GP suggests caution for new cancer detection blood test

Amanda Lyons

25/01/2018 3:00:23 PM

A new blood test may offer the possibility of early cancer detection in primary care – although it is still too early to be certain.

News teaser
Professor Jon Emery has raised concerns over the potential for false positives when screening asymptomatic people for cancers.

A research team led from the US and coordinated through three countries, including Australia, has developed a blood test called CancerSEEK that is designed to detect early signs of eight common cancers – ovary, liver, stomach, pancreas, oesophagus, bowel, lung and breast – by detecting mutated DNA and abnormal proteins that tumours release into the blood.
Although the initial research focused on only eight of the more common types of cancer, Professor Peter Gibbs, a clinician scientist at the Walter and Eliza Hall Institute in Melbourne who contributed to the research, believes CancerSEEK may have even wider application in the future.
‘It should detect all cancers, except for brain tumours, because all of the tumour types release these proteins and mutated DNA,’ he told newsGP.
Researchers hope CancerSEEK can eventually be used as a screening test in asymptomatic populations, starting at age 50, as well as in younger people who have high-risk family histories of cancer. If this vision is realised, the benefits to patients will be invaluable.
‘Firstly, we can screen for multiple tumours at one time, which is the first time we’ve had a test that can do that,’ Professor Gibbs said. ‘Secondly, the test screens for major killers like pancreas, stomach and liver cancer, that previously haven’t had any screening tests.’
While the possibilities from this research are exciting, Professor Jon Emery, GP and Herman Professor of Primary Care Cancer Research at the University of Melbourne and Western Health, advises caution at this early stage.
‘[The test] performs very well in a group of patients who already have a cancer diagnosis. That’s perfectly appropriate, as early research and shows this test has a lot of promise,’ he told newsGP. ‘But false positives become a bigger issue once you start testing in populations like you see in general practice, where the prevalence of cancer is quite low.
‘If you start screening thousands of people who are asymptomatic, when the next test might be a colonoscopy or a CT scan or an invasive biopsy, and you have high rates of false positives, then you start causing more harm than good.’
CancerSEEK has so far only been trialled on 1500 people with early diagnoses of non-metastatic cancers and 812 people in a ‘healthy’ control group.
Professor Emery cited the now-controversial prostate-specific antigen (PSA) test as an example of the potential issues related to such tests.
‘[The PSA test] was let loose before the large randomised controlled trials showed how much benefit it provided in terms of reducing number of deaths from prostate cancer, and how much that outweighed the harms from over-diagnosis and over-investigation of false positives,’ he said.
Professor Emery has also voiced concerns about CancerSEEK’s variable sensitivity to different cancers, which raises issues about its potential usefulness as a screening tool. However, he does agree that the research is very promising, even in its current form.
‘It might still help you with earlier detection, but in people with symptoms,’ he said.
‘It could have a place in general practice for the assessment of a patient with symptoms that might be due to cancer, but most of the time are not. It could be part of a diagnostic workup to determine whether the patient needs to be referred on for endoscopy or a CT scan or so on.’
The next step in CancerSEEK research will be focused on whether the test can detect asymptomatic cancers, as well as other questions about frequency and cost. But whatever the final results may be, they are still a long while away.
‘The next study is a 10,000-person study of 50–60 year-old people having an annual blood test,’ Professor Gibbs said. ‘There will be a control group as well. The data from that will be three to five years away, at least.
‘So this is really the beginning, but a very important beginning.’

cancer-blood-test cancer-detection cancerseek


 Security code