Scientists have used artificial intelligence to build a drug regimen for children with deadly brain cancer; the survival rate has not improved in 50 years. Diffuse internal pontine glioma is a rapidly growing and rare type of brain tumor in children.
Removal of these types of tumors surgically is quite risky. They are diffuse, which means that they do not have well-defined boundaries suitable for operations. Some children with DIPG have the mutation gene ACVR1, and there is currently no exact treatment for this mutation.
Using AI, in a new study, scientists from the ICR and the NHS Foundation Trust were able to find that the drug everolimus in combination with other vandetanib could increase the ability of vandetanib to treat cancer in the blood.
The combination was effectively tested on mice and four children. Experts hope to try the drug combination in the broader group of children in clinical trials. The study found that combining the two drugs increased the amount of vandetanib in the brains of mice by 56%.
The treatment was 14% more effective in mice than those undergoing standard control treatment. According to the research team, the study shows that artificial intelligence can find new ways to treat cancer by revealing a combination of existing medications.
Finding New Ways
Both drugs used in the study to treat other cancers have already been approved. According to Chris Jones, professor of pediatric brain tumors at the ICR, DIPG is a very aggressive and rare disease, and survival rates have hardly changed in the last 50 years. That is why finding new treatment is vital.
Research shows how much artificial intelligence can contribute to the discovery of drugs and offer new therapeutic combinations. The AI system used a combination of two existing medications to treat some children with DIPG.
The treatment increased the viability of the mouse model. Study participants began testing the method in a small number of children. Group members report that a full-scale clinical trial is necessary to assess whether it can treat children.
BenevolentAI is the company that built the AI drug discovery platform. The ICR researchers worked with representatives of BenevolentAI to use the platform to identify drugs that ACVR1 mutations could target in DIPG.
The platform contains all publicly available biomedical data and additional information used by the BenevolentAI system from scientific articles. Scientists have been able to study the information in the graph and discover the views that could not be found through human thinking alone.
According to a pediatric and adolescent neuro-oncologist, this study is encouraging and highlights the potential for artificial use. The current clinical results are too preliminary to draw firm conclusions. Researchers are working to make progress in clinical trials in children with ACVR1-mutant DIPG.
Professor Peter Richardson explained that enhanced AI approaches could effectively expand capabilities and find new strategies.