Introduction

Primary sclerosing cholangitis (PSC) is a rare, chronic cholestatic disease that is characterized by inflammation and fibrosis of the large and small bile ducts, which can lead to fibrosis involving the hepatic parenchyma and biliary tree, cirrhosis, and end-stage liver disease [1]. PSC is an immune-mediated disorder that is often associated with inflammatory bowel disease (IBD), with 60% to 70% of patients with PSC having coexisting IBD, most commonly ulcerative colitis (UC). PSC is associated with IBD in 70% to 80% of patients in the west as compared to 20% to 30% in Asia [2]. Notably, UC associated with PSC is unique and differs from the typical phenotype of UC, manifesting as pancolitis, with a right-sided predominance, rectal sparing, and backwash ileitis [3]. Although PSC was first described in medical literature in 1857, the etiology of PSC remains incompletely understood, which in part explains a lack of effective or proven medical therapy for this condition. Despite being an immune-mediated disease, treatment with immune suppressants does not appear to alter the course of disease or cure PSC, and liver transplantation is so far the only proven therapy to extend life expectancy [1].

Although there is no curative treatment for PSC, there is emerging evidence that targeted modulation of the gastrointestinal microbiome by antibiotic therapy appears to alter the natural course of PSC and delay the progression of disease. In our recently published systematic review and meta-analysis [4], we found in adult PSC patients with or without an associated IBD, short-term treatment with antimicrobial therapy, in particular with oral vancomycin (OV), was associated with significant improvement in cholestatic liver enzymes and PSC Mayo risk score (MRS). More recently, Dao et al. [5] found that OV was effective for the induction and maintenance of remission of UC in adults with UC-PSC, including PSC patients having undergone orthoptic liver transplantation (OLT). We describe here our experience in a cohort of well-characterized adult PSC patients (both pre- and post-OLT), with an associated IBD, who were treated with OV therapy for management of colitis, specifically having failed to respond to conventional medical therapy for management of IBD. Additionally, potential adverse effects, including the development of vancomycin-resistant enterococcus (VRE), were carefully monitored.

Methods

We conducted a retrospective audit of well-characterized cohort of consecutive adult patients with an established diagnosis of PSC and concomitant IBD, who were treated with OV at the Department of Gastroenterology and Hepatology at the Princess Alexandra Hospital between July 2015 and May 2021. Oral vancomycin was initiated in the IBD clinic, predominantly for pre-transplant patients with PSC and associated IBD who had failed to respond to conventional medical therapy for management of IBD and for post-transplant patients with PSC and associated IBD as a treatment option for managing colitis, prior to increasing immunosuppression. VRE testing was performed utilizing rectal swabs, employing standard clinical culture-based screening methods [6]. The aim of this study was not to focus on the liver-specific end points. The local ethics committee approved this study (HREC/2021/QMS/76269).

Results

The study included seven patients with PSC (four post-orthoptic liver transplantation and three pre-transplant) with concomitant UC (six with pancolitis and one with J-pouch). Their age ranged from 22 to 53 years, and out of the seven patients, two were females. All patients were treated for at least 6 months with OV (range 9 to 31 months, mean 32.1 months or cumulated vancomycin exposure 225 months).

All patients treated with OV achieved complete clinical remission of the concomitant UC, with mean reduction of fecal calprotectin by 634 μg/mg (87.3%) (Fig. 1, panel A), mean reduction in the C-reactive protein by 21.9 mg/L (74.2%), and mean reduction in the Mayo UC score by 9.3 (93.3%) (Fig. 1, panel B). With regard to the liver parameters, mean improvements of alkaline phosphatase and total bilirubin were 19.7% (48.7 U/L) and 19.6% (2.7 mg/dL), respectively. No patient developed VRE nor reported adverse events during treatment with OV. Six out of seven patients had pre- and post-OV treatment colonoscopies, and in three out of six patients, colonoscopies revealed endoscopic remission on OV with a Mayo endoscopic score of 0 and resolution of histological inflammation (Table 1). Furthermore, all patients treated with OV had clinical remission. No patient developed VRE or experienced any adverse effects related to this treatment.

Fig. 1
figure 1

Panel A indicates fecal calprotectin at baseline and at various intervals post-oral vancomycin therapy. Patient No. 5 had concomitant viral infection with Rota virus, identified in stool samples using quantitative real-time polymerase chain reaction, was managed conservatively with no further escalation in immunosuppression; no follow-up fecal calprotectin results are available for this patient. Panel B indicates ulcerative colitis Mayo score at baseline and at various intervals post-initiation of oral vancomycin therapy.

