Using Existing Drugs in New Ways to Treat & Cure Diseases of Brain & Body | Dr. David Fajgenbaum

Summary

This Huberman Lab interview follows David Fajgenbaum from repeated near-death episodes with Castleman disease to research on finding new uses for approved medicines. It presents drug repurposing as an evidence-integration problem: biomedical knowledge, patient observations, computation, laboratory work, clinical trials, expert review, and physician-patient judgment must converge before a plausible drug-disease match becomes care.

The episode also introduces Every Cure, a nonprofit attempting to rank approved drugs against diseases at scale. Its strongest systems claim is that generic-drug economics, fragmented knowledge, rare-disease coordination gaps, and slow diffusion can leave promising uses untested or unused; its central qualification is that case reports, mechanisms, or model rankings are hypotheses rather than treatment recommendations.

Key Claims

• Fajgenbaum reports identifying mTOR activation in his own samples and beginning physician-prescribed sirolimus, a transplant-rejection drug, after repeated relapses of Castleman disease; he says the treatment produced a remission lasting more than eleven years at recording.
• Approved drugs can affect multiple targets and may help diseases beyond their original indication, but indication-specific approval does not establish safety or efficacy for a new use.
• Weak commercial incentives are especially important when a drug is generic, a disease is rare, or neither party can readily capture the value of a new indication.
• The examples span strong and weak evidence: randomized or large clinical studies, observational findings, small mechanistic papers, individual rescues, and emerging associations should not be treated as equivalent.
• Every Cure combines biomedical knowledge mapping and machine-learning prioritization with laboratory studies, clinical trials, observational data, and expert review.
• Patient agency can improve navigation through disease organizations and specialist networks, but it supplements rather than replaces clinical care and evidence review.
• Knowledge diffusion is itself a bottleneck: the episode describes potentially useful treatment observations taking years to reach other clinicians and patients.
• Responsible repurposing requires attention to dose, formulation, route, population, interactions, adverse effects, mechanism, and outcome—not merely the fact that a drug is already approved.

Key Quotes

“unconditional love” — the words Fajgenbaum recalls from his mother after promising to pursue treatments for patients like her.

“hope, action, impact” — Fajgenbaum’s reinforcing circuit linking possibility, effort, and visible benefit.

Connections

• David Fajgenbaum - physician-scientist and patient whose Castleman disease experience anchors the episode.
• Every Cure - nonprofit building a systematic drug-repurposing pipeline.
• Drug Repurposing - central method and incentive problem.
• Castleman Disease - rare inflammatory disorder that motivated Fajgenbaum’s self-sampling and sirolimus hypothesis.
• Patient-Led Rare-Disease Infrastructure / 患者主导的罕见病基础设施 - disease organizations, expert networks, and patient participation that help fragmented knowledge travel.
• AI For Science - broader computational-science context for connecting dispersed biomedical evidence.
• AI Clinical Validation In Drug Discovery - validation boundary separating model prioritization from demonstrated patient benefit.
• Self-Experimentation - neighboring concept whose single-person evidence requires supervision and later validation.
• Food and Drug Administration - regulator underlying the distinction between an approved drug and a newly supported indication.
• Medical Risk Management - clinical boundary for off-label use, adverse effects, and physician-patient decisions.

Contradictions

• No settled contradiction with the existing wiki was found. The episode reinforces existing validation boundaries: computation and mechanistic plausibility can prioritize a hypothesis but do not replace laboratory, clinical, and safety evidence.
• The supplied document is a structured episode summary rather than a transcript, primary paper, prescribing label, or systematic review. Numerical claims—including drug and disease counts, lidocaine mortality reduction, angiosarcoma response, and remission duration—remain episode-attributed until checked against primary evidence.
• The document uses “DATA2,” while the recognized rare autoinflammatory vasculopathy is commonly written DADA2; the wiki preserves this as a likely transcription or summary error rather than creating a separate condition.
• The title spells the guest’s surname Fajgenbaum, while the body heading and prose sometimes use Fagenbaum; the canonical identity is David Fajgenbaum.
• Aspirin, lidocaine, thalidomide, pembrolizumab, DFMO, colchicine, sirolimus, TNF inhibitors, GLP-1 drugs, nicotine, lithium, and other compounds discussed have different evidence, indication, dose, and risk profiles. Their mention is not medical advice or support for self-directed use.