Why Fecal Transplants Work Unevenly in IBD and How Precision Microbiome Therapy Could Fix It

Inflammatory bowel disease (IBD), which includes ulcerative colitis (UC) and Crohn's disease (CD), is a chronic condition with no cure. Researchers believe it arises from altered gut microbiota interacting with genetic susceptibility. The success of fecal microbiota transplantation (FMT) in recurrent Clostridioides difficile infection raised hopes for microbiome-based treatment in IBD. This review asks what has been learned from IBD trials so far and how a more precise approach could be built.

This is a narrative review of preclinical and clinical studies, not a new trial. The authors argue that C. difficile infection and IBD are mechanistically different. In C. difficile, FMT mainly restores colonization resistance against a single pathogen, so almost any healthy donor works. IBD is not an infection but a complex interplay between diverse microbes and host susceptibility, so the specific strains and their functions matter. Several commercial live biotherapeutic products (LBPs) approved or successful for C. difficile have missed primary endpoints in UC, even when strain-level engraftment was documented.

On efficacy, a meta-analysis of 12 studies with 550 participants found higher clinical remission with FMT than control in UC (risk ratio 1.79, 95% CI 1.13–2.84). Only one randomized trial exists in CD, and it looked at maintenance of remission. Donor variability appears to be a major source of inconsistent results. In one single-donor UC trial, 4 of 8 donors accounted for all 6 remissions, while 4 donors used in 25 of 51 recipients produced none. Another trial with careful donor selection still failed. Pooled-donor strategies tend to improve efficacy but obscure which donor features matter. Responses were linked to taxa such as Odoribacter, Faecalibacterium, Eubacterium hallii and Roseburia inulivorans, and to the absence of Fusobacterium, Sutterella and E. coli. A lyophilized product from effective donors, given after antibiotic pre-conditioning in the LOTUS trial, also achieved remission.

Recipient and protocol factors also shape outcomes. Responders were more often those with mild, left-sided disease, lower fecal calprotectin and no prior biologic or advanced therapy. They also tended to have higher baseline species richness, although strain engraftment was easier in patients with lower diversity. Colonoscopic delivery appeared more successful than other routes, and the effect of antibiotic pre-conditioning is debated: vancomycin improved engraftment of the spore-based LBP SER-287, yet that product still failed clinically. High bacterial density alone did not restore donor efficacy. The authors call for standard pharmacokinetic and pharmacodynamic characterization to guide LBP dosing.

The review also frames IBD pathobionts, such as adherent-invasive E. coli, as therapeutic targets. The abstract points to diet, xenobiotics and strain-specific host interactions as further levers. Key caveats are that most FMT trials are small and heterogeneous, so effect sizes are likely overestimated, and the ideal "superdonor" criteria and LBP-host pairing rules remain unknown. The provided text was truncated partway through the pathobiont section, so later content is not covered here.