Galleri Multi-Cancer Signal Test · Blood

Most cancers have no screening test at all. You can get a mammogram for breast cancer, a colonoscopy for colon cancer, and a low-dose CT scan for lung cancer if you qualify. But for the vast majority of cancer types, pancreatic, ovarian, stomach, kidney, liver, head and neck, and dozens more, there is nothing to order until symptoms force the issue. By then, the cancer is often advanced.

The Galleri test is designed to change that equation. With a single blood draw, it scans for a shared cancer signal across more than 50 cancer types and, when it detects one, predicts the most likely organ of origin. It does not replace any existing screening you should already be doing. It adds a layer of detection for the cancers that no other screening can catch.

What the Test Actually Measures

Galleri does not measure a single protein, hormone, or chemical the way most blood tests do. Instead, it analyzes tiny fragments of DNA floating in your blood, called cell-free DNA (cfDNA). Every cell in your body sheds small bits of DNA as it ages and dies. When cancer is present, tumor cells shed DNA fragments that carry distinctive chemical tags called methylation patterns, essentially a unique fingerprint left by the cancer.

The test uses machine learning to compare the methylation patterns in your blood sample against patterns known to be associated with cancer. If the algorithm detects a cancer-consistent pattern, it reports "cancer signal detected" and predicts the tissue where the cancer is most likely originating. If no cancer-consistent pattern is found, the result is "cancer signal not detected." There is no numeric score or concentration level. The output is binary: signal found or signal not found, plus a predicted site of origin when positive.

How Accurate Is It?

The test's strongest feature is its specificity, meaning its ability to correctly identify people who do not have cancer. In a clinical validation study of 4,077 people (a mix of known cancer cases and cancer-free controls), specificity was 99.5%. That means out of every 200 cancer-free people tested, only about 1 would get a false alarm.

Sensitivity, the ability to catch cancers that are actually present, is more variable and depends heavily on how advanced the cancer is. In that same validation study, overall sensitivity across all cancer types was 51.5%. But the numbers shift dramatically by stage: the test caught only about 16.8% of stage I cancers, 40.4% of stage II, 77.0% of stage III, and 90.1% of stage IV. For 12 of the deadliest cancer types (including lung, colorectal, pancreatic, ovarian, and liver), sensitivity at stages I through III was about 67.6%.

When the test does detect a signal, it predicts the most likely organ of origin correctly about 85% to 91% of the time, depending on the study. This prediction is what guides your doctor toward the right imaging and workup, rather than requiring a head-to-toe scan.

Real-World Performance

Validation studies use carefully selected groups of people with and without known cancer. Real-world data, where the test is used in routine clinical settings, tells a slightly different story and is arguably more relevant to you.

What this means for you: if you take this test and it comes back positive, there is roughly a 1-in-2 to 3-in-4 chance (depending on the setting) that cancer will actually be found on follow-up. That is a much higher hit rate than many traditional cancer screening tests produce. But it also means a positive result does not guarantee cancer, and you will need additional imaging and possibly procedures to find out.

What a "Cancer Signal Not Detected" Result Means

This is where the biggest misunderstanding happens. A negative result means the test did not find a cancer-associated methylation pattern at the level its algorithm can detect. It does not mean you are cancer-free. Given the test's sensitivity profile, it will miss roughly half of all cancers overall, and the majority of very early stage I cancers. If you have a negative result and stop doing your routine colonoscopies, mammograms, or lung CT scans, you are making a dangerous trade.

The test is designed as a complement to existing screening, not a replacement. Think of it as casting a wider net: standard screening covers a few specific cancers very well, while this test adds surveillance for the dozens of cancer types that standard screening does not address at all. Both together are better than either alone.

Cancers It Is Especially Useful For

The test's greatest value is for cancers that have no other screening option. Pancreatic cancer, ovarian cancer, stomach cancer, kidney cancer, head and neck cancers, and many blood cancers currently have no recommended screening test for the general population. Case reports and cohort data show that the Galleri test has detected early-stage versions of these cancers in asymptomatic people, catching disease that would otherwise have gone unnoticed until symptoms appeared, often at a much later stage.

Research on cancers detected in symptomatic patients referred for urgent evaluation (the SYMPLIFY study of 6,238 people) found sensitivity of 66.3% and specificity of 98.4%, with a positive predictive value of 75.5%. When the test said cancer, it was right about three-quarters of the time. When it said no cancer, it was right about 97.6% of the time.

What Happens After a Positive Result

A positive result triggers a diagnostic workup directed by the predicted cancer signal origin. If the test predicts a signal from the gastrointestinal tract, for example, you would typically get imaging of the abdomen (CT or MRI) and possibly an endoscopy or colonoscopy. If the predicted origin is gynecologic, the workup would involve pelvic imaging and possibly biopsy. The predicted site of origin focuses the workup and avoids unnecessary whole-body scanning.

