Oral Vancomycin Is an Effective Therapy in Adult Patients With Inflammatory Bowel Disease and Primary Sclerosing Cholangitis.

Introduction
Primary sclerosing cholangitis (PSC) is an immune-mediated chronic liver disease characterized by inflammation and fibrosis of the biliary tree, resulting in strictures and eventually a sizeable percentage of patients progressing to liver cirrhosis requiring transplantation. It is also associated with a marked increased risk of hepatobiliary and colorectal malignancies.1,2 The etiology of PSC remains poorly understood.3 A strong link, however, exists between PSC and inflammatory bowel disease (IBD). Approximately 3-quarters of patients with PSC have concomitant IBD. PSC-IBD most commonly presents as ulcerative colitis (UC), though often with distinct features including rectal sparing, active right-sided colitis, and backwash ileitis.4,5
Bacterial translocation and dysbiosis have been observed in PSC pathogenesis. The gut microbiome has been shown to differ in patients with PSC compared to healthy controls and non-PSC-IBD.6,7 Currently, there is no approved therapy to improve the course or slow the progression of PSC.8 Furthermore, patients with PSC-IBD are at increased risk for chronic pouchitis following ileal pouch-anal anastomosis, and patients with PSC-associated pouchitis and enteritis may represent a unique phenotype that does not typically respond to conventional antibiotics, such as ciprofloxacin and metronidazole.9
Oral vancomycin, a nonabsorbable bactericidal glycopeptide antibiotic active against Gram-positive bacteria,10 has gained particular interest. Vancomycin has been shown to affect tumor necrosis factor-alpha (TNF-α) pathways as an immunomodulator, which is an important pro-inflammatory cytokine in the pathogenesis of IBD.11 In a propensity score–matched cohort from the Paediatric PSC Consortium, a large international retrospective cohort study of pediatric patients with PSC, gamma-glutamyltransferase, liver fibrosis, and transplant listing was similar at 1 year in vancomycin-treated, ursodeoxycholic acid (UDCA)-treated, and placebo cohorts.12 However, 2 small randomized controlled trials and a few cohort studies have reported beneficial liver effects,13,14 including successful treatment of recurrent PSC following liver transplantation.15 A recent supporting statement published by the American Association of the Study of Liver Disease did not make a recommendation for or against the use of oral vancomycin for PSC.16
Although many patients with PSC-IBD experience mild IBD-related symptoms, some may have disease that proves refractory to advanced therapies. The IBD phenotype found in approximately 80% of patients with PSC has unique features with an elevated risk of developing poorly controlled IBD or colorectal neoplasia necessitating colectomy.17 Even mild inflammation is clinically significant in PSC-IBD due to the increased risk of colorectal malignancy.18 A few case reports and case series have specifically examined vancomycin for treating PSC-IBD. While the observations in these studies include improvement and/or normalization of objective markers of inflammation including fecal calprotectin and endoscopic findings are positive preliminary findings, they are limited by the absence of a comparator or control group and small sample size.10,19, 20, 21, 22, 23, 24, 25 In this setting, we aimed to examine our institutional experience with the use of oral vancomycin on the clinical features and endoscopic outcomes of patients with PSC-IBD.

Methods

Patient Population
This retrospective study was approved by the Mayo Clinic Institutional Review Board. Using informatics tools to search the electronic medical record, all patients were identified across multiple Mayo Clinic sites (Rochester, Minnesota; Scottsdale, Arizona; and Jacksonville, Florida) using International Classification of Diseases-9 and International Classification of Diseases-10 diagnostic codes for IBD and PSC. Those who were treated with oral vancomycin between January 1992 and August 2024 comprised the study group. This was followed by manual review of individual patient charts as permitted by Minnesota Health Records Act, 144.295, which allows patients to opt out of retrospective research.
We then abstracted relevant demographic, clinical, endoscopic, and histologic outcomes. The date of first visit for IBD or PSC diagnoses (index) was recorded. We excluded patients without a confirmed IBD diagnosis; those who received vancomycin to treat Clostridioides difficile or other infections; those who were not prescribed oral vancomycin; and those who discontinued oral vancomycin prematurely (less than 1 week of use) due to side effects. This research was supported by internal Center for Clinical and Translational Science funding.

