Gut bacteria help predict whose fasting weight loss lasts

Prolonged fasting has a long history and a recent surge in popularity, with claims that several days without food can reset metabolism. Yet controlled data in healthy adults are limited, especially with follow-up long enough to show whether any benefit survives the return to normal eating. It is also clear that people respond very differently to the same fast, and nobody has been able to say in advance who will keep weight off. This trial asked both questions: what a 5-day fast does over 12 weeks, and whether baseline characteristics, including the gut microbiome, can predict individual response.

In the randomised, waitlist-controlled LEANER study, 38 healthy adults completed a 5-day fasting intervention and were followed for 12 weeks. The team tracked body mass index and body composition, gut microbiome composition, and metabolites in both plasma and stool. Changes were analysed with regression-based models and multivariate testing. They then trained a machine learning model on baseline microbiome and clinical data to predict 12-week weight response, and tested it in three independent fasting cohorts: people with metabolic syndrome, patients with multiple sclerosis exposed to repeated fasting, and healthy volunteers fasting for 6-12 days.

The fast reduced body mass index acutely, and the loss was driven predominantly by fat mass rather than lean tissue. Those improvements partially persisted at the 12-week follow-up. Fasting also produced marked shifts in gut microbiome composition and in plasma and faecal metabolites, and how microbial diversity changed depended on how diverse a person's microbiome was to begin with. The prediction model combining baseline microbiome and clinical variables forecast 12-week weight response, with its most prominent predictors being an unclassified Faecalibacterium species, Oscillibacter sp. 50_27, LDL cholesterol and systolic blood pressure. Crucially, the model generalised to all three external cohorts.

This is one of the few controlled trials of prolonged fasting in healthy adults with follow-up beyond the fast itself, so it fills a genuine gap. The finding that fat mass drove the acute loss, and that some benefit persisted at 12 weeks, is encouraging for fasting research, though the trial was small. The microbiome prediction result extends a broader theme in nutrition science: responses to the same intervention differ between people in ways gut bacteria may help explain. Larger trials are needed before baseline microbiome testing could sensibly guide anyone's fasting decisions.

If you have ever wondered why one person keeps weight off after a fast while another regains it all, this trial suggests part of the answer may be sitting in the gut before the fast even begins. It also offers a modestly reassuring picture of a supervised 5-day fast in healthy adults: the acute loss was mainly fat, and some improvement remained 12 weeks later. None of this means fasting suits everyone, and a study of 38 people cannot settle that. It is a reason to expect individual variation rather than a universal result.