P347 Identification of a cut-off for persistent anti-infliximab antibodies as a predictor of response to infliximab monotherapy
L. Del Nero*1, G. Bodini2, E. Giannini2, J. Anjali3, V. Savarino1, E. Savarino4
1University of Genoa, Internal Medicine Department, Genoa, Italy, 2University of Genoa, Department of Gastroenterology and Internal Medicine, Genoa, Italy, 3Prometheus Laboratories, Department of Research and Development, San Diego, California, United States, 4University of Padua, Gastroenterology Unit, Department of Surgery, Oncology and Gastroenterology, Padua, Italy
The clinical and predictive role of anti-Infliximab antibodies (AIA) presence and concentration are still debated, in both Crohn’s disease (CD) and ulcerative colitis (UC) patients. However, there is increasing evidence of their usefulness to improve the management of patients on biological treatment who experience a loss of response (LOR). AIA can be subdivided into 2 types, persistent and transient, on the basis of their occurrence on multiple samples and capability of interfering with infliximab trough levels (TL), and, therefore, persistent AIA seem to play a major role on treatment outcome.
The aim of our retrospective study was to evaluate the clinical relevance of persistent AIA in a single-centre cohort of inflammatory bowel disease (IBD) patients. We selected from our cohort of 56 IBD patients treated with IFX monotherapy who achieved clinical and biochemical remission after induction (IFX schedule: 5 mg/kg at week 0, week 2, and week 6), 18 patients (32.1%) who developed persistent AIA during 48 weeks follow-up. Blood samples were drawn at standardised time points (ie, baseline, 2 weeks, 6 weeks, and every 8 weeks) before IFX infusion. TL and AIA were measured using a homogenous mobility shift assay (HMSA; Prometheus Lab, San Diego, California, United States). Clinical disease activity was assessed both at week 14 (ie, after induction) and week 48 using the Harvey–Bradshaw Index (HBI, remission defined by HBI < 5) in CD patients and by the Mayo score for UC patients (remission defined by Mayo score <2). In addition, C-reactive protein and erythrocyte sedimentation rate (ESR) were measured
Eighteen patients (11 CD and 7 UC, 10M/8F, median age 39.5 years, range 18–69) developed persistent AIA at a median of 2 weeks (range 2–22) during 48 weeks follow-up. Amongst these patients, 12 (66.7%) experienced LOR during the follow-up period. Median AIA were significantly higher in patients who showed LOR as compared with patients who maintained remission (8.29 U/ml, range 0.62–30.52 U/ml, vs 1.41 U/ml, range 0.77–9.94 U/ml; p = 0.04). ROC curve identified a persistent AIA cut-off of 3.91 U/mL as the threshold with the highest accuracy for the identification of relapsers (AUROC = 0.799, specificity = 75.0%, and sensitivity = 83.3%).
The early occurrence of elevated persistent AIA serum concentrations during IFX monotherapy treatment is associated with high risk of LOR. Further, the use of an AIA concentration cut-off of 3.91 U/mL can be useful to accurately identify patients with LOR, although these results need to be confirmed in larger series.