Galleri test accuracy: what the numbers mean

Galleri is a blood test from GRAIL that looks for cancer DNA shed into the bloodstream and claims to detect a signal from more than 50 cancer types. Its accuracy splits into two very different numbers: how often a positive is right, which is good, and how many cancers it catches, which is modest. A positive deserves prompt follow-up; a negative cannot replace routine screening.

Why "99% accurate" is easy to misread: three numbers that matter

The headline figure you will most often see, around 99.5%, is specificity: the share of people without cancer who correctly get a negative result. Specificity tells you how rarely the test raises a false alarm. It tells you nothing about how many cancers the test catches.

Sensitivity is the share of people with cancer who get a positive result. Positive predictive value (PPV) is the share of positive results that turn out to be cancer. A screening test can have excellent specificity and poor sensitivity at the same time, and Galleri does.

Two kinds of sensitivity get mixed up. Test sensitivity comes from case-control studies, where researchers run Galleri on blood from people already known to have cancer; CCGA3 found 51.5% this way ¹. Episode sensitivity comes from real screening, where everyone is tested and followed for 12 months to count every cancer that shows up. Episode sensitivity is lower (about 39% in PATHFINDER 2) and is the number that applies to someone who feels well and is thinking about the test ³.

PPV sits well below specificity because cancer is uncommon in any single year. When fewer than 1 in 100 people tested has a cancer the test can see, even a 0.4% false-alarm rate produces a sizable share of the positives.

What the big studies found

Four studies carry most of the evidence. Each used a slightly different version of the test and a different group of people, so compare the numbers below side by side rather than averaging them.

The studies above used an earlier investigational version of the test. When stored PATHFINDER 2 samples were retested with the version GRAIL submitted to the FDA, the false-positive rate was 0.15% and episode sensitivity was 35.0%, against 0.36% and 39.3% in the published study ^{6 3}. Fewer false alarms came at the cost of slightly more missed cancers.

Out of 1,000 people tested

Scaling PATHFINDER 2's published counts (287 positives and 173 cancers among 32,007 people, with 39.3% episode sensitivity) to 1,000 adults aged 50 or older gives a clearer picture than percentages do ³. The counts below are our arithmetic from those figures, rounded.

A 39.3% sensitivity implies roughly 440 cancers in the PATHFINDER 2 year, so for every 5 cancers Galleri flagged, about 8 more came after a negative result. The NHS-Galleri trial phrases its own result the same way: about 1 in 100 people tested positive, about half of those had cancer, and about a third of people with cancer got a positive result ⁴.

Which cancers it finds, and which it mostly misses

Galleri reads methylation, the chemical tags on DNA, in fragments that tumors shed into the blood. Larger and more advanced tumors shed more DNA, so detection climbs steeply with stage ¹.

Cancer type matters as much as stage. In GRAIL's CCGA validation data, the cancers Galleri caught most often are those with no routine screening test, and the ones it caught least often are the ones routine screening already targets ¹.

Two caveats apply to that table. A cancer-type figure blends every stage, so 83.7% for pancreatic cancer does not mean Galleri finds 83.7% of stage I pancreatic cancers. And some cancer types had only small numbers of cases, which makes their figures less reliable.

How it compares with screening tests you already know

US screening guidelines cover four cancers (breast, cervical, colorectal and lung) and leave PSA testing for prostate cancer as an individual decision at 55 to 69. No US guideline recommends multi-cancer tests. Galleri adds a signal for cancers no guideline screens, and it is weaker than the guideline test for each cancer that has one.

Colon and cervical screening can find and remove changes before they become cancer. Galleri can only detect cancer that already exists, which is one reason the National Cancer Institute is explicit on where it fits ⁵:

Standard-of-care cancer screenings are recommended regardless of whether or not a person has an MCD test.National Cancer Institute, Questions and Answers about Multi-Cancer Detection Tests

If your result says "Cancer Signal Detected"

A "Cancer Signal Detected" result means the test found a methylation pattern that looks like cancer and predicts where in the body it likely started. It is not a diagnosis. In PATHFINDER 2, roughly 6 in 10 people with this result were diagnosed with cancer, and the test pointed to the right organ in 91.3% of those cases ³.

1. 1Book a doctor visit soonTake the full Galleri report, including the predicted signal origin, to the clinician who ordered it or to your primary care doctor.
2. 2Expect a targeted workupFollow-up is usually imaging aimed at the predicted organ, plus blood tests, and a biopsy if something is found.
3. 3Plan for the possibility of an invasive testIn PATHFINDER 2, 213 people had an invasive procedure such as a biopsy or endoscopy to check a positive result, including some whose result turned out to be a false alarm; no serious study-related harms were reported ³.
4. 4Allow for timeIn the first PATHFINDER study, people whose positive was a false alarm took a median of 162 days to reach a final answer, against 57 days for people who had cancer ².

