Why do some people lose weight quickly while others see little effect despite the same diet? Why do some maintain weight loss, whereas others regain it rapidly? Researchers from the School of Public Health at Huazhong University of Science and Technology (HUST), led by Professor Liu Gang, may have uncovered the answer hidden within each person’s unique gut microbiome.

Their study, Prediction of Weight Loss and Regain Based on Multiomic and Phenotypic Features: Results From a Calorie-Restricted Feeding Trial, was published in Diabetes Care on Jan 1, with PhD student Li Lin as the first author and Professor Liu as the corresponding author.

The study, based on the randomized controlled Low-Carbohydrate Diet and Time-Restricted Eating (LEAN-TIME) trial, is the first to show that pre-intervention gut microbiota and fecal metabolites can significantly influence both the success of dietary weight loss and the risk of weight regain. Using these insights, the team developed high-performance predictive models for weight loss outcomes and rebound, offering a scientific foundation for personalized dietary interventions.

With overweight and obesity now global public health challenges, and China elevating obesity prevention to a national strategy, sustainable weight management remains a critical concern. Although dietary interventions are increasingly refined, individual responses vary widely.

The team found that healthy low-carbohydrate diets and time-restricted eating affected body weight, composition, gut microbiota, and fecal metabolites, producing additional weight-loss benefits. Notably, 28 weeks after the intervention, reductions in body fat from the low-carb diet persisted.

By integrating multi-omics data, including metagenomics and fecal metabolomics, with machine learning algorithms, the team built predictive models that revealed substantial individual differences. During the intervention, weight changes ranged from a 10.2-kilogram loss to a 1.6-kg gain, and during the subsequent 28-week observation period, weight changes ranged from a 2.0-kg loss to a 14.2-kg regain. Baseline gut microbiota, fecal metabolites, dietary factors, and clinical metabolic traits were strongly correlated with weight loss outcomes and rebound.

Compared with traditional models, the multi-omics approach greatly improved predictive performance, effectively distinguishing participants likely to achieve clinically meaningful weight loss and accurately predicting continuous changes in weight, body fat, and lean mass. For weight regain, the model significantly improved predictions of changes in body composition.

Professor Liu highlighted the potential of the gut microbiota and fecal metabolites as tools for predicting individual risk of weight loss and rebound. The study opens new pathways for precision dietary interventions for obesity and has drawn international attention.

Diabetes Care also featured a commentary by Frank B. Hu, a member of the National Academy of Medicine in the United States, who noted that the study offers novel scientific strategies for precision nutrition.

In addition, Liu contributed as a principal drafter to the Guideline for population-based nutrition and health intervention study, advancing standardized, scientific practices in China’s nutrition research.

Source: School of Public Health, HUST