INTRODUCTION
The concurrent occurrence of autoimmune hepatitis (AIH) and primary sclerosing cholangitis (PSC) is referred to as AIH/PSC overlap syndrome or autoimmune sclerosing cholangitis (ASC) in children. Though not evidence-based, it is standard practice to treat the AIH component with corticosteroids (CS) and azathioprine, and the PSC component with ursodeoxycholic acid (UDCA). Antibiotics (AB) are increasingly being reported to have favorable effects in PSC, but their role in ASC is poorly evaluated. We retrospectively investigated the response to oral ABs as initial or rescue therapy in ASC children. A historical control group included ASC patients receiving the recommended treatment. We prospectively analyzed the gut microbiota and serum bile acid (BA) profile before and after AB therapy in children with either ASC or PSC alone, and evaluated whether changes in gut microbiota and/or BAs correlated with the therapeutic response.
METHODS
Our study included patients who had been diagnosed with ASC or PSC based on biochemical, histological, and radiological findings. In addition to their standard treatment, they received metronidazole (MTZ) or oral vancomycin (OV) for 14 days as induction or rescue therapy. When AB had been administered upon diagnosis, a CS-free induction regimen was chosen, while AB were administered as rescue therapy only if patients had not achieved biochemical remission with standard treatment. Stool and serum samples were prospectively collected before and after AB therapy. DNA isolation, amplification, and sequencing were performed in order to profile the microbiota composition using the bacterial 16sRNA, while serum BAs were assessed by ultra-performance liquid chromatography coupled to mass spectrometry. The beta-diversity measured the dissimilarity between each matched stool samples. The endpoints used to assess AB efficacy included decrease in liver enzymes, subsequent sustained biochemical remission, and, when AB were given upon diagnosis, CS avoidance.
RESULTS
Our study retrospectively included 11 patients with ASC, of whom seven had received oral ABs upon diagnosis. All patients showed a significant decrease in AST (-56%, p=0.006), ALT (-83%, p=0.004), and GGT (-54%, p=0.004) levels. All seven patients of the initial AB group did not need CS and continued to be in remission until last follow-up (351 days [216-888]). Among the four patients of the rescue AB group, two relapsed after stopping AB, while the remaining two showed a sustained biochemical remission until last follow-up. Compared to the historical control group treated with CS, a higher proportion of long-term biochemical remission (p=0.034) and long-term CS-free maintenance therapy (p=0.019) was observed among patients receiving AB, as well as a lower proportion of relapses (p=0.025) and a shorter therapy duration (p=0.0003). Prospectively, seven patients (four ASC, three PSC) were included. All patients showed a significant decrease in AST (-55%, p<0.025), ALT (-56%, p<0.025), and GGT (-41%, p<0.025) levels under MTZ therapy. Four children subsequently showed a sustained biochemical remission until last follow-up (375 days [119-502]). Among these four patients, and contrarily to relapsing patients, three exhibited a widely different microbial composition before and after MTZ therapy, as expressed by the beta-diversity variation. They showed an increased total serum BA concentration (+70%), primarily due to the large increment of UDCA and its glycine- and taurine conjugates, finally accounting for nearly 40% of the total serum BA pool.
CONCLUSION
We showed that short-term oral AB therapy could be an effective treatment of ASC and PSC by achieving a sustained biochemical remission when administered as an induction regimen but also as a rescue therapy in otherwise uncontrolled disease. In ASC patients, AB additionally impacted the AIH component by achieving a CS-free treatment, especially when given at diagnosis. Furthermore, intestinal microbiota and BAs may play a major role in these diseases, since sustained biochemical remission was associated with positive changes in gut microbiota communities and BA profile after AB therapy.