664. Are Thousands of Medical Cures Hiding in Plain Sight?

Episode Summary

Executive Summary: The episode argues that many existing drugs can be repurposed for new diseases at far lower cost and speed than developing new medicines. It follows David Fegenbaum’s life-saving Castleman disease breakthrough, explains EveryCure’s AI-driven approach to match drugs with diseases, and explores why weak market incentives—not scientific uncertainty—keep repurposed drugs underused.

Main Topics: Balamuthia and the promise of repurposed drugs (Priority: 5/5): The episode opens with the rare, often fatal amoebic brain infection balamuthia and a case where nitroxoline, an old European UTI drug, produced a remarkable survival after emergency FDA access. This sets up the core question: how many other existing drugs could save lives if matched to the right disease? David Fegenbaum’s personal path into drug repurposing (Priority: 5/5): Fegenbaum’s mother’s brain cancer pushed him toward medicine, and his own near-fatal Castleman disease led him to search for an existing drug that could treat a disease with no approved therapies. His self-directed investigation became the foundation for EveryCure. Discovering mTOR, rapamycin, and sirolimus for Castleman disease (Priority: 5/5): By analyzing his blood and stored lymph node tissue, Fegenbaum identified mTOR pathway overactivation and linked it to sirolimus/rapamycin, an existing transplant drug. The story shows how a novel biological insight can unlock a repurposed therapy. EveryCure’s AI platform for matching drugs to diseases (Priority: 5/5): EveryCure is building a large-scale system using biomedical knowledge graphs and AI to score all drug-disease pairs, looking for hidden matches across roughly 4,000 drugs and 18,000 diseases. Humans then validate the top-ranked possibilities. Why economics blocks generic drug repurposing (Priority: 5/5): The discussion emphasizes that generic drugs lack a business model for expensive trials because there is little or no patent protection or profit incentive. The science may be cheaper and faster, but the market fails to fund the work. Policy solutions: market-shaping and pull funding (Priority: 4/5): Economist Chris Snyder explains advanced market commitments and a proposed U.S. pull-funding model that would reward successful repurposing based on measured impact, potentially by agencies like Medicare or Medicaid. Adoption, CureID, and the long lag from evidence to practice (Priority: 4/5): Heather Stone discusses CureID, a registry for sharing repurposed-drug outcomes, and the broader problem that even effective repurposed therapies can take many years to diffuse into practice without commercial promotion.

Key Arguments: Repurposed drugs can produce major benefits because their safety, pharmacology, and manufacturing already exist, reducing risk and development cost dramatically. Fegenbaum’s Castleman story demonstrates that one person, armed with data and persistence, can identify a life-saving repurposed treatment. AI can scan the huge space of drug-disease combinations and surface likely matches that humans would never evaluate manually. Generic drugs are especially neglected because there is no scalable way to capture profits from new uses once patents expire. Market-shaping incentives can substitute for traditional pharmaceutical profit motives by paying for demonstrated public-health impact. Public sharing systems like CureID can help generate hypotheses and accelerate adoption, but they cannot replace rigorous trials. Repurposing research is underfunded relative to its potential societal value, despite costing roughly 1% as much as de novo drug development.

Data Points: Balamuthia fatality rate: about 90% - Used to illustrate how lethal and under-studied the rare amoebic infection is before nitroxoline treatment. FDA-approved treatments for all diseases: about one quarter of 18,000 known human diseases - Heather Stone cites the limited share of diseases with approved therapies. Drug development cost: $1 billion to $2 billion - Fegenbaum explains why developing a new drug from scratch was unrealistic. Drug development timeline: 10 to 15 years - Used to show why repurposing was the only feasible path for his rapidly fatal illness. Authors contacted by Fegenbaum: 10,000 to 15,000 emails - He emailed nearly everyone associated with papers mentioning Castleman disease to build a research network. Experts attending first Castleman meeting: 27 - Researchers who showed up to the 2012 American Society of Hematology meeting Fegenbaum organized. Patients treated with seven chemotherapies: 7 - Doctors used a broad, last-resort regimen to control Fegenbaum’s flares and Castleman disease. Patients surviving after sirolimus in initial series: 4 of 4 - Fegenbaum and three other patients responded strongly before a fifth patient did not. Observed effectiveness of sirolimus: 20% to 25% of patients - Later estimate of how often the treatment works for Castleman disease. Number of drugs found for off-label diseases: 14 - Fegenbaum’s lab identified 14 repurposed drugs across diseases. Lives saved with repurposed drugs: well over 1,000 - Total impact claimed by Fegenbaum’s lab through repurposing work. Drug-disease space analyzed by EveryCure: 4,000 drugs x 18,000 diseases - Scale of the AI-driven search for hidden therapeutic matches. Lidocaine breast cancer trial size: 1,600 patients - A randomized trial of peri-tumor lidocaine injection before surgery. Mortality reduction with lidocaine: 29% - Outcome from the breast cancer surgical trial cited as a repurposing example. CureID clinician-submitted cases: about 700 - Cases entered into the registry by clinicians. CureID patient/caregiver-submitted cases: about 650 - Cases entered directly by patients or caregivers. Cases extracted from literature: about 5,000 - Published case reports and studies added to CureID. CureID electronic medical record cases: 115,000 - COVID-19-specific cases extracted into the registry. EveryCure funding raised: over $100 million - Funding from federal government, foundations, and philanthropists to scale repurposing. Conditions EveryCure expects to treat: 15 to 25 in five years - Projected impact from current funding. Repurposing cost relative to new drug development: about 1% - Snyder and Fegenbaum argue repurposing is dramatically cheaper than novel drug discovery. Expected reward size in pull-funding proposal: roughly $1 billion per successful opportunity - Estimate for a government incentive program to encourage generic repurposing trials. Pneumococcal vaccine AMC fund: $1.5 billion - Example of an advanced market commitment used to stimulate vaccine development. Lives saved by second-generation pneumococcal vaccine: 700,000 children under five - Estimated impact credited to the AMC-supported vaccine rollout. Average adoption lag for medical advances: 17 years - Heather Stone notes the long delay before new treatments become standard practice. Repurposed-drug research share of medical R&D: fraction of 1% - Stone’s estimate of how little funding currently goes to drug repurposing.

Pivotal Quotes: "How many more things are there that we could potentially uncover to save lives today?" — Heather Stone: Introduces the episode’s central question after the balamuthia/nitroxoline example. "I realized that if I wanted any chance to survive, I would have to really turn my hope into action and start trying to find a drug that could save me." — David Fegenbaum: Explains the moment he shifted from patient to self-directed drug hunter. "We think this would cost roughly $1 billion per successful opportunity." — Saren Chethik: Describes the proposed pull-funding mechanism for generic drug repurposing.

Implications: Drug repurposing could unlock major health gains quickly if policy creates incentives to test and adopt old drugs for new uses. Without better funding and diffusion systems, many life-saving therapies will remain hidden and unused.