Geroscience’s boldest claim is that aging itself is modifiable — that one drug could postpone many diseases at once. Two molecules dominate that conversation. Here’s the state of the evidence, hype removed.
Rapamycin: the strongest animal signal
Rapamycin inhibits mTOR, a central nutrient-sensing pathway. In the NIH’s rigorous Interventions Testing Program it extended lifespan in mice — repeatedly, in both sexes, even started late in life. No other molecule’s animal data comes close.
In humans, we have hints, not answers: low-dose mTOR inhibition improved flu-vaccine response in older adults, and the small PEARL trial found weekly dosing safe over a year with modest signals. What we don’t have is a human outcome trial — nobody yet knows the right dose, schedule, or who benefits. Off-label use today is an informed bet, with real unknowns: lipid shifts, glucose effects, immune trade-offs.
Metformin: cheap, safe, and contested
Metformin’s longevity fame began with observational data suggesting diabetics on it outlived expectations. Newer, better-controlled analyses have largely shrunk that effect. It remains an excellent, safe diabetes drug — but for healthy people, trials suggest it may blunt some benefits of exercise, which is a strange trade for a longevity strategy. The TAME trial, if it runs, is designed to settle the question properly.
Neither drug is part of a default LUMA protocol. The honest ranking hasn’t changed: training, sleep, ApoB and blood-pressure control, and not smoking are worth more than every gero-drug combined — today. We’ll update this page the moment the evidence updates us.
How to follow the field
Watch for human randomized trials with clinical endpoints — not biomarker-only studies, not mouse headlines. When those land, this article changes. That’s the deal science offers: strong opinions, loosely held, re-tested on schedule.
Key references: ITP rapamycin lifespan studies (Nature 2009 onward); PEARL trial 2023; TAME trial design; UK Biobank metformin analyses. Concept brand — not medical advice; neither drug is part of a default LUMA protocol.
