In 2023, the most rigorous lifespan-testing program in the world published its most exciting result in 14 years. A supplement available at any pharmacy — astaxanthin — had extended male mouse lifespan by 12% [1].
Supplement companies immediately translated that into marketing. One brand, AX3, started claiming their product could deliver a "potential human lifespan increase of up to 9 years."
Meanwhile, another supplement — calcium alpha-ketoglutarate, a form of AKG — was being sold for $80 a month with a claim that it reduces biological age by 8 years. The company's 67-year-old founder even put out a press release saying he had achieved "longevity escape velocity" [2]. For context, longevity escape velocity means biological age is dropping faster than calendar age is rising. His evidence for this was an epigenetic test with a four-year margin of error. More on that shortly.
Last month, the same interventions testing program published brand new results. Astaxanthin at lower doses? Nothing. In females, it actually shortened lifespan. AKG? Nothing. For the second time [3].
And this isn't unusual. It's the pattern. Over the last 20 years, this program has tested compound after compound that the supplement industry sells as "anti-aging" breakthroughs. Almost all of them fail.
Here's what that program is, why it matters, and what — if anything — has actually survived rigorous independent testing.
Table of Contents
- The Program Nobody Talks About
- The One That Worked
- The Supplements the Industry Got Excited About
- The Supplement Machine
- The 2026 Verdict
- What This Means
- References
The Program Nobody Talks About
In 2002, the U.S. National Institute on Aging launched something unusual. They called it the Interventions Testing Program — the ITP [4].
The idea was simple but radical. Take any compound — a drug, supplement, or natural extract — that scientists believe might extend healthy lifespan, and test it. But not in one lab. In three, simultaneously. The Jackson Laboratory in Maine. The University of Michigan. The University of Texas in San Antonio.
Same mice. Same food, prepared centrally and shipped to all three sites. Same protocol. Three independent teams running the experiment in parallel.
The mice used are a mixed breed of different genetic backgrounds designed to capture genetic diversity, unlike inbred strains often used in research [5].
Why three sites? To avoid what happened with nicotinamide riboside, or NR. A study made the headlines in 2016 showing that NR extended lifespan and protected aging muscle in mice [6]. The result exploded into mainstream hype. NR was positioned as an accessible, evidence-based way to "boost NAD+ and fight aging" without the need for extreme diets or drugs. Supplement companies built major businesses around NR products.
But there was just one problem. The benefits seen in that initial trial didn't show up when NR was tested again in mice through the ITP [7]. One trial just isn't enough. The literature is full of results that have turned out to be unrepeatable. If a supplement extends lifespan at Michigan but not at Texas or Maine, that tells us something important. The result probably isn't real.
The ITP was designed to catch exactly that kind of false positive.
And here's the part that makes it truly unusual: they publish everything. Positive results, negative results — especially the negative results. If a compound fails, the ITP publishes that failure. That might sound obvious, but in the supplement world, negative results tend to disappear.
The ITP program screens five to seven new drugs a year. Most of them don't work. They tested the first compounds in 2004, and for the first five years, there was little to get excited about.
Then in 2009, they got their breakthrough.
The One That Worked
It had been a long time coming. Back in 1964, a team of Canadian scientists traveled to Easter Island. While there, Georges Nogrady's attention was drawn to something unusual. Despite walking barefoot in areas where risks were high, the local population was seemingly immune to tetanus. He suspected there might be some kind of natural antibiotic compound in the soil [8].
That compound was subsequently isolated from the soil and given the name "rapamycin" after the island's name in the local language. Rapamycin turned out to have diverse properties. It would eventually be approved as an immunosuppressant by the FDA in 1999. But it was its impact on a signalling pathway linked to cell growth — mTOR — that led to interest in its potential for extending healthy lifespan [8].
The ITP found that rapamycin extended lifespan by 9% in males and 14% in females — even when started at 600 days of age, the equivalent of about 60 human years. It was confirmed at all three sites [9]. Science magazine named it one of the top ten breakthroughs of 2009 [10].
This was the first drug ever shown to extend mammalian lifespan when given late in life. The ITP had proven its design worked. The three-site model could find real results and filter out false ones.
But rapamycin isn't a supplement anyone can buy. It's an immunosuppressant with real side effects. The search for something more accessible continued.
For the next fourteen years, the ITP tested dozens more compounds. Resveratrol — the supplement David Sinclair made famous? No effect on lifespan. Fish oil? No effect. NR? No effect. Fisetin, the senolytic that was generating huge excitement? No effect [11].
