Review explores deep phenotyping for personalized nutrition in metabolic disease

A review in Nature Reviews Endocrinology explores how deep phenotyping could inform personalized nutrition for the prevention and management of metabolic disorders. It considers data from large cohort studies, omics technologies and wearables, alongside clinical history, lifestyle and personal characteristics, to stratify individuals by metabolic variation and tailor dietary advice.

The review emphasizes that obesity is heterogeneous: excess adiposity does not uniformly translate into metabolic disease. Differences in adipose tissue biology, ectopic fat distribution and systemic inflammation contribute to distinct metabolic phenotypes, supporting assessment beyond BMI alone. Variation in individual dietary responses is presented as a limitation of population-level dietary guidance. The authors also discuss self-supervised foundation models trained on large-scale metabolic data as potential tools for predicting and simulating dietary responses.

For clinicians, the review outlines a possible framework for more individualized metabolic risk assessment and nutrition advice, rather than reporting a tested treatment protocol. The supplied subscription preview contains no sample sizes, quantitative outcomes or comparative efficacy estimates. It acknowledges challenges to widespread clinical and public health adoption, but does not detail them in the available text.