Multi-cancer early detection (MCED) tests are blood tests that screen for a cancer signal. Most read the cell-free DNA that dying cells shed into the bloodstream, and many also predict where in the body the signal originates. This guide explains how the tests work, what the clinical trial evidence shows, how to read the statistics, how the leading tests compare, and the status of regulation and reimbursement.
Educational information, not medical advice. This guide explains the science and current evidence around multi-cancer early detection for a general audience. It is not medical advice, not a diagnosis, and not a recommendation to obtain, avoid, or act on any test, screening, or treatment. Decisions about screening and care should be made by each person together with their own physician. See the full disclosures at the end of this page.
What this guide covers
What is multi-cancer early detection?
Conventional cancer screening looks for one cancer at a time (for example, a mammogram for breast cancer). Most cancers, however, have no recommended screening test for the general population, and many are found only after they have spread. Multi-cancer early detection is different in that a single blood draw is analyzed for a signal for many cancers simultaneously.
The biological foundation is cell-free DNA (cfDNA). As cells throughout the body die, they release fragments of their DNA into the blood. Those fragments carry the patterns of the cell they came from, and cancer cells carry distinctive, aberrant patterns. By reading the patterns on those fragments, a test can, in principle, detect that a cancer is present and predict the organ it most likely originated from.
Read the full chapter: How MCED Testing WorksHow the tests work
GRAIL's Galleri workflow is representative: blood is drawn into stabilized tubes, cfDNA is extracted from the plasma, the DNA is chemically treated so that methylation patterns become a readable sequence, a custom panel enriches for roughly 100,000 of the most informative genomic regions, the sample is sequenced on a high-throughput platform, and a machine-learning classifier trained on more than 15,000 participants decides whether the methylation pattern looks like cancer (and if so, where it most likely originated).
Not every company reads methylation alone. Some add circulating protein biomarkers, DNA mutations, fragmentomics (the size and positioning of cfDNA fragments), or signals from the immune system's response to a tumor.
Read the full chapter: How MCED Testing WorksHow to read the numbers
Three statistics carry most of the weight in any MCED discussion, and they are easy to conflate. Sensitivity is the share of true cancers the test catches. Specificity is the share of cancer-free people the test correctly clears. Positive predictive value (PPV) is the probability that a person with a positive result has cancer (and, notably, is dependent on how common cancer is in the population being screened).
Read the full chapter: Statistical TermsThe evidence at a glance
The case for and against MCED testing rests on a handful of studies. Each answered a slightly different question, and the figures below should be read in that context. Case-control validation studies and prospective screening studies are NOT directly comparable. The deeper "Journal Club" chapters work through the design and caveats of each.
| Study | Test | Design / size | Key reported figures |
|---|---|---|---|
| CCGA | Galleri (GRAIL) | Case-control validation, ~15,000 participants | 99.5% specificity; 88.7% cancer-signal-origin accuracy; sensitivity rose sharply with stage |
| PATHFINDER | Galleri (GRAIL) | Prospective interventional | Of 92 positive results, 57 were false positives; a first look at real-world workup |
| PATHFINDER 2 | Galleri (GRAIL) | Prospective; 32,007 analyzable (35,878 enrolled) | PPV 60.3% (ESMO 2025: 61.6%); 69.8% sensitivity for the 12 deadliest cancers (ESMO 2025: 73.7%) |
| DETECT-A | CancerSEEK / Cancerguard (Exact) | Prospective, 10,006 women aged 65-75 | 27.1% sensitivity, an early prototype; this figure rose in subsequent presentations reporting the performance of the test |
| Shield MCD | Shield (Guardant) | Case-control validation, 10 cancer types | 98.5% specificity; 59.7% overall sensitivity |
Comparing the tests
Three names dominate the field. GRAIL's Galleri has the most clinical data and the longest track record and is built on methylation. Guardant's Shield test (for colorectal cancer, only) is FDA-approved and is being extended into multi-cancer detection (Shield MCD) using methylation and fragmentomics. Exact Sciences' Cancerguard (Exact Sciences was acquired by Abbott in 2026) combines methylation, mutational analysis, and protein biomarkers.
