Serum calprotectin is useful to confirm inflammatory bowel disease activity but not to predict relapse

Aim

Serum calprotectin (SC), a novel biomarker of inflammatory bowel diseases (IBD), has been recently investigated with conflicting results. The purpose of this study was to assess the ability of SC to predict relapse in IBD patients treated by biologic therapies, and to evaluate the correlation between SC, clinical and endoscopic relapse and other biomarkers as fecal calprotectin (FC) and C-reactive protein (CRP).

Methods

All consecutive IBD patients in deep remission (clinical, endoscopic or imaging remission) were followed 12 months in this prospective study. Blood and stool samples were collected for SC, serum CRP and FC. SC was measured the day of inclusion (baseline, D0), 3 months (M3) and at 6 months (M6) or during the study period for clinical relapse. Relapse was defined as clinical, biomarkers, or endoscopic/imaging activities. Evolution of SC was quantified before relapse to analyze a predicting value of loss of response (LOR). SC for patients with active IBD and those with symptoms without inflammation were also compared.

Results

Among the 119 patients included, 54 (46.4%) patients experienced a disease relapse during follow-up. Median SC levels did not increase in patients with clinical relapse (3.15 µg/ml at baseline, 3.38 µg/ml at M3, 3.33 µg/ml at M6 and 3.99 µg/ml in case of relapse (p = 0.63)). SC were compared during relapse in patients with endoscopic remission but clinical symptoms defined as secondary Irritable Bowel Syndrome (IBS). SC levels were higher in active IBD and similar between the groups of patients with IBS or deep remission (3.05 µg/ml IBS vs 2.99 µg/ml remission vs 5.1 µg/ml for clinical relapse, p = 0.04). In patients with clinical symptoms, SC presents a good predictive value for relapse (AUROC 0.764, IC95: 0.68–0.88), with a sensitivity of 72%, a specificity of 77%, using a cut-off value of 4.45 µg/ml. A weak, but significant correlation was found between SC and FC levels (r = 0.35, P = 0.001). A combined score with CRP, FC and SC is not efficient to improve IBD diagnostic.

Conclusion

SC was significantly higher in patients with clinical relapse compared to those with endoscopic remission with or without clinical symptoms. SC allow to discriminate patients with active IBD or with IBS but failed to predict relapse.