Autophagy Explained: How It Works and Why It Matters

Autophagy is one of the few longevity concepts with a Nobel Prize behind it, and one of the most confidently oversold — a carefully studied recycling pathway reduced to a number, sixteen hours, repeated everywhere and supported almost nowhere.

Context sits in our hallmarks of ageing explainer, our longevity supplements guide and the anti aging supplements range. For the NAD⁺ compounds routinely confused with this pathway — they are adjacent, not the same — see NMN benefits, side effects and dosage, the NMN supplements collection and NMN Pro 1000.

What follows covers what autophagy is, how the switch is controlled, why it matters for ageing, and what genuinely influences it — including the two things most articles omit: nobody can tell you what your autophagy level is, and more of it is not automatically better.

What autophagy actually is

The word, and the Nobel Prize

"Autophagy" means self-eating, from the Greek. In practice it describes a controlled recycling system: the cell wraps unwanted material — misfolded proteins, protein aggregates, damaged organelles, invading bacteria — in a double-membrane vesicle called an autophagosome, delivers it to the lysosome, and breaks it down into components that can be reused [4].

The process was recognised in the 1960s, but the mechanism stayed obscure for decades. In 1993 Yoshinori Ohsumi identified 15 genes essential for autophagy in budding yeast, then cloned several and worked out what the encoded proteins did [4]. He was awarded the 2016 Nobel Prize in Physiology or Medicine for those discoveries [1]. That work is why we can talk about the pathway at a molecular level at all — and why claims about "activating autophagy" deserve to be held to a molecular standard.

The three types

Mitophagy: the subtype that matters most for ageing

Mitophagy is selective autophagy of mitochondria. Because mitochondrial dysfunction is itself a separate hallmark of ageing [2], the machinery that identifies and removes failing mitochondria is of particular interest — and it is the one target where a consumer compound has actually reached randomised human trials, discussed below. See urolithin A in the UK for the detail.

How the switch is controlled

Nutrient sensing: mTOR and AMPK

Autophagy is governed largely by two opposing nutrient sensors. mTOR is active when nutrients — particularly amino acids — are plentiful; it promotes growth and suppresses autophagy. AMPK is active when cellular energy is low; it promotes autophagy. In simple terms, abundance signals "build", scarcity signals "recycle".

This is also why deregulated nutrient sensing and disabled macroautophagy appear as separate but linked hallmarks in the current framework [2]. They are two views of the same control system.

Why insulin is the brake

Insulin signalling sits upstream of mTOR. When you eat, insulin rises and autophagic activity is suppressed. This is the mechanistic basis of the entire fasting-and-autophagy conversation — and it is genuinely sound as far as it goes. Where the reasoning breaks down is in the leap from "eating suppresses autophagy" to "therefore sixteen hours without food produces a meaningful, measurable increase in a human being".

Why it matters for ageing

In the 2023 update to the hallmarks of ageing framework, disabled macroautophagy was added as one of the twelve hallmarks — alongside genomic instability, telomere attrition, epigenetic alterations, loss of proteostasis, deregulated nutrient sensing, mitochondrial dysfunction, cellular senescence, stem cell exhaustion, altered intercellular communication, chronic inflammation and dysbiosis [2].

The logic is straightforward. If the disposal system slows, damaged material accumulates. Autophagic capacity declines with age, and disruption of the pathway is implicated in neurodegenerative disease, where toxic protein aggregates are not properly cleared, as well as in liver, kidney, heart and inflammatory bowel conditions [3,5].

What does not follow is that boosting autophagy in a healthy adult produces a corresponding benefit. That is a hypothesis, and it is the hypothesis the supplement market is currently selling as a conclusion. For related context, see biological age vs chronological age and how to lower your biological age.

The measurement problem nobody mentions

This is the single most important thing to understand before you change anything.

Autophagy is a flux — a rate of turnover, not a substance with a level. Measuring it properly requires comparing marker proteins with and without a blocker of lysosomal degradation, usually in tissue. That is straightforward in cultured cells, feasible in a laboratory mouse, and largely impractical in a living human outside research settings. There is no validated consumer blood test.

The consequences are worth stating plainly:

• Nobody can tell you whether your fast "triggered autophagy".
• No wearable, app or blood panel currently measures it.
• Any product marketed with an autophagy score or a percentage increase is extrapolating from cell or animal work.
• Human trials of autophagy-targeting compounds therefore measure downstream outcomes — muscle endurance, memory scores, biomarkers — rather than autophagy itself.

