Last updated 2026-07-27
TL;DR
Both rapamycin (sirolimus) and acarbose extend median lifespan in the NIA Interventions Testing Program, roughly 9-26% for rapamycin and 8-11% for acarbose depending on sex and dose. Neither has a completed human longevity trial. Acarbose is FDA-approved for type 2 diabetes with a gentler side-effect profile (mostly GI); rapamycin is approved only as an immunosuppressant and carries real immune and metabolic risks off-label.
What are sirolimus and acarbose, and why are they compared for longevity?
Sirolimus (also called rapamycin, the same molecule under two names) is an mTOR inhibitor first approved by the FDA in 1999 to prevent organ transplant rejection [1]. Acarbose is an alpha-glucosidase inhibitor approved by the FDA in 1995 to treat type 2 diabetes by slowing carbohydrate digestion in the gut [2]. Neither drug was designed with aging in mind. Both ended up in the same conversation because both showed up as lifespan-extending hits in the same government-funded screening program. That program is the National Institute on Aging's Interventions Testing Program (ITP), which tests candidate compounds in genetically diverse mice across three labs (Jackson Laboratory, University of Michigan, and University of Texas Health San Science Center) [3]. Rapamycin and acarbose are two of the very small number of compounds that have replicated a lifespan effect across multiple cohorts and both sexes. That track record is why longevity researchers keep putting them side by side. It is not because anyone has shown either one extends human life. Nobody has run that trial for either drug. The comparison matters for a practical reason too: some people in the longevity space actually take both, on the theory that an mTOR inhibitor and a glucose-absorption blunter hit different mechanisms and might stack. That's a reasonable hypothesis. It is not backed by any combination trial in humans or, as far as the public ITP record shows, in mice tested as a stacked pair at standard doses.
How do rapamycin and acarbose compare in the NIA Interventions Testing Program mouse data?
| ITP median lifespan increase, males | ~9-23% depending on dose/start age [1] | ~11% [4] | |
|---|---|---|---|
| ITP median lifespan increase, females | ~13-26% depending on dose/start age [1] | Smaller in early cohorts, improved at higher doses in later cohorts [4] | |
| Effective starting age in mice | As late as 600 days (~human 60s) still effective [1] | Started around 4 months in most ITP cohorts [4] | |
| FDA approval status | Approved for transplant rejection (1999); Fyarro and Hyftor approved for specific rare conditions [1] | Approved for type 2 diabetes (1995) [2] | |
| Completed human longevity trial | None [5] | None | The honest read: rapamycin's ceiling effect is bigger in mice. Acarbose's effect is smaller but was achieved at a dose and formulation already sold as an approved diabetes drug for three decades, which is part of its appeal to some people who want to stay inside approved-drug pharmacology even while using it off-label for aging. |
Rapamycin has produced the largest and most consistent lifespan effects of anything the ITP has tested. In the original 2009 report, mice started on rapamycin late in life (about 600 days old, roughly human 60s equivalent) still saw median lifespan extended by 9% in males and 13% in females [1]. Later cohorts starting rapamycin earlier and at higher doses pushed the effect further: one analysis reported up to 23% (male) and 26% (female) increases in median lifespan at higher doses. Acarbose's effect is real but smaller. In ITP reporting, acarbose extended median lifespan by about 11% in males and a smaller amount in females, with more recent higher-dose cohorts showing improved effects, including data suggesting stronger benefit in females at higher doses than earlier cohorts implied [4]. The sex-asymmetry is a real pattern in the acarbose data and one of the open questions in the field: nobody has fully explained why males respond more consistently at standard doses. Here's a side-by-side of the headline mouse numbers: | Measure | Rapamycin (sirolimus) | Acarbose |
Is there any human lifespan data for either drug?
No. Neither rapamycin nor acarbose has a completed randomized controlled trial measuring human lifespan or all-cause mortality as a primary aging outcome. This is the single most important fact in this whole comparison and it applies equally to both drugs. What does exist for rapamycin: short human trials on surrogate markers. A 6-week low-dose rapamycin trial in older adults (average age 74) found it was "safe and slowed several immunological measures of aging," using an mTOR-inhibitor drug combination (RTB101 plus everolimus) and showing improved response to influenza vaccination, per the study published in Lancet in 2018 [6]. The PEARL trial, a small human study of low-dose rapamycin in adults, reported some quality-of-life and bone-density signals over roughly 48 weeks, but was not powered or designed to show a lifespan effect . These are surrogate-marker studies, not survival trials. For acarbose, the human evidence is almost entirely about diabetes and cardiovascular risk reduction, not lifespan. The STOP-NIDDM trial found acarbose reduced progression from impaired glucose tolerance to type 2 diabetes and reduced the risk of cardiovascular events in that population . That is a genuinely useful clinical result. It is not a longevity trial in healthy people, and it says nothing directly about whether acarbose extends lifespan in someone without prediabetes. So when you see acarbose or rapamycin discussed as "longevity drugs," the correct mental model is: strong, replicated mouse data plus zero completed human lifespan trials for either one. Anyone telling you otherwise is overselling animal data as if it were clinical proof.
