Cancer is something many of us — or our loved ones — will have to face at some point, with one in two people affected in their lifetime. Getting an early diagnosis and starting treatment is key.
That’s why there’s been so much excitement about a new test currently being trialled by the NHS, for both those with symptoms of cancer, and those who don’t have any obvious signs of it.
And in June it was revealed that this new test could be offered to a million people from next summer.
The suggestion is that it could ‘transform cancer care’.
The test in question, in case you’ve missed all the headlines, is the Galleri test: it looks for DNA markers of 50 types of cancer in a single blood sample.
The Galleri test: it looks for DNA markers of 50 types of cancer in a single blood sample (stock image)
However, as I’ve discovered by looking more closely at the evidence, it isn’t as simple as the hype around it would have you believe.
And while it may be better than some current cancer tests, at this stage you wouldn’t want to rely on it as a standalone diagnosis — and it could cause real problems if you don’t actually have the disease.
To paraphrase Charles Dickens, it’s a ‘tale of two tests’.
Let me explain. Take the patient I saw who had come into A&E because of severe pain in her back after a minor fall. A scan showed that she had cancer in her bones, which had spread from a previously undiagnosed breast cancer.
A few years earlier, she’d actually noticed a lump and the skin tethering. She also had a family history of breast cancer.
She was petrified, and also scared of hospitals and doctors. But as she already had a routine appointment arranged for a mammogram (as part of the NHS screening programme), she thought, ‘this will tell me the answer anyway’. And when the scan results came back as normal, she celebrated.
But a couple of years later, she lost weight, felt tired, and then had the fall which led to her diagnosis in A&E.
The original mammogram had been reviewed by an expert breast radiologist, but they had no knowledge of her symptoms and risks and would have just thought it was a routine screening test.
Did this matter? In fact, context does matter when assessing test results: the fact that she had symptoms, and a family history, meant she needed to be seen in a breast cancer clinic even if the cancer was not picked up by the scan.
This is very different from another of my patients, an American who was on a business trip in the UK when he developed severe pain and couldn’t pass urine. The cause? A needless check for prostate cancer.
A healthy man in his 40s, with no symptoms, his job as a banker meant he had a yearly medical which included a PSA (prostate-specific antigen) test for prostate cancer.
It came back as positive and he had two biopsies. But, while his results, thankfully, proved negative, he started to experience problems getting an erection — quite important, especially as he’d started a new relationship.
Just before arriving in England, he developed pain on passing urine, then, during a meeting, the pain got worse and he was rushed to A&E. His symptoms were caused by an infection as a result of the biopsies, and he needed antibiotics and a catheter to help him urinate.
These two stories highlight the flaws in testing for cancer, and why the new ‘wonder’ test may not be all it’s cracked up to be.
In the first case, the patient was told she had a negative test when in fact she did have cancer (‘a false negative’).
The NHS is running a trial on more than 140,000 people to see if using this blood test helps pick up more cancers at an early stage (stock image)
In the second case, the PSA test came back as positive when in fact he didn’t have cancer — a ‘false positive’.
The best tests have as few of these misdiagnoses as possible and, as techniques improve, the error rates will reduce.
But no test will ever be 100 per cent perfect. We use the term ‘sensitivity’ as a measure of the chances of a test missing a cancer diagnosis if you’ve actually got cancer.
So if a particular test is 100 per cent sensitive, you can be 100 per cent sure that you don’t have cancer if the test comes back negative. Sadly, while mammograms are pretty good, they are not 100 per cent sensitive, they’re 85 per cent. So, if your result comes back as negative for cancer, there is still a small chance you do have the disease. This is the scenario many of us would find most worrying.
Another measure, ‘specificity’, is the chances of testing negative if you do not have the condition — in an ideal world, a test would be 100 per cent specific: so if you then tested positive, it would mean that you do definitely have cancer. Sadly, for my American patient, the PSA test is not 100 per cent specific; it’s more like 80-90 per cent, which meant ‘cancer’ was wrongly flagged up and he had unnecessary biopsies and subsequent complications. And all this is why we need to be less gung-ho about the Galleri test.
In an Oxford University-led study, 5,461 people in England and Wales who were referred to hospital by their GP with suspected cancer were given the Galleri test, as well as the other routine test they would normally get, such as a colonoscopy for suspected bowel cancer
The results, presented to the American Society of Oncology annual conference in May, showed that the test correctly identified two-thirds of cancers.
The Guardian newspaper reported that this test ‘has the potential to spot and rule out cancer in people with symptoms’ — imagine how transformative this would be. But this suggestion was based on a worrying misunderstanding because the Galleri test emphatically can’t ‘rule out’ cancer in people with symptoms.
The original studies on the Galleri test, published in 2021 in the Annals of Oncology, showed its specificity was 99.5 per cent — in other words, 99.5 per cent of people without cancer tested negative — which is very impressive.
But its sensitivity was only 51.5 per cent, i.e. just over half of people with cancer tested positive — missing the other half completely.
PROFESSOR ROB GALLOWAY: If I had symptoms of cancer, I would consider asking for the test as an additional measure to help the doctors (stock image)
And the numbers who tested positive were much lower (16.8 per cent) if they had early stages of cancer — before it has spread and when it’s easier to cure.
So the Galleri might well have missed my patient with breast cancer and I, for one, would not want to use this test to rule out a cancer in a patient with symptoms such as blood in their stool without other further tests including a colonoscopy.
But where it could be helpful is in screening. The NHS is running a trial on more than 140,000 people to see if using this blood test helps pick up more cancers at an early stage. Its specificity of 99.5 per cent means that my American patient, if he’d had it, may well have been spared the biopsy and infection that the less accurate PSA test led to.
Used alone without any other test, the Galleri would miss actual cancers — around 84 per cent of those in the early stages of cancer and, overall, around half of all cancers. So it can never be used as the only test for cancer for those with symptoms.
However, where it can be helpful in screening, is in picking up some patients without symptoms who otherwise would not get a diagnosis at all. These people could have their lives saved.
But at what cost? The test costs roughly $900 (about £700) in the U.S. and on a national scale, that would mean a lot of money spent to pick up a few extra cases of cancer.
And it’s not only the financial cost, there are those other people who’ll be harmed by the false positive test results that need further investigations.
We also need to consider that some of those with cancer who tested negative will assume that they have not got cancer and ignore their symptoms, whereas without the test, they would have sought medical help earlier.
If I had symptoms of cancer, I would consider asking for the test as an additional measure to help the doctors.
But, if I were in charge of the NHS budget, I’d need much more convincing that this was the best way of spending vital resources to reduce cancer deaths.
- Twitter: @DrRobgalloway
What’s in a name?
This week: Tailor’s bottom
Pain in the buttocks after prolonged sitting is known medically as ischiogluteal bursitis — but as it ‘traditionally affects people sitting on a hard surface for long hours, its more common name is weaver’s or tailor’s bottom’, explains Sampada Bhide, an advanced practice physiotherapist at Kingston Hospital NHS Foundation Trust, Surrey.
Tailor’s bottom is the more common name for pain in the buttocks after prolonged sitting. It is known medically as ischiogluteal bursitis
Sitting in the same position for a long period can put pressure on the bursae (the fluid-filled sacs that act as a cushion between the muscles and sitting bones), causing inflammation and pain.
It’s more common in older, more frail people, and it’s also seen in cyclists who use a hard saddle. Anti-inflammatories and a well-placed cushion can help, says Sampada, as can doing sit-to-stand squats using a chair, or bridges (where you lie on your back with your knees bent before raising your hips off the floor, keeping them in line with your knees, and holding for a few seconds before relaxing).