Treatment & Procedure

Fasting & autophagy

Biohacking

Intermittent fasting shifts mealtimes; autophagy is considered its most important promise. In the longest comparative study, both groups lost considerable weight – but the time window no more than the equally large calorie deficit, and an increase in autophagy in humans has so far only been hinted at in one exploratory analysis.

What it is about

Intermittent fasting means planned breaks from eating, from a daily time window such as 16:8 to fasting for several days. The expectation: the body switches from growth to clean-up and starts cell recycling, autophagy.

Two questions need to be examined: does the time window bring more than the amount of calories, which drops anyway? And does autophagy measurably start up in humans?

What autophagy measurements in humans show

Autophagic flux, that is, the actual degradation throughput in the cell, has so far been measured in humans under fasting in a single randomized trial, and in blood cells. In 121 people with obesity, the change after 6 months differed between the fasting group and the control group (p = 0.04). The analysis was post hoc, there was no difference after 2 months, and within the fasting group the flux did not rise significantly (Bensalem 2025). Otherwise, only surrogate markers are available, laboratory values that are supposed to reflect the process.

The most cited older measurement captured the mRNA expression of LC3A in whole blood cells, that is, a reading of genetic information and not degradation itself.

What has been tested in humans

A systematic search in ClinicalTrials.gov and Europe PMC found no randomized trial with mortality or cardiovascular events as an endpoint. All of the evidence concerns weight or surrogate values.

The recurring finding: intermittent fasting works about as well as classic calorie restriction, not better.

What is well supported

Both approaches led to considerable weight loss. In the 12-month study by Liu 2022 (n=139, obesity), the group with a time window from 8 a.m. to 4 p.m. lost −8.0 kg (95% CI −9.6 to −6.4), the group with calorie restriction alone −6.3 kg (−7.8 to −4.7); both groups were prescribed 1,200 to 1,800 kcal. So intermittent fasting works – the net difference of −1.8 kg (−4.0 to 0.4; p=0.11) only shows that it does not work better than an equally large calorie deficit. These are two different statements.
The decisive comparative question has been robustly answered. The study ran over twelve months, was randomized, had a predefined primary endpoint, achieved 84.9 percent follow-up and was publicly funded, i.e. free of manufacturer interests. Whether the time window contributes anything beyond the amount of calories is therefore no longer an open question but a tested one.
A short-term metabolic effect has been measured. In the crossover study by Jamshed 2019, 24-hour glucose in eleven overweight people was 4 mg/dl lower over four days (standard deviation 1; p=0.0003). The trial is small and short, but the measurement is clean and the effect clear.

What the studies show

Twelve months: the time window adds nothing on top

In a randomized trial with 139 people with obesity (Liu 2022), time-restricted eating from 8 a.m. to 4 p.m. plus calorie restriction was compared with calorie restriction alone over 12 months, both with 1,200 to 1,800 kcal. The primary endpoint, weight change, was not reached: the net difference was −1.8 kg (95 percent confidence interval −4.0 to 0.4; p=0.11). The time-window group lost −8.0 kg (−9.6 to −6.4), the comparison group −6.3 kg (−7.8 to −4.7), with 84.9 percent follow-up. Waist circumference, BMI, body fat, blood pressure and metabolic risk factors did not differ either. The study was publicly funded.

TREAT: no weight effect, but loss of lean mass

In the TREAT study (Lowe 2020), 116 people received either a 16:8 time window without calorie targets or the control condition over 12 weeks. The group difference in weight was not significant: −0.26 kg (95 percent confidence interval −1.30 to 0.78; p=0.63). A safety signal stood out: in the subgroup of 50 people whose body composition was measured on site, the appendicular lean mass index, i.e. the muscle mass of the arms and legs relative to body height, fell by −0.16 kg/m² (−0.27 to −0.05; p=0.005) – to the disadvantage of the time-window group.

Autophagy: only gene expression, only 11 people, only 4 days

The crossover study cited as evidence of autophagy included 11 overweight people over 4 days. What was measured was LC3A mRNA in whole blood cells, not autophagic flux. The signals were also inconsistent: LC3A and SIRT1 raised in the morning (p less than 0.04), MTOR likewise raised in the evening (p=0.007). 24-hour glucose was 4 mg/dl lower (standard deviation 1; p=0.0003). Gene expression in blood does not allow conclusions about degradation throughput or about processes in the liver, muscle or brain.

