UK Edition
Universal cancer test to pick up 95% of disease on horizon
A universal cancer test which could pick up dozens of different diseases before they become deadly could be widely available within ten years.
Scientists from universities across Britain have trawled through 23,000 studies published over the past five years to compiled a list of 750 ‘biomarkers’ which indicate cancer is present in the body, even when there are no symptoms.
They hope to whittle down the number to around 50, and create a single blood test which could check for not only the most common diseases such as breast, lung and bowel cancer, but also rarer conditions.
It is hoped that such a test could spot cancer early, before it has spread, and when it can be cured quickly with surgery or drugs.
The team also plans to use artificial intelligence software to increase the accuracy of the test over time. As more people are screened, computer learning would compare results to outcomes to avoid false positives.
“When we looked at all the papers we were quite surprised to find that nobody had yet created a test from any of the biomarkers that have, so far, been discovered,” said Professor Ian Cree, Molecular Pathologist at University Hospitals Coventry and Warwickshire.
“Our vision is that the screen will pick up even the small amounts of these biomarkers that might be in the blood at an early stage of the cancer, without necessarily identifying which cancer they relate to.
“Patients would then be referred for more specific tests, that could narrow down the tumour type. In theory, we could have a test ready within three years for use in high risk groups.
“It would be great if we could pick up 95 per cent of all cancers. But we need to make this available to the general population so it has to be accurate. We don’t want to worry people unnecessarily or give false reassurance.”
The work was carried out on behalf of the Early Cancer Detection Consortium, a group of nearly 40 organisations, including universities, hospitals and commercial companies and funded by Cancer Research UK.
The next step will be to look in detail at the research behind each biomarker, to check that it
is robust and then begin a clinical study, using samples from cancer patients and healthy controls, to check how effectively they identify the presence of cancer.
The biomarkers which work successfully in the study will be taken forward into a clinical trial, to see if the screening test works in practice and is cost-effective.
The team has already selected eight clinics in the UK to carry out initial testing and are hoping to recruit thousands of patients in the next few years.
Prof Cree added: “If everything went swimmingly, then we would probably be looking at rolling this out in general practice within 10 years.
“But we think that within a few years genetic screening will be able to show those who are most at risk and those high risk groups could be screened earlier.”
Cancer charities said early detection was vital.
Fiona Osgun, Cancer Research UK’s senior health information officer, said: "Early detection is a vital area of research, and that’s why Cancer Research UK is investing in projects exploring its potential.
"A blood test to detect different types of cancer is an exciting idea, but before a test like this could be used, there needs to be strong evidence that it’s accurate, could save lives, and its benefits would outweigh the risks – such as the possibility of diagnosing cancers that would never have caused harm.”
Lead researcher, Dr Lesley Uttley, from Sheffield’s School of Health and Related Research, said: “Our data mining approach allowed us to take in all relevant research findings from the five-year period, which meant we could map the full range of potential blood-based biomarkers that are particularly relevant for early detection of cancer.”
The Consortium has set up a company, Pinpoint Cancer Ltd, to take the research forward.