FDA Advisers Vote on Galleri’s Claim of Early Detection

FDA advisers meet September 23, 2026, to vote on GRAIL, Inc.’s Galleri blood test, a $949 screen the company says can detect up to 50 cancers. The Molecular and Clinical Genetics Panel sits from 9 a.m. to 6 p.m. Eastern at the agency’s White Oak campus in Silver Spring, Md.

The file in front of them is a first: no multi-cancer blood test has gone through this kind of safety and accuracy review. The sharpest fight is a single adjective. Staff asked whether the data support calling Galleri an early detection test at all.

The Word Early Is on the Table

Galleri looks for chemical marks on bits of DNA that tumors shed into blood, then names a likely organ if a signal appears. GRAIL wants it sold by prescription to adults 50 and older, as an add-on to mammograms, colon checks, and other usual screens. A positive result is not a diagnosis. A negative result does not clear someone of cancer. GRAIL’s own safety sheet says both of those things out loud.

That is why the word early is doing so much work. In the first year of the UK trial, among cancers found within 12 months, Galleri caught 13.6 percent of Stage I tumors and 60.0 percent of Stage IV tumors. Detection rose as tumors got larger, which is the opposite pattern an early-detection brand is built to promise.

NHS-GALLERI, FIRST YEAR, BY STAGE

Of 474 Stage I or II cancers in that round, the test flagged 97, or 20.5 percent. Of 820 new primary cancers, it flagged 259. Staff told the panel to look at that stage mix and decide whether early belongs in the intended use, or whether the claim should shrink to multi-cancer detection without it.

Missed Goal in the 142,826-Person NHS Trial

NHS-Galleri enrolled 142,826 adults in England. Investigators assigned 71,122 to yearly testing plus usual care and 71,128 to usual care alone. The primary goal after three rounds was fewer Stage III and Stage IV diagnoses among 12 pre-set cancers. The groups did not separate. The incidence rate ratio was 1.03 (95 percent CI 0.92 to 1.14, P=0.63), which is no drop in combined Stage III or IV cancer.

In this trial, we did not observe a lower incidence rate of stage III or IV cancers across 12 prespecified cancers after three screening rounds with multicancer early-detection testing plus usual care than with usual care alone.

NHS-Galleri investigators, New England Journal of Medicine

A pre-set secondary look at Stage IV alone was kinder to the test. The ratio after three rounds was 0.86 (95 percent CI 0.74 to 1.00), a 14 percent cut whose upper bound sits on 1.00. GRAIL says that cut deepened in later rounds, to 22 percent in the second round and 26 percent in the third, among those 12 cancers. Less than 1 percent of people had trial-related side effects, and none were serious. The trial was not built to prove fewer deaths. Those reads are slated later.

WHERE EXPERTS DISAGREE

• The miss as a technicality: Cristian Tomasetti, who directs early detection and monitoring at City of Hope, says three years is a short window for mortality and that the first rounds of any new screen will surface cancers that were already there, including Stage III disease.
• The miss as the result: The trial’s own conclusion is that combined late-stage incidence did not fall, and the authors say more follow-up is needed before anyone treats the Stage IV dip as a lasting shift.
• Stage III as a find, not a failure: Nima Nabavizadeh, a radiation oncologist at Oregon Health & Science University who led a U.S. site for PATHFINDER 2, argues that a first-round sweep of silent Stage III tumors is what a new blood test should be expected to do, and that many of those tumors can still be treated with the aim of cure.

Tomasetti, who once helped build a rival blood-test method now sitting at Abbott, still thinks the agency looks ready to approve. Oncologist Tom Beer, chief medical officer for Abbott Cancer Diagnostics, sees patients each week with tumors that have no screening program at all. He says 70 percent of cancer diagnoses and deaths come from those diseases. Neither man is on the panel. Neither is voting.

How Accurate Is the Galleri Blood Test?