That is not a reason to dismiss the field. It is a reason to judge products by their clinical endpoints rather than their mechanism story, exactly as we argue in longevity supplements that probably don't work and is NMN a scam?

What actually triggers autophagy, ranked

Exercise: the one nobody sells

Exercise is a recognised physiological trigger of autophagy [3], and it is the only item on that list that improves cardiorespiratory fitness, muscle mass, glucose control and mood whether or not the autophagy story holds up. UK guidance is at least 150 minutes of moderate activity weekly plus muscle-strengthening on two or more days [6]. See zone 2 training for longevity and creatine for longevity.

Sleep

Clearance of metabolic waste from the brain is substantially sleep-dependent, and chronic short sleep is associated with worse outcomes across essentially every hallmark. It costs nothing and it is far less discussed than fasting, largely because it is harder to sell. If shift patterns are the obstacle, see supplements for shift workers and apigenin vs magnesium for sleep.

Fasting and the 16-hour claim

Where the number came from

It is not clear that it came from anywhere in particular. The figure appears to be a loose extrapolation from rodent studies, repeated until it acquired the appearance of a finding. The Cleveland Clinic's position is that animal studies suggest autophagy may begin somewhere between 24 and 48 hours of fasting, and that not enough research has been collected on the ideal timing to trigger autophagy in humans [3].

The rodent translation problem

Mice have a far higher mass-specific metabolic rate than humans and deplete liver glycogen much faster. A 24-hour fast in a mouse is not biologically equivalent to a 24-hour fast in a person, and mapping mouse hours onto human hours is exactly the kind of reasoning that produces confident numbers with nothing underneath them.

What is fair to say

• Autophagy runs continuously at a baseline level. It is a dimmer, not a switch.
• Eating suppresses it via insulin and mTOR; not eating removes that suppression.
• Different tissues respond differently and at different rates — liver early, skeletal muscle more resistant.
• The honest answer to "how long do I need to fast for autophagy?" is that nobody knows the human threshold, and anyone who gives you a precise hour figure is going beyond the evidence.

If you do eat in a compressed window, practical questions about what to take and when are covered in supplements on an empty stomach, NMN with food and the supplement timing chart.

The supplement claims, assessed

We sell these products, so this section is written to be checkable.

Urolithin A: the best of the evidence, with a caveat

Urolithin A is a gut-microbiome metabolite of ellagitannins found in pomegranate, studied specifically as a mitophagy activator. In a randomised placebo-controlled trial of 66 adults aged 65–90, 1,000 mg daily for four months significantly improved muscle endurance in both hand and leg muscles and reduced plasma acylcarnitines, ceramides and C-reactive protein [7]. However, the primary endpoint — six-minute walk distance — did not reach statistical significance, which most marketing omits. A later randomised trial in middle-aged adults reported improvements in muscle strength and mitochondrial biomarkers [8].

Only around 40% of people appear to produce urolithin A from dietary sources unaided, which is the argument for supplementing it directly. Compare with the NAD⁺ route in urolithin A vs NMN.

Spermidine: the largest trial was negative

Spermidine is the compound most often marketed explicitly as an autophagy inducer, on the strength of animal work. The best test to date is the SmartAge trial: 100 older adults with subjective cognitive decline, randomised to a wheat-germ-derived spermidine supplement or placebo for 12 months. It found no significant beneficial effect on memory performance or on secondary outcomes compared with placebo [9].

The authors noted the supplement increased daily spermidine supply by only around 10%, and exploratory analyses hinted at possible effects on verbal memory and inflammation requiring validation at higher doses [9]. So the fair reading is "not demonstrated at the doses commonly sold" rather than "definitively useless". That is still a long way from what the labels imply. See buying spermidine in the UK and spermidine vs NMN.

Rapamycin and metformin: not supplements

Rapamycin inhibits mTOR directly and is the most potent autophagy inducer discussed in longevity circles. It is also an immunosuppressant prescription medicine with a meaningful side-effect profile.

The most relevant human data is the PEARL trial — a 48-week randomised, double-blind, placebo-controlled study of low-dose weekly rapamycin in 114 healthy adults aged 50–85. It found the regimen was generally well tolerated and reported improvements in lean tissue mass in women, but the primary endpoint, visceral adiposity, was not met [10]. A peer-reviewed appraisal of off-label rapamycin concludes that clinical evidence for low-dose mTOR inhibitors extending lifespan or delaying age-related disease in healthy adults remains unestablished [11].

Metformin is a prescription diabetes medicine under investigation for geroprotective effects; the large TAME trial designed to test this has not reported. Neither drug is appropriate for self-directed experimentation, and neither is available as a supplement in the UK. Background only: rapamycin and longevity, metformin and longevity, berberine vs metformin.