In the PATHFINDER trial, most participants with a positive result had their diagnostic evaluation completed within about 79 days. For those whose positive result turned out to be a false alarm, the median time to full resolution was 162 days, roughly five months of follow-up tests, imaging, and waiting. Qualitative interviews with participants in the NHS-Galleri trial found that the period between a positive result and diagnostic resolution was often anxiety-provoking, and some participants felt they had to advocate for themselves within healthcare systems unfamiliar with the test.

If you get a positive result, you should be seen by an oncologist or a physician experienced with this test. The predicted site of origin should guide targeted diagnostic imaging, not a random battery of tests. If the initial workup does not find cancer, a second round of evaluation may be warranted, and annual retesting should continue.

Performance Across Demographics

In a study of 4,077 participants from the CCGA (Circulating Cell-free Genome Atlas) cohort, the test's specificity remained very high (98.1% to 100%) and sensitivity was broadly similar (approximately 44% to 63%) across racial and ethnic groups, with sensitivity improving by stage in all groups. The NHS-Galleri trial is collecting detailed demographic, socioeconomic, and behavioral data to further evaluate whether the test performs differently in specific populations, but those results are not yet available.

When Results Can Be Misleading

The single most important thing to understand about this test is that sensitivity drops sharply at early stages. A cancer that sheds very little DNA into the bloodstream, which is typical of small, stage I tumors, may not produce enough of a methylation signal for the test to detect. This means the test is better at finding cancers that are somewhat more advanced, and a negative result provides less reassurance for very early disease.

No published data currently addresses whether common medications (statins, metformin, steroids, thyroid drugs, or others) affect the test's accuracy, or whether acute events like illness, surgery, or intense exercise in the days before testing can distort results. The NHS-Galleri trial collects medication data, suggesting researchers consider this relevant, but no results on medication-related interference have been reported. Until more is known, there are no specific pre-test restrictions beyond following the lab's blood draw instructions.

Understanding the Test's Result Format

Unlike most blood tests, Galleri does not produce a numeric value with a reference range. There is no "optimal level" or "borderline zone." The result is either "cancer signal detected" (with a predicted tissue of origin) or "cancer signal not detected." This binary format means the standard framework of low, normal, and high does not apply.

Because the result is a yes-or-no answer rather than a number on a scale, there are no research-based reference tiers like those you see for cholesterol or blood sugar. Instead, the relevant numbers are the test's accuracy metrics: 99.5% specificity (very few false alarms), 51.5% overall sensitivity (catches about half of cancers, more at later stages), and 85% to 91% accuracy in predicting where the cancer is located when a signal is found.

What We Still Do Not Know

The single biggest unanswered question is whether using this test actually helps people live longer. No completed randomized trial has yet shown that Galleri screening reduces cancer deaths. The NHS-Galleri trial, which enrolled approximately 140,000 adults aged 50 to 77 in England, is designed to answer exactly this question. Its primary goal is to see whether people screened with Galleri have fewer late-stage cancer diagnoses compared to those receiving usual care. Results are expected in the coming years.

Modeling studies project that annual screening in adults aged 50 to 79 could reduce late-stage diagnoses by roughly 39% to 49% and meaningfully lower cancer mortality. But models are projections, not proof. Cost-effectiveness analyses suggest the test may be economically reasonable for adults 50 and older, but this depends on assumptions about test performance translating from validation studies into the real world.

Tracking Over Time

Because this test produces a binary result rather than a continuous number, "tracking a trend" works differently here than it does for cholesterol or blood sugar. The value of annual retesting is not about watching a number drift upward. It is about repeated surveillance: a cancer that was too small to detect this year may shed enough DNA to trigger a signal next year. Each round of testing is a fresh opportunity to catch something early.

Modeling data suggests annual screening provides more benefit than biennial (every-two-year) screening in reducing late-stage diagnoses and deaths within five years. If you decide to use this test, annual retesting is the cadence most supported by the available modeling. A single negative result provides a snapshot, not a guarantee. The real power is in consistent, repeated screening year after year.

If Your Result Is Positive: A Decision Pathway

A "cancer signal detected" result should prompt a focused, efficient diagnostic workup. The predicted cancer signal origin narrows the field. Your next steps should include a visit with an oncologist or physician familiar with multi-cancer detection testing, targeted imaging of the predicted site (CT, MRI, ultrasound, or endoscopy as appropriate), and standard diagnostic procedures for the predicted cancer type.

If the initial workup finds cancer, treatment follows standard oncology protocols for that cancer type and stage. If the workup does not find cancer, this does not necessarily mean the result was wrong, sometimes cancers are difficult to locate or are too small for imaging to detect. Continued monitoring and a repeat of the test in 6 to 12 months is reasonable. If you receive a "cancer signal not detected" result, continue all your standard screening (mammograms, colonoscopies, Pap smears, lung CT if eligible) and plan to retest annually.