How do the mechanisms of action differ?
Rapamycin inhibits mTOR (mechanistic target of rapamycin), a master nutrient-sensing pathway that regulates cell growth, protein synthesis, and autophagy. Chronic mTOR activity is thought to drive some of the metabolic and proliferative changes associated with aging; rapamycin dials that pathway down, which is also exactly why it works as an immunosuppressant in transplant patients: it blunts T-cell proliferation [1]. Acarbose works entirely differently and much more locally. It inhibits alpha-glucosidase enzymes in the small intestine, which slows the breakdown of complex carbohydrates into absorbable glucose. The practical effect is a flatter post-meal (postprandial) blood glucose and insulin spike [2]. It is not a systemic immune-modulating drug. Its longevity hypothesis rests more on blunting glycemic and insulin-signaling load over decades than on any direct action on core aging pathways like mTOR. This mechanistic difference is the whole reason people talk about combining them: an mTOR inhibitor and a glucose-absorption blunter are, in theory, non-overlapping levers. But "non-overlapping in theory" is a hypothesis, not evidence of combined benefit, and no published ITP cohort as of this writing has reported a rapamycin-plus-acarbose combination arm with lifespan data.
What are the real side effects and risks of each drug?
Rapamycin's risk profile is the more serious one, because it is an immunosuppressant at FDA-approved transplant doses and still has immune effects at the lower, intermittent doses used off-label for longevity. Mouth ulcers (stomatitis) are the most commonly reported side effect in people using low-dose, intermittent rapamycin protocols. Other documented effects at approved doses include increased infection risk, impaired wound healing, elevated blood lipids, and in some patients, new-onset high blood sugar. The FDA label for sirolimus carries warnings about increased susceptibility to infection and the possible development of lymphoma or other malignancies with chronic immunosuppression . That label reflects transplant-level continuous dosing, not the lower, intermittent regimens (commonly once weekly) that longevity users take, but the underlying pharmacology, mTOR inhibition and T-cell suppression, is the same mechanism at a smaller dose. Acarbose's side effects are almost entirely gastrointestinal: flatulence, abdominal pain, and diarrhea, caused by undigested carbohydrate fermenting in the colon. The FDA label for acarbose describes these GI effects as the most frequently reported adverse events in clinical trials, and notes rare cases of elevated liver enzymes with high doses, which is why liver function monitoring is recommended during the first year of use . Acarbose does not suppress the immune system and does not carry a malignancy warning. So the risk comparison is not close. Acarbose's downside is mostly digestive discomfort that people often adjust to over weeks, or that resolves by lowering the dose. Rapamycin's downside touches immune function, and that is a meaningfully different category of risk, one worth discussing with a prescriber before starting, especially for anyone with a history of frequent infections, upcoming surgery, or plans for vaccination.
How do dosing protocols compare?
Acarbose for diabetes is typically dosed at 25-100 mg taken with the first bite of each meal, up to three times a day, titrated up slowly to reduce GI side effects, per FDA labeling . People using it off-label for metabolic/longevity reasons often use lower or less frequent dosing than the diabetes label, sometimes just with the largest meal of the day, though there's no standardized off-label protocol validated by any trial. Rapamycin's off-label longevity dosing looks nothing like its FDA-approved transplant dosing. Transplant patients take it daily, continuously, often alongside other immunosuppressants, at doses calibrated to keep blood trough levels in a specific therapeutic range for graft survival. The intermittent protocols people use off-label instead take a single larger dose, commonly in the 5-10 mg range, once weekly or once every 1-2 weeks, on the theory (supported by pharmacokinetic and some mouse data) that pulsed dosing gets mTOR inhibition benefits while giving the immune system recovery time between doses. This is a meaningfully different exposure pattern than continuous transplant dosing, and it has not been validated by a completed human aging trial either. If you're mapping out an actual rapamycin protocol, the practical questions (how much, how often, how to prepare and inject if using an injectable form, and how long a cycle should run) are covered in more detail in our guides on Sirolimus Rx dosage, the Sirolimus Rx dosage calculator, and Sirolimus Rx cycle length. Acarbose dosing questions are better directed to a physician managing it as a diabetes or prediabetes medication, since that's the only indication with real trial support.
Can rapamycin and acarbose be combined?
