$GRAL — Galleri at the FDA: a missed endpoint, a Stage IV signal—and a test that could redraw cancer screening
GRAIL’s blood test is approaching the most consequential regulatory debate in multi-cancer early detection. We separate what the data proved, what they did not, and what the September panel could mean for approval, reimbursement and the equity story.
Next confirmed catalystFDA advisory committee on September 23, 2026, 9:00 a.m.–6:00 p.m. ET. The hybrid meeting is public; the webcast and docket are already available.
The decision questionCan a favorable Stage IV signal and strong test accuracy support approval after the 140,000-plus-patient NHS trial missed its prespecified primary endpoint?
Follow the catalyst, not the noise.
Join the English MerlinTrader channel for biotech event updates.
Sep. 23FDA advisory committee date
>142Kparticipants in the NHS-Galleri randomized trial
99.6%specificity in PATHFINDER 2
$861.6Mcash and short-term securities at June 30
1Executive thesis: approval and clinical proof are not the same milestone
The September panel is not a referendum on whether Galleri has already proven that it saves lives. It is a regulatory assessment of whether the submitted analytical and clinical evidence supports a safe, effective, prescription-only screening test for adults aged 50 and older—while the ultimate mortality question remains open.
What is genuinely compelling
- A single blood draw can identify a cancer signal across more than 50 cancer types and predict its origin.
- Specificity is consistently near 99.5%–99.6%, a critical property in asymptomatic populations.
- PATHFINDER 2 demonstrated a 60.3% positive predictive value and a substantial increase in screen-detected cancers.
- The randomized NHS trial showed fewer Stage IV diagnoses in a prespecified secondary analysis.
What remains unresolved
- The NHS trial failed its prespecified primary endpoint for combined Stage III/IV incidence.
- Stage IV reduction does not automatically equal a reduction in cancer mortality.
- False positives, false negatives, diagnostic procedures and overdiagnosis remain population-scale questions.
- FDA approval would open a reimbursement path, not guarantee broad coverage or adoption.
MerlinTrader view: the stock is now an evidence-interpretation trade. The decisive issue is whether regulators regard the totality of evidence—accuracy, clinical workflow, safety and stage distribution—as sufficient despite the failed primary endpoint. A favorable panel would reduce regulatory uncertainty; it would not erase reimbursement, utilization or outcomes risk.
2What Galleri does—and what it does not do
Galleri analyzes cell-free DNA in whole blood using next-generation sequencing and machine learning. It looks for cancer-associated methylation patterns: chemical marks that help distinguish cancer-derived DNA from normal DNA. A positive result also predicts a cancer signal origin, guiding the clinician toward the part of the body that should be investigated first.
The proposed FDA use is a qualitative, prescription-only in vitro diagnostic for adults aged 50 or older. A positive result is not a cancer diagnosis; it triggers professional diagnostic workup. A negative result does not exclude cancer. Galleri is intended to complement—not replace—recommended breast, cervical, colorectal, lung and prostate screening.
| Step | What happens | Critical limitation |
|---|---|---|
| 1. Blood draw | Cell-free DNA is isolated from a routine blood sample. | Some cancers shed little DNA, especially at early stages. |
| 2. Methylation analysis | The assay searches for patterns associated with malignancy. | Biological and technical noise can still create false results. |
| 3. Cancer signal origin | A machine-learning classifier predicts the most likely tissue of origin. | Localization is directional, not definitive. |
| 4. Diagnostic resolution | Imaging, endoscopy, biopsy or other specialist testing confirms or rules out cancer. | The workup can create cost, anxiety and procedural harm. |
Current status: Galleri is commercially available in the United States as a CLIA laboratory-developed test, but it is not FDA-approved. The list price is $949, with some provider programs offering self-pay access at $799 or less. Medicare and most insurers do not currently cover it.
