A Blood Test Can Flag Many Cancers. The FDA Panel Still Questioned the Word “Early.”

On September 23, 2026, a panel of independent advisers to the U.S. Food and Drug Administration considered a question that could reshape cancer screening:

Can one blood test meaningfully screen for many different cancers at once?

The device under review was Galleri, a prescription blood test developed by GRAIL. It looks for cancer-associated methylation patterns—chemical modifications found on fragments of DNA circulating in the bloodstream—and, when it detects a cancer signal, predicts where in the body that signal may have originated.

The panel’s answer was cautiously favorable.

Its members voted 10–0 that the available evidence provided reasonable assurance of safety, 6–4 that it provided reasonable assurance of effectiveness, and 7–2, with one abstention, that the test’s benefits outweighed its risks for the proposed population of adults aged 50 and older.

But those votes require careful interpretation.

An advisory-panel recommendation is not FDA approval. The FDA will make the final regulatory decision, and it is not legally required to follow the panel.

More importantly, several advisers who voted favorably did so conditionally. They supported removing—or substantially reconsidering—the word “early” from the proposed indication, strengthening the labeling, and requiring further studies after any approval.

That tension is the real story.

Galleri appears capable of finding cancer signals that would otherwise go undetected. It also missed most cancers diagnosed during the follow-up period, detected some cancer types much better than others, and has not established that screening with the test reduces cancer mortality.

The technology may be important.

The clinical claim still matters just as much.

What the test actually measures

Cancer cells can release fragments of DNA into the bloodstream. Galleri analyzes methylation patterns on cell-free DNA, using machine-learning classifiers to distinguish a possible cancer signal from the background DNA circulating in someone without a detectable cancer signal.

The result is qualitative:

• Cancer Signal Detected
• No Cancer Signal Detected

When a signal is detected, the test also reports a predicted cancer signal origin—the anatomical location where the signal most likely came from.

This is not the same as diagnosing cancer.

A positive result requires additional diagnostic evaluation, which may include imaging, endoscopy, laboratory testing, biopsy or other procedures selected by a qualified healthcare professional. A negative result does not rule out cancer.

The proposed FDA indication is for prescription screening in adults aged 50 or older. The test is intended to supplement—not replace—established screening such as mammography, cervical screening, colorectal screening and low-dose CT for eligible people.

Galleri has already been commercially available in the United States as a laboratory-developed test. That availability should not be confused with FDA premarket approval. Laboratory certification establishes that a laboratory can reliably perform its testing processes; it does not independently establish that screening improves patient outcomes.

The September panel meeting was the FDA’s first-of-its-kind review of a premarket approval application for a multi-cancer screening test.

The performance numbers tell two different stories

The FDA reviewed evidence from two major studies: PATHFINDER 2 in the United States and NHS-Galleri in England.

In the FDA’s analysis of PATHFINDER 2, Galleri had:

• 35.0% overall sensitivity
• 99.85% specificity
• 77.0% positive predictive value
• 99.07% negative predictive value

In the NHS-Galleri analysis, it had:

• 31.6% overall sensitivity
• 99.74% specificity
• 66.2% positive predictive value
• 98.89% negative predictive value

At first glance, the specificity and positive predictive value look impressive.

Specificity describes how often the test correctly returned a negative result among people who were not diagnosed with cancer during the defined follow-up period. Positive predictive value describes the proportion of positive results that were followed by a cancer diagnosis.

In PATHFINDER 2, the estimated false-positive rate was 0.15%. Among positive results included in that analysis, approximately three out of four were associated with a cancer diagnosis within 12 months.

Those figures suggest the test is not simply generating large numbers of indiscriminate cancer alarms.

But sensitivity tells the other half of the story.

A sensitivity of 35% means that the test detected approximately one-third of the cancers diagnosed during the following 12 months—and did not detect approximately two-thirds.

In NHS-Galleri, sensitivity was again close to one-third.

