1, p<0.0005). In terms of extreme symptom load, 27% of <50 year old patients with poor QoL had between 8 and 10 (the maximum possible) significant symptom domain scores compared with 16% of the over 60s with poor QoL. In contrast to symptom load, UDCA non-response did not predict poor QoL in either age
group (>60 years: CS 2.4, OR 1.68 (0.9-3.2), p=0.12; <50 years: CS 1.1, OR 1.3 (0.7-2.1), p=0.3). Social dysfunction symptoms were a particularly discriminating feature in young patients with poor QoL compared to Trichostatin A clinical trial good QoL (OR for association between QoL status and social symptom status 423 [95% CI 58-3078], p<0.0001). Amongst younger patients with poor QoL, social dysfunction symptoms correlated find more particularly strongly with depression, fatigue and cognitive symptoms (r=0.67, 0.56, and 0.8 respectively, all p<0.0001). Discussion The UK-PBC Study has shown that there are marked phenotypic differences in PBC patients presenting at a younger age with worse perceived QoL and significantly increased symptom burden. Social
dysfunction symptoms are a specific feature of younger patients and associate strongly with depression, fatigue and cognitive symptoms. Offering psychological support and targeting specific symptoms in young PBC patients offer a potential approach to life quality improvement. Disclosures: Richard N. Sandford – Advisory Committees or Review Panels: Otsuka; Grant/ Research Support: Intercept David Jones – Consulting: Intercept The following people have nothing to disclose: Jessica K. Dyson, Laura Griffiths, Samantha J. Ducker, George F. Mells Background: The pathophysiology of PSC remains unclear, but a close association with IBD is overt. We sought to document changes in the gut microbiota in PSC and IBD by characterising gut adherent bacteria in patients with PSC and IBD, IBD alone and healthy controls. Methods: We collected pan-co-lonic biopsy samples from 9 controls, 10 IBD and 11 PSC-IBD patients, undergoing colonoscopy. Gut microbiota
were characterised using 16s rRNA based analysis of the V3 – V4 region (Illumina MiSeq). The sequences were clustered into operational taxonomic units using Uparse and analysed using Qiime and Nitroxoline the Vegan package in R. Results: We identified little difference in richness and complexity (Simpson’s index) of the microbiota between conditions. However an analysis of variance showed a significant difference in the composition of the microbiota between conditions, irrespective of biopsy site (p = 0.001). This was confirmed by constrained ordination, which resulted in clear separation between the three groups (Fig 1). However there was no difference in microbiota between sites. Indeed sites from the same patient were highly similar and clustered together. PSC-IBD and IBD showed reduced levels of Prevotella and Roseburia (a butyrate producer).