Full size image
Table 1 Characteristics of patients with primary sclerosing cholangitis and associated inflammatory bowel disease treated with oral vancomycin
Full size table

Discussion

There is cumulating evidence that in patients with PSC, targeted modulation of the gastrointestinal microbiota with antimicrobial therapy alters the course of disease [4] and delays or even stops progression. In 1959 Rankin et al. [7] for the first time successfully used tetracycline (500 mg once daily for 10 months) in five patients with PSC (then characterized as chronic, progressive pericholangitis) and associated UC caused by portal bacteremia with biochemical and clinical improvement. Since then, several antibiotics including azithromycin, vancomycin, metronidazole, minocycline, tetracycline, and rifaximin have been used as potential therapeutic option for PSC [4]. Among these, OV is a promising treatment, with benefits on the both the liver and the colon in patients with PSC. Cox et al. [8] were the first to use OV in three children with PSC and associated IBD, with improvement in their liver enzymes and symptoms, which recurred with discontinuation of vancomycin. Since then, several studies have shown OV to be effective for treating both PSC and associated IBD [4].

Vancomycin is a glycopeptide antibiotic, which is poorly absorbed after oral administration with bactericidal effect against Gram-positive bacteria and inactive against Gram-negative bacilli, mycobacteria, or fungi [9]. Thus, OV has minimal systemic absorption and concentrates in the intestine. In PSC and associated IBD, the potential mechanism of actions includes [3] (1) selective effect on Gram-positive species due to its relatively narrow antibiotic spectrum; (2) reduction of the hydrophobic secondary bile acids, which are potentially responsible for the right-sided colitis and after absorption through the enterohepatic circulation for mediating the bile duct injury; and (3) immunomodulatory effect of vancomycin via the tumor necrosis factor (TNF)–alpha inflammatory pathways and/or downstream Treg induction.

However, development of VRE remains a major concern and can have a major impact on public health. Based upon the available literature, multiple genes are necessary to generate vancomycin resistance in enterococci and acquisition of VRE colonization does not occur through mutations in susceptible enterococci in the intestinal tract [10]. Thus, treatment with OV does not directly cause VRE, but selective pressure exerted by OV may facilitate an increase in the concentration of exogenously acquired VRE. Indeed, available studies report that de novo vancomycin resistance is very rare in patients treated with vancomycin [10]. None of our patients developed VRE and similar findings were reported in a pediatric cohort study from the Queensland Children’s Hospital, Australia. In this cohort study of 17 pediatric PSC patients with UC who received vancomycin on average for >8 months, none of the patients developed VRE [11].

The results of our cohort study confirm that in patients with PSC and concomitant IBD, treatment with OV is associated with induction and maintenance of clinical remission in all UC patients who had previously failed to respond to conventional therapies of IBD. More than half of the patients treated with OV attained endoscopic and histologic remission of the associated UC. Given none of the patients developed VRE nor experienced any adverse effects with this treatment, OV might be considered in immunosuppressed patients (e.g. after OLT) since this treatment does not add to the burden of systemic immunosuppression in post-OLT PSC patients.

However, some of the limitations of this study need to be acknowledged. This is a retrospective study, with PSC patients both pre- and post-liver transplant with a relatively small sample size. All post-OLT PSC patients were on systemic immunosuppression, which included low-dose (≤10 mg prednisolone) with or without cyclosporine or tacrolimus. Additionally, OV was not used as a first-line therapy for induction or maintenance of remission of the colitis, and was used as an adjuvant therapy when conventional treatment options for management of IBD had failed.

Thus, the patient characteristics with the most favorable response, dosage, formulation, duration, and long-term impact of OV in patients with PSC with or without an associated IBD need to be determined by prospective, longer-term, randomized controlled trials. In addition, studies should aim to identify characteristics of gastrointestinal microbiota in PSC patients responding to antibiotic therapy and remain in remission after discontinuation of antibiotic therapy. Characterization of the interdependence between microbiota, immune function, liver function, and the antibiotic-induced clinical remission will allow to individualize therapy by defining patients who are likely to respond to antibiotic therapy (and who stay in remission). The findings also will enable the development of intervention strategies that specifically target microbes potentially involved in the pathophysiology of PSC.