Galleri’s FDA path brings hope and caution into the same clinic visit
Grail’s Galleri blood test is edging toward what could become the first FDA-approved multicancer early detection assay, and the debate around it has sharpened into a familiar clinical tension: how much disease can a screening test find before it creates new diagnostic problems of its own. The test analyzes DNA fragments in blood for a shared cancer signal and then predicts the likely tissue of origin. A positive result is not a diagnosis; it is a trigger for imaging, biopsy, and the kind of downstream workup that can either clarify a malignancy or expose a false alarm.
The regulatory stakes are high. An FDA advisory committee backed the test 6-4, while also signaling caution about what the assay can and cannot do for early cancer detection. Grail says the test should be used alongside guideline-recommended single-cancer screening, not in place of it, and the intended population is people age 50 and older. The list price is $949, and a new law creates a pathway for Medicare coverage beginning in 2028.
What the available trials show, and what they still cannot prove
The strongest recurring theme from cancer specialists is not whether the test can detect cancer, but how well it performs across different cancers and clinical settings. Dr. Elizabeth O’Donnell pointed to the positive predictive value as one of the first numbers that matters in real-world screening. In the NHS-Galleri study, the positive predictive value was just over 50%; in PATHFINDER 2, it was 62%. In plain terms, a positive test result now has a meaningful chance of representing true cancer, but it is still far from diagnostic certainty.
One sentence keeps returning to the center of the discussion.
The episode sensitivity remains relatively low, at roughly 30% to 40% in the capped 12-month clinical trial setting.
That matters because a screening test can only change outcomes if it catches enough cancers early enough to alter treatment. O’Donnell emphasized that the field is still missing cancers, even as specificity remains high and the false-positive burden stays low. In clinical trial data, 0.9% of tested participants had a positive result, which keeps the total number of people sent for follow-up relatively small. For primary care clinicians, that is not a trivial operational detail; it determines whether a screening program becomes manageable or swamps practice workflows.
Why tissue of origin may help clinicians, but cannot close the diagnostic loop
Karthik Giridhar framed the test’s appeal around three numbers: low false-alarm rates, decent positive predictive value, and tissue-of-origin prediction that helps doctors decide where to look next. In PATHFINDER 2, specificity reached 99.6%, meaning only about 1 in 280 people without cancer received a positive result. The test predicted the cancer’s location correctly more than 90% of the time, which usually steered follow-up toward scans and blood work rather than immediate invasive procedures. Only 0.6% of everyone tested underwent an invasive procedure after a positive result.
Those figures explain why some specialists see practical value in the assay, even before mortality data arrive. A blood test that points clinicians toward the right organ system can shorten the diagnostic trail, especially when the tumor type is one that currently lacks routine screening. In Giridhar’s account, the test performed best in cancers that shed more DNA into the bloodstream, including liver and bile duct cancers, colorectal cancers, pancreatic and gallbladder cancers, head and neck cancers, and lymphomas. In those groups, detection within a year ranged roughly from 75% to 90%.
Its weak point is just as important. It caught fewer than 1 in 10 prostate cancers and fewer than 1 in 4 breast cancers. That asymmetry is the reason the test cannot replace established screening pathways for malignancies where mammography, colonoscopy, and cervical screening already perform well.
The cancers it misses are the cancers already watched most closely
Parijatham Thomas focused on the most consequential limitation of all: uneven performance across tumor types. Galleri does not detect every cancer equally well, and that includes some of the most common ones in routine practice. Breast and prostate cancer sensitivity remains very low, which means patients still need mammography and, for men, individualized discussion about PSA screening. In the NHS-Galleri study, the pre-specified endpoint was reduction in stage 3 and 4 cancers across 12 primary cancer types after three annual rounds of screening, and the trial did not meet that primary endpoint. The trend moved in the desired direction, but it was not statistically significant.
That result keeps the test from being framed as a wholesale substitute for current screening. It may extend detection to cancers that currently arrive late and symptomatic, but it cannot be assumed to improve every category equally. In practice, the strongest argument for multicancer testing is not in replacing mammography or colonoscopy. It is in finding diseases such as pancreatic or ovarian cancer before symptoms develop, when curative treatment is still plausible.
Overdiagnosis, anxiety, and the burden of workup remain real clinical costs
Arthur Burnett pushed the discussion toward the downstream consequences of widespread adoption. A positive multicancer blood test may identify a malignancy, but it may also uncover findings that are ambiguous, low-risk, or not immediately actionable. If the assay is used too broadly, overdiagnosis could lead to extra imaging, extra procedures, and avoidable anxiety. The test may tell clinicians that something is wrong; it cannot, by itself, establish the precise threat level.
Burnett also raised a practical concern that extends beyond oncology: what happens when a positive result enters the medical record and a full workup fails to find a clear malignancy? That question is especially sensitive in a screening field that already lives close to genetic testing, insurance uncertainty, and long-tail consequences of documentation. The test’s role will need careful boundaries if clinicians are to avoid turning a screening signal into a permanent source of distress without clinical payoff.
Commercial uptake is ahead of the evidence on mortality
Galleri is already in circulation. Giridhar said Grail has sold more than 595,000 commercial tests through June 30, 2026, and FDA approval would almost certainly accelerate use. Yet the evidence that matters most to patients — whether screening actually reduces cancer deaths — is still pending. The long-term data have not matured to the point where mortality benefit can be claimed.
For now, the test sits in a narrow but important clinical space. It appears to be strongest where current screening is weakest, especially for aggressive tumors that shed enough DNA to be detected in blood. It is weakest where established screening already works well, and it cannot exclude cancer when the result is negative. A negative test does not rule out disease, and a positive test still demands confirmation.
The next phase will depend less on enthusiasm than on coverage decisions, follow-up pathways, and whether ongoing studies can show that earlier detection changes outcomes in a durable way.