3The PMA file and the September 23 meeting
GRAIL submitted its premarket approval application on January 29, 2026. Galleri received FDA Breakthrough Device designation in 2018, which can facilitate interaction with the agency but does not lower the approval standard. FDA has now scheduled its Molecular and Clinical Genetics Panel to review the application.
| Event detail | Confirmed information |
|---|---|
| Date and time | September 23, 2026, 9:00 a.m.–6:00 p.m. ET |
| Format | Hybrid, open to the public, with an FDA webcast |
| Docket | FDA-2026-N-8004 |
| Comment timetable | Comments submitted by September 8 are expected to reach the committee; the docket closes September 16 |
| Panel role | Independent expert advice; the vote is influential but not binding on FDA |
The original PMA package emphasized one-year follow-up in 25,490 consented PATHFINDER 2 participants, prevalent-round data from more than 70,000 NHS intervention-arm participants, and a bridging analysis to the updated assay version. The final three-round NHS results arrived after submission and now shape the public debate, but investors should not assume that every late-breaking analysis was part of the original application.
4NHS-Galleri: the primary endpoint failed
The NHS-Galleri trial randomized more than 142,000 asymptomatic adults aged 50–77 to three annual Galleri screens or usual care. Its prespecified primary endpoint was the incidence of Stage III or IV cancer across 12 selected cancers. The result was unambiguous: the endpoint was not met.
| NHS-Galleri outcome | Result | Interpretation |
|---|---|---|
| Primary: Stage III/IV, 12 cancers | IRR 1.03 (95% CI 0.92–1.14); p=0.6324 | Missed No reduction; the point estimate was 3% higher. |
| Stage IV, all rounds | IRR 0.86 (0.744–0.998) | Secondary 14% reduction, narrowly excluding 1. |
| Stage I/II, 12 cancers | +16% | Directionally consistent with earlier detection. |
| Screen-detected cancers | 4× higher | More cancers were found through screening. |
| Clinical/symptom detection | 21% fewer | Supportive evidence of a route-of-diagnosis shift. |
| Emergency presentations | 25% fewer | Supportive sensitivity analysis, not the primary endpoint. |
Do not blur the hierarchy: the statistically negative primary outcome carries more evidentiary weight than a favorable secondary endpoint. The Stage IV signal is clinically important, but multiplicity, stage migration and the timing of diagnoses must be examined before treating it as definitive proof of benefit.
5The Stage IV signal strengthens over repeated rounds
The most constructive NHS finding is the pattern across annual screening rounds. The estimated Stage IV reduction rose from 9% in the first, prevalent round to 22% in the second and 26% in the third. Only the third-round confidence interval excluded no effect.
GRAIL argues that the first screen uncovered a backlog of previously undiagnosed advanced cancers, while some cancers may have shifted from Stage IV to Stage III—muting a combined Stage III/IV endpoint. It also argues that longer follow-up may reveal cancers later in the control arm. Those explanations are biologically plausible, but they remain interpretations. The panel will decide how much weight to give them.
Across three rounds, 1,801 participants (0.91%) had a cancer signal detected and 937 were diagnosed with cancer, for a 0.48% detection rate. Positive predictive value was 52.0%, specificity 99.55%, false-positive rate 0.45%, and cancer signal origin accuracy 92.5%. Episode sensitivity was 54.7% for the 12 prespecified cancers and 30.7% across all cancers.
6PATHFINDER 2: strong operating performance in clinical practice
PATHFINDER 2 answers a different question from NHS-Galleri: how does the test perform when returned to patients and clinicians in a prospective US screening workflow? The full 35,878-participant dataset supports a high-specificity, actionable test—but it is not a randomized mortality study.