And then, in late 2023, something changed.
The Supplements the Industry Got Excited About
The ITP published a result that made the supplement world sit up. Astaxanthin showed a 12% lifespan extension of median lifespan in males when started at 12 months of age at a target dose of 4,000 parts per million [1].
This was significant. Astaxanthin is a carotenoid — the pigment that makes salmon pink — and a potent antioxidant available at any pharmacy for about $12 a month.
A 12% lifespan extension was the biggest result the ITP had produced in 14 years, for a compound that is widely available and safe enough for everyday use.
But the excitement wasn't limited to astaxanthin. Around the same time, another compound was generating serious buzz: AKG.
In 2020, researchers at the Buck Institute published a study showing AKG extended lifespan by up to nearly 17% and reduced frailty by 41–46% in mice [12]. The senior author, Gordon Lithgow, said something that captures why this result mattered: "The nightmare scenario has always been life extension with no reduction in disability. In this study, the treated middle-aged mice got healthier over time" [13].
David Sinclair tweeted that AKG "reduces hair greying, frailty & extended median lifespan 10-16%" [14].
And a company called Ponce de Leon Health launched a supplement called Rejuvant — calcium alpha-ketoglutarate.
The press release where the founder claimed he had achieved "longevity escape velocity" is worth examining. What does the evidence behind that claim actually show?
The Supplement Machine
Ponce De Leon Health published a study to back up its "8-year biological age reduction" claim. Here's what they did. They gave 42 people — 28 men, 14 women, average age about 63 — their AKG supplement and measured biological age using an epigenetic clock called TruAge [15].
The result: an average 8-year reduction in biological age. Sounds incredible. But there are two problems.
First: there was no placebo group. Everyone in the study knew they were taking the supplement. Without a placebo control, it is impossible to attribute the result to the supplement rather than to chance, lifestyle changes, or the placebo effect itself. The authors themselves acknowledged this, writing that "continued testing, particularly in a placebo-controlled design, is required" [15].
Second: the epigenetic clock they used — TruAge — measures only 9 CpG sites across 3 genes. CpG sites are like specific addresses within DNA where signs of aging are detected. The test has a median absolute error of about ±4 years [15]. Compare this to more robust approaches like Horvath's clock, which has a median error of ~2.7–2.9 years and assesses 353 CpG sites [16].
So the TruAge measurement tool itself has an error margin of 4 years. The claimed improvement was 8 years. When a measuring instrument wobbles by half the size of the result being claimed, that is a serious methodological problem.
This is not rigorous evidence. This is marketing material dressed up as science.
And if this pattern feels familiar, it should. We've seen it before.
In 2006, David Sinclair published a study in Nature showing resveratrol extended lifespan in mice on a high-fat diet [17]. The excitement was enormous. In 2008, GSK paid $720 million to acquire Sirtris Pharmaceuticals, which was built around resveratrol [18].
Then the problems started. Other labs found that the mechanism Sinclair proposed was likely caused by the fluorescent dye used in the experiment [19]. And a clinical trial of SRT501, a concentrated form of resveratrol, was halted after 5 of 24 subjects developed kidney damage [20]. By 2013, GSK shut down Sirtris entirely [21].
Matt Kaeberlein — a researcher who is an author on the 2026 ITP paper — put it bluntly, calling resveratrol "the most debunked longevity intervention of all time" [22].
And yet, today, resveratrol supplements are still for sale. The $720 million company is gone, but the products remain on shelves.
The pattern is always the same. Promising mouse study. Press release. Product launch. Years of sales. And then, quietly, the follow-up data comes in. But by then, nobody's paying attention.
Which brings us to the 2026 results.
The 2026 Verdict
The ITP published the results from their 2022 cohort. They tested eleven interventions — some new, some follow-ups on previously promising compounds. The title says it all. None of the eleven compounds extended lifespan [3].
Starting with the one people are paying $80 a month for: AKG was tested beginning at 7 months of age. No effect. This was actually the second time the ITP tested AKG. In a previous cohort, they started it at 18 months — closer to the protocol in the original Buck Institute study. No effect then either [3].
Two independent ITP tests. Two different starting ages. Same result: nothing.
Recall Gordon Lithgow's "nightmare scenario" — life extension with no reduction in disability. The ITP found something worse: no life extension and no reduction in disability. Rejuvant, the branded version of AKG, is still selling at $79.95 a month.
What about the compound that had actually worked before? Astaxanthin — the supplement that previously extended male lifespan by 12% — was tested again, this time at a lower dose of 880 parts per million instead of 4,000, starting at 11 months and at 16 months. No effect [3].
