Despite intensive research, pancreatic cancer remains one of the deadliest forms of cancer. Although only around four percent of all new cancer cases affect the pancreas, this type of tumor is the fourth leading cause of cancer death in Germany. At the German Cancer Congress 2026, German Cancer Aid is now presenting three large-scale projects that it is funding as part of its “German Alliance for Pancreatic Cancer” funding priority. In the funded project PrepAIred, researchers from the University Hospital Bonn (UKB) and the University of Bonn, in cooperation with the University of North Carolina (USA), want to use AI-based protein design to specifically combat pancreatic cancer. German Cancer Aid is funding the project with a total of 1.99 million euros over a period of four years.
Pancreatic cancer, known in medical terms as pancreatic ductal adenocarcinoma (PDAC), is one of the deadliest forms of cancer, with a five-year survival rate of only around ten percent. “Conventional therapies often fail due to the dense, hard-to-penetrate tumor tissue and the strong immunosuppression within the tumor environment,” says PD Dr. Gregor Hagelüken from the Institute of Structural Biology at the UKB. This is where the PrepAIred research consortium, which he heads together with Prof. Dr. Michael Hölzel, Director of the Institute for Experimental Oncology at the UKB, comes in. Both are members of the ImmunoSenstation3 Cluster of Excellence and the Transdisciplinary Research Area (TRA) “Life and Health” at the University of Bonn. The aim of the project is to use artificial intelligence to develop small proteins that can better overcome these tissue-specific barriers while simultaneously recognizing and binding to tumor cells in a targeted manner.
AI-generated mini antibodies combat hard-to-reach tumors
To achieve this, the project team is using groundbreaking AI tools that can predict protein structures and functions mathematically and design them specifically. “The AI minibinders developed in this way are expected to be smaller, more stable, and more effective than conventional antibodies, enabling them to reach even hard-to-access tumor regions,” says Prof. Hölzel. An interdisciplinary team of structural biologists, oncologists, surgeons, genome researchers, and pathologists is working together to develop, optimize, and preclinically test these novel molecules. The researchers are using patient-derived organoids and innovative 3D models to test the efficacy and safety of AI minibinders in a realistic manner without animal testing. “In the long term, this research could pave the way for new diagnostic and therapeutic procedures, for example in the form of imaging probes or immunotherapeutic agents,” says Hagelüken, and Hölzel adds: “The PrepAIred project is a prime example of the use of AI to improve cancer medicine.”
German Cancer Aid wants to make progress accessible to everyone
With the “German Pancreatic Cancer Alliance” (Deutschen Allianz Pankreaskarzinom), German Cancer Aid brings together the most renowned scientists in the field of pancreatic cancer research under one roof. At the core of the alliance are three research consortia based at a total of 23 university locations in Germany and three universities each in the US and the Netherlands. Together, they are working to better understand pancreatic cancer at the molecular level and to develop new, personalized therapy concepts. Modern technologies such as artificial intelligence are to help better characterize pancreatic cancer clinically.
Beyond funding individual research projects, German Cancer Aid is deliberately focusing on structural development. It will soon publish a separate call for proposals for a higher-level research and coordination platform, the aim of which is to bring together all experts researching pancreatic cancer in Germany within the alliance and to establish close international networks.
Contact
PD Dr. Gregor Hagelueken
Institut für Strukturbiologie
Universitätsklinikum Bonn
ImmunoSensation3 &TRA „Life & Health“, Universität Bonn
Tel. +49 228 287-51200
E-Mail: hagelueken@uni-bonn.de
