🧬 Aging Wiki

rapamycin

Properties1
aliasessirolimus, rapamune, AY-9944, AY9944, RAPA

14 min read

Rapamycin (Sirolimus)

A macrolide natural product isolated from the soil bacterium Streptomyces hygroscopicus, first discovered in a sample from Easter Island (Rapa Nui, hence “rapamycin”). Approved by the FDA in 1999 as sirolimus (INN) for prevention of organ-transplant rejection. The most extensively validated longevity intervention in mammals — extending lifespan in genetically heterogeneous mice across multiple independent sites 1. Works by forming a gain-of-function inhibitory complex with fkbp12, which then binds and allosterically inhibits mtor complex 1 (mTORC1). Aging-focused dosing regimens are substantially lower than transplant doses; intermittent schedules are increasingly favored to limit mTORC2 suppression and its metabolic side effects.

Identity

  • PubChem CID: 5284616
  • ChEMBL ID: CHEMBL413
  • DrugBank ID: DB00877
  • InChIKey: QFJCIRLUMZQUOT-HPLJOQBZSA-N
  • CAS: 53123-88-9
  • Molecular formula: C51H79NO13
  • Molecular weight: 914.2 g/mol
  • Class: macrolide lactone; mTOR inhibitor
  • Origin: natural product (Streptomyces hygroscopicus, Easter Island soil, 1972)
  • FDA approval: 1999 (renal transplant rejection prophylaxis; subsequent approvals: tuberous sclerosis complex [TSC], lymphangioleiomyomatosis [LAM], coronary stent coating)

Mechanism of action

Rapamycin does not bind mTOR directly. Instead 2:

  1. Rapamycin enters the cell and binds fkbp12 (FK506-binding protein 12 kDa) with high affinity.
  2. The rapamycin-FKBP12 complex then docks onto the FKBP12-rapamycin binding (FRB) domain of mTOR, at a site adjacent to — but not within — the catalytic cleft.
  3. This allosteric interaction destabilizes the mTORC1 complex (mTOR + Raptor + mLST8 + PRAS40 + Deptor), reducing its kinase activity toward downstream substrates S6K1 and 4E-BP1.
  4. The net effect is inhibition of cap-dependent translation, ribosome biogenesis, and lipid synthesis, and de-repression of autophagy.

mTORC2 sensitivity: mTORC2 (mTOR + Rictor + mSin1 + mLST8) is acutely resistant to rapamycin because the Rictor subunit occludes the FKBP12-rapamycin binding site 2. However, chronic or high-dose rapamycin can suppress mTORC2 activity — evidenced by reduced AKT S473 phosphorylation — in many cell types 3. This is supported by Arriola Apelo 2016, which showed that daily rapamycin in male C57BL/6J mice robustly suppressed AKT S473 (to ~40% of vehicle), while weekly dosing did not cause a statistically significant suppression at days 3 or 7 post-injection 4. This mechanism explains why daily rapamycin causes insulin resistance (mTORC2 → Akt S473 → insulin signaling) while intermittent dosing largely spares mTORC2 function and glucose homeostasis 4.

Rapalogs — everolimus (RAD001), temsirolimus (CCI-779), ridaforolimus — share the same FKBP12/FRB mechanism but differ in solubility, half-life, and prodrug status. See everolimus (implicit stub) for aging-relevant data.

Effects on aging hallmarks

HallmarkEffectEvidence
deregulated-nutrient-sensingPrimary target — direct mTORC1 inhibition suppresses anabolic sensing of amino acids and growth factors1, 2
disabled-macroautophagymTORC1 inhibition de-represses ULK1, initiating autophagosome formation2
cellular-senescenceReduces SASP in senescent cells by inhibiting mTOR-dependent translation of SASP factors; some evidence for reduced senescent cell accumulationneeds-replication
chronic-inflammationReduced mTOR→NF-κB signaling lowers cytokine output; in Mannick 2014, everolimus reduced PD-1+ exhausted T cells and improved vaccine response in elderly5
stem-cell-exhaustionLow-dose rapamycin rejuvenates aged hematopoietic stem cells and intestinal stem cells in miceneeds-human-replication

Mammalian lifespan data

The Harrison 2009 ITP study is the foundational mammal lifespan experiment 1:

