Faecal microbiota transplantation treatment efficacy correlates with specific donor microbiota bile acids

Background Clostridioides difficile infection (CDI) is a major cause of hospital-acquired infections. Faecal microbiota transplantation (FMT) is a safe and effective treatment to prevent CDI recurrence, but treatment efficiency depends on patient and donor characteristics. To investigate how donor microbiome and metabolite profiles are associated with FMT treatment efficiency in patients with CDI, data from a Danish cohort where faeces samples were collected from 47 donors and 102 patients with CDI before FMT, and one and eight weeks after FMT, were analysed. Analyses included microbiome composition analysis, alpha diversity, beta dispersion, and bile acid concentrations. Based on the weighted balance of four bile acids, a predictive model was created to predict variance in treatment efficiency using coda4microbiome . Results CDI resolution in patients increased with higher abundances of Clostridia and Clostridiales in donor faeces, was positively associated with glycocholic acid, and negatively associated with 3-oxo-chenodeoxycholic acid in donor faeces. Using our predictive model, a linear regression was found where the predicted value explained 86% of the variance in efficiency score. When applying our model to the dataset with alternative filtering, it predicted a donor efficiency score (S e ) with an average treatment efficiency of 73% for all donors, 83% for high-S e donors and 67% for low-S e donors. Conclusions Donor-specific microbiome and metabolite compositions were linked to higher FMT success rates in the treatment of CDI, suggesting that predictive modelling can help select FMT donors. Our findings highlight a potential method for donor selection or optimization in FMT treatment of CDI.