P214. Faecal calprotectin assay after induction with anti-TNF alpha agents in IBD patients: prediction of clinical response and mucosal healing at one year
L. Guidi1, M. Marzo1, G. Andrisani1, C. Felice1, D. Pugliese1, I. De Vitis1, A. Papa1, G.L. Rapaccini1, F. Forni2, A. Armuzzi1, 1Internal Medicine and Gastroenterology Unit, Complesso Integrato Columbus, Catholic University, Rome, Italy, 2Clinical Laboratory Unit, Complesso Integrato Columbus, Catholic University, Rome, Italy
Faecal calprotectin (FC) is a protein present in the cytosol of neutrophil granulocytes and of macrophages that is released in the biological fluids under inflammatory conditions. FC levels seem to be correlated with active inflammation in inflammatory bowel diseases (IBD), Crohn's Disease (CD) and Ulcerative Colitis (UC). Non-invasive markers of response to biological therapy in IBD are needed to limit current routine investigations that are expensive and invasive. The aim of our study is to evaluate the role of FC as a non invasive marker of inflammation in order to monitor and predict the effect of therapy with TNF-antagonists in IBD.
63 IBD patients (44 CD, 19 UC) provided samples for FC assay, before and after the induction course of anti-TNF therapy (Infliximab n = 42, Adalimumab n = 18, Certolizumab Pegol n = 3). Clinical activity, measured by clinical indexes (CDAI, CAI), was assessed in all patients before and after induction treatment. For mucosal healing assessment (defined as CDEIS <3 or Mayo Endoscopic Score ≤1) all patients performed colonscopy before and after 1-year of anti-TNF treatment.
Clinical response at 1-year was obtained in 36/63 (57%, 12 UC and 24 CD) of the patients with a median FC value of 106 ug/g after induction therapy, while the 27 patients not in clinical response at 1-year had a post-induction median FC value of 300 ug/g (p = 0.0002). The ROC curve analyses showed that FC levels after anti-TNF induction have a sensitivity of 83% and a specificity of 74%, with a cut-off of 168 ug/g, for predicting clinical response at 1-year (p = 0.0001). In CD we detected a sensitivity of 83% and a specificity of 75%, with a cut-off of 125 ug/g, for predicting clinical response at 1-year (p = 0.0001), while in UC we detected a sensitivity of 100% and a specificity of 79%, with a cut-off of 121 ug/g, for predicting mucosal healing at 1-year (p = 0.0001). Survival analysis showed for patients with FC above 168 ug/g after anti-TNF induction course a probability of being in clinical response at 1-year significantly lower than for patients with FC less than 168 ug/g (HR 0.20, p = 0.0001). The same analysis conducted in the UC patients showed a HR of 0.16 (p = 0.037), while for CD patients with FC above 125 ug/g after anti-TNF induction course the HR was 0.22 (p = 0.0008).
Our preliminary results suggest that faecal calprotectin can be used as a marker to estimate anti-TNF agents's efficacy in IBD and to predict clinical response and mucosal healing at 1-year.