  • Design: genetically heterogeneous UM-HET3 mice (CB6F1 females × C3D2F1 males: BALB/cByJ × C57BL/6J F1 × C3H/HeJ × DBA/2J F1); feeding started at 600 days of age (~20 months, late middle age); total cohort n=1,960 (674 controls; 317–328 per treatment arm)
  • Dose: microencapsulated rapamycin in chow at 14 mg/kg food (= 2.24 mg/kg body weight/day)
  • Sites: three independent NIA ITP sites (The Jackson Laboratory, University of Michigan, University of Texas Health Science Center)
  • Result (mean lifespan, pooled): females +13%, males +9%. Result (90th-percentile / maximal lifespan): females +14%, males +9%; effect replicated at all three sites independently (P<0.0001 pooled log-rank)
  • Note: disease spectrum in treated vs control mice did not significantly differ (necropsy of 71 mice); lifespan benefit was not attributed solely to reduced cancer
DimensionStatus
Pathway conserved in humans?yes
Phenotype conserved in humans?in-progress
Replicated in humans?in-progress (PEARL trial)

Miller 2014 ITP dose-escalation extended this 6:

  • Three doses tested in chow: 4.7, 14, and 42 ppm, started at 9 months of age in UM-HET3 mice (n=156 males and n=136 females per rapamycin group; n=300 male controls, n=280 female controls, pooled across 3 sites)
  • Highest dose (42 ppm): median lifespan extended 23% (males) and 26% (females) — the largest rapamycin effect sizes yet observed in this model; significant at all three doses for females (P<0.0001), significant at 14 and 42 ppm for males
  • Effect is dose- and sex-dependent; females consistently show larger extension and higher blood rapamycin levels at the same chow dose
  • Rapamycin-treated mice showed elevated fasting glucose and altered insulin levels; these metabolic changes are distinct from those seen in dietary-restricted (DR) mice, and gene expression of xenobiotic-metabolizing enzymes also differs between rapamycin and DR, suggesting independent mechanisms

Cross-vertebrate meta-analysis (Ivimey-Cook 2025). A 2025 meta-analysis of 911 effect sizes from 167 papers across 8 vertebrate species found that rapamycin produces statistically significant vertebrate lifespan extension at a magnitude comparable to dietary restriction, while metformin does not 7. The cross-species replicability — rapamycin matches DR effect size in pooled vertebrate data — is the strongest non-mouse-ITP signal currently available. High heterogeneity and significant publication bias caveats are noted by the authors.

needs-human-replication — All mammalian lifespan data are from mouse models. No human lifespan endpoint trial exists or is in progress. Human trials target healthspan surrogates.

Human evidence

Mannick 2014 — immune function in elderly 5

  • Design: randomized, placebo-controlled; n=218 elderly volunteers (≥65 years)
  • Intervention: everolimus (RAD001) 0.5 mg/day, 5 mg/week, or 20 mg/week for 6 weeks prior to influenza vaccination
  • Key finding: all three dosing schedules improved influenza vaccine response by ~20% (higher titers) and reduced PD-1+ CD4/CD8 T cell frequency (a marker of immune senescence) — suggesting mTOR inhibition can partially reverse immunosenescence
  • Adverse effects: dose-dependent; the two lower doses were generally tolerable; mouth sores at higher doses
  • Caveat: everolimus is not identical to rapamycin; results likely generalize but should be confirmed with sirolimus

needs-replication — Single trial; not yet independently replicated with rapamycin itself.

PEARL trial (NCT04488601) — aging-focused RCT 8

  • Design: 48-week decentralized, double-blinded RCT; n=114 completers (40 received 5 mg/week compounded rapamycin, 35 received 10 mg/week, 39 placebo); 11 additional participants discontinued prior to completion and were excluded from analysis; 35.1% of participants were women overall
  • Intervention: compounded rapamycin 5 mg/week or 10 mg/week. Important caveat: compounded rapamycin used in this trial had approximately 1/3 the bioavailability of commercial formulations; effective doses were ~66% lower than labeled doses (equivalent to ~1.67 mg and ~3.33 mg of commercial sirolimus, respectively)
  • Primary endpoint (visceral adiposity, VAT): did not change significantly in either dose group (η²=0.001, p=0.942)
  • Key secondary findings: lean tissue mass (LTM) improved significantly for women in the 10 mg group at both 24 and 48 weeks (48w: η²=0.202, p=0.013); self-reported pain improved significantly for women in the 10 mg group (48w: p<0.001); emotional well-being improved for all genders in the 5 mg group (η²=0.108, p=0.023); general health improved for all genders in the 5 mg group (η²=0.166, p=0.004)
  • Safety: adverse events and serious adverse events were similar across all groups; GI symptoms more frequent in rapamycin users; one case of anemia in 5 mg group (resolved with transfusion)
  • Limitations: small n, compounded-rapamycin bioavailability confound, short duration (48 weeks), no lifespan endpoint, no validated biological-age composite; participants were health-conscious at baseline (lower BMI, healthier lifestyle), which may have limited detectable VAT changes; largely exploratory

RAPA-EX-01 (ACTRN12624000790549) — exercise-combination RCT 9

The first trial designed to test the rapamycin “cycling hypothesis” — that alternating mTORC1 inhibition (rest days) and activation (training days) could amplify exercise-induced adaptation while preserving autophagy.

  • Design: double-blind RCT, n=40 sedentary adults aged 65–85 (mean 72.2 yr; 47.5% female); 1:1 randomization
  • Intervention: sirolimus 6 mg orally once weekly for 13 weeks vs. matched placebo; both arms performed standardized home-based exercise (chair-stands + exercycle, 3×/week)
  • Primary endpoint (30-s chair-stand reps, ITT): adjusted Δ −2.13 reps (95% CI −4.61 to 0.34, p=0.089) — favors placebo, not significant at 0.05
  • Sensitivity analyses (prespecified) reach significance favoring placebo: complete-case (n=16/19) Δ −2.46 reps (p=0.045); per-protocol (n=15/16) Δ −3.44 reps (p=0.007, Cohen’s d −0.90)
  • All secondary functional outcomes (grip, 6MWD, SF-36 physical/mental) directionally favored placebo, none significant
  • CRP +4.26 mg/L in sirolimus arm (p=0.152), driven by two outliers (17 + 50 mg/L); excluding outliers reduces between-arm difference to <1 mg/L
  • Four epigenetic clocks (PCGrimAge, SystemsAge, OMICmAge, DunedinPACE) showed mixed, non-significant trends; PCGrimAge alone showed a directionally favorable trend (Δ −2.28 yr, p=0.098) but contradicted the functional-outcome signal
  • Safety: AE incidence 85%/85% but burden was higher in sirolimus arm (99 vs. 63 total events; 35% vs. 15% adjudicated drug-related). One SAE in sirolimus arm: community-acquired pneumonia after a single 6 mg dose in a 65-yr-old participant; required overnight hospitalization with IV antibiotics; resolved with study withdrawal. Authors note immunosuppressive contribution cannot be excluded
  • Authors’ interpretation of the null result: the rapamycin half-life (~62 h) means a 6 mg weekly dose does not produce the on-off pulse the cycling hypothesis requires — drug persists across the training stimulus, suppressing mTORC1 during exercise and blunting adaptation. The result does not falsify the cycling hypothesis in principle but rules out 6 mg/week as the right schedule for combined rapamycin-exercise interventions in older adults
TrialNCT/registrationPhasenDosePrimary outcomeStatus
PEARLNCT04488601Phase 2114 completers5 or 10 mg/week (compounded; ~1/3 bioavailability)Visceral adiposity — null (η²=0.001, p=0.942)Published 2025
RAPA-EX-01ACTRN12624000790549Phase 2 (exploratory)406 mg/week (commercial sirolimus)30-s chair-stand — null ITT, sensitivity analyses favor placebo (PP p=0.007)Published 2026
Mannick (everolimus)N/APhase 2a2180.5–20 mg (everolimus, various schedules)Influenza vaccine response — positive (~20% titer increase)Published 2014

Pattern note (R34, 2026-05-08): The two recent dedicated aging-rejuvenation rapamycin RCTs (PEARL on body composition, RAPA-EX-01 on functional capacity) both reported null/negative on their prespecified primary endpoints at 5–10 mg/week. The earlier positive trial (Mannick 2014) used everolimus, not sirolimus, with a different mechanism endpoint (immunosenescence reversal in vaccine response). The mouse ITP signal at 14–42 ppm continuous in chow is unaltered — but the human aging-rejuvenation thesis at the n=1 biohacker dose range now has two negative high-quality data points and one positive (everolimus, immune-aging only). contradictory-evidence — the discrepancy between mouse continuous-feeding lifespan extension and human weekly-pulse rejuvenation null results is unresolved.

long-term-unknown — No trial exceeds one year of follow-up in healthy aging adults. The safety profile at aging-relevant (low, intermittent) doses over 5–10 years is not established.

