AI tool used to find a life-saving medicine for rare disease

Imagine being told you have no options left, that hospice care is the only path forward. That was the reality for one patient battling idiopathic multicentric Castleman's disease (iMCD), a rare and deadly condition with few effective treatments. But then, artificial intelligence stepped in. By sifting through thousands of existing drugs, an AI-powered system pinpointed a treatment that not only worked — but saved his life. Nearly two years later, he remains in remission.

This breakthrough represents a beacon of hope for countless others facing rare and often overlooked diseases.

How does it work?

So, what is the magic behind this life-saving discovery? Drug repurposing — using an existing medication for a new, unforeseen purpose. A team of researchers at the Perelman School of Medicine at the University of Pennsylvania fed an AI system with thousands of known drugs, searching for one that could target iMCD. The system zeroed in on adalimumab, a monoclonal antibody typically prescribed for arthritis and Crohn's disease.

Lab tests soon confirmed that adalimumab inhibits tumor necrosis factor (TNF), a protein found at abnormally high levels in severe iMCD cases.

As Dr. David Fajgenbaum, a senior researcher on the study, put it, "The patient in this study was entering hospice care, but now he is almost two years into remission. This is remarkable not just for this patient and iMCD, but for the implications it has for the use of machine learning to find treatments for even more conditions."

Why does it matter?

Rare diseases often fall through the cracks of medical research. There are simply too many of them and too few patients for pharmaceutical companies to invest in costly drug development. But AI changes the game. By leveraging machine learning, researchers can analyze astronomical amounts of medical data, uncover hidden connections between diseases, and find existing treatments that might work in unexpected ways.

For Fajgenbaum, this isn't just science — it's personal. He has iMCD himself and found his own life-saving treatment through a similar repurposing process over a decade ago. His experience led him to co-found Every Cure, a nonprofit aiming to unlock AI's full potential in drug discovery.

"There are probably a few hundred patients in the United States and a few thousand worldwide who, each year, are in the midst of a deadly flare-up like this patient had been experiencing," he explained. "More research is needed, but I'm hopeful that many of them could benefit from this new treatment."

The context

iMCD is a brutal disease. It triggers a cytokine storm — an out-of-control immune response where inflammatory proteins flood the body, attacking organs and causing multi-organ failure. Until now, treatment options have been limited, and survival rates bleak. That's why this AI-driven breakthrough is so significant.

The patient in the study had exhausted all conventional treatments. His body was failing. Then came one last shot: adalimumab. Against all odds, the drug reversed his symptoms, offering not just a second chance but a new frontier for medical research.

This success story is just the beginning. The research team is now preparing a clinical trial for another AI-identified repurposed drug, a JAK1/2 inhibitor, to treat iMCD. With AI's ability to comb through existing drugs at lightning speed, the future of rare disease treatment may no longer be a matter of luck — it may simply be a matter of looking in the right place.