The application rests on two studies, and on a bridge from an older lab version of the assay to the Galleri kit in the file. PATHFINDER 2 is a single-arm North American study in adults 50 and older with no suspected cancer. NHS-Galleri is the randomized UK trial. For the PMA, reviewers used 12 months of follow-up from PATHFINDER 2 and the first screening round in Britain.

Those two slices, using the Galleri analysis in the briefing, sit close together on false alarms and farther apart on what a positive result is worth.

WHAT THE FDA FILE SHOWS AT 12 MONTHS

The live U.S. run used an investigational version of the test, and those figures are not the same numbers. In the PATHFINDER 2 results in 35,878 people, 32,007 were evaluable at one year. A signal showed up in 287 of them, and cancer was confirmed in 173, a 60.3 percent positive predictive value. Overall sensitivity was 39.3 percent, rising to 69.8 percent for a group of 12 cancers that cause about two-thirds of U.S. cancer deaths. Specificity was 99.6 percent.

Adding the blood test to recommended screens for breast, cervical, colorectal, and lung cancer produced a 6.5-fold jump in cancers found by screening, GRAIL said. Adding it to those four plus prostate screening nearly tripled screen-detected cancers. Among Galleri-detected cancers in the FDA analysis, 49 percent were Stage I or II and 69 percent were Stage I through III.

Performance was uneven by tumor. In the PATHFINDER 2 Galleri analysis, 12-month sensitivity was 62.5 percent for colorectal cancer, 47.1 percent for lung, 26.4 percent for breast, and 10.7 percent for prostate. For cancers that already have a guideline screen, the test did better in people who were not eligible for that screen than in people who were. Median time to a workup result was 48 days. Invasive procedures followed 0.6 percent of people in the safety set (213 of 35,335), and 90.5 percent of those procedures were not surgery. Five study-related adverse events were logged, all in people who did have cancer. One serious event tied to the workup turned up after the data lock.

Three Votes, None of Them an Approval

Members will not approve or reject the test. They will answer three yes-or-no questions, then talk through labeling, warnings, and whether GRAIL should be made to run more studies after a yes from the agency. The FDA usually follows its panels and does not have to. GRAIL has said it still expects a decision in the first part of 2027.

WHAT THE PANEL IS ASKED TO ANSWER

• Safety: Is there reasonable assurance Galleri is safe for adults 50 and older under the proposed use, including the scans, biopsies, and anxiety that follow a positive result.
• Effectiveness: Is there reasonable assurance it works for that group, measured by 12-month detection, false alarms, and the mix of stages it actually finds.
• Balance: Do members believe whether its benefits outweigh its risks for people who would get the test as labeled.

Those three answers can split. A member can accept that the assay detects a cancer signal and still reject the early-detection claim, or accept the claim only with hard limits in the label. Staff also want language that a negative result does not rule out cancer and that a positive result is not itself a diagnosis. They asked how to explain, to both doctors and patients, that some tumors are caught far more often than others.

If the product is cleared on this package, another large randomized trial against usual care gets harder to enroll. People who can buy the test will not want to sit in a control arm. That is the bind the missed NHS primary endpoint leaves behind: the study that would settle deaths may be least doable after a blessing.

THE ROAD TO THE SEPTEMBER 23 VOTE

1. 2018: The FDA gives Galleri Breakthrough Device status.
2. 2021: GRAIL begins selling the test as a prescription lab-developed product, without FDA approval.
3. January 29, 2026: GRAIL files the premarket approval application.
4. February 3, 2026: The Medicare MCED coverage law is signed.
5. September 21, 2026: Briefing documents post. GRAIL shares close at $107.97, up 33.7 percent.
6. September 23, 2026: The genetics panel meets and votes.

Staff told the panel they had no leftover questions on lab performance or on the main safety analyses. They left effectiveness, the early-detection wording, and the benefit-risk call for the room.