Dosing — transplant vs aging

ContextDoseScheduleGoal
Transplant (immunosuppression)2–5 mg/dayDaily, long-termmTORC1 + mTORC2 inhibition; immunosuppression
Aging (mouse ITP high dose, Miller 2014)42 ppm in chow (approx. 6–7 mg/kg body weight/day)Daily in chowmaximal lifespan extension
Aging (PEARL trial)5–10 mgOnce weeklymTORC1 inhibition; mTORC2 sparing
Aging (common n=1 biohacker)2–6 mgOnce weeklymTORC1 inhibition; mTORC2 sparing

Rationale for intermittent dosing: In male C57BL/6J mice, weekly rapamycin (1×/7 days at 2 mg/kg s.c.) did not impair glucose tolerance, whereas daily dosing caused a 20–116% increase in blood glucose and a 71% increase in AUC over a glucose tolerance test. Weekly and every-5-day dosing also spared fasting glucose, insulin levels, and beta-cell function compared to daily dosing. mTORC1 (S6 S240/244 phosphorylation) was inhibited at day 1 post-injection in both weekly and daily groups; mTORC2 (AKT S473) was robustly suppressed only with daily dosing — weekly dosing showed a trend toward suppression on day 1 but this was not statistically significant at days 3 or 7. The “hit-and-run” rationale is that transient mTORC1 inhibition can extend lifespan without sustained mTORC2 suppression, thereby avoiding glucose intolerance and dyslipidemia 4. Note: this study used male C57BL/6J mice only; generalizability to heterogeneous genetic backgrounds and females has not been directly tested in this design.

dose-response-unclear — The optimal human dose and schedule for aging benefit (vs immunosuppression risk) has not been established in any adequately powered trial. Current clinical evidence is consistent with a favorable window at 5–10 mg/week but effect sizes on aging outcomes are small and endpoints heterogeneous.

Pharmacokinetics

  • Oral bioavailability: ~14% (highly variable; affected by food, CYP3A4 and P-gp status)
  • Half-life: ~62 h (range 46–78 h) in healthy adults
  • Protein binding: ~92% (primarily to erythrocytes and plasma proteins)
  • Metabolism: extensive hepatic CYP3A4/5; also P-gp substrate
  • Drug interactions: strong CYP3A4 inhibitors (ketoconazole, grapefruit) markedly elevate levels; inducers (rifampin) decrease levels
  • Renal excretion: <3% of dose excreted unchanged; primarily fecal

Rapalog family

CompoundINNKey difference from rapamycinPrimary approved use
EverolimusEverolimusHydroxyethyl group; shorter half-life (~28 h); better oral PKTSC, cancer, transplant
TemsirolimusTemsirolimusIV prodrug (converted to rapamycin in vivo)Renal cell carcinoma
RidaforolimusRidaforolimusPhosphonate-substituted; IV/oral; investigationalCancer (trials)

See everolimus for aging-specific everolimus data.

Adverse effects and safety

At transplant doses (daily, 2–5 mg/day), well-documented adverse effects include:

  • Impaired glucose tolerance / insulin resistance (mTORC2 suppression → reduced Akt S473 phosphorylation) 3
  • Dyslipidemia (elevated triglycerides, LDL)
  • Impaired wound healing
  • Mouth ulcers (mucositis)
  • Immunosuppression → increased infection risk
  • Interstitial pneumonitis (rare)
  • Potential impairment of testicular function / male fertility (high dose)

At aging-targeted intermittent doses (5–10 mg/week), PEARL trial found the profile tolerable with fewer metabolic side effects. Mouth sores were the most common complaint at higher doses (Mannick 2014). long-term-unknown — extended safety data in healthy aging adults is lacking.

Classification

  • mtor — pathway page with detailed mTORC1/mTORC2 biology and verified ITP lifespan numbers
  • autophagy — primary downstream process activated by mTOR inhibition
  • ampk — parallel energy-sensing pathway; partial functional overlap with mTOR inhibition
  • caloric-restriction — shares mTOR suppression as one mechanism; mechanistically and metabolically distinct from rapamycin 6
  • fisetin — senolytic with distinct mechanism; parallel geroprotector literature
  • fkbp12 — direct rapamycin-binding partner (implicit stub)
  • everolimus — rapalog; aging immune data from Mannick 2014 (implicit stub)
  • deregulated-nutrient-sensing — primary hallmark targeted

