Metabolic modeling reveals determinants of synbiotic efficacy in a human intervention trial

Synbiotic interventions show variable effects across individuals, likely driven by ecological interactions with the endogenous microbiota and the host diet. Rationally predicting individual-specific success or failure of probiotic and prebiotic interventions remains an outstanding challenge. In this study, we leverage microbial community-scale metabolic models (MCMMs) to predict probiotic engraftment and shifts in microbiota-mediated short-chain fatty acid (SCFA) production in response to a synbiotic intervention. Using data from a placebo-controlled synbiotic intervention trial, involving a cocktail of five probiotic strains and the prebiotic inulin, we validate model engraftment predictions with quantitative PCR (qPCR), demonstrating that MCMMs accurately predict probiotic engraftment outcomes in the treatment group with over 85% accuracy. Engraftment varied by species, withAkkermansia muciniphilaandBifidobacterium infantisdisplaying higher engraftment rates thanClostridium beijerinckii, Anaerobutyricum hallii, andClostridium butyricum. Furthermore, MCMMs predicted significant increases in butyrate and propionate production following synbiotic treatment. MCMM-predicted changes in propionate production in the treatment group were negatively associated with changes in C-reactive protein (CRP), a blood marker of systemic inflammation, from baseline to 12 weeks after the synbiotic intervention. Finally, we explore MCMM-predicted responses to a wider range of synbiotic combinations in a larger observational cohort, suggesting that personalized prebiotic selection can augment probiotic efficacy. These findings highlight the potential of metabolic modeling to inform precision microbiome therapeutics.