David Fajgenbaum was hospitalized without a diagnosis for nearly seven weeks as a medical student, facing a disease doctors could barely explain, let alone treat. He would nearly die in a hospital five times before he found his own way out — and that fight is now the engine behind an AI-powered nonprofit racing to find hidden cures for thousands of diseases that have none.
Fajgenbaum was diagnosed with idiopathic multicentric Castleman disease, a rare hematologic illness that causes cytokine-driven inflammation, dangerously low blood counts and potentially fatal damage across multiple organs, according to jci.org. As NPR reported, he was diagnosed with the condition while still in medical school, and nearly died in a hospital five times over the course of his illness. Speaking Sunday at the Purdue Northwest Sinai Forum Lecture Series on PNW's Westville campus, Fajgenbaum told the Indiana audience he had nearly died four times before yet another near-death episode struck, according to the Chicago Tribune.
During his initial episode, Fajgenbaum remained hospitalized without a diagnosis for nearly seven weeks and later spent 4.5 of the previous six months hospitalized. He missed his entire third year of medical school before returning to finish his training, and he later attended business school at Wharton. When his Castleman disease symptoms returned, doctors treated him with seven chemotherapy drugs that were never designed for his condition, a grueling regimen detailed by Science.org, which reports he left the hospital on June 2, 2012.
A Protein Discovery Led to His Own Remission
Fajgenbaum told the Tribune he lacked the resources to invent an entirely new drug, so he looked instead for something that already existed. By analyzing blood samples and datasets, he found increased mTOR signaling in his body, according to the Tribune's reporting. That discovery pointed him toward sirolimus, which inhibits mTOR, and he began remission after taking it.
He has said he has been in remission on sirolimus for over 10½ years, according to aamc.org. The science behind that recovery was later documented in a small study of three patients whose Castleman disease had not responded to standard IL-6 blocking treatments; sirolimus produced clinical benefit in all three, with remissions lasting 66, 19 and 19 months, according to findings published on jci.org. Before IL-6-targeting treatments existed, Castleman disease therapy had no real standard of care and often meant steroids, rituximab or multi-agent chemotherapy, the same research notes.
Turning a Personal Fight Into Every Cure
Fajgenbaum is an associate professor of medicine at the University of Pennsylvania. He is also founding director of its Center for Cytokine Storm Treatment & Laboratory, according to his Penn faculty profile. He also cofounded the Castleman Disease Collaborative Network, which works from the premise that drugs already approved for one condition might also treat rare diseases with little or no treatment available, according to the Castleman Disease Collaborative Network. The Tribune also reports he started a support network called Students of AMF — Actively Moving Forward — after his mother was diagnosed with glioblastoma in 2004, while he was a sophomore at Georgetown.
That personal history runs through Every Cure, the nonprofit Fajgenbaum founded to use artificial intelligence to match existing drugs against diseases that currently have none, per the Tribune. The group could not raise money in its first year and now uses AI to scan roughly 4,000 existing drugs, as reported by CNBC. Fajgenbaum has said pharmaceutical companies typically do not chase new uses for drugs already on the market because there is little financial incentive to do so, according to the Tribune, leaving that gap for Every Cure's AI matching to fill.
Thousands of Diseases, Still Waiting
The scale of the problem is stark: about 20,000 diseases and 3,000 approved drugs, with many diseases lacking an approved therapy. Every Cure has identified 10 active programs in its drug-repurposing portfolio since the nonprofit's founding in late 2022, CNBC reports. Its portfolio lists an active program studying the drug DFMO, also known as eflornithine, for a condition called Bachmann-Bupp syndrome.
Every Cure has backing to match its ambitions: it secured a $60 million commitment from TED's Audacious Project and has access to more than $130 million across two rounds of ARPA-H funding, per CNBC's reporting. The same report notes that AI has slashed the time needed to generate the group's drug-disease match list from 100 days down to about 17 hours. The teams' repurposed drugs have saved thousands of lives.
Old Drugs, New Jobs
Viagra is one example of drug repurposing, per the Tribune.
Fajgenbaum, who has written about his ordeal in his book Chasing My Cure, has two children. For a man who once lacked the resources to invent a new drug, the pitch he now makes to rooms like the one in Westville is simpler: the cures may already be sitting on pharmacy shelves, waiting to be matched to the right disease.