People do combine them, reasoning that mTOR inhibition and glucose-absorption blunting are different enough mechanisms to be complementary rather than redundant. That reasoning is not crazy on its face. It is also not supported by any published combination trial in humans, and the public ITP dataset does not include a rapamycin-plus-acarbose arm reporting a combined lifespan number as of this writing. The honest position: if you're already taking rapamycin under medical supervision and you separately have a metabolic reason to be on acarbose (prediabetes, elevated postprandial glucose), there's no known pharmacokinetic interaction that should block combining them, but you should tell your prescriber you're on both so blood work and side effect monitoring account for both drugs. If your only reason for adding acarbose is "the mouse data says it stacks," recognize that's an inference from two separate single-drug datasets, not a tested combination. This is also a place where a provider-reviewed source matters more than internet dosing charts. Sirolimus Rx exists to connect people who want to pursue rapamycin under medical supervision with providers who can review labs, medical history, and drug interactions, and prescriptions are filled through a licensed fulfilling pharmacy partner, not compounded or manufactured by Sirolimus Rx itself.
Which one has stronger evidence behind it right now?
For mouse lifespan extension, rapamycin has the stronger, larger, and more replicated effect. It has been retested across more ITP cohorts, at more doses, and starting at more different ages than acarbose, and its effect size ceiling (up to the 23-26% range at higher doses in later cohorts) is larger than acarbose's [4]. For human clinical evidence, acarbose is ahead, but on a different outcome. It has an actual completed outcome trial in humans, STOP-NIDDM, showing reduced diabetes progression and reduced cardiovascular events in people with impaired glucose tolerance . Rapamycin's human data is limited to small, short trials on immune biomarkers (the 2018 Lancet mTOR-inhibitor trial [6]) and quality-of-life measures (PEARL ), not hard outcomes like mortality or cardiovascular events. So "stronger evidence" depends entirely on what you're asking. If the question is "which drug's animal data for lifespan extension is more convincing," it's rapamycin. If the question is "which drug has actually been shown, in a real human outcome trial, to reduce a real health risk," it's acarbose, just not for the lifespan-extension use case, for diabetes prevention and cardiovascular risk in people with impaired glucose tolerance. Neither answer gets you to "proven to extend human lifespan." That trial doesn't exist for either drug.
What does it cost to use each one?
Acarbose is a decades-old generic. Cash prices for a month's supply of generic acarbage at standard diabetes doses commonly run in the range of $15-40 at U.S. retail pharmacies without insurance, though prices vary by pharmacy and region; check GoodRx or a local pharmacy for current pricing since generic drug cash prices shift often. Rapamycin/sirolimus is also generic and has been for years, but off-label longevity use typically requires a telehealth or in-person prescriber visit (since no doctor is required to prescribe it for diabetes the way acarbose commonly is), plus the drug cost itself. Generic oral sirolimus tablets can run anywhere from roughly $20 to over $200 a month depending on dose, pharmacy, and whether GoodRx-type discounts are used; injectable or compounded formulations used in some off-label protocols carry their own separate cost structure. There is no single "longevity price" for rapamycin because dosing protocols (weekly vs. biweekly, 5mg vs. 10mg) vary so much between prescribers. The bigger cost difference isn't the pill, it's the medical oversight. Acarbose for diabetes is something most primary care doctors will prescribe without much friction. Off-label rapamycin for longevity generally requires finding a prescriber willing to work outside FDA-labeled use, which is exactly the gap that provider-reviewed telehealth services for rapamycin are built to close.
What should someone actually do with this information?
If you have prediabetes or impaired glucose tolerance, acarbose is an approved, well-studied option with real outcome data behind its primary indication (STOP-NIDDM ), and any interest in it for longevity is essentially free-riding on a drug that already has a legitimate reason to exist in your medicine cabinet. Talk to your doctor about it as a diabetes-prevention drug first; the longevity angle is a bonus hypothesis, not the reason to start it. If you're interested in rapamycin for longevity, go in clear-eyed: this is off-label use of an immunosuppressant, the human trial data is limited to short surrogate-marker studies [6] , and the strongest evidence you're relying on comes from mice [1]. That doesn't mean don't do it. It means don't do it without a prescriber who will order baseline labs (lipids, kidney function, blood counts), watch for mouth ulcers and infection risk, and adjust dose and frequency based on your actual response, not a forum protocol. If you want the practical mechanics, start with Sirolimus Rx dosage and, if you're using an injectable formulation, how to reconstitute Sirolimus Rx, Sirolimus Rx how to inject, and Sirolimus Rx injection sites. Either way, the honest summary to carry forward is short: strong mouse data, no human lifespan trial, real and different risk profiles, and a decision that should involve a doctor who knows your labs, not a spreadsheet from a longevity forum.
Frequently asked questions
Is rapamycin the same as sirolimus?
Yes. Sirolimus is the generic drug name; rapamycin is the original name derived from Rapa Nui (Easter Island), where the soil bacterium producing it was discovered. They are the same molecule, and the terms are used interchangeably in both clinical and longevity literature [1].
