Research tools

Intervention Compare.

Compare two to four interventions, keeping the claim, population and endpoint in view.

These are different research questions, not interchangeable treatments. Read across evidence dimensions rather than choosing an overall winner.

Intervention 1

Intervention 2

Intervention 3

Intervention 4

Default comparison: rapamycin and metformin.

Put the effects in context →

Claim-specific intervention comparison
Evidence question	Rapamycin	Metformin
Claim assessed	Extending human lifespan	Delaying multiple age-related outcomes in people without diabetes
Current classification	Frontier	Frontier
Potential impact	Possible ageing modification	Possible ageing modification
Human evidence	Mixed trials — Short studies; inconsistent functional benefits.	Small mechanistic trials — Clinical use in diabetes answers a different question.
Animal evidence	Lifespan demonstrated — Replicated mouse work; species and exposure matter.	Dose-dependent survival — Benefits are not uniform across exposures.
Mechanism	Well characterised — mTOR signalling connects growth and maintenance.	Multiple pathways — Metabolism and stress signalling are intertwined.
Safety evidence	Known risks — Clinical use is informative; longevity exposure remains uncertain.	Known clinical risks — Kidney function, B12 and tolerability matter.
Longevity evidence	No human proof — Neither PEARL nor RAPA-EX measured lifespan.	No human lifespan proof — Geroprotection remains a clinical hypothesis.
Primary mechanism	mTOR signalling connects growth and maintenance.	Metabolism and stress signalling are intertwined.
Research maturity	Mixed trials	Small mechanistic trials
Major limitation / verdict	Rapamycin extends mouse lifespan. Human trials have not demonstrated lifespan extension, and recent functional results are mixed.	Metformin’s ageing hypothesis is credible enough to test. Existing human molecular studies do not establish slower ageing or longer life.
Most important unresolved question	An independently replicated, adequately powered trial showing less disability, fewer age-related clinical events or durable improvement in function would materially strengthen the human case. A change in a methylation score alone would leave a larger inference to make. The most valuable next trial would measure target engagement, predefine its primary outcome, report withdrawals and harms, and examine whether benefits persist. Human lifespan extension remains a separate, unanswered test.	A sufficiently powered randomized trial in the intended population, with a prespecified clinical composite, would be materially more informative than another molecular signature. Components of the composite should also be reported separately: delaying one diagnosis is not equivalent to improving every aspect of ageing. Disability, treatment discontinuation and interactions with training belong in that assessment.