**Background:** Primary sclerosing cholangitis (PSC) is a rare, progressive liver disease characterized by inflammation and stricturing of the bile ducts, with a median transplant-free survival of 21 years. No medical therapy has been demonstrated to alter disease course. An aetiopathological link between PSC and gut microbial dysbiosis has been suggested, supported by consistent findings of altered gut microbial composition in PSC patients, including reduced colonic alpha diversity and shifts in taxa such as increased Veillonella and Fusobacterium, independent of IBD status. The gut-liver axis, where portal vein transports microbial products directly to the liver, is increasingly recognized as a therapeutic target.
**Methods:** A comprehensive scoping review of PubMed and the Cochrane Library was performed from inception to December 2021, identifying all studies evaluating interventions that manipulate the gastrointestinal microbiome for the treatment of PSC. All eligible papers including RCTs, observational studies, case series, case reports, and letters to the editor were incorporated. Paediatric studies were included.
**Key Results:** Vancomycin has been the most widely studied antibiotic. Two RCTs were identified: Rahimpour et al. (n=29) reported a significant reduction in Mayo Risk Score with vancomycin vs placebo at 12 weeks (−322.03%, p=0.026 vs −45.45%, p=0.337), but no significant change in ALP (−44.49%, p=0.112). Tabibian et al. (n=35) found significant ALP reductions in both low-dose (−46%, p=0.03) and high-dose (−40%, p=0.02) vancomycin groups at 12 weeks. A retrospective propensity-matched study in 88 children found no significant difference in GGT between vancomycin (44 U/L), UDCA (46 U/L), and observation (58 U/L) groups after 1 year (p=0.657). Metronidazole was evaluated in two RCTs: Farkkila et al. (n=80) found a significantly larger mean ALP reduction with metronidazole+UDCA (−337 IU/L, 95% CI [−283, −391]) vs placebo+UDCA (−214 IU/L, 95% CI [−164, −264]) at 36 months (p<0.05), with improved histological stage (34.4% vs 14%, p=0.047) and grade (43.8% vs 16.6%, p=0.014). However, metronidazole was poorly tolerated with increased side effects (18 vs 7, p<0.05). Tabibian et al. found no significant ALP reduction with metronidazole at 12 weeks. Rifaximin (n=16, open-label) showed no meaningful change in ALP (+0.9%, p=0.47). Minocycline (n=16) showed a significant ALP decrease (−330 U/L, p=0.04) but only 2/12 (17%) achieved >50% reduction, and 56% experienced adverse effects. A single probiotic RCT (n=14) found no improvement in ALP or symptoms. FMT was evaluated in one open-label pilot trial (n=10): 3/10 (30%) achieved ≥50% ALP reduction at 24 weeks, and alpha-diversity increased in all patients. A single case report of exclusive enteral nutrition showed complete normalization of liver enzymes in a 13-year-old girl.
**Clinical Implications:** Despite promising early results, particularly for vancomycin, no microbiome-targeted therapy has been demonstrated to improve transplant-free survival in PSC. The lack of validated surrogate biomarkers, small sample sizes, short follow-up periods, and heterogeneity in study designs limit current evidence. Antibiotic therapies may perpetuate dysbiosis and carry risks of resistance and intolerance. FMT and dietary approaches that enhance microbial diversity may have theoretical advantages but require larger studies. The authors conclude that innovative, well-designed clinical trials of microbiome-targeted therapies with long-term follow-up are urgently needed for this orphan disease.