That does not make the test useless. Screening for cancers that lack any routine screening option may still have value even if sensitivity is incomplete.

But it means a negative result cannot function as an “all clear.” It also means the meaning of a high negative predictive value must be interpreted in the context of cancer’s relatively low prevalence in an asymptomatic screening population.

When the condition being screened for is uncommon, most people will not have it. A test can therefore produce a very high negative predictive value while still missing a meaningful proportion of the cancers that are present.

PATHFINDER 2 also estimated that approximately 202 people would need to be screened to detect one cancer signal associated with a cancer diagnosis during the study’s 12-month assessment period.

One test does not perform like one test across every cancer

The phrase “multi-cancer test” can create the impression of a single, stable level of performance across cancer types.

That was not what the data showed.

In PATHFINDER 2, the FDA reported substantial differences in sensitivity by cancer type. The test detected 62.5% of colorectal cancers and 47.1% of lung cancers diagnosed during follow-up, but only 26.4% of breast cancers and 10.7% of prostate cancers.

Some estimates were based on very small numbers of cases, producing wide confidence intervals and substantial uncertainty.

Across all cancers with existing guideline-recommended screening, combined sensitivity in PATHFINDER 2 was 22.9%. For a prespecified group of 12 cancers responsible for a large share of U.S. cancer deaths, sensitivity was higher at 61.3%.

That may be where multi-cancer testing offers its most compelling potential advantage: cancers such as pancreatic, ovarian, liver and esophageal cancer generally lack broadly recommended population-screening programs.

The FDA panel agreed that Galleri’s greatest potential clinical significance was for cancers without an established screening pathway. It saw more limited value for cancers already covered by evidence-based screening programs and emphasized that Galleri should not replace them.

But “can sometimes detect a cancer for which no routine screening exists” is not the same claim as “reliably detects cancer early.”

That distinction became the central issue at the meeting.

The word “early” carries a higher evidentiary burden

The sponsor’s proposed indication described Galleri as a test for the “early detection” of multiple cancers.

The panel generally doubted that the submitted evidence supported that wording across the test’s broad intended use.

Galleri was more sensitive for many later-stage cancers than for stage I disease. In the NHS-Galleri analysis cited during the FDA review, only a minority of stage I cancers produced a detected signal. Performance also varied sharply by cancer type.

That is biologically understandable.

Larger and more advanced tumors may release more detectable material into the bloodstream. A small, localized tumor may shed little circulating DNA, making the cancer that would benefit most from early detection one of the hardest to find.

This creates a paradox for blood-based cancer screening: the test may become more sensitive as a cancer becomes easier to detect, while the largest potential benefit may come from detecting it before that point.

An “early detection” claim therefore needs more than evidence that a test can find some cancers before symptoms appear.

It needs evidence that the test meaningfully changes when cancers are diagnosed, reduces the incidence of advanced disease, enables beneficial treatment, or ultimately improves outcomes without producing disproportionate harm.

The NHS-Galleri randomized trial was designed to examine part of that chain.

Its primary objective was to determine whether adding Galleri to usual care reduced diagnoses of stage III and IV cancers. The trial did not meet that primary endpoint: there was no significant overall difference between the test group and control group when stage III and IV cancers were combined.

There were encouraging secondary findings. The test group had more stage I and II diagnoses, fewer stage IV diagnoses, and fewer cancers diagnosed through emergency presentations.

Those observations may prove important. But a favorable secondary finding does not erase a missed primary endpoint, and a stage shift does not by itself prove that screening reduces cancer deaths.

The trial is continuing to collect outcome information, including mortality data.

Detection is an intermediate outcome

The intuitive case for early detection is powerful:

Find cancer sooner, treat it sooner, and improve the chance of cure.

That logic is often correct. It is not automatically correct for every cancer or every screening test.

Screening studies can be distorted by several well-known effects.

Lead-time bias makes survival after diagnosis appear longer simply because the diagnosis occurred earlier, even if the patient does not live longer overall.

