An international team of researchers, including members from the Medical University of Vienna, has developed an artificial intelligence (AI) tool that can classify brain tumors using genetic material found in cerebrospinal fluid. The findings were published in Nature Cancer.
The new method uses an AI algorithm called M-PACT (Methylation-based Predictive Algorithm for CNS Tumors). It analyzes cell-free DNA fragments released by cancer cells into the cerebrospinal fluid. These fragments carry molecular patterns unique to different types of brain tumors, allowing for reliable classification even with very small amounts of DNA.
Until now, diagnosing brain tumors typically required tissue samples obtained through neurosurgery, which carries risks and is not always possible. The new approach instead relies on analyzing cell-free tumor DNA from the cerebrospinal fluid. Researchers report that M-PACT was able to classify brain tumors accurately even when only limited tumor-associated DNA was present. The tool also allows tracking genetic and epigenetic changes over time, enabling non-invasive monitoring of treatment response and detection of relapses or secondary tumors.
"This could make a decisive difference, especially for children with tumors that are difficult to access or in the early stages of the disease," said Gojo. "In the long term, this technology opens up the possibility of diagnosing brain tumors from a cerebrospinal fluid sample before surgery and monitoring the course of the disease closely and less invasive."
The study included analysis of samples from multiple international centers and showed strong agreement between AI-based classifications and established methods based on tissue samples. However, authors note that more clinical studies are needed before this technique can be widely adopted in routine practice.
