If you’ve ever played The New York Times game Connections, you know the feeling. You’re staring at a grid of words, knowing the solution is there, but unable to see how the pieces fit together. All you can do is work with the words in front of you. There are no extra clues, no new information coming. The only option is to shuffle, to look at the same information in a different arrangement until patterns begin to appear.
Nothing about the problem changes. Then something about how you see it does.
In 2014, a third-year medical student named David Fajgenbaum checked himself into the emergency room mid-exam. He felt off. By the time anyone understood why, he was in the ICU with multiple organ failure from a disease so rare it wasn’t taught in medical school: Castleman disease. The only approved drug didn’t work. A priest came to his bedside and read him his last rites. He was 25.
Fajgenbaum survived that relapse, and four more after it. As he recounted in a recent episode of NPR’s Radiolab, he understood that chemotherapy was keeping him alive without curing him, and that waiting for a new drug to be developed (a process that typically takes 10 to 15 years and billions of dollars) wasn’t an option he had. So he did something unusual. He started asking his doctors to save his blood samples, and he ran experiments on himself.
What he found was that a specific signaling pathway in his immune system, mTOR, was in overdrive. When he searched the existing pharmacological literature for something that could block it, he found an answer that had been sitting in pharmacies for 25 years. Sirolimus, a drug approved in 1999 to prevent organ transplant rejection, had never been used for Castleman disease. The biology of his disease hadn’t changed. The drug had always existed. The connection simply hadn’t been made.
He took it. It worked. He has been in remission for over a decade.
The detail worth holding onto isn’t the drug or the disease. It’s the instinct. Fajgenbaum didn’t wait for new knowledge to arrive. He looked differently at what already existed.
The Data Was Always There
That experience reshaped how Fajgenbaum thought about medicine. “This drug was out there,” he said later. “How many more are out there that could save other lives, if only people were looking?” Through his nonprofit Every Cure, he began applying computational approaches to systematically scan across thousands of existing drugs and diseases, looking for matches that the traditional drug discovery system, with its financial incentives, siloed research and massive resource requirements, was structurally unlikely to find on its own.
An initial pilot identified 106 promising drug repurposing opportunities across 147 diseases. The treatments were already approved. The patients were already out of options. The connections just hadn’t been made.
Fajgenbaum’s question — how many answers already exist that nobody has thought to look for — isn’t unique to drug repurposing. It runs through drug discovery more broadly. A researcher working on a difficult target faces a version of the same problem every time they ask whether a compound is actually doing what they think it’s doing inside a living cell. The answer might already be there. The question is whether they can see it.
The Same Limitation, Smaller
Drug discovery pipelines are full of compounds that were deprioritized or abandoned, not because they failed, but because researchers couldn’t confirm whether they were actually engaging their intended target inside a living cell. The compound existed. The target existed. What didn’t exist was a reliable way to see what was happening between them. Without that visibility, a candidate that might have worked looks indistinguishable from one that wouldn’t, and it gets set aside.
NanoBRET® Target Engagement technology was built around exactly that gap. Rather than inferring whether a compound has bound to a protein through indirect measurements, NanoBRET® detects the interaction directly in living cells. A compound that might otherwise be abandoned for lack of evidence can now be evaluated on its actual merits. Not because the compound changed, but because the visibility did.
What the Puzzle Looks Like Now
Fajgenbaum ended his Radiolab interview with a question nobody can yet fully answer: how many breakthroughs are still out there, waiting to be seen? For drug discovery researchers working on difficult targets, the NanoBRET® technology is a version of the same shift: a way to see what was always there.
That’s not a story about luck. It’s the same feeling as staring at a grid of words and suddenly seeing how they fit. The breakthrough was there all along.
References
Nasser, L. (Host). (2025, August 22). The medical matchmaking machine [Audio podcast episode]. In Radiolab. WNYC Studios. https://radiolab.org/podcast/the-medical-matchmaking-machine
Every Cure. (2022). Our story. https://everycure.org/our-story/
Fajgenbaum, D.C., et al. (2019). Identifying and targeting pathogenic PI3K/AKT/mTOR signaling in IL-6 blockade-refractory idiopathic multicentric Castleman disease. Journal of Clinical Investigation, 129(1), 423–436. https://doi.org/10.1172/JCI122836
Elise Johnson
Latest posts by Elise Johnson (see all)
- What Counts as Evidence - May 7, 2026
- The Filter You Didn’t Choose - April 30, 2026
- The Body Already Knows - April 23, 2026