| Metric | PATHFINDER 2 result | Why it matters |
|---|---|---|
| Cancer signal positives | 287; 173 cancer diagnoses | Shows the real-world diagnostic funnel. |
| Positive predictive value | 60.3% | Roughly three in five positive results were confirmed. |
| Specificity / false-positive rate | 99.6% / <0.4% | Limits unnecessary workups in a low-prevalence population. |
| Episode sensitivity | 69.8% for 12 lethal cancers; 39.3% for all cancers | A negative test still misses a meaningful share of cancers. |
| Cancer signal origin accuracy | 91.3% | Helps focus diagnostic evaluation. |
| Median time to resolution | 48 days | Defines the practical burden after a positive result. |
| Stage distribution | 53.0% Stage I/II; 70.9% Stage I–III | Supports earlier-stage detection, without proving survival benefit. |
| Incremental detection | 6.5× more screen-detected cancers when added to USPSTF A/B screening | Suggests utility beyond established single-cancer programs. |
Only 0.6% of all safety-analyzable participants underwent an invasive procedure after a positive result, and 90.5% of diagnostic procedures were nonsurgical. Five study-related adverse events occurred during diagnostic evaluation, all in people ultimately diagnosed with cancer. Anxiety rose temporarily after a positive result and returned to baseline at 12 months. One serious adverse event related to diagnostic workup was identified after data lock, underscoring why post-market surveillance and clear workup protocols matter.
7What the FDA panel is likely to interrogate
FDA’s briefing documents and final voting questions will define the exact debate. Until they are posted, the most defensible framework is to focus on the issues inherent to a first-of-kind population screening device.
Effectiveness
- Is stage distribution an adequate intermediate measure of clinical benefit?
- How should the failed primary endpoint constrain interpretation of Stage IV results?
- Are PATHFINDER 2 and NHS populations representative of the intended US population?
- Does assay bridging preserve performance across versions?
Safety and use
- Can labeling prevent Galleri from replacing established screening?
- Are diagnostic pathways sufficiently standardized?
- How will false negatives, false positives and incidental findings be communicated?
- What post-approval evidence is needed on mortality, disparities and longitudinal use?
The regulatory middle ground matters: a favorable outcome could still include restrictive labeling, required post-approval studies, physician education, a registry or defined diagnostic-management expectations. “Approved” is not a single economic outcome.
8Finding cancer earlier is not yet the same as saving lives
Cancer screening can look successful while failing to reduce mortality. Lead-time bias makes survival from diagnosis appear longer simply because the clock starts earlier. Length bias preferentially finds slower-growing tumors. Overdiagnosis finds cancers that would never have caused harm. Stage migration can improve the apparent distribution without changing the underlying course of disease.
That is why the National Cancer Institute emphasizes randomized evidence and cancer mortality. There are currently no FDA-authorized multi-cancer detection tests, no routine recommendations from major medical societies or the US Preventive Services Task Force, and no broad insurance coverage. Mortality follow-up from NHS-Galleri is expected later and will be far more consequential than any single detection metric.
Investor discipline: Galleri may be useful before the mortality dataset matures; regulators routinely use intermediate endpoints where the totality of evidence supports benefit. But the absence of mature mortality evidence should be priced as uncertainty, not waved away as a technicality.
9The population-scale risk equation
A false-positive rate below 0.5% sounds small. Applied annually to millions of asymptomatic adults, it can still generate tens of thousands of diagnostic episodes. Conversely, limited sensitivity means reassurance after a negative test must be carefully framed.
| Risk | Evidence so far | What could mitigate it |
|---|---|---|
| False positives | 0.45% in NHS; below 0.4% in PATHFINDER 2 | High specificity, accurate origin prediction and disciplined workup pathways |
| False negatives | All-cancer episode sensitivity 30.7% in NHS and 39.3% in PATHFINDER 2 | Clear labeling: continue standard screening and evaluate symptoms |
| Procedural harm | Low invasive-procedure rate in PATHFINDER 2; one post-lock serious event | Specialist coordination, registries and post-market monitoring |
| Overdiagnosis | Not fully quantifiable yet | Long-term outcomes and tumor-specific natural-history analysis |
| Access disparities | Self-pay model limits reach | Coverage, diverse evidence and equitable diagnostic networks |
10FDA approval is necessary for the reimbursement thesis—not sufficient
The current self-pay price restricts adoption. A February 2026 US law created a Medicare benefit category that could allow coverage of FDA-approved multi-cancer early detection tests as early as January 1, 2029, initially for Medicare beneficiaries aged 50–65 and expanding by one age-year annually. GRAIL says it intends to pursue that pathway.
