From One-Size-Fits-All to Precision Nutrition: Metabotype-Based Stratification of Gut Health Responses

Precision nutrition aims to optimise health by tailoring dietary interventions to individual biological characteristics. In ageing populations, gut microbiome composition and intestinal barrier integrity are key modulators of immune and metabolic function, yet both decline with age, contributing to chronic low-grade inflammation and disease risk. Nutritional interventions such as omega-3 fatty acids, plant bioactives, and postbiotics can improve these parameters, but inter-individual variability limits consistent benefit. Metabotyping—classifying individuals by metabolic phenotype—offers a strategy to personalise interventions based on metabolomic fingerprints reflecting diet–microbiome–host interactions (1) . This study applied a metabotyping framework to integrate urinary metabolomics, gut microbiome sequencing, and gut barrier biomarkers in older adults, to identify metabolic phenotypes predictive of response to nutritional supplementation. A randomised, double-blind, placebo-controlled trial (ISRCTN75484092) was conducted in healthy adults aged ≥60 years (n = 60). Participants received 12 weeks of either Cerbella™ (EPA, DHA, ginseng, green tea extract, phospholipids) or PoZibio™ (heat-inactivated Lactobacillus paracasei D3.5) or placebo. Biological samples collected at baseline and post-intervention included stool (whole-genome microbiome sequencing), plasma (short-chain fatty acids; SCFAs), and urine (untargeted metabolomics by FIE-MS). Metabolomic data underwent preprocessing, normalisation, and unsupervised clustering using k-means and principal component analysis to define metabotypes. Associations between metabotypes, microbiome α-diversity, and gut barrier biomarkers (zonulin, sCD14, mucin, calprotectin) were assessed using ANOVA and multivariate regression. Machine-learning models (random forest) were trained to predict intervention response based on baseline metabolic signatures. Preliminary analyses identified distinct metabolic clusters among participants, each defined by unique metabolomic fingerprints linked to diet and microbiome activity. Early integration with microbiome data indicates that these metabotypes differ in bacterial diversity and in taxa associated with SCFA and polyphenol metabolism. Participants with plant- and fibre-associated metabolic profiles appear to show more favourable gut barrier biomarker patterns (e.g., lower zonulin and sCD14), though validation is ongoing. Initial modelling suggests that baseline metabotypes may predict responsiveness to Cerbella™ and PoZibio™ supplementation, particularly in relation to microbiome composition and gut integrity. These interim findings support the feasibility of metabotyping as a stratification tool for personalised nutrition in older adults. This study provides preliminary evidence that metabotyping can stratify older adults according to gut microbiome composition, barrier integrity, and response to nutritional supplementation. Integration of untargeted metabolomics, microbiome sequencing, and biomarker data identified reproducible metabolic phenotypes predictive of intervention outcomes. By incorporating metabotyping into nutrition research, personalised dietary strategies can be developed to enhance efficacy, reduce non-responder rates, and support healthy ageing. These results advance the translation of multi-omics science into actionable tools for precision nutrition (1,2) .