AB0114 IMPROVEMENT OF GUT MICROBIOTA DYSBIOSIS IN PATIENTS WITH AXIAL SPONDYLOARTHRITIS AFTER ONE YEAR OF BIOLOGICAL TREATMENT

Abstract

BackgroundEmerging evidence suggests that dysbiosis of the gut microbiota is involved in the initiation and perpetuation of spondyloarthritis (SpA). Biological disease-modifying antirheumatic drugs (bDMARDs) are a successful treatment to improve symptoms and reduce structural damage occurring in SpA; however, non-responders are frequent and few predictive factors for clinical response have been identified. Whether or not a patient responds to treatment could be related to gut microbiota composition.ObjectivesTo investigate the gut microbiota changes in patients with radiographic axial SpA (r-axSpA) after receiving one year of treatment with bDMARDs and identify potential microbial biomarkers predictive of treatment response.MethodsPatients with r-axSpA were recruited between 2015 and 2019 in an extension of the prospective GErman SPondyloarthritis Inception Cohort (GESPIC) before beginning bDMARD therapy. All patients had high disease activity (BASDAI >=4 and/or ASDAS >=2.1) despite previous treatment with nonsteroidal anti-inflammatory drugs, and had not received treatment with bDMARDs for at least three months before enrollment in the study. The choice of bDMARD was left to the discretion of the clinical rheumatologists in accordance with standard practice. Disease activity measures (BASDAI, CRP and ASDAS) and fecal samples were assessed at baseline prior to treatment and after one year of treatment. Patients with back pain negative for inflammatory disease served as a control group. Microbiota composition was determined by 16S rRNA gene sequencing, followed by taxonomic profiling with the SILVA138 database. Response to bDMARD therapy was defined as a clinically important improvement of ASDAS (>=1.1).ResultsA total of 99 patients with r-axSpA and 63 control individuals were included based on the availability of clinical and microbiome samples. Average age (mean±SD) was 36.4±10.4 years and 64 patients were males. The prevalence of HLA-B27 was 89.9% among r-axSpA patients compared to 7.9% among control individuals. Simpson indices showed an increase in alpha diversity between baseline and year 1 in r-axSpA patients which was statistically insignificant (paired Wilcoxon p=0.154) but brought the r-axSpA cohort nearer to controls. Likewise, Bray-Curtis dissimilarities to measure beta diversity showed a qualitative normalization to healthy individuals after treatment when visualized in principal coordinate space.At the genus level, patients were mainly depleted in Lachnospiraceae taxa such as Blautia, Roseburia, and Fusicatenibacter, and enriched in Collinsella compared to the control group at baseline. After one year of treatment, most SpA patients exhibited increased abundances of these taxa, most notably Blautia. Patients also exhibited depletions in Bacteroides and Faecalibacterium, which was strongly enriched in HLA-B27+ individuals at baseline (adjusted Wilcoxon p<0.001). Collinsella showed a very slight median increase after one year of treatment, with no significant difference between responders and non-responders (adjusted Wilcoxon p=0.33). Shifts in highly abundant Prevotella and Bacteroides were strongly correlated with the change in ASDAS after one year when controlling for intra-individual variance and overall changes in alpha diversity.ConclusionThe gut microbiota composition of r-axSpA patients who underwent treatment with bDMARDs for one year more closely resembled the controls. The unique enrichment of Collinsella in r-axSpA patients remained stable across time and treatment, suggesting it may be a disease biomarker.Figure 1.a) Flowchart of axSpA patients summarizing the main clinical and disease activity parameters of the cohort. b) Taxa with the most pronounced shifts in median relative abundance in patients with axSpA after receiving biological treatment for one year. c-d) Alpha and beta diversity analyses, respectively, of axSpA patients before and after treatment compared to control individuals. Labeled points in d represent group means.Disclosure of InterestsNone declared