Prediction of Weight Loss and Regain Based on Multiomic and Phenotypic Features: Results From a Calorie-Restricted Feeding Trial
Authors' conclusion
Baseline multiomic and phenotypic data effectively predict weight and body composition loss and regain, offering insights for personalized weight management.
Study context
- Population
- Adults with overweight/obesity in the LEAN-TIME trial; 12-week calorie-restricted weight-loss phase and 28-week weight-regain phase79 completed regain phase
- Participants
- 88 adults
- Comparison
- Multiomic and phenotypic models vs phenotype-only models
Methodological detailscalorie-restricted feeding trial, prediction modelling (multivariable regression, LASSO)
- Calorie-restricted feeding trial
- Prediction modelling (multivariable regression, LASSO)
Primary result
As reported in the abstract
Prediction of changes in weight, BFM and SLM during weight loss · 12-week weight-loss phase
| Group | R2 weight | R2 body fat mass (BFM) | R2 soft lean mass (SLM) | RMSE weight (kg) | RMSE BFM (kg) | RMSE SLM (kg) |
|---|---|---|---|---|---|---|
| Multiomic and phenotypic model | 0.49 | 0.61 | 0.54 | 1.59 | 1.41 | 0.98 |
Multiomic and phenotypic model
R2 weight0.49
R2 body fat mass (BFM)0.61
R2 soft lean mass (SLM)0.54
RMSE weight (kg)1.59
RMSE BFM (kg)1.41
RMSE SLM (kg)0.98
- Multiomic and phenotypic model vs Phenotype-only model: P < 0.05 — multiomic models significantly outperformed
Further findings
Binary classification of clinically meaningful weight loss (≥5%) · weight-loss phase (12 weeks)
- AUC of the multiomic and phenotypic model
- 0.95
- Sensitivity
- 94.12%
- Specificity
- 86.79%
Prediction of changes in weight, BFM and SLM during weight regain · weight-regain phase (28 weeks)
- R2 for weight, BFM and SLM
- 0.72, 0.73, 0.66
- RMSE for weight, BFM and SLM
- 1.40, 1.62, 0.73 kg
Shared baseline predictors across both phases
- Several key baseline predictors, primarily gut microbes and fecal metabolites (e.g. N-acetyl-l-aspartic acid, Ruminococcus callidus, Bifidobacterium adolescentis), were shared for weight and body composition changes in both phases
Original publication
Prediction of Weight Loss and Regain Based on Multiomic and Phenotypic Features: Results From a Calorie-Restricted Feeding Trial. Diabetes Care. Published 30 June 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →