Rapamycin (sirolimus): the monograph

Rapamycin, sold as sirolimus (brand name Rapamune), is an FDA-approved oral immunosuppressant: approved in 1999 for kidney-transplant rejection prophylaxis, and since 2015 for the rare lung disease lymphangioleiomyomatosis (LAM) ^3,1. It is also, separately, the drug with the strongest animal longevity data in pharmacology: the NIA's Interventions Testing Program extended mouse lifespan with it in both sexes, at three independent sites, even when started late in life ¹².

Those two facts power an off-label prescribing economy in longevity medicine, and this page keeps them in separate columns. The mouse data get full credit at mouse grade. The human record gets read as it actually is: transplant and LAM trials that prove the drug works for its approved uses, immune-aging trials that mostly tested a different rapalog (everolimus), and one completed longevity-population trial (PEARL) that missed its primary endpoint ^21,28. No human trial has tested whether rapamycin extends human life ^28,23.

Sourced fact chips render from the facts array: class, FDA status, longevity status, half-life, boxed warning, monitoring, key interaction, approved doses, best animal result, and the compound-labeling caveat on the famous human trials.

Yes, with a precise boundary. Rapamune oral solution was approved September 15, 1999 under NDA 021083; tablets under NDA 021110; generic sirolimus tablets and solution followed ^3,4. The label indication is prophylaxis of organ rejection in renal transplant patients aged 13 or older, plus treatment of LAM, added by an efficacy supplement approved May 28, 2015 ^1,3.

Longevity is not on that list. Taking sirolimus to slow aging is off-label prescribing of a real drug, not a supplement purchase, and every legitimate route to it runs through a prescriber and a licensed pharmacy ¹. Other sirolimus products exist for other uses: HYFTOR, a topical gel for facial angiofibroma in tuberous sclerosis; FYARRO, an IV protein-bound sirolimus for a rare sarcoma; and SCOMARA, a topical cream ^4,5.

In its approved lane, sirolimus earns its keep with hard numbers. In the two pivotal renal-transplant trials it roughly halved biopsy-confirmed acute rejection: 16.9 to 12.0 percent versus 29.8 percent against azathioprine, and 24.7 to 19.2 percent versus 41.5 percent against placebo ^24,25. In the MILES trial, it stopped the lung-function decline of LAM while patients stayed on it: FEV1 slope +1 mL per month on drug versus -12 mL per month on placebo, with the decline resuming after discontinuation ²⁷.

The off-label longevity lane borrows the same molecule at lower, usually weekly doses, on the theory that mTOR inhibition will do for humans what it does for mice. That theory is coherent, actively researched, and unproven in any human outcome trial ^28,31.

The animal side deserves its reputation. The NIA Interventions Testing Program, the most rigorous drug-testing framework in aging biology, fed rapamycin to genetically heterogeneous mice starting at 600 days old (late life) and extended lifespan in both sexes at all three independent sites: +14 percent for females and +9 percent for males at the 90th-percentile age ¹². Started at 9 months, median survival rose 10 percent in males and 18 percent in females ¹³. At a threefold higher dose, +23 and +26 percent ¹⁴. A 3-month transient course in middle-aged mice raised remaining life expectancy up to 60 percent ¹⁵. The same program found resveratrol, simvastatin, fish oil, and metformin alone did nothing ^13,16. TOR inhibition also extends lifespan in yeast, worms, and flies ¹⁰. In 24 companion dogs, a 10-week course produced no clinical side effects and echocardiographic improvements ¹⁹. This is the best preclinical longevity dossier any drug owns, and this table grades every row of it at its species ³¹.

The human side is a different table. Zero completed randomized trials, of sirolimus or any rapalog, have tested lifespan, mortality, or disease incidence as a primary endpoint in a healthy aging population ^28,23. The mouse rows in our study table say mouse. The human rows say what was actually measured: vaccine response, infection rates, DXA scans, safety labs. Where the marketing says 'rapamycin is proven to reverse aging,' the graded table is the reply.

Read the compound_studied column before the headline. The famous elderly immune trials, the ones quoted as proof that 'rapamycin' improves human aging, tested other rapalogs: the 2014 trial that improved flu-vaccine response by about 20 percent used everolimus (RAD001) ²¹; the 2018 trial that cut reported infections used dactolisib plus everolimus ²²; and when RTB101 (dactolisib) advanced to a 1,024-person phase 3, it failed its primary endpoint outright, 26 percent symptomatic respiratory illness on drug versus 25 percent on placebo ²³. Everolimus is a real, approved drug (Afinitor for oncology, Zortress for transplant) with a 30-hour half-life; it is not sirolimus ^6,7.