That legislation improves strategic optionality, but it does not automatically set price, coverage criteria or utilization. Commercial insurers may demand outcomes data, health-economic evidence and real-world diagnostic-cost estimates. A future USPSTF A or B recommendation could materially broaden coverage, but such reviews can take years.
Economic bottleneck: FDA can validate the product’s benefit-risk profile. CMS and private payers determine whether a broad eligible population can actually access it. The equity story needs both gates to open.
11Commercial momentum, cash and valuation context
GRAIL is not a pre-revenue binary. Galleri volume and revenue are growing, giving the company a commercial feedback loop before approval. The trade-off is a large loss base and continued financing risk.
| Q2 2026 / balance-sheet metric | Reported value | Context |
|---|---|---|
| Total revenue | $44.7M | +26% year over year |
| Galleri revenue | $42.6M | +24% year over year |
| Quarterly test volume | >61,000 | +35% year over year |
| First-half Galleri revenue / volume | $82.5M / >117,000 | +30% / +42% year over year |
| Net loss / adjusted EBITDA | $(110.2)M / $(90.3)M | Investment intensity remains high |
| Cash + short-term securities | $861.6M | At June 30, 2026 |
| H1 operating cash use | $167.7M | Annualized mechanically: about $335M |
| Shares outstanding | 44.67M | +10.7% versus December 31, 2025 |
Samsung invested $110 million in June at $70.05 per share, purchasing 1.57 million shares. GRAIL also had approximately $189.3 million available under its at-the-market program at quarter-end. These are strategic and funding resources, but the rising share count is a concrete reminder that dilution belongs in the valuation.
Illustrative valuation snapshot—not company guidance
At the August 7 close of $69.48 and 44.67 million shares, the simplified equity value is approximately $3.10 billion. Annualizing Q2 revenue produces about $178.7 million, implying roughly 17.4× equity value/revenue. Subtracting reported cash and short-term securities gives a simplified enterprise-value proxy of $2.24 billion, or about 12.5× annualized revenue. These shortcuts ignore debt-like items, working capital, seasonality and future financing.
Annualizing first-half operating cash use suggests crude cash coverage of roughly 2.6 years. That is not a management runway forecast: regulatory spending, scale-up, reimbursement investment and financing decisions can change the trajectory substantially.
12A first-mover lead, not a permanent monopoly
GRAIL’s strongest competitive asset is the combination of a large prospective evidence base, a randomized NHS program, a scaled commercial laboratory and the first PMA review for a broad multi-cancer detection test. Competitors are nevertheless advancing quickly.
| Company / test | Status | Competitive relevance |
|---|---|---|
| GRAIL / Galleri | US LDT; PMA under FDA review; $949 list price | Evidence depth, commercial scale and first regulatory review |
| Exact Sciences / Cancerguard | US LDT launched in 2025; $689 self-pay price | Lower price and established cancer-screening commercial infrastructure |
| Guardant / Shield MCD | Breakthrough Device; selected for NCI Vanguard; launched in parts of Asia | Liquid-biopsy platform and payer relationships; distinct from FDA-approved colorectal Shield |
| ClearNote / Avantect | Selected for NCI Vanguard | Alternative epigenomic approach under prospective federal evaluation |
Notably, NCI selected Guardant’s and ClearNote’s assays—not Galleri—for the Vanguard study that precedes a larger US randomized trial. That is not a verdict on Galleri, but it prevents investors from treating the category as a winner-take-all market.
13Bull, middle and bear pathways
Bull pathway
The panel judges total evidence favorable, with manageable diagnostic harms. FDA approval follows with commercially workable labeling. Volume accelerates, health-economic evidence matures and the 2029 Medicare pathway becomes credible.