Limitations and gaps

  • Human evidence is surrogate-endpoint only — and the two recent dedicated aging-rejuvenation RCTs were null/negative on their primary endpoints. PEARL 2025 (n=114; visceral adiposity) and RAPA-EX-01 2026 (n=40; functional capacity) both used the n=1-biohacker-typical 5–10 mg/week intermittent regimen and both failed their primary endpoints. The mouse ITP continuous-feeding evidence (Harrison 2009, Miller 2014) remains robust, but the dose × schedule × outcome combination that produces human aging-rejuvenation benefit at sub-immunosuppressive exposure has not been demonstrated. #gap/contradictory-evidence between mouse continuous-dose lifespan extension and human weekly-pulse rejuvenation null results.
  • Dose-response in humans is poorly characterized. The optimal weekly dose, treatment schedule, and age-at-initiation for aging benefit remain unknown. dose-response-unclear
  • Sex dimorphism unexplained. Females consistently show larger lifespan extension than males in mouse ITP studies. The mechanism (sex-dependent pharmacokinetics, hormonal interactions) is unknown. no-mechanism
  • mTORC2 suppression threshold uncertain. The dose/duration boundary at which intermittent rapamycin begins to suppress mTORC2 in human tissues has not been measured. dose-response-unclear
  • Long-term immunosuppression risk at low intermittent doses has not been quantified in healthy aging cohorts followed for years. long-term-unknown
  • Cancer risk: mTOR inhibition is antiproliferative (reducing some cancers) but also immunosuppressive (raising others). The net cancer-risk balance at aging doses is unresolved. contradictory-evidence

Footnotes

Footnotes

  1. harrison-2009-rapamycin-lifespan-itp · n=1,960 total (674 controls; 317–328 rapamycin/group) · in-vivo · p<0.0001 · model: UM-HET3 genetically heterogeneous mice, rapamycin 14 mg/kg food started at 600 days · doi:10.1038/nature08221 · locally available ↩ ↩2 ↩3

  2. doi:10.1038/nature11861 · review · Johnson SC, Rabinovitch PS, Kaeberlein M · Nature 2013 · not_oa — no local PDF ↩ ↩2 ↩3 ↩4

  3. lamming-2012-rapamycin-mtorc2-insulin · n=~10–20/group · in-vivo + in-vitro · model: mouse, MEFs · doi:10.1126/science.1215135 · download failed (not_oa Science paper) — no-fulltext-access for primary verification of mTORC2 claims attributed to this source ↩ ↩2

  4. arriola-apelo-2016-intermittent-rapamycin · n=9–11/group (GTT experiments); n=6/group (islet experiments) · in-vivo + ex-vivo · model: male C57BL/6J mice, multiple rapamycin dosing schedules · doi:10.1111/acel.12405 · locally available ↩ ↩2 ↩3

  5. mannick-2014-everolimus-immune-aging · n=218 · rct · p<0.05 · model: human elderly ≥65 yr · doi:10.1126/scitranslmed.3009892 · not_oa — no local PDF; no-fulltext-access for full verification ↩ ↩2

  6. miller-2014-rapamycin-dose-sex-itp · n=156 males / 136 females per rapamycin dose; 300 male / 280 female controls (pooled 3 sites) · in-vivo · p<0.0001 (42 ppm) · model: UM-HET3 mice, 3-dose rapamycin (4.7/14/42 ppm) started at 9 months · doi:10.1111/acel.12194 · locally available ↩ ↩2

  7. doi:10.1111/acel.70131 · Ivimey-Cook ER, Sultanova Z, Maklakov AA · Aging Cell 2025;24(9):e70131 · meta-analysis · 911 effect sizes from 167 papers across 8 vertebrate species · rapamycin produced statistically significant lifespan extension at a magnitude matching dietary restriction; metformin did NOT produce significant vertebrate lifespan extension across the pooled dataset · high heterogeneity and significant publication bias caveats noted · OA gold; PMC12419861 · verified-scope: PubMed efetch abstract only ↩

  8. pearl-2025-rapamycin-healthspan · n=114 completers (40/5 mg, 35/10 mg, 39/placebo) · rct · p<0.05 primary (VAT ns); LTM significant in women at 10 mg · model: healthy normative-aging adults, 48 weeks · doi:10.18632/aging.206235 · locally available ↩

  9. stanfield-2026-rapa-ex-01 · n=40 (20/20) · rct (double-blind, placebo-controlled) · primary ITT chair-stand Δ −2.13 reps (95% CI −4.61 to 0.34, p=0.089); per-protocol Δ −3.44 reps (p=0.007, favors placebo) · model: sedentary adults 65–85 yr, sirolimus 6 mg/week × 13 weeks combined with home exercise · doi:10.1002/jcsm.70274 · PMC13082878 (OA) · ACTRN12624000790549 ↩

Graph View

Backlinks