A new blood test that incorporates AI into its diagnostic techniques to detect more than 50 types of cancer will be trialled by the NHS next year.
The Galleri test, developed by California-based biotech company Grail, screens for cell-free nucleic acids (cfNAs), small fragments of DNA and RNA that are thought to be cast off by tumours in the body. These cfNAs reflect the genomic profile of the tumours where they originated, allowing not only the potential presence of cancer to be detected, but also its location within the body. Galleri uses artificial intelligence to help distinguish cancerous cfNA’s from the masses of other genomic material floating in the bloodstream.
Grail claims a US clinical study using an earlier iteration of Galleri successfully detected over 50 types of cancer with a low false-positive rate of less than one per cent through a single blood draw. The company’s own modelling indicates the test has the potential to decrease the number of cancers diagnosed at late stage by nearly half, which could reduce the total number of cancer deaths in the UK by approximately one-fifth.
Commencing in 2021, the NHS trial of the technology will involve approximately 165,000 people divided into two groups. The first will include 140,000 people over the age of 50 without any suspicion of cancer, while the second will include 25,000 people aged 40 and above with suspicious signs or symptoms of cancer. Based on data from this initial trial, access to the test could be expanded to around one million people across 2024 and 2025.
While the potential for the Galleri test is significant, many within the scientific community have warned that it is not a magic bullet and pointed to limitations in its ability to detect early-stage cancers
“Results so far, from studies outside of the UK, have been promising, and the number of cancer types that it seems to be able to pick up is large,” said Jodie Moffat, head of Early Diagnosis at Cancer Research UK.
“But the sample sizes, particularly for some cancer types, have been very small and so it needs to be tested in a much larger sample, and with longer follow up of patients not testing positive with the blood test to understand where it is missing cancers. Based on the evidence we have seen, the test is not currently that good at picking up stage I cancer, where it is small and hasn’t spread to other parts of the body.”
If the blood test can back up Grail’s claims and does turn out to be an effective tool to screen for early-stage cancer, it could dramatically improve survival rates for a disease that claims almost 200,000 lives in the UK each year.
“Detecting cancers including common and rare ones at an earlier stage should increase the cure rate,” said Prof Justin Stebbing, NIHR Research Professor of Cancer Medicine and Medical Oncology at Imperial College London.
“Blood tests, especially those that look at cancer DNA in the blood, that are touted to detect cancer are being increasingly studied, but their real world applicability, for example in the NHS, is still unproven. Investigating this with the possibility of potentially helping millions of people and integrating this into the NHS would be very worthwhile.”