They're often mentioned together as geroprotector candidates, but they work differently and have very different evidence behind them in this Atlas. Rapamycin directly and potently inhibits mTORC1 (its designed mechanism) and has 38 evidence-graded studies here, including a Tier A systematic review. Metformin inhibits mTORC1 only indirectly — mainly by activating AMPK, though it also acts on mTORC1 through AMPK-independent routes — and has just 3 studies in the Atlas, none of them a randomized trial for a longevity endpoint.
Rapamycin binds FKBP12 and directly blocks mTORC1's Raptor-dependent activity — a precise, well-characterized mechanism. Metformin's route to mTORC1 is murkier and is itself an open scientific question: it activates AMPK, an energy-sensor that inhibits mTORC1 upstream, but studies in the Atlas show metformin still suppresses hepatic gluconeogenesis in AMPK-null and LKB1-null mouse liver, and still inhibits mTORC1 in AMPK-null cells via the Rag GTPases — meaning at least part of its action bypasses AMPK entirely.
Metformin has decades of safety data as an FDA-approved diabetes drug, is inexpensive, and is already prescribed off-label by some longevity-focused physicians — its safety profile is a real practical advantage. But the Atlas's evidence for metformin specifically extending healthy lifespan rests on one retrospective observational study, not a randomized trial. Rapamycin's mechanism is more direct and its dose-response in animal models better characterized, but comes with the mTORC2-linked side-effect profile covered on the side-effects answer page.
