Revolutionary AI study: Better understand and treat cancer metastases!

Revolutionary AI study: Better understand and treat cancer metastases!

A new, promising research project to combat cancer metastasis took place on March 1, 2025. The project called DECIPHER-M, led by Prof. Dr. med. Jakob N. Kather at the Else Kröner Fresenius Center for Digital Health aims to decipher the complex mechanisms of metastasis through the use of artificial intelligence (AI). Loud Mainz University Medical Center The main actors in the project are an interdisciplinary team from medicine, computer science and biotechnology.

Metastasis is considered the main cause of death among cancer patients. To better predict the development and location of metastases, DECIPHER-M uses AI multimodal foundation models to analyze different types of data. This includes radiological images, pathology reports and genetic information. This allows to answer fundamental questions about metastasis, including the identification of features associated with continuity and prognosis, as well as the development of a functionally validated risk model. The aim is to make more precise predictions about the individual risk of metastasis and to develop personalized treatment strategies to improve the survival rates of cancer patients, such as Helmholtz Munich explained.

Interdisciplinary collaboration

The project brings together several renowned institutions. In addition to the Mainz University Medical Center, the RWTH Aachen University Hospital, the Else Kröner Fresenius Center of the Technical University of Dresden, the University Hospitals in Essen, the German Cancer Research Center in Heidelberg, the Helmholtz Center Munich and the Technical University of Munich are also involved. This creates an extensive network that is intended to promote innovative approaches to cancer research, reports Health research BMBF.

In order to advance the implementation of this complex research, a total of around 9 million euros in funding will be made available. The first funding period, which lasts three years, is supported with around 5.5 million euros from the Federal Ministry of Education and Research (BMBF). These funds will be used to further develop AI technologies and create the “Cancer Foundation Model,” which combines various medical data sources. A significant aspect of this technology is the combination of image and text analysis through the combination of Vision Transformers and Large Language Models.

Objectives and outlook

With DECIPHER-M, the research institutions involved are pursuing the goal of sustainably improving the effectiveness of treatments for cancer patients. By identifying tumor biological signaling pathways and scientifically testing these principles, a comprehensive contribution to reducing metastasis should be made. The development of practical tools for adapting screening and treatment methods is also a central part of the project.

The emergence of a risk model that could be essential for the individualized treatment of future cancer patients shows the relevance of this research. The interdisciplinary team in Mainz is taking a decisive step in cancer research to further improve the understanding of metastasis and increase the quality of life of patients.