction is that these are three research reports from two clinical studies, two commentaries, and one consensus statement. In particular, the Sasieni and Neal papers describe different aspects of the same NHS-Galleri trial, not independent trials.
1. Sasieni et al., NEJM: The randomized trial’s clinical results
In England, 142,250 participants were randomized to usual care with or without Galleri screening across three annual rounds. The trial missed its primary endpoint: reducing the incidence of stage III and IV cancers combined across 12 prespecified cancer types.
Stage IV incidence alone was approximately 14% lower with screening, suggesting possible benefit. However, because the primary endpoint failed, the prespecified statistical testing sequence did not permit formal significance testing of secondary endpoints. Longer follow-up will help determine whether this encouraging signal develops into established benefit.
Reference: Sasieni P, Johnson P, Round T, et al. “Effect of Screening with Multicancer Early-Detection Test on Late-Stage Cancer Diagnosis.” New England Journal of Medicine. Published September 22, 2026. Read article — DOI: 10.1056/NEJMoa2505723.
2. Hunter, NEJM: Op Ed, A clear “not yet” on population screening
Hunter praises the trial’s scale, rigor, and socioeconomic representation, but concludes that Galleri is not yet ready for population-wide screening. Its limited sensitivity for early-stage cancers remains a central weakness; detecting cancer at stage III rather than IV is valuable, but generally less valuable than finding it at stage I or II.
His conclusion is not “never.” Better early-stage sensitivity, further follow-up, and rigorous cost–benefit assessment could change the judgment.
Reference: Hunter DJ. “A Step in the Search for the Elusive Holy Grail of Early Detection of Cancer.” New England Journal of Medicine. Published September 22, 2026. Read editorial — DOI: 10.1056/NEJMe2611312.
______________________________
3. Neal et al., Nature Medicine: How the test performed within NHS-Galleri
This companion report examines test performance in the trial’s screened participants. Specificity remained approximately 99.5–99.6%, and cancer-origin predictions were correct in approximately 91–94% of cases.
However, positive predictive value declined from 58% to 50% to 46% across screening rounds. Sensitivity for all cancers diagnosed within the relevant 12-month periods ranged from approximately 27–37%. These findings characterize detection performance; they do not independently establish improved health outcomes.
Reference: Neal RD, Dolly S, Johnson P, et al. “Performance of a multi-cancer early detection test in the randomized controlled NHS-Galleri trial.” Nature Medicine. 2026. Read article — DOI: 10.1038/s41591-026-04652-8.
This separate, prospective U.S. study enrolled approximately 35,900 participants; approximately 32,000 were evaluable for performance. Galleri detected cancer in 0.54%, with 60.3% positive predictive value, 99.64% specificity, and 39.3% sensitivity for cancers diagnosed within 12 months.
No serious study-related adverse events were reported at this analysis. The study supports feasibility and provides reassuring short-term safety findings, but its nonrandomized design cannot establish that screening reduces advanced cancers or deaths.
Reference: Nabavizadeh N, McDonnell C, Kurbegov D, et al. “Performance and safety of a multi-cancer early detection test: the PATHFINDER 2 study.” Nature Medicine. 2026. Read article — DOI: 10.1038/s41591-026-04618-w.
This commentary connects the three research reports. The authors see stronger evidence that MCED testing can find cancers outside existing screening programs, while emphasizing that finding more cancers does not yet establish population-level benefit.
They also stress context: some additional early-stage detection involved colorectal cancer, so Galleri’s incremental value may differ between health systems with different existing screening practices. Their tone is more exploratory than Hunter’s, but both identify an incomplete case for population screening.
Reference: Marinac CR, Rebbeck TR, O’Donnell EK. “Some answers, more questions for multi-cancer early detection tests.” Nature Medicine. 2026. Read commentary — DOI: 10.1038/s41591-026-04688-w.
__________________________
6. Etzioni et al., Cancer: What should count as persuasive evidence?
The American Cancer Society workshop’s Peachtree Consensus addresses trial design rather than reporting another Galleri study. It supports considering late-stage incidence as a primary endpoint, while acknowledging that it is not a universally validated substitute for cancer mortality.
Recommendations include cancer-specific definitions of “late stage,” comparable staging across trial arms, adequate follow-up, and reporting absolute as well as relative effects. A significant, clinically meaningful late-stage reduction could justify implementation-focused demonstration studies while mortality follow-up continues—not automatic population-wide adoption.
Reference: Etzioni R, Kessler L, Schrag D, et al. “Recommendations of the American Cancer Society workshop on design, conduct, analysis, and reporting of multicancer early detection trials with late-stage incidence end points and post-trial ongoing evaluation—The Peachtree Consensus.” Cancer. 2026;e70628. Read consensus statement — DOI: 10.1002/cncr.70628.
