Have you been diagnosed with a brain tumour? Order your free information pack.

‘Noah’s Ark’ language test could help diagnose brain tumours earlier

Verbal fluency test to name animals found to identify those with persistent headaches who are more likely to have a brain tumour, and could help GPs decide who needs rapid imaging.

A researcher, a young person of colour wearing a lab coat, blue latex gloves and red, The Brain Tumour Charity-branded bandana, investigates samples over a lightbox.

A simple language test – which asks people to name as many animals as they can in 60 seconds – could help GPs identify people with common symptoms such as headaches who are most likely to have a brain tumour, new research suggests.

In a study of 270 people, funded by The Brain Tumour Charity, those with a brain tumour were found to have significantly lower ‘verbal fluency task’ scores than those presenting with headaches who did not have a brain tumour.

The findings, published in BMC Neurology, suggest the triage test could provide a valuable addition to clinical assessments to help GPs identify patients in need of an urgent brain scan, while providing reassurance that a period of further observation may be appropriate for others.

The test, which could be quickly administered by GPs, is already used in assessing cognitive function for patients with neurological conditions including brain tumours – but the study represents the first time it has been investigated as a way to speed up the diagnosis of brain tumours.

While further research is needed to validate and optimise use of the test, experts at the University of Edinburgh hope the tool could in future help reduce time to diagnosis for people with brain tumours – enabling prompt access to treatment and maximising their quality of life. Over 12,000 people are diagnosed with a brain tumour in the UK each year. While more people than ever are surviving cancer in the UK, progress on brain tumours has continued to lag behind improvements in early detection and survival seen in other cancer types.

Just 12% of adults survive for five years following diagnosis with a malignant brain tumour, and overall the disease continues to reduce life expectancy by 27 years on average – the highest of any cancer.

Early detection and timely diagnosis remains a unique challenge in brain tumours. Less than 2% of all brain tumours are diagnosed via the suspected cancer pathway (two week wait) following a referral by a GP, compared to 54% in breast cancer – with around 40% of all brain tumour patients being diagnosed as an emergency presentation in England, many of whom may have presented to a medical professional previously.

The first signs experienced by patients can be vague; common to multiple conditions and many of the most frequent symptoms, such as persistent headaches or co-ordination problems, can have a variety of other far more likely causes.

With no effective screening tools to support GPs to identify those who need urgent imaging, fewer than 1% of all people currently referred by a GP for further investigation with a suspected brain tumour are ultimately found to have one.

In a new study led by Dr Paul Brennan at the University of Edinburgh, researchers assessed the performance of a test known as a verbal fluency task (VFT) as a potential new triage tool for patients presenting with persistent headaches and other possible brain tumour symptoms.

The study, conducted at Western General Hospital in Edinburgh, recruited 270 people over the age of 16:

  • 180 patients with a newly-diagnosed brain tumour (via MRI) who had presented with or without headache in addition to non-headache symptoms. For these patients, the verbal fluency task (VFT) was conducted at their pre-surgery appointment
  • 90 people with headache who had been referred by a GP for a scan with a suspicion of a brain tumour, whose final diagnosis excluded a brain tumour All participants were asked to name as many different animals as they could within 60 seconds, with repetitions discounted. Using a receiver operating characteristic (ROC) curve analysis, a score of 14 or more animals in one minute was found to differentiate ‘good’ from ‘poor’ performance in this group.

The researchers analysed the results for any association between VFT score and the presence of a brain tumour – finding that 87.5% of people with a poor score had a brain tumour, while 48.1% of people with a good score did not have a brain tumour.

Those who scored 14 or higher were nearly eight times more likely not to have a brain tumour – while a poor performance more than tripled the likelihood of a brain tumour being present.

The researchers also assessed how brain tumour grade, location and size affected verbal fluency performance. They found that a reduction in performance was most marked in patients with one of the three most aggressive brain tumours: high-grade gliomas, cerebral lymphomas and cerebral metastases.

Patients with high-grade glioma had significantly lower VFT scores compared to meningioma brain tumours and other intracranial tumours. Having a tumour in the frontal lobe or having a larger tumour were both also found to be associated with a lower VFT score.

The researchers suggest that in patients where a GP has a low suspicion of a brain tumour and the patient has a high VFT score, the test may provide reassurance that period of further observation may be appropriate – however a poor VFT score would be of concern and doctors may consider lowering their threshold for referral for urgent imaging.

Phonemic verbal fluency (based on letters) was also assessed, asking participants to name as many different words as they could beginning with the letter P – however this was not found to be as effective as semantic fluency (naming animals).

The researchers also argue that adjusting scores for educational level may help provide an even more accurate differentiator and suggest that further studies should test this hypothesis, as well as examining other patient factors that may have an impact on performance, such as a history of alcohol or drug abuse.

Lead author, Dr Paul Brennan, Honorary Consultant Neurosurgeon at the University of Edinburgh and NHS Lothian, said:

“The first symptoms experienced by patients with a brain tumour are often non-specific, such as headache, where a non-tumour diagnosis is much more likely. For example, for every 1,000 people presenting to a GP with headaches, just one or two will have a brain tumour.

“It remains difficult to determine which patients with these non-specific symptoms are most at risk of having a brain tumour. This unfortunately contributes to significant delays in diagnosis.

“Our study showed that a simple language test, which can be quickly and easily administered, could help GPs decide who is most likely to have a brain tumour. Symptomatic patients with low scores could be prioritised for rapid imaging, whilst other patients with high scores could be monitored as they are more likely to improve.

“Our findings are important in demonstrating proof-of-concept and we now need to validate this test in a larger group of patients to prove whether it could help guide referral for suspected brain cancer. Reducing time to diagnosis for people with a brain tumour remains critical as it means more rapid access to treatment, which is crucial to maximising patients’ quality of life.”