New Weapon Against Cancer: Artificial Intelligence Creates Drugs Faster Than Ever
Artificial intelligence could transform the way we fight cancer. Scientists have developed a platform that designs precise protein molecules that direct immune cells directly to cancer cells. A process that previously took years can now be completed in a matter of weeks.
A team of scientists from the Technical University of Denmark (DTU) and the Scripps Research Institute in the US has developed an artificial intelligence platform that designs proteins capable of precisely recognizing cancer cells. These "molecular keys," as the researchers call them, allow T cells to be redirected to target only the tumor, bypassing healthy cells.
“We are essentially creating a new set of eyes for the immune system,” says Dr. Timothy P. Jenkins, assistant professor at DTU and lead author of the study published in Science.
Until now, designing immunotherapy has been a tedious, individual process, taking up to several years. The new platform shortens this time to 4–6 weeks. In their research, the team focused on the well-known tumor antigen NY-ESO-1, present in many types of cancer. AI designed so-called minibinders—proteins that bind to cancer cells. After inserting them into T lymphocytes, new IMPAC-T cells were created, which effectively destroyed cancer in the laboratory.
“It was incredibly exciting to see how these mini-binders, created entirely on a computer, work in the lab,” comments Kristoffer Haurum Johansen from DTU.
Importantly, the team also used the platform to design proteins for a patient with metastatic melanoma—proving the method's suitability for personalized immunotherapy. To avoid complications, the researchers conducted a virtual safety check, eliminating proteins that could potentially bind to incorrect targets in healthy tissues.
“By predicting and excluding cross-reactions at the design stage, we have managed to reduce the risks associated with the designed proteins and increase the likelihood of developing a safe and effective therapy,” explains Prof. Sine Reker Hadrup.
The team anticipates that the first clinical trials in humans will be possible within five years. Ultimately, the entire treatment process will resemble CAR-T therapy, currently used for leukemia, with the difference that target design will be based on data from an AI platform.
"This is just the beginning. But if this technology lives up to the expectations placed in it, we could be talking about a real breakthrough in cancer treatment," the authors conclude.
politykazdrowotna
