P575 Vedolizumab trough levels predict clinical outcomes in inflammatory bowel disease

Abstract

BackgroundVedolizumab is an α4β7 integrin antagonist for the treatment of inflammatory bowel disease (IBD). The role of drug monitoring, based on the assessment of Vedolizumab trough levels (VTL) and anti-Vedolizumab antibodies (AVA) has not been clarified. In this study we investigated the correlation between VTL and AVA and clinical outcome.MethodsWe prospectively enrolled consecutive IBD patients starting VDL. Each patient underwent 300 mg infusion at Weeks 0, 2, 6, and 14; additional doses at Week 10 and then every 4 weeks were given to non-responders at Week 6. We assayed VTL and AVA by ELISA (Theradiag, Marne La Vallee, France) at Weeks 6 and 14. Limits of detection for VTL and AVA were 2μg/ml and 35 ng/ml, respectively. Clinical response was defined as at least 30% reduction of Harvey Bradshaw Index (HBI) and partial Mayo score (pMayo) from baseline while remission was defined as HBI <5 or pMayo < 2. Statistics was performed by Mann–Whitney test, Spearman’s rho, receiver operating characteristics (ROC) curve analysis.ResultsWe included 66 patients (mean age 46, 1 year; male 60%) with Crohn’s disease (CD, n = 34) and ulcerative colitis (UC, n = 32). Forty-seven (71%) IBD patients had previous anti-TNFα. Patients were followed up to a median of 36 week. Median VTL (IQR) at Weeks 6 and 14 were 32.4 (20.2–48.7) and 16.9 (10.6–22.7) μg/ml, respectively. We detected higher median VTL in responders/remitters at all concomitant and subsequent analysed time points (Table 1). Week 6 VTL were inversely correlated with CRP (rho −0.32, p = 0.011). By ROC curve analysis we identified a value for VTL at Week 6 of 40.3 μg/ml indicative of response at Week 6 (AUC 0.714, p = 0.0009). Remission at Week 14 was predicted by a Week 6 VTL of 24.3 (AUC 0.682, p = 0.03), remission at Week 22 by a Week 6 VTL of 44.3 (AUC 0.713, p = 0.008) and remission at Week 36 by a Week 6 VTL of 52.9 μg/ml (AUC 0.666, p = 0.04), respectively. We also identified a cut-off of Week 14 VTL for remission at the different time points: 18 μg/ml (AUC 0.683, p = 0.01, AUC 0.729, p = 0.0049, AUC 0.714, p = 0.03, respectively, for remission at Week 14, Week 22, and Week 36). AVA were detected in 1.5% at Week 6 and in 3% at Week 14 and were not correlated with clinical outcome.   Median Week 6 VTL (IQR) (μg/ml)  Median Week 14 VTL (IQR) (μg/ml) p-value  Responders at Week 6 41.3 (26.5–53.9)§ — § 0.003 vs. non-responders Non-responders at Week 6 26.9 (17.2–38.3) — — Remitters at Week 14 45.6 (28.4–55.3)* 20.6 (18.6–36.6)$ * 0.03 vs. non–remitters; $ 0.009 vs. non–remitters Non-remitters at Week 14 27.9 (19.4–42.1) 15.1 (10–20.2) . Remitters at Week 22 46 (28.8–58)@ 20.6 (18.6–36.6)^ @ 0.012 vs. non–remitters; ^ 0.035 vs. non–remitters Non-remitters at Week 22 28.3 (19.8–40) 15.4 (10.3–20.9) — Remitters at Week 36 40.2 (27.1–57.3)# 19 (17.3–32.9)* # 0.047 vs. non–remitters; * 0.03 vs. non–remitters Non-remitters at Week 36 28.8 (19.8–42.1) 15.6 (10–20.8) —   Median Week 6 VTL (IQR) (μg/ml)  Median Week 14 VTL (IQR) (μg/ml) p-value  Responders at Week 6 41.3 (26.5–53.9)§ — § 0.003 vs. non-responders Non-responders at Week 6 26.9 (17.2–38.3) — — Remitters at Week 14 45.6 (28.4–55.3)* 20.6 (18.6–36.6)$ * 0.03 vs. non–remitters; $ 0.009 vs. non–remitters Non-remitters at Week 14 27.9 (19.4–42.1) 15.1 (10–20.2) . Remitters at Week 22 46 (28.8–58)@ 20.6 (18.6–36.6)^ @ 0.012 vs. non–remitters; ^ 0.035 vs. non–remitters Non-remitters at Week 22 28.3 (19.8–40) 15.4 (10.3–20.9) — Remitters at Week 36 40.2 (27.1–57.3)# 19 (17.3–32.9)* # 0.047 vs. non–remitters; * 0.03 vs. non–remitters Non-remitters at Week 36 28.8 (19.8–42.1) 15.6 (10–20.8) — ConclusionsOur data suggest that Week 6 VTL is correlated with clinical response and could predict clinical remission at Week 14, Week 22, and Week 36. Week 14 VTL correlate with clinical remission and could predict clinical remis ion at Weeks 22 and 36. Immunogenicity of VDL is low in our patients.