What about NAD⁺ precursors?

NMN and NR are not autophagy compounds. They target NAD⁺ availability, which supports mitochondrial function and DNA repair enzymes — an adjacent mechanism, not the same one. If a product is sold to you as doing both, check which claim the trials actually tested. See our NMN pillar guide and the NAD supplement guide. On verifying what you buy, see third-party tested supplements and how to read a certificate of analysis.

When more autophagy is not better

This is the section most articles skip, and it is the one with actual clinical relevance.

The cancer paradox

Autophagy plays a dual role in cancer. It can help prevent cancer developing by clearing damaged cellular components — but it can also help established tumour cells survive under stress [3]. This is why some cancer research investigates autophagy inhibitors rather than activators. "More autophagy" is therefore not a coherent universal goal, and anyone undergoing cancer treatment should not add autophagy-targeting supplements without oncology input.

Muscle is also recyclable

The recycling system does not only dispose of things you want gone. Aggressive or prolonged energy restriction in an older adult risks losing muscle mass, which drives frailty and loss of independence — precisely the outcome a longevity protocol is meant to prevent. Adequate protein and resistance training matter more with age, not less.

The mechanism-first trap

Autophagy is a mechanism, not an outcome. The compounds that reached randomised human trials were judged on endurance, strength and memory — and produced mixed results. A plausible mechanism is a reason to run a trial, not a reason to buy a product.

A practical, low-risk approach

If you are assembling a broader routine, see the longevity stack overview, the minimal supplement stack, how to start longevity supplements and how long supplements take to work. Since autophagy itself cannot be measured, track what can be: a HbA1c blood test, a cholesterol blood test, grip strength, and a single panel such as the Full Body MOT Health Check or Welzo Well-Human advanced blood test.

When to speak to a doctor

Speak to your GP before any fasting beyond a normal overnight pattern if you have diabetes or take medication affecting blood glucose, take blood pressure medication, are pregnant, breastfeeding or trying to conceive, are under 18 or underweight, have a history of disordered eating, or have kidney, liver or adrenal conditions. Medication timing frequently needs adjusting around fasting, and that is a clinical decision.

Speak to your GP before starting an autophagy-targeting supplement if you take prescription medication, have an existing condition, or are undergoing cancer treatment — where the dual role of autophagy makes oncology input essential [3].

Seek prompt medical assessment rather than continuing a protocol if you experience dizziness or fainting, palpitations, confusion, unintended weight loss, or persistent unexplained fatigue. Never stop or reduce prescribed medication to pursue a fasting or supplement protocol.

Frequently asked questions

What is autophagy in simple terms?

It is the cell's recycling system. Damaged proteins, protein aggregates and worn-out organelles are enclosed in a vesicle, delivered to the lysosome, broken down, and the components reused. Yoshinori Ohsumi received the 2016 Nobel Prize in Physiology or Medicine for identifying the genes and mechanisms behind it [1,4].

How long do you have to fast for autophagy?

Nobody knows the human threshold. Animal studies suggest meaningful activation somewhere between 24 and 48 hours of fasting, and the Cleveland Clinic states there is not enough research to determine the ideal timing in humans [3]. The commonly repeated "16 hours" figure is an extrapolation rather than a finding. Autophagy also runs continuously at a baseline level rather than switching on at a fixed hour.

Does 16:8 intermittent fasting trigger autophagy?

Probably to some degree, since eating suppresses autophagy through insulin and mTOR and not eating removes that suppression — but the evidence for a large effect at that duration in humans is thin [3]. A 16:8 pattern is a reasonable way to structure eating if it suits you. Treating it as a verified autophagy protocol goes beyond what is known.

Can you measure your own autophagy?

No. Autophagy is a rate of turnover rather than a level, and measuring it properly requires tissue analysis with and without a lysosomal blocker. There is no validated consumer blood test, wearable or app. Any product offering an autophagy score is extrapolating from cell or animal work.

What is mitophagy and how is it different?

Mitophagy is the selective autophagy of mitochondria specifically. It matters because mitochondrial dysfunction is its own hallmark of ageing [2], and it is the one target where a consumer compound — urolithin A — has reached multiple randomised human trials [7,8].

Do autophagy supplements work?

Judge them on clinical endpoints, not mechanism. Urolithin A improved muscle endurance and mitochondrial biomarkers in a randomised trial of older adults, though it missed its primary endpoint of six-minute walk distance [7]. Spermidine, the compound most explicitly marketed for autophagy, showed no significant effect on memory or secondary outcomes in a 12-month randomised trial of 100 older adults [9].

