AI tool could help tailor treatment for most common adult brain tumour  

Scientists at the University of Toronto and Mayo Clinic have pioneered a new tool that uses AI to analyse images of tumour tissue, providing doctors with reliable information in less than an hour.    

The tool requires only a laptop to run. It could help identify meningiomas that have a high risk of growing back quickly after surgery and may therefore benefit from earlier, more aggressive treatment.   

We funded the study with the Canadian Institutes of Health Research. Findings have been published in The Lancet Digital Health. 

What did the study find? 

Meningiomas are usually low grade, but some grow quicker than others and may come back after treatment. 

In this study, researchers applied AI to images of surgically removed meningioma tissue from more than 600 patients. These images are routinely available following surgery. 

Results showed the AI tool could reliably classify meningiomas into molecular groups, achieving up to 97% accuracy. A growing body of work shows that these groups can help doctors give a more accurate prognosis and personalised treatment plan than more traditional approaches. 

Why is an AI tool important? 

Significantly, these new findings suggest that AI can provide the same level of insight into meningioma biology and behaviour as expensive molecular tools.  

This is important because molecular tools are only available at select institutions. It can also take several weeks to get the results. 

Previously, expensive genetic profiling was needed to reliably predict how these tumours would behave, and how well they’d respond to treatment. With this new study, such information becomes available to all patients with a meningioma. Our main hope is that it will pave the way for better treatment decisions.”

Dr Alexander Landry, neurosurgery resident and PhD candidate at the University of Toronto and first author 

Meningioma research 

Dr Gelareh Zadeh, who is one of our Quest for Cures grant holders, was a senior author on the paper. Her work is driving new understanding of meningiomas, the most common adult brain tumour, and how best to treat them. Find out more 

This study underlines the huge potential of AI to support meningioma diagnosis and treatment. By helping us overcome the cost and resource limitations associated with molecular tools, our AI tool could make a real difference.”

Dr Gelareh Zadeh, neurosurgeon at Mayo Clinic and senior author

The models developed in this study are now being made publicly available. Researchers hope they can be a useful decision support tool for clinicians, while validation for wider use is ongoing. 

Welcoming the new AI tool

Andy Tudor, a retired defence systems engineer from Waterlooville in Hampshire, was diagnosed with a grade 2 meningioma in 2017. He had successful surgery to remove the tumour – which measured around 7cm in diameter. The tumour recurred five years later, and Andy had Gamma Knife radiotherapy to kill the cells.   

Andy, now an Involvement Champion at The Brain Tumour Charity, welcomed the research:  

“It’s fantastic to see technologies like AI being used in such a positive way. I very much welcome its potential to help tailor meningioma treatment, which could prove a massive benefit for people like me in the future.”