Has acarbose been shown to extend human lifespan?
No. Acarbose's human evidence comes from the STOP-NIDDM trial, which showed reduced progression to type 2 diabetes and reduced cardiovascular events in people with impaired glucose tolerance [10]. That is a real clinical benefit, but it is not a lifespan trial, and no completed human study has measured acarbose's effect on all-cause mortality or lifespan directly.
Why do longevity researchers compare rapamycin and acarbose specifically?
Both are among the small number of compounds that have shown replicated, statistically significant lifespan extension in the NIA Interventions Testing Program across multiple mouse cohorts and both sexes [3][4][6]. That shared track record in the same standardized mouse-testing pipeline is why they get compared, not any shared mechanism or human trial.
Can I take rapamycin and acarbose together?
Some people do, reasoning the mechanisms (mTOR inhibition vs. glucose absorption blunting) are complementary, but no published trial has tested the combination for lifespan or safety. If combining, tell your prescriber about both drugs so labs and side-effect monitoring cover each one; there's no known dangerous interaction, but there's also no data showing added benefit.
What are the main side effects of low-dose rapamycin?
The most commonly reported side effect in off-label intermittent-dosing users is mouth ulcers (stomatitis). Other risks tied to rapamycin's immunosuppressant mechanism include increased infection susceptibility, impaired wound healing, and elevated blood lipids; the FDA label also warns about malignancy risk with chronic immunosuppression at approved transplant doses [11].
What are the main side effects of acarbose?
Almost entirely gastrointestinal: flatulence, abdominal discomfort, and diarrhea, caused by undigested carbohydrates fermenting in the colon. The FDA label lists these as the most common adverse events in clinical trials, and recommends liver enzyme monitoring during the first year at higher doses due to rare elevated liver enzyme cases [12].
Is rapamycin FDA-approved for longevity?
No. Sirolimus is FDA-approved only for preventing organ transplant rejection, and related formulations (Fyarro, Hyftor) are approved for specific rare conditions like malignant PEComa and tuberous sclerosis skin lesions [1]. Any use for aging or lifespan extension is entirely off-label and not FDA-reviewed for that purpose.
How much does rapamycin lifespan extension research show in mice?
In NIA Interventions Testing Program cohorts, rapamycin extended median lifespan by roughly 9-13% in mice started on the drug late in life (around 600 days old) [4], with later cohorts using higher doses and earlier starts showing effects up to roughly 23% in males and 26% in females [5].
Is acarbose available over the counter?
No, acarbose is a prescription drug in the United States, approved by the FDA in 1995 for type 2 diabetes [2]. It requires a doctor's prescription regardless of whether it's being used for its approved diabetes indication or off-label for metabolic/longevity reasons.
Which drug is cheaper, rapamycin or acarbose?
Acarbose is typically cheaper and simpler to obtain since it's commonly prescribed by primary care doctors for diabetes, with generic cash prices often in the $15-40/month range. Rapamycin is also generic but usually costs more in practice due to the added step of finding a prescriber willing to prescribe it off-label, plus variable formulation and dosing costs.
Do I need a prescription to get rapamycin for longevity use?
Yes. Sirolimus requires a prescription in the United States for any use, on-label or off-label. Provider-reviewed telehealth routes exist specifically because most primary care doctors won't prescribe it off-label without a longevity-focused evaluation; a licensed pharmacy still fills the actual prescription.
What's the biggest gap in the evidence for both drugs?
Neither rapamycin nor acarbose has a completed randomized controlled trial measuring human lifespan or all-cause mortality as a primary outcome. All the lifespan-extension evidence for both drugs comes from mouse studies; human data is limited to short trials on biomarkers, immune function, or disease-specific outcomes like diabetes progression.
Sources
- National Institute on Aging, Interventions Testing Program: The ITP tests candidate longevity compounds in genetically diverse mice across three research centers
- Harrison DE et al., Nature, 2009: Rapamycin extended median lifespan in mice even when started late in life, by 9% in males and 13% in females in the original ITP report
- Miller RA et al., Aging Cell / NIA ITP reporting: Higher-dose rapamycin cohorts showed increased median lifespan up to roughly 23% in males and 26% in females
- Mannick JB et al., Lancet, 2018: A short trial of an mTOR-inhibitor combination in older adults improved influenza vaccine response and was reported safe
- PEARL trial, AgelessRx / Kaeberlein lab: A small human trial of low-dose rapamycin (PEARL) reported quality-of-life and bone-density signals over about 48 weeks without measuring lifespan
- Chiasson JL et al., STOP-NIDDM Trial, Lancet, 2002: Acarbose reduced progression to type 2 diabetes and reduced cardiovascular events in people with impaired glucose tolerance