Fewer than a dozen trials: the whole human evidence base for longevity rapamycin

Rapamycin extends lifespan in mice with heterogeneous genetic backgrounds, in both males and females, across multiple studies from independent laboratories, at a range of doses and dosing regimens, and even when started in elderly animals [s2]. The finding has been replicated in the Interventions Testing Program, which reported consistent extensions of median and maximum lifespan in male and female mice [s1]. That is about as robust as a preclinical result gets, and it is why some physicians now prescribe the drug off-label to healthy adults.

A review published in Aging in August 2025 asked what the human evidence for that practice actually consists of. The answer: fewer than a dozen known trials, examining biomarkers such as immune function, protein synthesis and haematological parameters — and no outcome data at all [s1].

The strongest finding is about vaccines, not ageing

The most cited human result is not from rapamycin but from everolimus, a related mTOR inhibitor. In a trial of 218 healthy older adults, low-dose everolimus given at 0.5 mg daily or 5 mg weekly and stopped two weeks before influenza vaccination produced a 20% increase in immune titres, alongside a decline in PD-1-positive CD4 and CD8 T cells relative to placebo — a pattern associated with better T-cell function [s1]. Benign mouth ulcers were significantly more common in the treatment arm [s1].

The follow-up work is messier than the headline suggests. A phase 2 trial in 264 healthy adults did not detect a significant difference in annualised rates of respiratory tract infection, though it may have been underpowered [s1]. A phase 2b trial randomised 652 participants to 5 mg/day RTB101, 10 mg/day or placebo; the 5 mg dose did not differ from placebo, and in a second part the remaining 473 participants were randomised to 10 mg/day, 20 mg/day, 10 mg/day plus everolimus, or placebo, with only the 10 mg/day dose reducing infection rates [s1]. The phase 3 study did not replicate the phase 2b result — complicated by a change of endpoint, at the FDA's request, from laboratory-confirmed to patient-reported infections [s1]. The review's own verdict is that the evidence here is compelling but not convincing [s1].

The one detailed healthy-ageing trial found no clear benefit

The most detailed study of low-dose rapamycin in healthy older people enrolled 25 adults aged 70 to 95 and gave 1 mg daily for eight weeks [s1]. It reported no significant improvement in metabolic parameters, alongside several unwelcome changes: a significant fall in plasma albumin, increases in triglycerides and HbA1c, and a near-significant rise in VLDL at p = 0.06 [s1]. Inflammatory markers broadly increased rather than decreased, including a significant elevation in TNF-alpha [s1]. Walking speed declined markedly in the control group while the rapamycin group maintained gait speed, but handgrip strength was unchanged [s1]. The investigators concluded that rapamycin caused no significant adverse outcome in the short term, and that no signal of clear benefit was identified [s1].

Muscle results conflict. One study found 16 mg of rapamycin blunted the post-exercise rise in protein synthesis; another, at the same dose, found no change in skeletal muscle synthesis rates — though it measured only basal, non-post-exercise conditions [s1]. In 22 healthy young men given everolimus over 15 days, IL-2 and IL-10 both fell, and lower and medium doses significantly increased self-reported anxiety [s1]. IL-10 is a key anti-inflammatory cytokine whose expression has been associated with increased longevity [s1].

Two numbers in the review do not reconcile

The review closes with a worked example: it fed the group-average biomarker values from that 25-person trial into the PhenoAge calculator to estimate what happened to biological age. It reports the placebo group moving from 78.32 to 78.47 years, a change of +0.15 years, and the rapamycin group from 81.34 to 77.38 years, described as a net change of −3.96 years [s1].

Its own table then presents those numbers as gaps between phenotypic and chronological age, and the gaps do not follow from the figures given. Taking the table's rapamycin post-treatment phenotypic age of 77.38 against its chronological age of 80.4 gives a gap of −3.02 years, not the −3.96 the table records on that line; −3.96 is instead the difference between the pre- and post-treatment gaps [s1]. The control column has the mirror problem: its listed gaps of −2.28 and −1.93 years differ by 0.35 years, while the row beneath states the control difference as 0.15 years — the raw phenotypic age change [s1]. The table also gives the control group a chronological age of 80.6 years before the study and 80.4 after [s1].

None of this changes the paper's conclusion, and the authors say plainly that statistical significance could not be determined because subject-level data were unavailable, and that two inputs — C-reactive protein at 1.75 mg/L and lymphocyte percentage at 19% — were imputed rather than measured [s1]. But the arithmetic is worth flagging, because a "−3.96 years of biological age" figure is exactly the kind of number that travels without its caveats.

A second unit inconsistency sits in the same paper. It reports the circulating rapamycin level achieved in the eight-week trial as a mean of 7.2 ng/dL, while describing the drug's biological threshold as about 5 ng/mL and greater toxicity above 15 ng/mL [s1] — measures that differ by a factor of a hundred.

What is missing

There is no literature reporting cancer incidence in healthy, non-immunocompromised people taking low-dose rapamycin, so nothing can be said about that risk in this population [s1]. There is insufficient human outcome data to comment on cardiovascular effects [s1]. And there is no established dose-response curve for healthspan extension, despite once-weekly 5 to 7 mg or biweekly 10 to 15 mg being commonly recommended to patients [s1].

The longest trial to date, PEARL, found no change in its primary endpoint of visceral fat over 48 weeks [s2]. The mechanistic case for intermittent dosing — that inhibiting mTORC1 does the good and mTORC2 does the harm — remains a working model whose optimal schedule in humans is unknown [s3].

Rapamycin is a prescription immunosuppressant. This article reports what has been studied; it is not medical advice, and no dose here should be read as guidance.

Sources

1. [s1] What is the clinical evidence to support off-label rapamycin therapy in healthy adults? Aging (Albany NY), published online August 7, 2025. https://doi.org/10.18632/aging.206300
2. [s2] Influence of rapamycin on safety and healthspan metrics after one year: PEARL trial results. Aging (Albany NY), published online April 4, 2025. https://doi.org/10.18632/aging.206235
3. [s3] Testing the Translational Potential of Rapamycin on Healthy Aging. Innovation in Aging, published online December 31, 2025. https://doi.org/10.1093/geroni/igaf122.406