P136 Faecal calprotectin as therapeutic target in patients with Crohn’s disease treated with anti-TNF: a meta-analysis of individual data

A. Mami1, S. Nancey2, D. Pugliese3, P. Molander4, G. Boschetti2, I. Nogueira5, B. Pereira6, A. Buisson1

1CHU Estaing, Department of Digestive and Hepatobiliary Medicine- Department of Gastroenterology, Clermont-Ferrand, France, 2CHU Lyon-Sud, Gastroenterology, Lyon, France, 3Fondazione Policlinico A. Gemelli IRCCS- Rome- Italy, IBD Unit Presidio Columbus, Rome, Italy, 4Helsinki University Hospital- Meilahti Hospital, Hus, Helsinski, Finland, 5Federal University of São Paulo–UNIFESP-, Gastroenterology, Sao Paulo, Brazil, 6CHU, Biostatistics unit, Clermont-Ferrand, France


Faecal calprotectin (Fcal) could be a non-invasive alternative to endoscopy to assess therapeutic efficacy in patients with Crohn’s disease (CD). We performed a systematic review and meta-analysis of individual data to assess the performances of FCal after anti-TNF induction therapy to predict mid-term steroids-free clinical remission (CFREM) in CD.


We included all the studies reporting patients with CD receiving anti-TNF agents with Fcal measurement before and after induction therapy (from 8 to 14 weeks) with follow-up > 6 months, no therapeutic intervention based on FCal during the follow-up and mid-term clinical evaluation between week 32 and week 54. CFREM was defined as CDAI < 150 and continuation of the same anti-TNF agent without steroids.


We identified five articles including 165 patients with CD (mean age = 34.5 ± 12.8 years, median CD duration 5 years [2–12], female gender = 52.7%, active smokers = 33.3%, prior bowel resection = 25.6%, CD location (L1 = 32.1%, L2 = 15.8%, L3 = 52.1%), perianal lesions = 22.4%). The patients were treated with infliximab (50.9%), adalimumab (47.3%) or certolizumab (1.8%), and received concomitant immunosuppressants in 48.5% of the patients. Mid-term CFREM was achieved in 57.3% of the patients. Fcal level after anti-TNF induction therapy was lower in patients who achieved mid-term CFREM (100 µg/g [48–267] compared with those who did not (491 µg/g [165–856]; p < 0.001). The relative decrease of Fcal between before and after induction therapy was also associated with mid-term CFREM (p = 0.022). In multivariable analysis adjusted on CDAI, CRP and disease duration, FCal level (p = 0.014) and relative decrease of anti-TNF (p = 0.025) after induction were associated with mid-term CFREM. Cut-off values between 200 and 400 µg/g showed close performances to predict mid-term CFREM. FCal <250µg/g predicted mid-term CFREM with: Se = 73.3%[63.0%-82.1%], Spe = 67.7%[54.9%-78.8%], PPV = 75.9%[65.5%-84.4%] and NPV= 64.7%[52.2%-75.9%]. A relative decrease ≥ 50% between baseline and the end of induction was the best threshold to predict mid-term CFREM (Se = 59.6%[48.6–69.8%], Spe = 78.5%[66.5–87.7%], PPV = 79.1%[67.4–88.1%] and NPV = 58.6% [47.6–69.1%]). We found a significant study effect (ICC =0.36 ± 0.18) due to the variability according to FCal assay. The likelihood of being in CFREM at mid-term (PPV) according to the combination of CDAI, CRP and Fcal was presented in Figure 1.


The improvement of Fcal after anti-TNF induction therapy is predictive of mid-term CFREM in patients with IBD. The combination of clinical remission, CRP normalisation and FCal improvement should be a therapeutic target in patients with CD. Our work highlighted the need of international standardisation of Fcal measurement.