Background Calprotectin is a major S100 leucocyte protein, associated with disease activity in rheumatoid arthritis (RA) patients. Calprotectin is a potentially biomarker more sensitive of disease activity than conventional acute-phase reactans.
Objectives To evaluate the performance of calprotectin serum levels in stratifying disease activity in RA patients treated with TNF inhibitors (TNFi) compared with acute phase reactants.
Methods Cross-sectional study, including consecutive RA patients (ACR 1987 criteria) from our arthritis unit receiving etanercept, adalimumab or infliximab. DAS28, SDAI, CDAI, joint counts, ESR, CRP and calprotectin serum trough levels were measured. Associations between calprotectin, ESR and CRP and articular indices were analysed by correlation and linear regression.
Results 87 patients were included; mean duration of biological treatment was 82±46 months. 47 patients were in remission/low disease activity (DAS28≤3.2), and 40 had moderate/high disease activity (DAS28>3.2). Serum calprotectin levels were higher in active patients than in those in remission/low disease activity (2.21±1μg/mL vs. 4.28±2μg/mL, p≤0.001). Calprotectin, CRP and ESR distinguished between patients in remission/low disease activity and those with moderate/high disease active disease, according to all indices assessed. Calprotectin levels were significantly-lower in patients in remission compared to those with low disease activity according to all articular indices analysed, whereas ESR discriminated between disease states only according to DAS28, and CRP only according to SDAI. Calprotectin, but not CRP or ESR, strongly correlated with all composite indices and the 28 SJC/TJC (all r coefficients over 0.50). In patients in remission/low disease activity, calprotectin but not CRP or ESR, distinguished between patients with no swollen joints and those with ≥1 swollen joints (1.95±1μg/mL vs. 3.07±1μg/mL, p=0.010). Adjusted analysis showed a significant association between serum calprotectin levels and DAS28 according to different covariates (combined therapy, reduced dose, use of glucocorticoids, disease duration, autoantibody status and erosive disease). Backward selection of variables did not substantially modify the association between calprotectin and DAS28. The accuracy analysis with “activity by DAS28≤3.2” as the reference variable showed an AUC of 0.846 (95% CI 0.763 to 0.930, p≤0.001) with a cut-off calprotectin value of ≥3.2. μg/mL. The cut-off had a sensitivity of 77.5% and a specificity of 81.9%.
Conclusions Calprotectin is an accurate biomarker of disease activity in RA patients receiving TNFi therapy and show a better correlation with all disease states, including remission and low disease activity, than acute phase reactants (ESR and CRP).
Disclosure of Interest None declared