Does exercise trigger autophagy?

Exercise is a recognised physiological trigger [3], and it is the most sensible option in practice because it delivers well-evidenced benefits to fitness, muscle mass, glucose control and mood whether or not the autophagy effect is large in humans. UK guidance is 150 minutes of moderate activity weekly plus strengthening work twice weekly [6].

Does rapamycin boost autophagy, and should I take it?

It inhibits mTOR directly and is the most potent autophagy inducer discussed in this field, but it is a prescription immunosuppressant, not a supplement. The PEARL trial of low-dose weekly rapamycin in 114 healthy adults over 48 weeks did not meet its primary endpoint of visceral adiposity, though it reported acceptable tolerability and some lean-mass benefit in women [10]. Peer-reviewed appraisal concludes the case for off-label use in healthy adults remains unestablished [11]. Not something to self-prescribe.

Is more autophagy always better?

No. Autophagy has a dual role in cancer — it can help prevent cancer developing but can also help established tumour cells survive stress, which is why some research targets autophagy inhibitors [3]. Aggressive energy restriction can also cost muscle mass in older adults. Maximising an unmeasurable process is not a coherent health goal.

Does coffee or black tea break autophagy?

Black coffee and unsweetened tea contain negligible calories and are unlikely to meaningfully raise insulin, so they are generally considered compatible with a fasting window. Adding milk, sugar or anything protein-containing does provide calories and amino acids. Given that no one can measure the effect either way, this is not worth agonising over.

Is autophagy the same as detoxing?

No, and the confusion is worth clearing up. Autophagy is a specific, Nobel-recognised intracellular pathway that recycles the cell's own damaged components [1,4]. "Detox" in commercial wellness usually refers to removing external toxins, a function handled by the liver and kidneys, and products sold on that basis rarely have anything to do with autophagy.

References

1. The Nobel Prize in Physiology or Medicine 2016 — Yoshinori Ohsumi, "for his discoveries of mechanisms for autophagy". NobelPrize.org. https://www.nobelprize.org/prizes/medicine/2016/ohsumi/facts/
2. López-Otín C, Blasco MA, Partridge L, Serrano M, Kroemer G. Hallmarks of aging: An expanding universe. Cell. 2023;186(2):243–278. https://www.cell.com/cell/fulltext/S0092-8674(22)01377-0
3. Cleveland Clinic. Autophagy: Definition, Process, Fasting & Signs. https://my.clevelandclinic.org/health/articles/24058-autophagy
4. The 2016 Nobel Prize in Physiology or Medicine — Advanced Information. NobelPrize.org. https://www.nobelprize.org/prizes/medicine/2016/advanced-information/
5. Frake RA, Rubinsztein DC. Yoshinori Ohsumi's Nobel Prize for mechanisms of autophagy: from basic yeast biology to therapeutic potential. Journal of the Royal College of Physicians of Edinburgh. 2016;46(4):228–233. https://pubmed.ncbi.nlm.nih.gov/28504774/
6. NHS. Physical activity guidelines for adults aged 19 to 64. https://www.nhs.uk/live-well/exercise/exercise-guidelines/physical-activity-guidelines-for-adults-aged-19-to-64/
7. Liu S, D'Amico D, Shankland E, et al. Effect of urolithin A supplementation on muscle endurance and mitochondrial health in older adults: a randomized clinical trial. JAMA Network Open. 2022;5(1):e2144279. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2788244
8. Singh A, D'Amico D, Andreux PA, et al. Urolithin A improves muscle strength, exercise performance, and biomarkers of mitochondrial health in a randomized trial in middle-aged adults. Cell Reports Medicine. 2022;3:100633. https://www.sciencedirect.com/science/article/pii/S2666379122001586
9. Schwarz C, Benson GS, Horn N, et al. Effects of Spermidine Supplementation on Cognition and Biomarkers in Older Adults With Subjective Cognitive Decline: A Randomized Clinical Trial (SmartAge). JAMA Network Open. 2022;5(5):e2213875. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2792725
10. Zalzala S, Cheng KA, Sharp A, et al. Influence of rapamycin on safety and healthspan metrics after one year: PEARL trial results. Aging (Albany NY). 2025. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12074816/
11. What is the clinical evidence to support off-label rapamycin therapy in healthy adults? A critical appraisal of low-dose rapamycin studies. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12422820/

All product photography in this article is owned by Welzo and served from the Welzo media library. Research positions correct at the time of publication.