Questions worth asking at that visit:

• Which tests will check the predicted organ first?
• What happens if the imaging is clear?
• Who pays for the workup? The Galleri price does not include follow-up testing.
• Should any finding change my regular screening schedule?

If your result says "No Cancer Signal Detected"

A negative Galleri result does not rule cancer out. In PATHFINDER 2, more than half of the cancers diagnosed within a year came after a negative test ³.

• Guideline screening stays in place: mammogram, colon screening, cervical screening, lung CT for eligible smokers, and the PSA conversation for men aged 55 to 69. Ask your doctor which of these you are due for.
• Galleri does not measure inherited risk; a negative says nothing about BRCA or Lynch syndrome.
• GRAIL suggests repeating the test yearly, but no guideline sets an interval because no study has shown which interval helps.

What the NHS-Galleri trial showed about saving lives

NHS-Galleri is the only large randomized trial so far. It assigned 142,250 people aged 50 to 77 in England to usual care or to three yearly Galleri tests ⁷. The main goal was fewer cancers diagnosed at stage III or IV among 12 deadly types, and the trial did not reach it: late-stage diagnoses were about as common in both groups (about 3% more in the Galleri group, rate ratio 1.03, 95% CI 0.92 to 1.14, a difference small enough to be chance) ⁷.

The National Cancer Institute's summary still stands: there are no definitive trials showing that multi-cancer tests reduce cancer deaths, and these tests may not separate slow cancers that would never cause harm from dangerous ones ⁵.

Galleri vs single tumor markers like CEA, CA-125 and PSA

Galleri reads a pattern across many cancer types to screen people who feel well. Single protein tumor markers work differently: most were built to track a known cancer during and after treatment, and they can rise for benign reasons too. PSA is the partial exception, used for prostate screening by individual choice. A raised marker on a routine panel calls for a doctor's read and is a different signal from a Galleri result. The tumor markers guide covers each one, including the benign causes.

Other multi-cancer blood tests

Galleri has the most published data of any multi-cancer early detection (MCED) test. Other companies, including Exact Sciences and Guardant Health, are developing or selling their own. None is in any US screening guideline, and the NCI's Vanguard study is now testing MCED approaches before any larger trial ⁵.

People often search for Galleri vs Shield. Guardant's Shield is a different kind of test: an FDA-approved blood test that screens for colorectal cancer only, not a multi-cancer test. For colorectal cancer, colonoscopy and stool tests remain the standard guideline options. For how Galleri sits among consumer lab options, see the at-home blood test comparison.

Who it's for, cost and FDA status

• Who GRAIL offers it to: GRAIL's website lists adults with elevated cancer risk, such as people 50 or older. It is not offered during pregnancy, at age 21 or under, or during active cancer treatment.
• Price: GRAIL's website lists it at about $949. Insurance usually does not cover it, and follow-up scans and biopsies after a positive result are billed separately.
• How it's ordered: by prescription from a clinician, with a standard blood draw.
• FDA status: Galleri is sold as a laboratory-developed test, meaning one lab runs it under lab rules without full FDA review of the test itself. GRAIL applied for full FDA approval (premarket approval) in January 2026, and on 23 September 2026 an FDA advisory panel voted 7 to 2, with one abstention, that its benefits outweigh its risks ⁶. The panel advises; the approval decision rests with the FDA.

Is the Galleri test worth it?

Galleri is always an add-on to guideline screening, never a replacement. Whether it adds enough depends on your age, risk and budget.

It adds the most if you:

• Are already up to date on guideline screening.
• Are older: cancer risk rises with age, which also makes a positive more likely to be right.
• Want some chance of catching cancers no screen covers, such as liver or pancreatic cancer.
• Can absorb the cost of a workup after a positive result.

Reasons to wait:

• No trial has yet shown fewer cancer deaths.
• A workup after a false alarm can take months.
• The price, which insurance rarely covers.

A Galleri result should never be the reason to skip a mammogram or a colonoscopy. Galleri needs a prescription, so this is a decision to talk through with your doctor, who can weigh it against your own history and the screening you are already due for. If you are thinking about it, first check you are up to date on mammogram, colon and cervical screening (and lung CT if you smoked), then ask whether Galleri adds anything for your age and risk.

For the wider picture of what routine blood work can and cannot pick up, see what a blood test can detect.

Free, no accountDrop in your lab report and see every marker explained in plain words, placed on the body, against your lab's range. It's read on your device: nothing is uploaded and nothing is kept.

Read my results Download the app