Where the data stop

The time window brings no additional benefit beyond the calorie deficit. Over 12 months, neither weight nor the metabolic risk factors differed between the time window and calorie restriction alone.
Autophagy has barely been measured in humans. The only flux measurement in blood cells shows only a post-hoc group difference after 6 months. The rest of the data base is gene expression in blood in 11 people over 4 days – and even that came out contradictory. Processes in the liver, muscle or brain have not been measured in humans.
No hour threshold is supported. The claim that autophagy begins after a certain fasting duration has no basis in humans.
No data on mortality or cardiovascular events. Such endpoints have not been studied in randomized trials of intermittent fasting.

Status, approval and legal

Intermittent fasting is not an approved treatment and not a product, but an eating pattern. The evidence reviewed contains no regulatory assessment.

The funding is noteworthy: the 12-month study was publicly funded, i.e. free of manufacturer interests – and came out negative. The strongest objection to an additional benefit of the time window thus comes from the most independent source.

Safety

The clearest signal concerns muscle: in TREAT, the appendicular lean mass index fell measurably under the 16:8 time window without calorie targets in the subgroup of 50 people measured on site; there, a relevant part of the weight loss was due to muscle.

Intermittent fasting is not suitable for pregnant women, breastfeeding women, children, underweight people and people with a history of eating disorders. Fasting for several days requires medical supervision. This text contains no usage recommendation and does not replace medical advice.

BK-Score Supported, with caveats

Human evidence		6
Mechanism		5
Safety data		4
Hype gap		4
Track record of use		8

Plenty of evidence for surrogate markers, none for hard endpoints: an umbrella review of 23 meta-analyses (Sun et al., eClinicalMedicine 2024) found high certainty of results for only 10 of 103 significant associations – including waist circumference, fat mass, fasting insulin and blood lipids. Mortality and cardiovascular events were not studied at all. The recurring finding: intermittent fasting works about as well as classic calorie restriction, not better. The popular claim that autophagy switches on after a certain number of hours is not supported in humans.

The score rates the state of knowledge, not the substance. “Safety data 9” means well studied – not harmless. “Track record of use 9” means used long and widely – that is not proof of efficacy.
Subjective assessment by Biohacking Kompakt based on published scoring rules – not a scientific rating and not a medical recommendation. Rules and all ratings (German)

Frequently asked questions about fasting & autophagy

Does intermittent fasting bring more than eating less?

Not according to the longest comparative study. Over 12 months, a randomized trial with 139 people with obesity compared an eating window from 8 a.m. to 4 p.m. plus calorie restriction with calorie restriction alone. The net difference in weight was minus 1.8 kg and was not significant. Waist circumference, body fat and blood pressure did not differ either.

After how many hours does autophagy begin?

This number does not exist for humans. Autophagic flux under fasting has so far been measured in only one study, in blood cells and without comparing different fasting durations. The commonly cited hour figures are based on model organisms and indirect markers. A threshold at which autophagy demonstrably sets in in humans cannot be derived from the available data.

Has autophagy ever been measured in humans?

Barely, and only in blood cells. In a randomized trial with 121 people with obesity, the change in autophagic flux in blood cells after 6 months differed between the fasting and control groups; the analysis was post hoc, and within the fasting group the value did not rise significantly. The older, most cited study with 11 people over 4 days measured only LC3A gene expression in blood cells, and the signals were contradictory.

Do you lose muscle mass with intermittent fasting?

In one study, exactly that was the case. In TREAT, 116 people received a 16:8 window without calorie targets over 12 weeks. Weight did not differ significantly. In the subgroup of 50 people whose body composition was measured on site, the appendicular lean mass index fell by 0.16 kg per square meter more under the time window than in the comparison group. So a relevant part of the loss was muscle.

Does intermittent fasting lower the risk of death?

There are no data on this. A systematic search of trial registries and literature databases found no randomized controlled trial that studied mortality or cardiovascular events as an endpoint. All of the available evidence concerns weight or laboratory values. Claims of life extension through intermittent fasting are therefore unsupported in humans.

Who is intermittent fasting not suitable for?

It is not suitable for pregnant women, breastfeeding women, children, underweight people and people with a history of eating disorders. Fasting for several days requires medical supervision. Added to this is the finding on muscle loss under a time window without calorie targets. This page gives no usage recommendation and does not replace medical advice.

The podcast episode (in German)

Episode 59