GRAIL

This is the first randomised answer to whether a multi-cancer blood test changes outcomes, and for its main question the answer was no: adding the test to NHS care for three years did not lower the number of cancers diagnosed at stage III or IV. The stage IV result is a secondary finding with a confidence interval that touches no effect, and deaths have not yet been reported, so the trial continues to follow participants.

These are the numbers GRAIL's US application rests on. Six in ten positive results were cancer and the predicted organ was usually right, but six in ten cancers diagnosed in the year were not flagged, and the study had no control group, so it cannot say whether testing changes outcomes.

For someone offered the test, these are the numbers that describe what a result means in an NHS population: about 1 in 100 tests came back positive each year, between 46 and 58 in 100 positives were cancer, and a negative result left about 1 in 100 with an undetected cancer within the year. The test found between a quarter and a third of all cancers diagnosed in a screening round, and about half to two thirds of the 12 cancer types it was designed to find. Whether finding them this way reduces late-stage diagnoses is the primary question, and this paper says only that the answer was no; the primary-endpoint paper had not appeared on Europe PMC by 23 September 2026.

Outcomes are known for fewer than half of the positives, so the share of positives that were cancer in this group cannot be read as the test's positive predictive value, and cancers missed by the test are not counted at all.

In people who already have symptoms the test is more often right when positive, because cancer is commoner, but it still misses most stage I cancers, so a negative result cannot cancel an urgent referral.

Whether a cancer is found depends mostly on how much tumour DNA it sheds into the blood, which is why small early cancers are often missed by every approach.

Methylation-based screening is real but weakest exactly where pancreatic cancer needs it, at stage I, which is why high-risk surveillance rather than population screening carries the field here.

Set beside the result (35 cancers among 92 positives, positive predictive value 38%), it shows the test found fewer cancers than its makers had modelled.