Fat burning starts early. The body's bigger rewrite waits.

Twelve volunteers who went without food for seven days showed significant changes in more than 1,000 blood proteins, with the broad, system-wide response becoming evident only after roughly three days. The study reveals a second stage of adaptation beyond the body's early switch toward burning stored fat, giving researchers a detailed map of pathways that could have both useful and harmful effects. ^{1 2}

The experiment, published in Nature Metabolism on March 1, 2024, followed five women and seven men through a water-only fast. Researchers measured roughly 3,000 proteins circulating in their blood before the fast, each day during it and afterward. They identified nine distinct patterns of response, with changes involving multiple organs and a striking degree of similarity across volunteers. ²

That timing is the central finding. The body began moving from glucose toward stored fat as a major energy source during the first two or three days. The broader changes in circulating proteins appeared later. Running low on incoming fuel and reorganising the body's molecular activity were related parts of the fast, but they were not the same event. ^{1 2}

The volunteers were healthy, and this was a study of their biological response, not a trial showing that a week without food improves health. Its disease-related findings came from further analyses using genetic and population data. Those produced leads for research, not demonstrated treatment effects. ²

Two changes, on different clocks

The study was conducted by scientists from Queen Mary University of London and the Norwegian School of Sports Sciences, including researchers Maik Pietzner and Claudia Langenberg. Their approach was to follow the fast as it unfolded rather than compare only a starting point with an endpoint. Daily measurements made it possible to distinguish early responses from those that took several days to emerge. ^{1 2}

That distinction matters because a fast has more than one clock. Changes in the body's energy supply can begin before the wider protein response becomes clear. In this experiment, the shift toward stored fat was already underway during the first two or three days, while the system-wide protein changes became evident after about three days of complete calorie restriction. ^{1 2}

Langenberg said the findings showed effects beyond weight loss, but that these became visible only after three days without calories, later than previously thought. Her point was about the timing of the measured response, not proof that every change appearing after that point was a benefit. ¹

The practical implication is that the duration and completeness of the fast are part of the result. These volunteers consumed no calories for a full week. The experiment does not establish that shorter intermittent-fasting schedules produce the same pattern. Nor does it establish a universal switch that flips at precisely the same hour in every person. The finding is a broad response emerging after roughly three days in this group. ^{1 2}

That leaves a clear scientific distinction. A study showing an early change in energy use cannot, on its own, establish the later protein changes found here. And finding those later changes does not, on its own, establish better health. The researchers measured a sequence of adaptations. What those adaptations accomplish is the next question.

More than a number on the scales

Participants lost an average of 5.7 kilograms during the seven-day fast, with a reported standard deviation of 0.8 kilograms. That is a substantial change over a short period, but it is not the most revealing number in the experiment. The molecular measurements show that the response extended beyond weight loss. ^{1 2}

Nor should the 5.7-kilogram figure be read as a measurement of fat lost. The reported figure is body-weight loss. The study separately describes the early shift toward using stored fat, but those two findings do not make every kilogram lost a kilogram of fat. ^{1 2}

Proteins offer a different kind of view. They perform many jobs in the body, and measuring them in blood can provide clues to changes in biological pathways and tissues. By tracking about 3,000 of them, the researchers could look for a coordinated response rather than rely on a single marker. More than 1,000 responded significantly, and the authors described the overall pattern as remarkably conserved across the volunteers. ^{1 2}

The nine response profiles also matter. They show that the measured proteins did not all follow one trajectory through the week. A single before-and-after reading would have been less informative about when different parts of the response emerged. Here, time was part of the measurement. ²

Among the prominent findings were changes in extracellular matrix proteins, which help form the structural material around cells. The researchers also recorded a marked change in tenascin-R, a protein associated with the brain. These observations widen the scope of the experiment beyond an account of fuel use or a falling weight measurement. ²

The tenascin-R finding is striking, but it is still a finding about a protein measured in blood. It does not establish improved brain function, a neurological treatment effect or damage to the brain. Its value is as a clue to a biological response that deserves closer investigation, not as a verdict on what fasting does to the mind. ²

The same distinction applies across the map. A protein level can change clearly and consistently without the experiment establishing whether that change helps, harms or simply allows the body to adapt to a week without calories. The strength of this study is the breadth and timing of the measurements. It is not a ranking of fasting's benefits.

From molecular signals to disease research

To explore possible health consequences, the researchers combined the fasting-related protein measurements with genetic and population data. They estimated potential implications for 212 proteins across about 500 outcomes, looking beyond the volunteer experiment to ask which signals might matter for disease. ^{1 2}

The authors reported possible beneficial and adverse associations. These included putative links involving the protein SWAP70 and rheumatoid arthritis, and HYOU1 and heart disease. The analyses did not demonstrate that fasting treated either condition. They proposed connections that can guide further work. ²

There are therefore two layers of evidence in the paper. The first is direct observation: the volunteers fasted, researchers measured their blood proteins, and many levels changed. The second is an estimate of what some of those changes might imply for health, using information from other datasets. Both are useful, but they answer different questions. ²

The first layer establishes a molecular response in the people studied. The second helps select targets worth testing. It cannot substitute for studies that directly measure whether patients become healthier, whether symptoms change or whether adverse effects outweigh any benefit.

This is where the work's most promising possibility lies. The authors said the molecular findings could help identify pathways that might eventually be targeted without requiring patients to undergo prolonged fasting. Pietzner said an extended fast may not be an option for people with ill health, and that the findings could help inform treatments they could use instead. ^{1 2}

That is a more useful ambition than treating every adaptation to food deprivation as something to copy. The task is to isolate a pathway with a demonstrated benefit, understand its costs and test whether it can be activated safely. A whole-body response is a starting point for that work, not a finished therapy.

What the next experiment must settle

The paper places prolonged fasting in a long history of religious practice and attempts to treat disease, including epilepsy. It also identifies a gap: the systemic human response to extreme calorie restriction has been poorly understood. This experiment addresses that gap by mapping what changed during a full week without food. ²

Its reach is nevertheless defined by its participants. Twelve healthy volunteers can reveal a shared response worth investigating, but they cannot establish how it varies across patients, ages, medications or other health circumstances. Those differences matter especially if the eventual goal is treatment rather than understanding adaptation. ²

The measurements after the fast offer another direction for investigation: recovery. Researchers need to establish how much of the response reverses when eating resumes and how long any remaining changes persist. A signal that appears during deprivation may have a different significance from one that continues afterward.

Safety is also unresolved in the available material. It does not establish what medical supervision and safety measures accompanied the volunteer fast, or what adverse events, if any, were recorded. It therefore provides no basis for declaring a seven-day water-only fast safe or appropriate for a particular person.

The next scientific step is not merely to repeat the weight-loss result. It is to test which protein changes have meaningful effects, verify the proposed disease links against clinical outcomes and determine whether useful pathways can be reproduced without prolonged food deprivation. The study has supplied a timetable and a set of targets. Researchers now have to find out which targets are worth hitting.