Read the full chapter: The Competitive LandscapeThe case for and against
The strongest argument for MCED testing is that most cancer deaths come from cancers we cannot currently screen for, and a test that shifts even a fraction of those diagnoses earlier could be enormously valuable. The strongest arguments against are serious. There is, as of 2026, no randomized evidence that MCED screening makes people live longer; sensitivity is weakest precisely for the early-stage cancers the tests are meant to catch; and every false positive sends a healthy person into weeks (or months) of scans, biopsies, and anxiety.
Read the full chapter: The Case Against the Blood TestRegulation, FDA, & insurance
The regulatory picture is changing, perhaps most obviously through The Nancy Gardner Sewell Medicare MCED Screening Coverage Act, which was signed into law on February 3, 2026 and will make way for Medicare coverage to begin January 1, 2029. The downstream cost of the diagnostic workups triggered by positive results (and who pays for them, especially after an elective, out-of-pocket test) is a largely unanswered question. As of 2026, the FDA has not approved any test, although GRAIL applied for premarket approval (PMA) to the FDA.
Read the full chapter: Regulation and ReimbursementWhat is next
Several developments will shape the next few years. The NHS-Galleri trial has now presented its primary results at ASCO in 2026. The primary endpoint, a reduction in stage III and IV cancers, was not met, though the secondary endpoint, a reduction in stage IV cancer, was. The full peer-reviewed manuscript and, eventually, the mature mortality data are still awaited and remain the most important readouts on the horizon. Galleri's PMA application is pending at the FDA, and an approval would make it the first FDA-approved MCED test. Whether the FDA and CMS move from diagnostic-yield endpoints toward mortality endpoints will set the evidentiary bar for other tests, and how professional societies (NCCN, USPSTF, ASCO) treat MCED testing as data accumulate could change everyday practice more than any single regulatory or commercial event.
Frequently asked questions
What is a multi-cancer early detection (MCED) test?
An MCED test is a blood test that screens for a cancer signal. Most analyze cell-free DNA (cfDNA) for abnormal patterns characteristic of cancer, and many also predict where in the body a detected signal is coming from.
Is the Galleri test FDA-approved?
As of 2026, no MCED test has full FDA approval as a multi-cancer screening test. GRAIL's Galleri is available as a laboratory-developed test with a premarket approval (PMA) application pending at the FDA.
How accurate are MCED tests?
Reported accuracy varies widely by cancer type and stage. In published research, specificity has been high (around 99.5% for Galleri in the CCGA study), while overall sensitivity is more modest and is lower for early-stage cancers than it is for advanced-stage cancers. Positive predictive value depends heavily on how common cancer is in the population studied. These figures come from research studies and do not predict how a test will perform for any individual.
Does Medicare or insurance cover MCED testing?
The Nancy Gardner Sewell Medicare MCED Screening Coverage Act was signed into law on February 3, 2026, with Medicare coverage scheduled to begin January 1, 2029. Private coverage remains limited, and most people currently pay out-of-pocket. Who pays for what happens after a positive result is a separate, unsettled question.
What does a "cancer signal detected" result mean?
According to test makers and published study protocols, a "cancer signal detected" result is not, by itself, a cancer diagnosis. In studies, it has generally been followed by further diagnostic evaluation, which may include imaging studies and a biopsy, and some positive results turn out to be false positives. What any particular result means is a question for the person undergoing a test and their physician.
How do MCED tests relate to mammograms or colonoscopies?
Test makers and current guidance describe MCED tests as a potential complement to established screening, such as mammography for breast cancer and colonoscopy for colorectal cancer, rather than a replacement for established screening. Whether and how to use any screening test is a decision for each person together with their physician.
Which companies make MCED tests?
The most prominent are GRAIL (Galleri), Guardant Health (Shield and Shield MCD), and Exact Sciences (Cancerguard, with Exact Sciences acquired by Abbott in 2026). Other companies are also developing tests.