Is a pilot project of a multi-cancer detecting blood test a revolution in the making? Or is the project wildly premature given the state of the evidence?
The Galleri test, produced by the California-based company GRAIL, can detect more than 50 types of cancer, from lung cancer to lymphoma, based on a simple blood draw. The UK’s NHS announced last week that it will pilot the test in 165,000 patients.
“Early detection – particularly for hard-to-treat conditions like ovarian and pancreatic cancer – has the potential to save many lives,” said NHS chief executive Sir Simon Stevens. “This promising blood test could therefore be a game-changer in cancer care, helping thousands more people to get successful treatment.”
The NHS Long Term Plan includes a goal of increasing the proportion of cancers detected early from half to three quarters by 2028. In theory, the Galleri test, which analyses blood for cell-free DNA that is shed from tumour cells, could help reach that goal.
The pilot project, which will be funded by GRAIL, will include 140,000 participants aged 50 to 79 years who do not have any cancer symptoms, along with 25,000 with possible cancer symptoms.
But some experts are skeptical of the NHS announcement. “It is extremely difficult to determine the scientific basis for this so-called pilot study,” said Professor Paul Pharaoh, of Cambridge University, according to the British Society for Haematology. “There is no published evidence that this test is an effective test for screening for cancer, so one has to question the aims of the pilot.”
A study published earlier this year found the test had greater than 99% specificity, with a 0.7% false-positive rate. However, as Prof. Pharaoh pointed out, it has not been specifically tested in randomised trials as a screening test – which is what it will function as in the 140,000 participants with no cancer symptoms.
“The NHS should not be investing in such a test before it has been adequately evaluated in well-conducted, large-scale clinical trials,” he said.
In the validation study from earlier this year, the test’s accuracy improved with rising cancer stage; in stage I disease it may be too low to be effective.
“This will be a great breakthrough for cancer diagnosis if it can be validated,” said Professor Yong-Jie Lu, of Queen Mary University of London. “For a cancer screen test, it needs very high specificity (>99%), otherwise it may end up in a similar situation as the PSA test for prostate cancer or even worse.”
The NHS pilot will be by far the largest study of the blood test; the validation study had fewer than 7,000 patients. Results of the pilot are expected by 2023, when, if successful, it will expand to one million participants over the following two years.