Background/Purpose: Monitoring disease activity in giant cell arteritis (GCA) and polymyalgia rheumatica (PMR) remains challenging as standard inflammation parameters, i.e., C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR), are not specific for rheumatic disease activity. Furthermore, tocilizumab (TCZ), meanwhile a cornerstone of GCA treatment, inhibits interleukin-6 (IL-6)-dependent synthesis of CRP. Calprotectin (CALPR) and serum amyloid A (SAA) are promising alternatives, especially since they are at least largely independent of IL-6. However, there are few prospective data on large cohorts. Our aim was to evaluate the value of CALPR and SAA for indicating disease activity in GCA/PMR.

Methods: In a prospective cohort study, CRP, ESR, CALPR, and SAA were measured repeatedly at each visit in 98 patients with confirmed GCA (N=81) or PMR (N=17). A total of 576 visits (=measurements) were performed with a mean of 5.9 ± 2.7 visits for each patient. Patients were classified as “active disease” by clinical evaluation according to current guidelines at 131 visits (=22.7% of all visits). TCZ was administered at 140 visits. To test the predictive power of patients’ active status based on all measurements, we performed univariate logistic regression. For the resulting receiver operator curves (ROC), 99% confidence intervals of the area under the curve (AUC) were calculated based on a bootstrap approach (9999 replicates). Because patients were measured multiple times, we modeled the log value of all measurements using a mixed-effects model with the random factor “patient.” These values were modeled using the numeric factors BMI and age and the binary factors sex (male/female), prednisolone (yes/no), MTX (yes/no), diagnosis (GCA/PMR), and active (yes/no).

Results: ROC analysis of examinations without TCZ showed the best AUC for CRP (0.76; 0.69-0.83), followed by SAA (0.74; 0.66-0.82), ESR (0.7; 0.63-0.78), and CALPR (0.66; 0.59-0.73). For visits with TCZ, SAA showed the best AUC (0.73; 0.5-0.9), then CRP (0.58; 0.41-0.77), CALPR (0.55; 0.32-0.74), and ESR (0.48; 0.31-0.68). In the adjusted model for visits without TCZ, CRP, ESR, and SAA showed highly significant mean change in active disease (p >0.001), CALPR showed significant change (p < 0.05). For visits with TCZ, no parameter showed significant mean change with active disease.

Conclusion: SAA shows better AUC than the other seromarker in patients treated with TCZ. In patients without TCZ treatment, the AUC of CRP and SAA are almost equal. However, after multivariate regression, no significant effect is detectable for any of the tested parameters.

« Back to ACR Convergence 2023

ACR Meeting Abstracts - https://acrabstracts.org/abstract/calprotectin-s100a8-s100a9-and-serumamyloid-a-are-not-superior-to-c-reactive-protein-and-erythrocyte-sedimentation-rate-in-monitoring-disease-activity-in-giant-cell-arteritis-treated-with/