Length bias makes screening more likely to detect slower-growing cancers because they remain detectable for longer.

Overdiagnosis occurs when screening finds a cancer that would never have caused symptoms or shortened the person’s life. Once discovered, however, that cancer may lead to further testing, treatment and long-term surveillance.

And some aggressive cancers may progress despite being found earlier.

This is why detection rate, test accuracy and stage shift are important—but incomplete—outcomes.

The highest bar is whether a screening program reduces serious illness or death while producing an acceptable burden of false positives, invasive workups, overdiagnosis, cost and anxiety.

The National Cancer Institute states that no definitive clinical trial has yet shown that multi-cancer detection screening reduces overall cancer mortality. Large randomized studies are needed because the benefit belongs to the entire screening pathway, not the blood test alone.

The safety question extends beyond the needle

The panel voted unanimously that Galleri provided reasonable assurance of safety under the proposed conditions of use.

The blood draw itself is low risk.

But screening safety cannot be evaluated only by what happens when the blood is collected.

A positive signal begins a diagnostic process. In PATHFINDER 2, the median time to diagnostic resolution after a positive result was 46 days. For people whose positive result was not followed by a cancer diagnosis, the median was 75 days.

During that period, a person may undergo imaging, procedures, repeat testing and significant uncertainty.

The FDA’s summary reported relatively few study-related adverse events during diagnostic evaluation, and no device-related adverse events. That is reassuring.

Still, physical adverse events are only one category of potential harm. Other questions include:

• How often will a positive signal lead to an invasive procedure?
• What happens when the predicted tissue of origin is wrong?
• How long should surveillance continue if no cancer is found?
• Will a negative result cause some people to delay conventional screening or ignore symptoms?
• Will repeated annual testing increase cumulative false-positive findings?
• Who pays for the diagnostic workup?
• Will access to high-quality follow-up differ by geography, income or insurance coverage?

The panel specifically recommended clearer education for patients and clinicians, resources to navigate diagnostic workups, continued monitoring of false positives and false negatives, and longer-term follow-up of patient outcomes.

What the vote does—and does not—mean

The September 23 vote means a majority of the advisory panel believed Galleri’s benefit-risk profile could be favorable for its proposed population, with important qualifications.

It does not mean:

• the FDA has approved the test;
• Galleri has been proven to reduce cancer mortality;
• one negative result rules out cancer;
• one positive result establishes a diagnosis;
• the test performs equally well across cancers or stages;
• it can replace mammography, cervical screening, colorectal screening, low-dose CT or other recommended care;
• every adult should obtain multi-cancer testing.

Several panelists conditioned their favorable effectiveness votes on changing the proposed “early detection” language and conducting additional studies. Those conditions are not editorial details. They define the claim the evidence can currently support.

FDA approval, if eventually granted, would mean the agency concluded that the device met its regulatory standard for a defined intended use.

It would not turn every unresolved clinical question into a settled one.

A more useful way to evaluate a cancer blood test

The question is not simply, “Can it detect cancer?”

A more complete evaluation asks:

1. What signal does the test measure?
   Galleri measures cancer-associated methylation patterns in circulating cell-free DNA.

2. What population was studied?
   Performance in asymptomatic adults aged 50 and older cannot automatically be generalized to younger people, symptomatic patients or every higher-risk group.

3. How often does it find cancer when cancer is present?
   Overall sensitivity was approximately one-third in the two pivotal analyses reviewed by FDA.

4. How often is a positive signal followed by a diagnosis?
   Positive predictive value was 77.0% in PATHFINDER 2 and 66.2% in the FDA’s NHS-Galleri analysis, although these estimates depend on the study population and follow-up definitions.

5. Which cancers and stages does it detect well—or poorly?
   Performance varied substantially. A single overall sensitivity number conceals that variation.

6. What happens after the result?
   Screening has value only if the downstream diagnostic and treatment pathway is timely, accurate and accessible.