Medicare Already Set a 2029 Coverage Clock

Cash payers already buy Galleri. Most insurers, including Medicare, do not. Screening tests sit outside ordinary Medicare coverage unless Congress writes them in. In February, lawmakers did.

The Nancy Gardner Sewell Medicare coverage law, folded into the 2026 spending package, creates a benefit for FDA-cleared or FDA-approved multi-cancer tests that read cell-free nucleic acids in blood, or tests the Secretary finds comparable. Coverage can start January 1, 2029, for people ages 50 to 65, with the upper age rising by one year each year. Medicare would pay for one qualifying test every 11 months, at the same rate it pays for a multi-target stool DNA screen, through the end of 2030. After that, payment is the lower of that peg and the usual lab fee. CMS still has to find the test reasonable and necessary through a national coverage decision. Usual screens for breast, cervix, colon, lung, and prostate stay in place.

That calendar is why Wednesday’s wording fight is not academic. A label that says early detection is the product private payers and Medicare will be asked to buy. A narrower detection claim is a different product, with a slower path into yearly wellness labs. Jefferies analyst Tycho Peterson told clients a positive vote would take risk out of the whole multi-cancer category, including tests from Caris Life Sciences and Guardant Health. The panel does not set prices. It does set the claim those prices will sit under.

More Than 595,000 Tests Sold Without FDA Approval

Galleri has been on the U.S. market as a lab test since 2021. Function Health and Hims & Hers are among the cash channels. The list price is $949. In the second quarter of 2026, GRAIL’s average selling price was about $700, pulled down by employer deals and digital-health volume.

GALLERI’S CASH MARKET

• Tests sold: More than 595,000 commercial tests through June 30, 2026, including 117,000 in the first half of 2026 and more than 185,000 in 2025.
• U.S. screening revenue: $136.8 million in 2025, up 26 percent, inside $147.2 million of total company revenue.
• Latest quarter: 61,000 tests in the second quarter of 2026, up 35 percent, for $42.6 million of screening revenue.
• Prescribers: More than 19,000 as of June 30, 2026, in a market that still has almost no insurance coverage.

Josh Ofman, GRAIL’s chief executive, has told investors the company still expects approval in the first part of 2027. Shares had fallen more than 20 percent after February, when the NHS primary endpoint miss became public. The September 21 jump followed a staff memo that declined to pick a fight on the lab work. It did not settle the word early, and it did not rewrite the UK ratio of 1.03.

Abbott and Guardant Need the Same Kind of Nod

Galleri is the first multi-cancer blood test in this room. It is not the only one in clinics. Abbott is pursuing FDA review of CancerGuard, built on work Tomasetti once helped develop. Caris has already launched a test. Guardant is building one. GRAIL tells doctors to keep using mammograms and colonoscopies with the blood draw. The Sewell law says the same thing about Medicare’s older screens.

Only four cancers have a broad U.S. Preventive Services Task Force A or B screen: breast, cervix, colon, and lung in a high-risk group, with a C grade for prostate. Beer’s clinic point still stands. Most lethal tumors have no poster and no appointment. A blood test that finds some of those tumors, even some of them late, is the reason this file exists. A blood test that misses most Stage I disease is the reason the file is this hard.

The panel’s yes or no will move a stock and a sales force. The sentences members attach to the early-detection claim are what doctors, ads, and, in 2029, Medicare will inherit. GRAIL is already selling the kit. The agency is being asked to put its name on the word that sells it.

Disclaimer: This article is news reporting on an FDA advisory meeting and related trial results. It is for information only and is not medical advice, a screening recommendation, or an instruction to buy or skip any test. Readers should talk with a qualified physician, oncologist, or other licensed clinician before using a multi-cancer blood test or changing any cancer screening plan. Figures, study results, panel votes, and coverage rules reflect the cited company filings, trial papers, and agency materials as of the dates on those documents and may change as the FDA, CMS, and GRAIL update them.