What validates it: supportive briefing documents, clean safety discussion, limited post-market burden and improving recurring-screen data.
Middle pathway
The panel supports availability but demands narrow labeling, extensive post-approval evidence or tighter clinical management. Approval remains possible, yet adoption and payer negotiations progress more slowly.
What validates it: split votes, benefit-risk support paired with strong reservations, or conditions that raise cost and complexity.
Bear pathway
The failed primary endpoint dominates. The panel concludes that a Stage IV secondary signal cannot establish benefit, or that diagnostic and overdiagnosis risks are insufficiently characterized. Delay, another study or a complete response follows.
What validates it: FDA skepticism on endpoint hierarchy, assay bridging, intended-use population or clinical utility.
Asymmetry warning: GRAL closed August 7 at $69.48 after a 9.5% gain. With a multi-billion-dollar equity value and revenue still below $200 million on a simple annualized basis, the market already capitalizes a meaningful probability of regulatory and commercial success. Positive news can still re-rate the story, but the downside from a material delay is not cushioned by current earnings.
14The event-day watchlist
- FDA briefing tone: does the agency frame Galleri as an acceptable first step with post-market obligations, or as an unproven screening paradigm?
- Voting questions: separate safety, effectiveness and benefit-risk votes reveal where disagreement sits.
- Primary-endpoint hierarchy: listen for how panelists treat the negative combined Stage III/IV result.
- Stage IV credibility: focus on multiplicity, annual-round trend, cancer-type heterogeneity and stage-migration analysis.
- Diagnostic burden: look for questions about workup standardization, serious adverse events and false-positive management.
- Labeling: age range, screening interval, contraindications and explicit language that Galleri cannot replace standard screening.
- Post-approval plan: mortality follow-up, registries, diverse populations, repeat testing and real-world harms.
- Commercial read-through: any restriction that narrows the eligible population or increases diagnostic friction changes revenue potential.
Written comments intended for the committee should be submitted by September 8; the docket closes September 16. The meeting begins at 9:00 a.m. ET and is scheduled to end at 6:00 p.m. ET.
15Bottom line
Galleri has produced something rare in diagnostics: a genuinely scalable test, strong specificity, prospective clinical workflow data and randomized evidence suggesting fewer Stage IV cancers—alongside a failed primary endpoint that cannot be edited out of the story.
A favorable FDA pathway would establish an entirely new device category and give GRAIL a valuable first-mover position. It would also move the debate from “can this be approved?” to harder questions about mortality, repeated use, reimbursement and economics. An unfavorable pathway would expose the danger of building a premium valuation around a secondary endpoint.
The investable conclusion is therefore not “Galleri works” or “Galleri failed.” It is more precise: the assay has demonstrated detection performance and a promising Stage IV signal; it has not yet demonstrated the prespecified population-level stage endpoint or a mortality benefit. September 23 will show whether FDA advisers believe that gap can be managed through labeling and post-market evidence—or must be closed before approval.
16Primary sources and further reading
- FDA — September 23, 2026 Molecular and Clinical Genetics Panel meeting
- GRAIL — FDA advisory committee announcement
- GRAIL — PMA submission and evidence package
- Journal of Clinical Oncology — NHS-Galleri ASCO 2026 abstract
- GRAIL — full NHS-Galleri results release
- ClinicalTrials.gov — NHS-Galleri, NCT05611632
- GRAIL — full PATHFINDER 2 results
- ClinicalTrials.gov — PATHFINDER 2, NCT05155605
- SEC — GRAIL Q2 2026 Form 10-Q
- SEC — GRAIL Q2 2026 earnings release
- National Cancer Institute — Q&A on multi-cancer detection tests
- Galleri — current price and coverage information
Related MerlinTrader research: Biotech Radar: liquid biopsy and multi-cancer detection · Biotech Stocks Hub & 2026 Catalysts
Biotech catalysts move fast.
Get the next research update in the MerlinTrader English channel.