Danish and American researchers have developed an AI platform that can modify protein components to arm a patient's immune cells in the fight against cancer. The platform was developed by a team from the Danish Technical University (DTU) and the American Scripps Research Institute. It aims to solve a major challenge in cancer immunotherapy by demonstrating how scientists can generate targeted treatments for tumor cells while preventing damage to healthy tissue.
According to the Danish Technical University, the new method, published in the scientific journal Science, demonstrates for the first time that it is possible to design proteins in a computer that can redirect immune cells to cancer cells via so-called pMHC molecules. This significantly shortens the process of finding effective molecules for cancer treatment: from years to 4-6 weeks. Peptide-MHC (pMHC) molecules are formed by MHC (Major Histocompatibility Complex) molecules and a peptide—small protein fragments that the body presents on the cell surface, where they are recognized by T cells.
New Perspective
"We are essentially creating a new perspective on the immune system," explains Timothy P. Jenkins, associate professor at DTU and co-author of the study. "Current methods for individual cancer treatment rely on finding so-called T cell receptors in the immune system of a patient or donor that can be used for treatment. This is very time-consuming and challenging. Our platform designs molecular keys to target cancer cells using the AI platform, and it does this at incredible speed, so that a new molecule—acting like a kind of molecular rocket—can be ready within 4-6 weeks."
Normally, T cells naturally identify cancer cells by recognizing specific protein fragments—peptides. These fragments are presented on the cell surface by pMHC molecules. However, the process of applying this knowledge to therapy is slow and challenging, often because the body's variability in T cell receptors complicates the development of personalized treatment.
Power of AI Platform
In the Danish-American study, the researchers tested the power of the AI platform on a known cancer target, NY-ESO-1, which occurs in a wide range of cancers. The team succeeded in designing a "minibinder" that binds strongly to the NY-ESO-1 pMHC molecules.
When the designed protein was introduced into T cells, it created a unique new cell product, dubbed "IMPAC-T" by the researchers, which effectively guided the T cells to kill cancer cells in laboratory experiments. "It was incredibly exciting to see these minibinders, which were created entirely on a computer, work so effectively in the laboratory," said postdoc Kristoffer Haurum Johansen, co-author of the study and researcher at DTU.
The researchers also applied the method to design binders for a cancer target identified in a patient with metastatic melanoma. Here too, they successfully generated binders for the target. Based on this, they concluded that the method could also be used for personalized immunotherapy against new cancer targets.
Critical Step
A crucial step in the study was the development of a "virtual safety check." The research team used an AI platform to screen and evaluate their designed mini-binders in relation to pMHC molecules found on healthy cells. This method allowed them to filter out mini-binders that could cause dangerous side effects before conducting experiments.
"Precision in cancer treatment is crucial," says Sine Reker Hadrup, professor at DTU and co-author of the study. "By predicting and eliminating cross-reactions at the design stage, we were able to reduce the risk associated with the designed proteins and increase the likelihood of developing a safe and effective therapy."
Five-Year Project
Co-author Patrick Jenkins expects it will take up to five years before the new method is ready for the first human clinical trials. By then, the treatment process will resemble current cancer treatments using genetically modified T cells (CAR-T cells), which are currently used to treat lymphoma and leukemia.
Cancer patients will first have blood drawn in the hospital, similar to a routine blood test. Their immune cells will then be extracted from this blood sample and modified in the laboratory to carry the AI-designed mini-binders. These enhanced immune cells will be returned to the patient, where they will act as targeted missiles that accurately detect and eliminate cancer cells in the body.
Using AI in Cancer Treatment
AI applications and algorithms have long been used in diagnosing and predicting cancer development. Drug development can also be accelerated. Earlier this year, researchers at The Institute of Cancer Research (ICR) in London developed an AI technology that can predict how cancer cells respond to new drugs based on changing cell shapes. This innovative approach enables faster and more efficient drug testing, meaning new treatments could be available to patients years sooner. The technology uses 3D imaging and deep learning to analyze a cell's characteristic "fingerprint." Until now, other methods could only use "flat" 2D images.
Publication in Science, July 24, 2025: 'De novo-designed pMHC binders facilitate T cell-mediated cytotoxicity toward cancer cells', K.H. Johansen et al.
