Pioneering blood test for non-invasive detection of Paediatric Brain Cancer

12 December 2024

Charlie Teo Foundation has invested over AUD$1 million to support scientists at Northwestern University to build Better Tools to treat brain cancer.

One of these courageous scientists is Craig Horbinski, M.D., PhD, director of Neuropathology at Northwestern Medicine and part of the team at Malnati Brain Tumor Institute at Northwestern.

Key Findings and Impact of this Research Grant

We’re excited to announce the latest progress from Prof. Craig Horbinski, a recipient of a $275,860 AUD Better Tools grant from the Charlie Teo Foundation. Prof. Horbinski is leading groundbreaking research to develop new ways to diagnose paediatric brain cancer using just a blood sample, reducing the need for invasive surgical biopsies. His team’s findings, recently published in the high-impact journal Acta Neuropathologica, bring hope for more accurate and less complex ways to detect brain tumours in children.

The team’s approach uses a methodology called a liquid biopsy, which works by searching for tiny bits of DNA that tumours shed into the bloodstream. Think of the tumour as a tree, and these fragments of cell-free DNA are like leaves that fall into a river. By collecting and analysing a blood sample, doctors can catch and study these “leaves” which can provide important clues about the tumour, all without needing to touch the brain.

In the study, Prof. Horbinski’s team examined blood samples from 77 children with brain tumours and compared them to blood samples from 16 children without tumours. They developed a blood test that can detect different types of brain tumours by identifying specific patterns in the DNA shed by the cancer cells—kind of like finding the fingerprints of a criminal at a crime scene. By developing machine learning algorithms to look at these patterns, researchers can help detect and potentially diagnose brain tumours without going inside the brain. Incredibly, they were able to distinguish children with brain tumours from those without with 83% accuracy from just a blood sample.

When it came to specific tumour types, the test worked particularly well. For circumscribed astrocytic glioma and glioneuronal tumours, it accurately distinguished these tumours from non-cancerous conditions in 86% and 83% of cases respectively.

To the best of our knowledge, this is the world’s first study to show that a blood test could aid in diagnosing children suspected of having a brain tumour. While the test demonstrated significant promise for certain tumour types, it was less effective for others, likely due to the complex and varied nature of brain tumours. For the two main subtypes—circumscribed astrocytic glioma and glioneuronal tumours—the test achieved a solid 80% accuracy. However, it had lower performance for rarer brain tumour types and may need further refinement and a larger sample size to provide statistically reliable results for these less common tumours.

Although the test showed strong potential, in the interim the researchers caution that it is not yet perfect and cannot replace traditional diagnostic tools such as brain scans or surgery. Nevertheless, this approach could be a valuable addition to presurgical evaluations, providing doctors with critical information to help guide treatment decisions.