P703 The real-world long-term effectiveness of vedolizumab in inflammatory bowel diseases: a single-centre observational study

Abstract

The effectiveness of vedolizumab (VDZ) in real-world practice is under evaluation. We aimed to report the effectiveness of VDZ on intestinal and articular symptoms after 10 and 52 weeks of treatment, and at the end of follow-up. All consecutive patients with moderate–severe Crohn’s disease (CD) or ulcerative colitis (UC) who started a treatment with VDZ from July 2016 to December 2017 were entered in a prospectively maintained database. The following clinical end-points were set at 10 and 52 weeks: steroid-free remission (Harvey–Bradshaw Index < 5 for CD and Mayo Partial Score < 2 for UC without steroids use), and response (absence of steroid-free remission, but reduction of Harvey–Bradshaw Index ≥3 for CD and Mayo Partial Score ≥2 for UC compared with baseline). Patients with steroid-free remission and response were deemed as having clinical benefit. In patients with active spondyloarthropathy (SpA) at baseline, the response on articular symptoms was defined as disappearance of objective signs of arthritis and resolution of pain. 169 patients (112 with CD and 57 with UC) were included. At Week 10, a steroid-free remission was achieved in 44 out of 169 patients (26.0%), and a response in 35 (20.7% - overall clinical benefit: 46.7%), while at 52 weeks a steroid-free remission was achieved in 30 out of 128 patients (23.4%), and a response in 18 (14.1% - overall clinical benefit: 37.5%). The median follow-up was 47.0 weeks (148.39 person-years), and the failure-free survival was 60.4% at 1 year. Semi-parametric Cox model showed that patients with CD had a higher risk of treatment failure compared with patients with UC (HR 2.06, 95% CI: 1.05–4.05, p = 0.036). After 10 weeks, a response on articular symptoms was reported in 12 out of 39 patients (30.8%) with active SpA at baseline, and in 12 out of 16 patients (75.0%) at Week 52. At Week 10, the only factor that was marginally associated with the articular response was the clinical benefit on intestinal symptoms (OR 5.07, 95% CI: 0.97–31.70, p = 0.055), while the coexistence of axial and peripheral SpA was associated with a reduced response rate compared with peripheral manifestations only (OR 0.13, % CI: 0.02–0.64, p = 0.021). Overall, 67 adverse events were reported (incidence rate: 45.2 per 100 person-years). Twenty (11.8% of patients) adverse events leading to treatment discontinuation were reported: 11 arthritic flares, 5 subjective perceptions of intolerance, 2 infusion reactions, one pneumonia, and one prostate cancer. In this large cohort, VDZ provided good effectiveness on intestinal symptoms, particularly in patients with UC. A subset of patients reported improvement also on articular symptoms, especially in cases of peripheral SpA.