7. Has it improved outcomes that matter to patients?
   No definitive mortality benefit has been established.

8. Does it complement or displace proven screening?
   The evidence and proposed labeling support addition, not substitution.

These questions apply not only to Galleri, but to the entire emerging category of multi-cancer detection tests.

Where the evidence goes next

The FDA must now decide whether to approve the application, reject it or require additional information. It may also narrow the indication, modify the labeling or require post-approval studies.

If approval occurs, the label will matter enormously.

Calling the test “multi-cancer detection” would describe what the assay does more conservatively. Calling it “early detection” implies something broader about stage, timing and clinical benefit.

Longer-term NHS-Galleri results will also matter, particularly whether the observed increase in earlier-stage diagnoses and reduction in stage IV disease translate into fewer cancer deaths.

Independent mortality trials remain essential. The National Cancer Institute is developing large randomized research infrastructure to study whether multi-cancer blood testing produces more benefit than harm across diverse populations.

And implementation will need its own evidence.

Even a technically strong test can fail to improve health if positive results lead to fragmented workups, delayed imaging, unaffordable procedures or inconsistent follow-through.

Cancer screening is not a specimen sent to a laboratory.

It is a system.

The Modern Bio Take

Galleri represents a genuine technological advance.

A single blood draw that can identify a plausible cancer signal and point clinicians toward its likely origin could become especially valuable for cancers that currently have no routine screening pathway.

But the FDA panel’s debate exposes the standard that advanced testing must meet.

A measurable signal is not a diagnosis.

Detection is not necessarily early detection.

Earlier diagnosis is not automatically longer life.

And regulatory approval, if it comes, will not be the same as proof that population-wide screening reduces cancer mortality.

The correct response is neither dismissal nor hype.

It is disciplined optimism: recognize what the test can already do, be explicit about what it misses, protect existing evidence-based screening, and demand outcome data strong enough to justify the medical cascade that follows a positive result.

The future of cancer screening may include one blood test for many cancers.

The future of evidence-based medicine still requires proving that finding more ultimately helps people live longer and better.

Educational note: This article is for general educational purposes and is not medical advice. Cancer screening decisions depend on age, symptoms, personal and family history, prior findings and other clinical factors. A multi-cancer blood-test result cannot diagnose or exclude cancer. Screening and diagnostic decisions remain the responsibility of the patient and their qualified healthcare practitioner.

Primary Sources and Editorial Fact-Check Notes

• U.S. Food and Drug Administration — September 23, 2026: Molecular and Clinical Genetics Panel meeting announcement and event materials. Confirms the proposed indication, intended population, prescription status and advisory nature of the meeting.
• U.S. Food and Drug Administration — September 23, 2026: 24-Hour Summary of the advisory committee meeting. Reports the 10–0 safety vote, 6–4 effectiveness vote and 7–2–1 benefit-risk vote, along with panel concerns about “early detection” wording, variability by cancer type and absent long-term outcome evidence.
• U.S. Food and Drug Administration — September 2026: FDA Executive Summary for the Galleri PMA review. Provides the FDA’s analyses of PATHFINDER 2 and NHS-Galleri, including sensitivity, specificity, predictive values, cancer-type performance, diagnostic-resolution time and proposed labeling.
• NHS-Galleri Trial — updated May 30, 2026: What the trial found. Reports that the randomized trial did not meet its primary endpoint for combined stage III and IV cancers, while noting more early-stage diagnoses and fewer stage IV and emergency-presenting cancers in the intervention group.
• National Cancer Institute: Questions and Answers About Multi-Cancer Detection Tests. Reviews the distinction between screening and diagnosis, potential harms, current evidence gaps and the absence of definitive evidence that multi-cancer screening reduces overall cancer mortality.
• National Cancer Institute: Vanguard Study. Describes the actively recruiting pilot intended to inform a larger randomized trial evaluating whether multi-cancer detection screening reduces deaths and provides net benefit.