The Sinclair stack, rated claim by claim

• Raises NAD+ levels in blood 4/5Strong
  Human biomarker

  Multiple randomised placebo-controlled trials show a dose-dependent rise in blood NAD+, largest at 600-900 mg. This part is not seriously disputed.

  What that leaves out. Blood NAD+ is not the same as NAD+ inside muscle, brain or liver, which is where the ageing argument actually lives. The trials measure the compartment that is easy to sample, not the one the theory needs.

• Improves walking and physical performance 2/5Limited
  Human outcome

  Some trials report longer six-minute walk distance and better self-reported sleep at 300-900 mg. A systematic review of physical-performance trials found improvements in several.

  What that leaves out. These are mostly secondary endpoints in small, short, often industry-funded trials. A 2024 meta-analysis of glucose and lipid outcomes across 12 studies found no consistent benefit, and its authors wrote that exaggeration of NMN benefits may exist in the field.

• Slows or reverses human ageing 1/5Minimal
  Mechanism only

  No trial has tested whether NMN changes how long people live, or how long they stay healthy. The argument runs from NAD+ decline with age, through sirtuin biology, to ageing itself.

  What that leaves out. Every step after the NAD+ rise is inference. A supplement that reliably moves a blood marker and has never been shown to change an ageing outcome is the ordinary result here, not a scandal — but it is not what the marketing says.

• Well tolerated at studied doses 3/5Moderate
  Human outcome

  Trials up to 900 mg/day, and a long-term trial at 250 mg/day, reported no significant adverse events.

  What that leaves out. Trials have run months, not decades. "No harm found in 12 weeks" is a much smaller statement than "safe to take daily for life", which is how it is used.

Status. The FDA excluded NMN from the dietary-supplement definition in November 2022 under the drug-preclusion clause, then reversed itself on 29 September 2025, concluding NMN had been marketed as a supplement before the drug investigation began. It is lawfully sold in the US again. Articles written between those dates say the opposite, and many have not been updated.

Reported dose. About 1 g each morning — reported, not recommended.

• Absorbed poorly and cleared fast in humans 4/5Strong
  Human biomarker

  Oral resveratrol is extensively metabolised on first pass; most is conjugated and excreted, leaving very low free concentrations in plasma.

  What that leaves out. This is the finding with the least dispute attached and the most awkward implications: it constrains every downstream claim, because the tissue never sees the concentration the cell experiments used.

• Directly activates SIRT1 1/5Minimal
  Mechanism only

  The original activation result came from an assay using a fluorophore-tagged substrate. A 2011 Nature paper argued the effect was an artifact of that tag and did not occur with native substrates.

  What that leaves out. This is a live scientific dispute, not a settled fraud finding, and later work proposes indirect routes. But the direct-activation claim that the popular story rests on is the specific thing that was contested.

• Extends lifespan in mammals 1/5Minimal
  Animal outcome

  Resveratrol improved health measures in mice on high-fat diets. It did not extend lifespan in normally-fed mice.

  What that leaves out. The distinction matters more than it looks: rescuing an unhealthy animal towards normal is not the same as extending a healthy lifespan, and only the second is the longevity claim.

• Improves clinical outcomes in people 2/5Limited
  Human outcome

  Human trials are numerous, small, and inconsistent, with conflicting results across cancer, cardiovascular and inflammatory endpoints.

  What that leaves out. GSK acquired Sirtris for $720M in 2008 on the strength of this class, halted trials — at least two for unexpected side effects — and closed the unit. That is the most expensive negative result the field has produced.

Status. Sold as a dietary supplement. No approved therapeutic indication.

Reported dose. About 1 g each morning, with a fat source — reported, not recommended.

• Effective for type 2 diabetes 5/5Very strong
  Human outcome

  First-line therapy worldwide, with decades of randomised evidence and regulatory approval.

  What that leaves out. This is not in question, and it is not the longevity claim. It is why the drug is available to argue about.

• Extends healthspan in people without diabetes 1/5Minimal
  Mechanism only

  The hypothesis comes largely from observational cohorts suggesting diabetics on metformin outlived non-diabetic controls. TAME, the trial designed to test it directly in 3,000 older adults, has been seeking full funding since 2015.

  What that leaves out. Researchers close to TAME say plainly that metformin is not ready for use as an ageing intervention by the general population. Benefits seen in metabolically unhealthy people may not transfer to healthy ones.

• Blunts muscle gains from resistance training 4/5Strong
  Human outcome

  The MASTERS randomised double-blind trial in older adults found the placebo group gained more lean mass and thigh muscle than the metformin group over a progressive training programme.

  What that leaves out. This is the rare case where the evidence against a longevity use is stronger than the evidence for it — and it is a plausible reason to pause the drug if you lift.

Status. Prescription-only. Requires a diagnosis and monitoring; not obtainable as a supplement.

