AI Matches Untested Drugs to Rare Diseases, Saves Lives

Kaila Mabus once collapsed on a grocery store floor, too exhausted from Castleman disease to stand. Doctors whispered about her dying when she wasn't in the room.

Today, the college nursing student feels fine. She credits an unexpected treatment discovered through artificial intelligence that matched a blood cancer drug to her rare lymphatic disorder.

Dr. David Fajgenbaum built the AI system after finding his own lifesaving treatment the hard way. In 2010, while in medical school, he developed Castleman disease and spent countless hours manually searching research papers for anything that might help.

Now his nonprofit Every Cure uses AI to analyze thousands of existing drugs against thousands of diseases simultaneously. The technology scores 75 million possible combinations, uncovering treatments hiding in plain sight.

"We utilize the world's knowledge, rather than relying only on what is in our brains," Fajgenbaum explained. The system has suggested surprising matches like Botox for depression and Lidocaine for breast cancer.

For Mabus, the AI recommended Ruxolitinib, a drug designed for blood cancer that the FDA never approved for her condition. Within weeks, her failing kidneys woke up.

"I don't think I would be here if this treatment was not found for me," said Mabus, who would have needed a kidney transplant without the intervention.

Meanwhile, Columbia University researchers use AI-powered microscopes to find single sperm cells among billions of fragments in seemingly empty semen samples. The technology helps men who thought they were hopelessly infertile become fathers.

"The AI is what allows us to see the sperm," said Dr. Zev Williams, who directs Columbia's Fertility Center. Humans attempting the same search faced mind-numbing, impossible odds.

Moderna and Merck recently announced breakthrough results for a personalized melanoma vaccine possible only through AI analysis. The technology examined each patient's unique tumor to create individualized treatments for 1,100 people.

The Ripple Effect

These AI systems accomplish in minutes what would take human doctors lifetimes to research. Every Cure's database grows daily, offering hope to patients with rare diseases who have exhausted traditional options. The personalized vaccine approach being tested for melanoma could eventually adapt to other cancers, multiplying its impact across millions of patients worldwide.

Dr. Williams believes we're entering medicine's golden age, where technology grants insights never before possible.