P532 Measuring endoscopic activity in patients treated with vedolizumab for ulcerative colitis

C. EMONT1, C. Gay2, A.L. Parmentier3, S. Koch4, S. Ambregna5, G. Claudé5, P. Sage5, P. Zhong4, C. Briot5, L. Vuitton4

1Department of Gastroenterology, Besançon University Hospital, Hospital of Haute-Saone, Besançon, Vesoul, France, 2Department of Gastroenterology, Lyon University Hospital, Lyon, France, 3Center of Clinical Methodology, Besançon University Hospital, Besançon, France, 4Department of Gastroenterology, Besançon University Hospital, Besançon, France, 5Department of Gastroenterology, Hospital of Haute-Saone, Vesoul, France

Background

Vedolizumab (VDZ) has proven its efficacy in ulcerative colitis (UC). The delay to obtain endoscopic remission with anti-α4β7 integrin is unclear. We aimed to assess the potential of VDZ to induce endoscopic remission and its characteristics in real-life settings.

Methods

We conducted a pooled analysis with prospective and retrospective data of patients who started VDZ for an active UC in two French hospitals. Patients underwent proctosigmoidoscopies at baseline and at weeks 14, 30 and 54 in the prospective cohort. A central blinded review of all endoscopy recordings was performed by two experts, pooled with some recordings from anti-TNFɑ treated patients. The primary endpoint was the proportion of patients in endoscopic remission (endoscopic Mayo score (eMs) ≤ 1) within 54 weeks. Secondary endpoints included the description of the pattern of mucosal healing and the evolution of UC activity. Characteristics of endoscopic remitters and non-remitters were compared and predictive factors of endoscopic remission were reported in univariate logistic regression.

Results

Forty-two patients were enrolled, 22 in the prospective cohort and 20 in the retrospective cohort. At baseline, 38 (90.5%) patients had previously experienced anti-TNFɑ, 14 (33.3%) were treated with corticosteroids and 15 (35.7%) with immunosuppressors. Twenty-six (61.9%) patients achieved endoscopic remission after a median time of 14 weeks. Endoscopic response and histologic remission occurred in 27 (64.3%) and 21 (50%) patients within 54 weeks, respectively. No specific pattern of mucosal healing was found with VDZ as compared with anti-TNFɑ treated patients. UC activity steady improved over time. Endoscopic remitters had a lower disease activity at baseline (total Mayo score (tMs) 8.1 vs. 9.5, p = 0.007; UCEIS 4.3 vs. 5.6, p = 0.0005; Nancy’s score 2.8 vs. 3.6, p = 0.004) and at week 14 (tMs 3 vs. 6, p = 0.001; eMs 1.3 vs. 2.5, p = 0.005; UCEIS 2.7 vs. 5.4, p = 0.0007; Nancy’s score 2.2 vs. 3.6, p = 0.001) and needed less treatment optimisation (11.5 vs. 81.3%, p < 0.0001) than non-remitters. In univariate logistic regression, predictors of no endoscopic remission were a high endoscopic activity at baseline (eMs > 2, OR 0.2 IC 95% 0.05–0.86, p = 0,021; UCEIS > 5 OR 0.22 IC 95% 0.06–0.84, p = 0.022) and treatment optimisation (OR 0.03; IC 95% 0.01–0.17, p < 0.001).

Conclusion

In our series, about two-thirds of UC refractory patients achieved endoscopic remission during the first year. Endoscopic healing was early reported, in half of patients at the end of the first trimester, actually suggesting a rather fast mechanism of action of VDZ on mucosal healing. Baseline UC activity impacted endoscopic remission.