16 hyped anti-aging interventions, 0 proven on a hard human endpoint — and why that should lower your prior
Of the 16 most-hyped anti-aging interventions — rapamycin, NMN, senolytics, young blood, caloric restriction, partial reprogramming — 0 have a proven human benefit on a hard clinical endpoint (lifespan/healthspan). The sharper number: only 1 (aspirin) was ever tested at that bar, and it failed; the other 15 are untested there, not disproven. The real takeaway is calibration — put a low prior on the next mouse headline reaching humans, and expect any real human effect to be far smaller than the mouse one.
Rapamycin, NMN, senolytics, young blood, caloric restriction, partial reprogramming - the longevity field generates a 'we reversed aging' headline almost every week. So I built a scorecard: the 16 flagship interventions, each one's mouse evidence, and where it actually stands in HUMANS on a hard clinical endpoint (lifespan or healthspan - not a biomarker). Here is the running tally. Scope note: this list is the hyped geroprotective candidates (NIA ITP positives + the most-hyped supplements) — not everything that extends life. Interventions with hard-endpoint human mortality data (exercise, GLP-1 and SGLT2 drugs, statins) are deliberately out of scope; the question here is whether the hyped longevity set has cleared the bar.
The informative number is not the zero — it is that only 1 of 16 has ever been put to a hard-endpoint human trial at all (aspirin), and it failed; the rest are untested at that bar. Proven human benefit on a hard endpoint: 0 of 16. Tested in humans and did NOT translate: 1 (aspirin). Human trial reported but safety/biomarker/null only: 5. Human trial pending, no result yet: 2. No meaningful human outcome data at all (mouse/cell/observational only): 8.
| Intervention | Mouse evidence | Human hard-endpoint status |
|---|---|---|
| ❌ Aspirin (ITP) | Lifespan ↑ in males | ASPREE RCT: no healthy-lifespan benefit; slight mortality ↑ (cancer-driven, primary prevention) |
| 🟡 Rapamycin (ITP) | Robust lifespan, both sexes | PEARL RCT: safety only; efficacy unproven |
| 🟡 Caloric restriction | Robust in rodents; primates split (UW yes / NIA no) | CALERIE 2-yr: biomarkers, no lifespan endpoint |
| 🟡 Senolytics (D+Q, fisetin) | Healthspan gains in mice | Phase-1 pilots null (n=5 Alzheimer's; STAMINA) |
| 🟡 NMN / NAD+ precursors | Mouse metabolic gains | RCT meta-analyses null (muscle, metabolism); blood NAD+ doesn't track age |
| 🟡 Resveratrol (ITP) | ITP: negative (no lifespan ↑) | Human trials null on longevity |
| ⏳ Metformin | ITP: null for lifespan | TAME trial designed, unreported; observational mixed |
| ⏳ Young plasma / GDF11 | Robust mouse parabiosis; GDF11 contested | Small Phase-1 plasma RCTs; no longevity endpoint |
| ⬜ Acarbose (ITP) | +~22% male / ~5% female median | Approved diabetes drug; no longevity RCT |
| ⬜ 17-α-estradiol (ITP) | Lifespan ↑ males only | No human longevity data |
| ⬜ Canagliflozin (ITP) | Lifespan ↑ males only | Approved diabetes drug; no longevity RCT |
| ⬜ NDGA (ITP) | Lifespan ↑ males (early) | No human longevity data |
| ⬜ Glycine (ITP) | Modest lifespan ↑ | No human longevity outcome |
| ⬜ Captopril (ITP) | +4–5% female | BP drug; no longevity RCT |
| ⬜ Partial reprogramming (OSK) | Mouse in-vivo | Human cells in vitro only |
| ⬜ FTL1 knockdown | Mouse, male, hippocampus (2025) | None (mouse-stage) |
Legend: ❌ tested, did not translate · 🟡 human trial: safety/biomarker/null only · ⏳ trial pending · ⬜ no human longevity-endpoint data (mouse/cell/observational only).
A few specifics, because the names matter:
- Rapamycin is the best-evidenced of all of them, and its main human trial (PEARL) reported essentially safety - no demonstrated anti-aging effect yet.
- NMN / NAD+ boosters reliably double a blood marker, but randomized-trial meta-analyses show no benefit to muscle or metabolism - and new 2026 work finds that blood marker doesn't even decline with age in humans. The premise under the whole supplement category is shaky.
- Senolytics (dasatinib + quercetin) showed nothing in early human pilots (a 5-patient Alzheimer's pilot, not powered for efficacy, showed no efficacy signal).
- Caloric restriction is the most robust in animals, yet the two primate trials disagree with each other, and the main human trial (CALERIE) moved biomarkers over two years - not lifespan.
- Aspirin is the cautionary tale: it extended male-mouse lifespan, but the large ASPREE trial found no healthy-lifespan benefit in older adults and a slight, cancer-driven increase in all-cause mortality (in healthy older adults taking it for prevention, not for established heart disease). That is what a real non-translation looks like.
Two honest caveats. First, '0 proven' is partly because hard-endpoint human trials are expensive and rarely run (though ASPREE shows a ~5-year healthspan RCT is feasible — and when it ran, the answer was negative); read it as 'no proven win yet,' not 'everything fails.' Aspirin is the only row that is a tested failure. Second, this is not pessimism about the biology; tissue-level aging mechanisms are real. The point is calibration: the right prior on the next mouse headline producing a proven human benefit within a decade is low, and the eventual human effect will likely be far smaller than the mouse one — and even some mouse 'wins' are statistically fragile: at least for age-localized survival effects, a runnable check flips between 'null' and 'works' on ~40% of identical datasets depending on the survival test used.
Method, in two sentences: the population is the rigorous NIA Interventions Testing Program positives plus the most-hyped non-ITP interventions; each was scored on whether a controlled human trial has shown a hard-endpoint benefit, with the key trial named per row. This isn't the first longevity tracker — Lifespan.io's Rejuvenation Roadmap already charts interventions by trial stage, and evidence-tier guides grade them by strength; what's different here is the bar (a single hard clinical endpoint, biomarkers excluded) paired with an explicit prior. That prior rests on standard preclinical attrition — roughly 90% of drugs entering human trials fail, mostly for lack of efficacy, and the NIA ITP exists precisely because single-lab mouse positives rarely replicate. We keep this as a calibration snapshot (the named trial per row above), updated as trials report.
What would change the number: a single positive hard-endpoint human result on any row (a positive metformin TAME trial, a rapamycin-successor efficacy signal, a young-plasma readout) moves it up, and we will say so.
FAQ
Do any hyped anti-aging interventions have proven human benefit? On a hard clinical endpoint (lifespan/healthspan), 0 of 16. The scorecard rates each flagship intervention's mouse evidence against where it actually stands in humans — and none has crossed the line.
What does the human evidence actually show? Of the 16: 1 was tested in humans and did not translate (aspirin); 5 have a human trial reporting only safety, biomarker, or null results; 2 have a trial pending with no result yet; the rest rest on animal or mechanistic data.
Which interventions are on the list? The flagships people cite most — rapamycin, NMN, senolytics, young blood, caloric restriction, partial reprogramming, and others — each scored on mouse evidence versus hard human endpoints.
What should a consumer take from this? Strong mouse or biomarker data is not proven human benefit. With 0 of 16 proven on a hard endpoint, treat any hyped longevity supplement or geroprotector claim as unproven until a human lifespan/healthspan trial reads out.