Inflammatory Bowel Disease Characteristics
IBD was diagnosed based on the clinical diagnostic criteria as per the treating gastroenterologist and review of the medical record for confirmation of endoscopic, radiographic, and/or histologic findings and then subcategorized as having Crohn’s disease or UC. The date of IBD diagnosis was defined by the earliest pathologic confirmation by our institution.
Specific disease characteristics including extent of involvement, disease behavior (for Crohn’s disease), presence of backwash ileitis, or perianal disease were abstracted in addition to endoscopic and histologic severity at the time of IBD diagnosis if this was known. IBD-related complications including prior surgical interventions were noted.
Prior medication use for the management of IBD was recorded, including prior exposure to corticosteroids, 5-aminosalicylates (5-ASA) products or sulfasalazine, immunomodulators (6-mercaptopurine, azathioprine, or methotrexate), TNF-α inhibitors (infliximab, adalimumab, certolizumab pegol, or golimumab), vedolizumab, anti-interleukin (IL)-12/23 and anti-IL-23 agents (ustekinumab or risankizumab), Janus kinase inhibitors (tofacitinib or upadacitinib), or sphingosine-1-phosphate agents (ozanimod or etrasimod).

Primary Sclerosing Cholangitis Characteristics
PSC was diagnosed based on clinical diagnostic criteria as per the treating gastroenterologist and review of the medical record for confirmation of radiographic and/or histologic findings.26 The date of PSC diagnosis was defined by the earliest radiographic and/or histologic confirmation by our institution.
Specific disease characteristics including the presence of autoimmune hepatitis overlap, duct type, and laboratory parameters at diagnosis including serum liver tests (alanine transaminase, aspartate aminotransferase, and alkaline phosphatase [ALP]), albumin, platelet count, and C-reactive protein were abstracted. PSC-related complications including cholangiocarcinoma and prior liver transplantation were noted.
Prior medication use for the management of symptoms related to PSC was recorded, including prior treatment with UDCA.

Vancomycin Use in Primary Sclerosing Cholangitis Inflammatory Bowel Disease
Treatment dates for initiation and discontinuation of oral vancomycin were recorded. Dose, route, and frequency of administration of oral vancomycin were recorded. Oral vancomycin use was subcategorized into total daily dose received (0–500 mg daily and >500 mg daily).
Clinical improvement was assessed by review of clinical notes for improvement in bowel habits, rectal bleeding, and abdominal pain. Laboratory parameters prior to and following discontinuation of oral vancomycin were noted including serum C-reactive protein and ALP, as well as fecal calprotectin when available in the medical record. Endoscopic and histologic activity of the colon, subcategorized as none/inactive, mild, moderate, or severe after review of endoscopic and pathology reports, was noted prior to and following initiation of oral vancomycin. Potential side effects related to oral vancomycin and reasons for discontinuation were recorded.

Statistical Analysis
Patient characteristics and clinical data are presented as median and interquartile range (IQR) (25th and 75th percentiles) for continuous variables and number and percentage for categorical variables. Associations between clinical outcomes and the 2 levels of total daily dose of oral vancomycin used a Wilcoxon rank sum test for continuous and ordered outcomes and a chi-square test for discrete outcomes.
Conditioning on disease activity prior to oral vancomycin initiation (inactive/mild and separately for moderate/severe), we compared the post-vancomycin disease activity, worsening and then improvement, between the 2 levels of total daily dose of oral vancomycin using a Fisher’s exact test.
The cumulative probability of oral vancomycin discontinuation was estimated using the Kaplan-Meier method, displayed as 100% minus the survival estimate. A Cox proportional hazards model was used to assess the association of dichotomized daily oral vancomycin dose (0–500 mg vs >500 mg) with discontinuation of oral vancomycin.
The alpha-level was set at 0.05 for statistical significance. All analyses were performed using SAS version 9.4.

Results
The initial data search identified 417 patients with suspected diagnoses of PSC-IBD who were treated with oral vancomycin and fulfilled the inclusion/exclusion criteria of the study. After manual review, 90 patients with PSC-IBD treated with oral vancomycin were included in the final analysis, and 327 were excluded due to the absence of IBD diagnosis, use of vancomycin to treat C difficile or other infections, lack of an oral vancomycin prescription, and premature discontinuation of oral vancomycin due to possible side effects (Figure 1).

