OP020 Recent anti-TNF exposure predicts lower vedolizumab trough concentrations in patients with Crohn's disease

Abstract

Background: Vedolizumab (VDZ) inhibits interaction of the α4β7 integrin with mucosal addressin cell adhesion molecule (MAdCAM)-1. We have previously shown that anti-TNF treatment influences MAdCAM-1 expression in gut biopsies. We studied the impact of recent anti-TNF exposure on VDZ trough concentrations (TC) and response in patients with Crohn's disease (CD) and ulcerative colitis (UC). Methods: From 75 patients (46 CD, 29 UC) starting VDZ therapy in a tertiary referral center, VDZ and anti-TNF serum concentrations (median [IQR]) were measured at trough during VDZ therapy at w2, w6, w10 (CD only), w14 and w22 using in-house developed ELISAs. Clinical response was evaluated by clinical symptoms and physician global assessment. Biological response and remission (based on CRP) were assessed at w6 and w22 in patients with CD who had baseline CRP >5 mg/L (n=25). Twenty-eight patients with CD underwent endoscopy after w22 to assess mucosal healing (MH) and all patients with UC received sigmoidoscopy at w10 (MH defined as Mayo endoscopic sub-score of 0/1). Results: Clinical response was achieved in 46% (21/46) and 66% (19/29) of the patients with CD and UC. Only in patients with UC, an exposure-response relation was found between VDZ TC up to w22 and clinical response (data not shown, p<0.0001). At w22, 48% (12/25) and 32% (8/25) of the patients with CD were in biological response and remission. Patients in biological remission had higher VDZ TC at w6, w10 and w22 (data not shown, p=0.02, p=0.04 and p=0.01). MH was achieved in 18% (5/28) of the patients with CD and in 66% (19/29) of the patients with UC. Patients with UC with MH had higher VDZ TC up to w22 compared to patients with no MH (p=0.02, p=0.006, p=0.03 and p=0.04 for w2, w6, w14 and w22). Patients with CD with MH had higher VDZ TC at w6 and w10 (p=0.006, p=0.03). Most patients (70/75) were previously exposed to anti-TNF. Of these, 30 had still detectable anti-TNF concentrations at the first VDZ infusion. Patients with CD who were recently exposed to anti-TNF (<16 weeks before the start of VDZ therapy, n=38) had lower VDZ TC at all time points, compared to patients with no recent anti-TNF exposure (e.g., w6: 16.8 μg/mL [12.5], n=18 vs. 28.5 μg/mL [18.2], n=28, p=0.006 and w22: 6.8 μg/mL [8.0], n=16 vs. 15.5 μg/mL [12.4], n=27, p=0.005). We observed numerical though non-significant lower response rates in patients with recent anti-TNF exposure (data not shown, p>0.1). Conclusions: An exposure-response relation was observed as early as w2 and up to w22, with impact of higher VDZ TC on meaningful outcomes. The inverse association between recent anti-TNF exposure and VDZ TC in patients with CD is intriguing and might be explained by a residual effect of anti-TNF treatment on MAdCAM-1 expression.