P345 The Effectiveness of Ustekinumab and Vedolizumab as Third-line Therapy in Patients with Refractory Crohn’s Disease

Abstract

Abstract Background There is no head-to-head trial comparing ustekinumab (UST) and vedolizumab (VDZ) in Crohn’s disease (CD). Recently, real-world studies evaluated the two biologics in patients refractory to anti-TNF therapy, i.e. as second-line agents. Conversely, no study specifically focused on the effectiveness of UST and VDZ as third-line agents, i.e. after failure with at least one TNF-α inhibitor plus failure with VDZ or UST. We performed a multicentre, real-world assessment of the effectiveness of UST and VDZ in these highly-refractory patients with CD. Methods Data of consecutive patients with CD treated with UST and VDZ as third-line therapy until October, 2021 were extracted from the cohort of the Sicilian Network for Inflammatory Bowel Disease (SN-IBD). The effectiveness was evaluated at, 8, 26, and, 52 weeks, and as treatment persistence at the end of follow up. The clinical outcomes were steroid-free clinical remission (SFCR - Harvey-Bradshaw Index <, 5 without steroid use) and clinical response (CR - reduction of the Harvey-Bradshaw Index ≥, 3 points with a concomitant decrease of steroid dosage compared with baseline). Multiple logistic regression models were build to find independent predictors of effectiveness, including the use of UST and VDZ as candidate predictors. Results 143 patients (UST: n=113; VDZ: n=30) were included. After, 8 weeks, SFCR was reported in, 38.1% of patients treated with UST and in, 43.3% of patients treated with VDZ (p=0.75), while CR was reported in, 61.9% of patients treated with UST and in, 60.0% patients treated with VDZ (p=1.00). After, 26 weeks, the rates of SFCR and CR were, 41.6% for UST and, 50.0% for VDZ (p=0.60), and, 61.4% for UST and, 66.7% for VDZ (p=0.81), respectively. After, 52 weeks, the rates of SFCR and CR were, 51.8% for UST and, 57.1% for VDZ (p=0.78), and, 65.9% for UST and, 71.4% for VDZ (p=0.77), respectively. Lack of difference in effectiveness between the two drugs at any time point was confirmed by multiple logistic regression models. After, 26 weeks, 88.7% of patients treated with UST and, 81.4% of patients treated with VDZ were still on treatment. Univariable Cox survival analysis showed a higher probability of treatment discontinuation for VDZ compared with UST (HR for VDZ:, 2.66; p=0.008), even if such difference was not confirmed at the multiple Cox proportional hazard regression model (HR for VDZ:, 1.94; p=0.08), where only age (HR, 0.98; p=0.04) and use of systemic steroids at baseline (HR, 3.29; p=0.003) were found to be independent predictors of treatment discontinuation. Conclusion This is the first real-world study assessing VDZ and UST as third-line therapy in patients with CD. Both drugs showed surprisingly high effectiveness, without significant differences between them.