Demographics and Clinical Characteristics
Baseline characteristics including demographic data are summarized in Table 1. In total, 90 patients with PSC-IBD who were treated with oral vancomycin were identified and stratified into 2 groups based on the total daily dose of oral vancomycin: 0–500 mg and >500 mg daily. A majority had UC (80.0%). Most had no prior smoking or tobacco use history (92.0%). The median age at IBD diagnosis was 16.4 years (IQR: 12.5, 24.0), with 40.0% diagnosed in adulthood (≥18 years). The median age at PSC diagnosis was 17.4 years (IQR: 13.8, 31.4).
Oral vancomycin was initiated at a median age of 22.7 years (IQR: 15.1, 36.4), corresponding to a median interval of 2.9 years (IQR: 0.7, 8.6) after IBD diagnosis and 1.7 years (IQR: 0.3, 6.0) after PSC diagnosis.
Prior medical therapies were common including UDCA (68.9%), corticosteroids (82.2%), 5-ASA (82.2%), and immunomodulators (47.8%). Biologic exposure was also common, with 42.2% receiving anti-TNF agents, 30.0% vedolizumab, and 22.2% IL-12/23 and IL-23 inhibitors. Use of Janus kinase inhibitors (4.4%) or sphingosine-1-phosphate agents (0%) was uncommon.
Patients receiving >500 mg/d of oral vancomycin (n = 32) were more likely to be female (31.3% vs 22.4%) and have UC (87.5% vs 75.9%) compared to those on ≤500 mg/d (n = 58). Median ages at IBD and PSC diagnosis were somewhat lower in the higher-dose group, and fewer were diagnosed with IBD in adulthood (25.0% vs 48.3%).

Outcomes With Use of Oral Vancomycin
Outcomes with the use of oral vancomycin are summarized in Table 2. For the entire cohort, the median duration of therapy with oral vancomycin was 1172 days (IQR: 392.5, 1900.5). The median time from obtaining fecal calprotectin was 148 days (IQR: 366.0, 23.0), and endoscopic evaluation was 144 days (IQR: 350.0, 44.0) prior to the start of oral vancomycin. The median time from the start of vancomycin to follow-up fecal calprotectin was 254.5 days (IQR: 129.0, 518.5), and follow-up endoscopic evaluation was 310.5 days (IQR: 161.0, 586.0).
Among 90 patients with PSC-IBD treated with oral vancomycin, 50.0% demonstrated clinical improvement of IBD, and 12.2% maintained clinical remission of IBD, with no significant differences between the low- (0–500 mg/d) and high-dose (>500 mg/d) groups (P = .72).
Endoscopic activity of the colon improved following oral vancomycin initiation. Overall, 60.3% of patients achieved inactive disease compared with 18.8% at baseline (P = .0113). The median change in endoscopic activity score was −1.0 (IQR: −2.0, 0.0), which did not differ by dose group (P = .19) (Table 2). Among patients with inactive or mild endoscopic activity prior to oral vancomycin use, 17.7% in the low-dose group (≤500 mg/d) worsened to moderate or severe activity compared with 0% in the high-dose group (>500 mg/d) (P = .26). Conversely, among those with moderate or severe endoscopic activity at baseline, the majority improved to inactive or mild disease following treatment, with similar rates in the low- and high-dose groups (73.3% vs 72.7%, respectively; P = 1.0) (Table 3). Histologic activity of the colon also improved, with inactive or mild disease observed in 86.2% post-treatment compared to 57.7% pretreatment. Patients receiving higher doses were more likely to achieve inactive disease after treatment, whereas the reduction in histologic activity score was greater in the lower-dose group (P = .03) (Table 2). Following initiation of oral vancomycin, clinical disease activity of IBD improved overall, with the proportion of patients in clinical remission (inactive disease) increasing from 9/54 (17%) pretreatment to 33/54 (61%) posttreatment. Histologic activity of the colon also decreased, with inactive histology increasing from 4/44 (9%) pretreatment to 27/44 (61%) posttreatment.
Biochemical and inflammatory markers decreased after oral vancomycin use. The median ALP declined from 238.0 U/L (IQR: 144.0, 401.0) pretreatment to 170.0 U/L (IQR: 108.0, 284.0) posttreatment, and median change in ALP was −48.5 U/L (IQR: −151.0, −3.5), though differences between dose groups were not significant (P = .08 for posttreatment values; P = .57 for change). Similarly, fecal calprotectin decreased from 485.0 μg/g (IQR: 224.0, 1000.0) to 67.7 μg/g (IQR: 31.3, 212.0), without significant differences between groups (P = .63 for posttreatment values; P = .66 for change) (Table 2).
Stratified by total daily dose, 11 and 19 patients in the 0–500 and >500 mg groups discontinued use of oral vancomycin during the study. Five-year cumulative probability of oral vancomycin discontinuation was 81.0% (95% confidence interval [CI]: 69.0, 94.1) among 0–500 mg patients and 70.2% (95% CI: 51.3, 93.9) among patients initially treated with >500 mg. There was no significant difference in discontinuation of oral vancomycin. A patient treated with >500 mg, relative to 0–500 mg, had approximately 60% increased likelihood of discontinuation (hazard ratio = 1.62; 95% CI: 0.64, 4.11; P = .31) (Figure 2).

