AI-Designed First‑In‑Class Small Molecule Inhibitor Shows Promise Against Pancreatic Cancer in Preclinical Studies

Mayo Clinic researchers have developed an AI-designed, first-in-class small molecule inhibitor targeting the PDZ domain of GIPC1, a protein that promotes pancreatic cancer growth and therapy resistance.

Preclinical studies show the candidate drug GIPCi slowed tumor growth, improved survival, and enhanced the effect of the chemotherapy drug gemcitabine in pancreatic cancer models, with no detectable toxicity.

The inhibitor was discovered using AI‑driven virtual screening of nearly 40,000 compounds and advanced molecular modeling, highlighting how AI can accelerate discovery of drugs against “undruggable” protein domains.

Compelling transcriptomic and spatial analyses indicate GIPCi remodels the tumor microenvironment, activates immune responses, and downregulates key tumor‑promoting pathways, suggesting strong potential for combination therapies.

The therapy remains experimental; further toxicology and regulatory work are ongoing before it can advance to clinical trials in humans.