Sirolimus itself has been tested in aging populations exactly twice in completed randomized trials. Kraig 2018: 25 adults aged 70 to 95, 1 mg daily for 8 weeks, a safety pilot that changed no efficacy measure and flagged mild red-cell decrements ²⁰. PEARL 2025: 129 enrolled adults, compounded 5 or 10 mg weekly for 48 weeks, run by the longevity telehealth company AgelessRx; its primary endpoint (visceral fat by DXA) did not move (p=0.942), while secondary analyses showed lean-mass and pain improvements in women on 10 mg and self-reported well-being gains at 5 mg ^28,32. The strongest human data for sirolimus remain its approved-use trials: transplant rejection, LAM lung function, and a skin-cancer-prevention trial in transplant patients that cut new squamous-cell carcinoma (RR 0.56) at the cost of quadrupled serious adverse events ^24,27,26.

The claim behind longevity dosing is specific: once-weekly low doses inhibit mTORC1 (the longevity target) while sparing mTORC2 (the complex whose disruption drives glucose problems). The mechanistic core is real mouse science: chronic rapamycin impaired glucose tolerance via mTORC2 loss, and intermittent schedules reduced the metabolic and immune side effects in mice ^11,18.

In humans, the schedule has safety data, not outcome data. PEARL ran 5 and 10 mg compounded weekly for 48 weeks with adverse events similar to placebo ²⁸. A survey of 333 off-label users, mostly on weekly regimens, reported initial safety signals through self-report ²⁹. Real-world blood-level work found compounded product delivers about one third the exposure of commercial per milligram, which means many weekly users absorb less drug than their milligram number implies ³⁰. What no source shows: that weekly dosing extends human healthspan or lifespan. The dosing folklore is ahead of its evidence, and the studied-doses table below keeps the two visibly separate ²⁸.

Sirolimus binds the intracellular protein FKBP-12; the complex inhibits mTOR, a kinase that integrates nutrient and growth signals, which suppresses T-lymphocyte proliferation (the immunosuppressant effect) and shifts cells toward autophagy and reduced protein synthesis (the aging-biology interest) ^1,9. The TOR genes were discovered in yeast in 1991 through rapamycin-resistance screens; the molecule itself was isolated in 1975 from Streptomyces hygroscopicus in an Easter Island soil sample and named for the island, Rapa Nui ^9,8.

The two-complex detail matters more than most mechanism trivia: mTORC1 inhibition tracks with the longevity effects in animals, while disruption of mTORC2 under chronic dosing tracks with insulin resistance and glucose intolerance ¹¹. In flies, the lifespan effect runs through TORC1, autophagy, and translation ¹⁰. Every dosing debate in the longevity community is ultimately an argument about how to hit complex 1 without complex 2, and no human outcome trial has yet tested whether any schedule achieves that ^18,28.

From the label, at transplant doses: the most common adverse reactions include peripheral edema, hypertriglyceridemia, hypertension, hypercholesterolemia, increased creatinine, constipation, abdominal pain, diarrhea, headache, fever, urinary tract infection, anemia, nausea, and arthralgia ¹. The warning sections add hyperlipidemia requiring treatment, impaired wound healing and fluid accumulation, interstitial lung disease and non-infectious pneumonitis, proteinuria, hyperglycemia and new-onset diabetes, skin-cancer risk with a sun-protection instruction, hypersensitivity including angioedema, and azoospermia or oligospermia ¹. In the LAM study population (2 mg daily, no cyclosporine), the most frequent reactions were stomatitis, diarrhea, abdominal pain, nausea, nasopharyngitis, acne, chest pain, peripheral edema, upper respiratory infection, headache, dizziness, myalgia, and hypercholesterolemia ¹.