Safety and Discontinuation of Oral Vancomycin
Seventeen patients (18.8%) discontinued oral vancomycin during the study period. The median time from vancomycin start to discontinuation (for those that discontinued) was 1181 days (IQR: 398.0, 2065.0). Reasons for discontinuation include lack of efficacy for PSC (35.3%) or IBD (17.6%); side effects included brain fog (11.8%), nausea (5.9%), post-liver transplant lymphoproliferative disorder (5.9%), unknown side effects (5.9%), cost (5.9%), and other unknown reasons (11.8%).

Discussion
In this cohort of patients with PSC-IBD, we observed that oral vancomycin was associated with meaningful improvements in clinical, endoscopic, and histologic disease activity of IBD involving the colon. Approximately half of patients achieved clinical improvement, and more than 60% demonstrated inactive endoscopic disease following oral vancomycin initiation. Histologic remission or improvement was also common, particularly in patients receiving higher doses of vancomycin. Serum biochemical and inflammatory markers, including ALP and fecal calprotectin, declined after treatment, further supporting the contributory effect of oral vancomycin. Importantly, Kaplan-Meier analysis demonstrated comparable oral vancomycin-free survival between patients treated with ≤500 mg/d and those receiving >500 mg/d, suggesting durable responses may occur across a range of doses.
Our findings align with prior studies demonstrating clinical and biochemical benefits of oral vancomycin in PSC-IBD.8,12, 13, 14,18,24,25 Consistent with recent data from the Paediatric PSC Consortium18 and retrospective analyses in adults,27,28 we found that oral vancomycin improved colonic inflammation, with parallel improvements in endoscopic and histologic scores. This effect may reflect oral vancomycin’s immunomodulatory properties, including expansion of regulatory T cells,11 as well as its impact on gut microbial composition and function.6,7 Notably, both low- and high-dose therapy resulted in similar clinical and oral vancomycin-free survival outcomes, though patients on higher doses had numerically greater histologic inactivity, suggesting dose optimization and individualized therapy may be warranted.20
PSC-IBD represents a distinct phenotype characterized by extensive colitis, right-sided predominance, and increased colorectal cancer risk.4,5 Traditional therapies for IBD, including 5-ASA, immunomodulators, advanced therapies, and UDCA, have limited efficacy in this population.1,8 Increasing observational and mechanistic evidence supports the role of oral vancomycin as a unique therapy targeting intestinal and likely liver.8,10,13,14,17,18 Case reports and series have demonstrated efficacy even in refractory IBD, post-liver transplantation, and pouchitis associated with PSC.9,15,19,21,23, 24, 25,29,30 Our study adds to this growing body of evidence, strengthening the rationale for formal evaluation of oral vancomycin in prospective randomized controlled trials.
Despite encouraging results, several limitations merit consideration. First, this was an observational study, and we cannot exclude confounding by disease severity, concomitant therapy, or selection bias in dosing strategies. Second, the lack of standardized oral vancomycin dosing regimens across studies remains a challenge, with some reports suggesting differential efficacy by dose or brand formulation.20 Third, although we did not identify vancomycin-resistant enterococci in this cohort, long-term antimicrobial safety requires ongoing vigilance.22 Fourth, because this was an observational study without a control group, we cannot exclude the possibility that unmeasured factors contributed to the observed associations. Prospective controlled studies are needed to confirm these findings. Finally, outcomes such as transplant-free survival and colorectal neoplasia risk were not assessed but represent critical endpoints for future studies.

Conclusion
Oral vancomycin was associated with improvements in clinical, endoscopic, histologic activity of the colon, and biomarker outcomes of the colon and liver in patients with PSC-IBD, with sustained oral vancomycin-free survival observed across both low- and high-dose groups. These findings support a potential role for oral vancomycin as a disease-modifying therapy in PSC-IBD, consistent with recent systematic reviews,31,32 and highlight the need for additional randomized, controlled, and long-term studies to establish efficacy, safety, and optimal dosing.