28 mrt 2025
Cyclomics and UMC Utrecht have received a Health Holland research grant to develop AI-driven software for ultra-rapid intraoperative brain tumor classification, leveraging real-time DNA methylation sequencing and advanced AI models to improve surgical decision-making and expand diagnostic applications across multiple cancer types.
Cyclomics and UMC Utrecht receive Health Holland Public-Private Collaboration (publiek-private samenwerking) research grant to develop AI-driven software for brain tumor classification.
Central nervous system (CNS) tumours represent one of the most lethal cancer types, particularly among children. Primary treatment includes resection of the tumour, which requires maximizing the extent of resection and minimizing risk of neurological damage and comorbidity. However, surgeons have limited knowledge of the tumour type prior to surgery. Therefore, Cyclomics - an Oncode Institute portfolio company - has joined forces with the research group of Cyclomics co-founder and Oncode Investigator Prof. Jeroen de Ridder, to develop software for ultra-rapid classification of brain tumors. The collaborative project now receives a major push through the subsidy provided by Health Holland.
The PPS project titled 'DISCUS: Development and Implementation of Software for Classification of brain tumours Using real-time methylation Sequencing' builds on research from the group of Prof. Jeroen de Ridder [https://www.nature.com/articles/s41586-023-06615-2].
This work demonstrated how advanced AI models can classify brain tumors from sparse DNA methylation data. Since such methylation data can be generated in real-time using Oxford Nanopore sequencing, brain tumors can be diagnosed during surgery (intra-operatively), which is a game-changer for the treatment of patients with brain cancer.
The DISCUS consortium will now develop improved versions of the AI-classification models for brain cancer, and also develop a software package - termed Epinn - with a user interface and a data reporting module for routine clinical use of the AI models.
"The support from Health Holland through this PPS grant strongly accelerates the development of advanced AI-driven diagnostic solutions at Cyclomics", says Cyclomics CEO Dr. Wigard Kloosterman. "The clinical value of ultra-rapid classification of brain cancers has already been demonstrated, and this initiative will pave the way for expansion of the technology to many clinical centers worldwide."
"Beyond brain cancer, our AI models can be used for many other cancer types for which classification is important for determining optimal treatment trajectories", adds Prof. de Ridder. "Methylation is a very strong biomarker to type cancers, and we are at the beginning of a new wave of diagnostic solutions that combine novel AI technologies with DNA and methylation profiling to radically improve the way we diagnose cancer."