Search in the Abstract Database

Abstracts Search 2018

OP007 Detection of mucosal healing with a serum marker panel in adalimumab-treated patients with ulcerative colitis

M. de Bruyn1,2*, R. Ringold3, E. Martens2, M. Ferrante1,4, G. Van Assche1,4, G. Opdenakker2, A. Dukler3, S. Vermeire1,4

1Translational Research Center for GastroIntestinal Disorders (TARGID), Chronic Diseases, Metabolism and Ageing (CHROMETA), Leuven, Belgium, 2Rega Institute for Medical Research, Microbiology and Immunology, Leuven, Belgium, 3Kepler Diagnostics, Inc., Simi Valley, USA, 4University Hospitals Leuven, Gastroenterology and Hepatology, Leuven, Belgium


Surrogate markers that accurately detect mucosal healing (MH) in patients with ulcerative colitis (UC) are urgently needed. After infliximab therapy, neutrophil gelatinase B-associated lipocalin in complex with matrix metalloproteinase-9 (NGAL-MMP-9), cathelicidin LL-37 and chitinase 3-like 1 (CHI3L1), together with C-reactive protein (CRP) and neutrophil count were significantly associated with MH. We investigated whether these five markers were also associated with MH after treatment with adalimumab (ADM).


Serum samples were obtained from 31 UC patients (58% female, median age 35.6 years) both at baseline and after a median (IQR) of 14 (10–25) weeks. All patients had endoscopic disease activity prior to treatment (Mayo endoscopic subscore/MES ≥2) and MH was defined at follow-up endoscopy as MES ≤1. NGAL-MMP-9, LL-37, and CHI3L1 were measured with ELISA. Binary logistic regression analysis was used to combine multiple markers. ROC analysis was used to test the performance of individual and combined markers. Non-parametric tests were performed, and p-values of <0.05 were considered significant.


Twenty-one patients (68%) achieved MH with ADM therapy. Compared with baseline, NGAL-MMP-9 levels significantly decreased in patients with MH (96.9 vs. 214.1 ng/ml, p = 0.002), whereas patients without MH had similar levels. Neutrophil count was significantly lower after ADM compared with baseline in patients with MH only (3.6 vs. 6.5 109/l, p = 0.023). CRP levels significantly decreased after ADM in patients with MH (15 to 1.1 mg/l, p = 0.008) and were higher at baseline in patients with MH compared with without MH (15 vs. 2.4 mg/l, p = 0.004). LL-37 levels significantly increased in patients without MH post-ADM compared with baseline (29 vs. 19.3 ng/ml, p = 0.034) and were significantly higher compared with patients with MH post-ADM (29 vs. 18.7 ng/ml, p = 0.031). CHI3L1 levels significantly decreased after ADM in patients with MH only (47.4 to 21.6 pg/ml, p = 0.049). The performance of NGAL-MMP-9 (AUC=0.70), neutrophil count (AUC=0.58), CRP (AUC=0.56), LL-37 (AUC=0.79), and CHI3L1 (AUC=0.69) to discriminate MH post-ADM significantly increased when all markers were combined (AUC=0.86). The 5-marker panel was discriminative for MH with 75% sensitivity, 89% specificity, 67% positive predictive value, and 92% negative predictive value.


Neutrophil-related markers (NGAL-MMP-9, LL-37, and CHI3L1) as well as CRP and neutrophil count were significantly associated with MH after treatment with ADM. These data are in concordance with our previous findings in infliximab-treated patients. We therefore suggest a broad clinical utility of this panel for monitoring MH in UC patients under anti-TNF treatment.