Reported dose. Previously ~800 mg; publicly described as paused — reported, not recommended.

• Induces autophagy 3/5Moderate
  Mechanism only

  Well characterised in cells and animals, and the mechanism is not seriously disputed.

  What that leaves out. Autophagy induction is a cellular process, not a health outcome. The step from one to the other is the whole question.

• Extends lifespan in animals 2/5Limited
  Animal outcome

  Lifespan extension shown in yeast, flies, worms and mice.

  What that leaves out. Lifespan extension in short-lived animals is where most failed longevity interventions also looked good.

• Improves memory or cognition in people 2/5Limited
  Human outcome

  Small human trials, mostly in older adults with subjective complaints, with mixed results.

  What that leaves out. Trials are small and short. Dietary intake is confounded with eating patterns that track with health generally.

Status. Sold as a supplement, usually as wheat-germ extract.

Reported dose. Discussed favourably; daily use not confirmed — reported, not recommended.

• Clears senescent cells in mice and extends their lifespan 2/5Limited
  Animal outcome

  A 2018 study found fisetin reduced senescent-cell burden and extended health and lifespan in mice, including when started late in life.

  What that leaves out. Mouse senolytic results have been unusually hard to reproduce in humans.

• Clears senescent cells in humans 1/5Minimal
  Mechanism only

  Despite trials planned and under way, no published data demonstrates senolytic activity of fisetin in humans. The Mayo Clinic AFFIRM phase IIb trial in older women is ongoing.

  What that leaves out. Only dasatinib plus quercetin, and one locally-delivered UNITY compound, have human data showing senescent-cell clearance — and dasatinib is a chemotherapy drug, not a supplement.

Status. Sold as a supplement. The pulse-dosing pattern comes from trial protocols, not from established practice.

Reported dose. Intermittent, high-dose "pulses" — reported, not recommended.

• Reduces senescent-cell burden when combined with dasatinib 3/5Moderate
  Human biomarker

  A phase 1 trial in diabetic kidney disease found a 3-day oral course of dasatinib plus quercetin reduced adipose senescent cells measured by biopsy. An open-label pilot in 14 pulmonary-fibrosis patients reported improved walking measures.

  What that leaves out. Both are tiny and uncontrolled, and both used dasatinib. The evidence belongs to the combination; quercetin alone has not been shown to do this.

• Works as a senolytic on its own 1/5Minimal
  Mechanism only

  No human trial has tested quercetin alone as a senolytic.

  What that leaves out. Buying the supplement gets you the half of the pair without the human evidence.

Status. Widely sold as a supplement. Dasatinib is prescription-only.

Reported dose. Intermittent, alongside fisetin — reported, not recommended.

• Corrects deficiency and supports bone health 5/5Very strong
  Human outcome

  Established, with clear clinical guidance and a measurable blood test to confirm both the problem and the fix.

  What that leaves out. The benefit is concentrated in people who are genuinely deficient.

• Extends lifespan in replete adults 1/5Minimal
  Mechanism only

  Large randomised trials in unselected adults have generally not found the mortality or disease benefits observational studies suggested.

  What that leaves out. A textbook case of correlation dissolving under randomisation: low vitamin D tracks illness partly because ill people go outside less.

Status. Ordinary supplement, and the one item here with a routine blood test behind it.

Reported dose. Adjusted to blood levels — reported, not recommended.

• Acts as a methyl donor 4/5Strong
  Mechanism only

  Established biochemistry; betaine is a recognised methyl donor and lowers homocysteine in trials.

  What that leaves out. True and largely beside the point unless the depletion premise holds.

• Prevents methyl depletion caused by NMN 1/5Minimal
  Mechanism only

  No human trial has shown that NMN at these doses causes clinically meaningful methyl depletion, nor that TMG prevents it.

  What that leaves out. A supplement taken to solve a side effect of another supplement, where neither the side effect nor the solution has been demonstrated in people.

Status. Ordinary supplement.

Reported dose. Not clearly confirmed as routine — reported, not recommended.

• Reduces cardiovascular events in at-risk people 5/5Very strong
  Human outcome

  One of the most thoroughly established results in medicine, across many large randomised trials.

  What that leaves out. The benefit depends on your risk profile, which is why it needs a prescriber rather than a protocol.

Status. Prescription-only.

Reported dose. Prescribed for personal cardiovascular risk — reported, not recommended.

• Prevents first cardiovascular events in healthy adults 2/5Limited
  Human outcome

  Large trials found the bleeding risk offsets the cardiovascular benefit for most people without established disease.

  What that leaves out. Guidance narrowed sharply on this. It is a good example of a popular preventive being walked back on evidence.

Status. Over the counter, but a genuine bleeding risk.

Reported dose. No longer emphasised — reported, not recommended.