AI-Designed Drug Targets an "Undruggable" Cancer Protein

Mayo Clinic researchers have developed an experimental drug with the help of artificial intelligence (AI) to target the "undruggable" PDZ-domain of GIPC1 protein that helps different types of cancer, including pancreatic cancer, to grow and resist treatment. In laboratory studies, the drug slowed tumor growth, improved survival and enhanced the effects of the chemotherapy drug gemcitabine.

The findings, published in Cell Reports, also demonstrate how AI could help develop a small molecule inhibitor of a target protein that can be less toxic and speed the discovery of new cancer treatments.

Pancreatic cancer is one of the most difficult cancers to treat because it is often diagnosed after it has spread and is resistant to many current therapies. While survival rates remain low, advances in research are creating new opportunities to develop more effective treatments. This study represents one of those promising advances, offering a potential new strategy for targeting pancreatic cancer.

Using AI, in collaboration with Sravathi AI Technology for IP Sharing, a company based in Bangalore, India, the research team screened nearly 40,000 potential compounds before identifying one that blocks GIPC1. The experimental drug also showed early signs of changing the environment around tumors in ways that could improve future combination therapies. Additional studies are needed before it can be tested in people.

"Pancreatic cancer has remained exceptionally difficult to treat because tumors rapidly adapt and become resistant to many available therapies," says Debabrata (Dev) Mukhopadhyay, Ph.D., senior author of the study and a cancer researcher at Mayo Clinic in Florida. "Our study demonstrates that AI can help us identify entirely new therapeutic opportunities against targets that have historically been considered undruggable. While these findings are preclinical, they provide a strong foundation for the next phase of research."

The therapy remains experimental, and additional studies are needed to evaluate its safety and determine whether it can advance to clinical trials.

Reference: Rachamala HK, Rao Nakka NM, Angom RS, et al. AI-driven discovery and validation of a GIPC1 PDZ domain inhibitor for pancreatic ductal adenocarcinoma. Cell Rep. 2026:117777. doi: 10.1016/j.celrep.2026.117777

This article has been republished from the following materials. Note: material may have been edited for length and content. For further information, please contact the cited source. Our press release publishing policy can be accessed here.