Concept art of artificial neural network

AI-generated drug regime for deadly childhood cancer in trials

Image credit: Dreamstime

Scientists from the Institute of Cancer Research in London have used an AI platform to suggest a drug regime for children with a deadly brain cancer, for which survival rates have not improved for 50 years. The treatment has been shown to be effective in mice and has been initially tested on four children.

Diffuse intrinsic pontine glioma (DIPG) is a rare, aggressive type of brain tumour which affects children. It occurs in a part of the brainstem which controls many of the body’s most critical functions. DIPG is difficult to treat with surgery because the cancer is diffuse; there is no well-defined border suitable for operation.

A quarter of children with DIPG have a mutation in a gene known as ACVR1; there are no approved treatments to target this mutation.

Scientists at the Institute of Cancer Research and the Royal Marsden NHS Foundation Trust have used AI to identify a potential treatment for the cancer. They found that combining the drug everolimus with another called vandetanib could enhance the latter’s capacity to pass through the blood-brain barrier in order to treat the cancer. Both drugs are already approved for treating other cancers.

So far, the combination has proved effective in mouse models – increasing the amount of vandetanib in the DIPG-afflicted mouse brains by 56 per cent and extending survival 14 per cent compared with a control treatment – and has been initially tested on four children. The drug combination’s next step is a clinical trial in a wider group of children.

“DIPG is a rare and aggressive childhood brain cancer, and survival rates have not changed over the past 50 years, so we desperately need to find new treatments for this disease,” said Professor Chris Jones, a paediatric brain tumour biologist at the institute. “Our study demonstrates just how much AI can bring to drug discovery for cancers like DIPG in proposing new treatment combinations that would not have been obvious to people.”

“The AI system suggested using a combination of two existing drugs to treat some children with DIPG: one to target the ACVR1 mutation, and the other to sneak the first past the blood-brain barrier.”

He explained that a full-scale clinical trial will assess whether the treatment can benefit children, but they had moved to this stage more quickly than would have otherwise been possible thanks to the platform.

The initial plan for the study came from BenevolentAI, a company that has built an AI drug discovery platform. Researchers at the Institute of Cancer Research worked with the company to identify drugs that could be used to target ACVR1 mutations in DIPG. The platform hosts a knowledge graph containing all publicly available biomedical data and further information from scientific articles.

The platform allowed them to explore the vast information spaces in the graph and uncover insights they would not have found using human thinking alone.

Professor Peter Richardson, pharmacology lead at BenevolentAI, said: “AI-enhanced approaches are already proving their value in expanding researchers’ capabilities to find innovative new treatment approaches – be it through uncovering new therapeutics or repurposing existing ones – not only in DIPG, but also other diseases in the future.”

Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.

Recent articles