AI-Designed Molecule "Apt1" Makes Pancreatic Cancer More Responsive to Chemotherapy

Researchers at the Italian Institute of Technology have used artificial intelligence to design a novel aptamer molecule called "Apt1" that could transform treatment for one of the deadliest forms of cancer. The AI-designed molecule disrupts DNA repair mechanisms in pancreatic cancer cells, making tumors significantly more sensitive to chemotherapy drugs like olaparib — even at lower doses.

Pancreatic cancer has a five-year survival rate of approximately 12%, making it one of the most lethal cancers. A major reason for its poor prognosis is that pancreatic tumors are notoriously resistant to chemotherapy, with cancer cells quickly repairing the DNA damage that chemotherapy drugs are designed to inflict.

“The AI-designed molecule disrupts DNA repair mechanisms in pancreatic cancer cells, making tumors significantly more sensitive to chemotherapy drugs like olaparib — even at lower doses.”

Apt1 works by interfering with this repair process. By blocking the cancer cells' ability to fix DNA damage, the molecule essentially removes their defense mechanism, leaving them vulnerable to much lower doses of chemotherapy. This dual approach — weakening the cancer's defenses while attacking it — showed superior results in preclinical models compared to either treatment alone.

Crucially, Apt1 demonstrated limited effects on healthy cells. This selectivity is one of the most promising aspects of the research, as traditional chemotherapy's inability to distinguish between cancerous and healthy cells is what causes many of its devastating side effects. A treatment that primarily targets cancer cells could dramatically improve patients' quality of life during treatment.

The use of AI in the molecule's design was essential to its development. Machine learning algorithms analyzed vast databases of molecular structures and protein interactions to identify the optimal aptamer configuration, a process that would have taken years using traditional experimental methods. This success demonstrates AI's growing role in accelerating drug discovery for diseases that have long resisted conventional approaches.