In females, astaxanthin was actually associated with shorter lifespans [3].
The dose was different. The original positive result was at 4,000 ppm; this test used 880 ppm. That's an important distinction, and it tells us something crucial — the effect, if real, may depend on a very specific dose window. A dose that's four-fifths lower didn't just fail to help. In females, it may have caused harm.
And it wasn't just astaxanthin. Three other compounds that previously showed promise — mitoglitazone, meclizine, and pioglitazone — also failed when tested at different doses or start ages. Pioglitazone and mitoglitazone actually shortened female lifespan. Pioglitazone was the worst: females lived 6.2% shorter lives [3].
There's one more detail worth noting. The control females at the Jackson Laboratory in this cohort lived unusually long — 3.13 standard deviations above the historical average, a statistical anomaly the researchers couldn't explain. So they re-ran the entire analysis excluding Jackson Lab data, using only Michigan and Texas. The harm results for pioglitazone and mitoglitazone held up [3].
That kind of transparency — flagging data anomalies and re-running the analysis — is what sets the ITP apart.
What This Means
The ITP has been running for over twenty years. In that time, they've tested dozens of compounds. The vast majority have failed — and the few that succeeded were prescription drugs, not supplements.
The compounds that have shown consistent, reproducible lifespan extension across all three sites include rapamycin, acarbose — a diabetes drug — 17-alpha-estradiol in males, and canagliflozin, another diabetes drug [11]. That's essentially it. None of these are supplements available over the counter. They are prescription medications with real side-effect profiles.
Does this mean astaxanthin is worthless? No. There is some early evidence from randomised controlled trials that astaxanthin at around 6 mg per day may improve skin hydration and elasticity [23]. But the existing trials are small, with unclear risks of bias — and those trials are a very different question from lifespan extension. The ITP data provide no basis for "anti-aging" marketing claims about astaxanthin.
And AKG? Two ITP cohorts. Two different starting ages. No effect in either. The $80-a-month supplement with the "8-year biological age reduction" claim is based on a study with no placebo group and an epigenetic clock with a ±4-year error margin.
Healthy scepticism toward any company marketing a "longevity" or "anti-aging" supplement is well warranted — particularly when the marketing precedes the independent evidence by years.
So how did the same program find that astaxanthin both extended and didn't extend lifespan? The 2023 result was real: 4,000 parts per million, starting at 12 months, in males. The 2026 result is also real: 880 parts per million, different start ages, different sexes. They're not contradictions. They're part of filling out the complete picture.
In aging biology, dose, timing, and sex are not footnotes. They're everything. A compound that works at one dose may do nothing at another. A compound that benefits males may harm females. And none of this can be known without testing properly — at multiple sites, in genetically diverse animals, across multiple doses and timepoints [3].
The ITP has been publishing result after result that quietly contradicts what the supplement industry is selling. The positive results go viral. The failures land quietly in academic journals that most people never read.
But the ITP does its job anyway. And the reason their list of successes is so short isn't because the program has failed. It's because aging is genuinely hard. That's not a failure of science. That's what honest science looks like.
References
1. https://link.springer.com/article/10.1007/s11357-023-01011-0
3. https://link.springer.com/article/10.1007/s11357-026-02201-2
4. https://www.nia.nih.gov/research/dab/interventions-testing-program-itp
5. https://www.nia.nih.gov/research/dab/interventions-testing-program-itp/about-itp
6. https://www.science.org/doi/10.1126/science.aaf2693
7. https://pmc.ncbi.nlm.nih.gov/articles/PMC8135004/
8. https://pmc.ncbi.nlm.nih.gov/articles/PMC9494524/
9. https://www.nature.com/articles/nature08221
10. https://www.science.org/doi/10.1126/science.326.5960.1600
11. https://www.nia.nih.gov/research/dab/interventions-testing-program-itp/supported-interventions
12. https://pmc.ncbi.nlm.nih.gov/articles/PMC8508957/
14. https://x.com/davidasinclair/status/1300868036023459841
15. https://www.aging-us.com/article/203736/text
16. https://www.mdpi.com/2673-8856/5/2/28
17. https://www.nature.com/articles/nature05354
19. https://pmc.ncbi.nlm.nih.gov/articles/PMC2832984
20. https://www.fiercebiotech.com/biotech/safety-concerns-force-glaxo-to-suspend-enrollment-srt501-trial
21. https://www.bioworld.com/articles/438084
22. https://www.levels.com/blog/matt-kaeberlein-longevity-science
