AI Predicts Brain Cancer Survivors within Eight-Months of Radiotherapy Using MRI Scan

Glioblastoma, a particularly challenging adult primary brain cancer to treat, has a low survival rate, with only one in four patients living beyond a year after diagnosis. Typically, patients undergo an eight-month course of chemotherapy following radiotherapy. Currently, regular and routine scanning is performed on patients with adult primary brain cancer to evaluate the effectiveness of chemotherapy. However, this process can result in some patients undergoing ineffective chemotherapy that not only fails to prolong life but also subjects them to detrimental side effects. Now, researchers have demonstrated that artificial intelligence (AI) can predict whether adult brain cancer patients will survive for more than eight months following radiotherapy. This groundbreaking use of AI in predicting patient outcomes could significantly guide clinicians in planning subsequent treatment stages and expedite referrals to potentially life-saving therapies. This marks the first instance of AI being used to distinguish between short-term and long-term survivors within eight months post-radiotherapy.

After radiotherapy, follow-up brain scan findings are often non-specific and oncologists cannot be certain whether a treatment is working or failing. A team from King’s College London developed a deep learning model to enhance the reliability and accuracy of predicting outcomes for patients with adult primary brain cancer. The AI was trained on tens of thousands of scans from a diverse range of brain cancer patients. Instead of trying to grapple with interpreting every non-specific follow-up brain scan, the AI can make an immediate and accurate prediction of which patients will not survive the next 8 months by simply looking at one routine scan after radiotherapy. This can empower clinicians and patients to make choices about their treatment. By providing an instant and accurate prediction from one routine MRI scan, the AI enables physicians to identify those patients unlikely to benefit from chemotherapy, allowing them to consider alternative treatments or enroll patients in clinical trials for experimental therapies.

“This is exciting and fundamental research for people living with a glioblastoma, for two reasons,” said Dr. Helen Bulbeck, Director of Services and Policy at brainstrust. “At its simplest level it demonstrates how AI can be used for patient benefit. More importantly however, it empowers patients and their caregivers to make choices about the clinical pathway and gives control back at a time when so much control has been lost. Patients will be able to make informed decisions about treatment choices and will be able to plan how they want to spend the time they have left so that they can live their best possible day, every day.”

Related Links:
King’s College London