At longevity doses, the honest answer is thinner and milder, and it is measured, not proven absent. Kraig 2018 (1 mg daily, 8 weeks, ages 70 to 95) saw facial rash, stomatitis, and GI complaints in a handful of subjects, plus statistically significant but clinically insignificant red-cell decrements, and no glucose deterioration on formal testing ²⁰. PEARL (5 or 10 mg compounded weekly, 48 weeks) reported adverse and serious adverse events similar to placebo, with blood biomarkers staying in normal ranges ²⁸. Mouth ulcers, the signature mTOR-inhibitor complaint, appear in the LAM label list as stomatitis and in the off-label user survey ^1,29. The mouse literature adds a mechanism-level caution: chronic dosing impairs glucose tolerance via mTORC2, the exact effect intermittent dosing is meant to avoid ^11,18.

The label recommends therapeutic drug monitoring for all patients: whole-blood trough concentrations, checked at least 3 to 4 days after loading, after dose changes, in hepatic impairment, and around interacting drugs ¹. In LAM the target trough is 5 to 15 ng/mL ¹. Because sirolimus sequesters in red cells (blood-to-plasma ratio 36), trough measurement uses whole blood, and assay methods differ, so the same sample can read differently by immunoassay versus chromatography ¹.

Beyond drug levels, the label's warning sections imply the lab panel: lipids (hyperlipidemia requiring treatment), glucose (hyperglycemia and new-onset diabetes), urinary protein (proteinuria monitoring is recommended), and blood counts ¹. The half-life arithmetic sets the tempo: at 62 hours, steady state arrives in roughly 11 to 13 days, which is why level checks days after a change are meaningful and same-day checks are not ¹. Off-label prescribers who order no labs at all are running a protocol no studied population ran: even PEARL monitored blood biomarkers throughout ²⁸.

The label states sirolimus can cause fetal harm based on animal studies and its mechanism, was embryo/fetotoxic in rats at sub-therapeutic doses, and pregnant women should be advised of the risk; effective contraception is expected before, during, and after therapy per the label's specific-populations text ¹. On the male side, azoospermia or oligospermia may occur, listed in the label as a male-infertility warning ¹. For a drug taken by healthy adults on a wellness theory, reversible sperm suppression and fetal-harm potential are not fine print.

The numbers that explain the dosing rules: terminal half-life about 62 +/- 16 hours in stable renal transplant patients, so once-daily dosing stacks and even once-weekly dosing never fully clears ¹. Oral solution bioavailability is about 14 percent, tablets about 27 percent higher, and the two are not bioequivalent (clinically equivalent only at 2 mg) ¹. The drug rides in red cells (blood-to-plasma ratio 36 +/- 18), is about 92 percent protein bound, and is cleared through CYP3A4 and P-gp, the door every major interaction walks through ¹. In real-world longevity users, blood levels peaked about 2 days after a weekly dose, and compounded product delivered roughly a third of commercial exposure per milligram ³⁰.

Approved dosing is precise: renal transplant at low-to-moderate risk gets a 6 mg loading dose then 2 mg daily; high-risk gets up to 15 mg loading then 5 mg daily; LAM starts at 2 mg daily and titrates to a 5 to 15 ng/mL trough ¹. Studied longevity-adjacent dosing is thinner: 1 mg daily for 8 weeks (Kraig pilot) and compounded 5 or 10 mg weekly for 48 weeks (PEARL) ^20,28. Real-world users take commercial 2 to 8 mg or compounded 5 to 15 mg weekly ³⁰.

The mouse doses do not convert. The ITP dosed in parts per million of chow (4.7 to 42 ppm) with exposure verified by mouse blood levels; no validated method turns those into a human milligram number, and this site does not invent one ¹⁴. The comparison table below puts every studied dose next to the folklore so the gap is visible instead of implied ^1,28.

Oral solution: refrigerated at 2 C to 8 C (36 F to 46 F), protected from light, used within one month of opening; brief room-temperature storage is allowed per the label. Tablets store at controlled room temperature ^1,2. A slight haze can develop in refrigerated solution; the label describes handling for it. Compounded capsules follow the compounding pharmacy's labeling, one more way a compounded product is not the studied article ³⁰.

Rendered from limits_text; the honest boundary of this page.

Rendered from the faq array; answers are html-canonical with cite superscripts, and each carries a 320-character direct answer for the machine layer.

Numbered references render from citations[] in n order; the same array ships as the machine-readable citation manifest.

Comparisons render from comparisons[]: sirolimus vs everolimus (the compound the famous trials actually used), and studied doses vs folklore doses. Related fleet reading: metformin discourse is addressed in the FAQ with the ITP's own metformin rows.