Background Dissociation between the responses of a number of acute-phase proteins (APP) is well known. C reactive protein (CRP) levels have been used extensively to model disease activity in rheumatoid arthritis (RA). However, biologic therapy reduce CRP levels, even without associated reductions in RA disease activity. Serum amyloid A (SAA) also models RA disease activity and has been suggested as a better biomarker for RA disease activity.

Objectives To determine if SAA levels had better correlation with conventional clinical and serological assessments in RA than CRP and ESR.

Methods A cross-sectional study was performed. Samples were analyzed from RA patients under biologic therapy of a reference hospital in northern Portugal. We compared the correlation between SAA, CRP and ESR levels with swollen and tender joints counts (SJC and TJC), DAS28-4v, SDAI, CDAI, HAQ and visual analogue scale for pain (VAS-P). Correlation was calculated using the Spearman rank correlation (r). P-values <0.05 were regarded as significant.

Results 173 patients were evaluated, 86.7% (n=150) were women. The mean (SD) age was 56 years (10.75). Mean disease duration was 16.31 years (8.87). Most of patients had positive serology (rheumatoid factor and/or anti-CCP antibody) (n=146, 84.4%). Ninety eight patients (56.6%) were under TNF antagonists, 23.7% (n=41) were under rituximab and the remaining under tocilizumab (TCZ).

SAA levels were moderately correlated with CRP levels (r=0.49, p<0.001) but there was no statistically significant correlation with ESR (r=0.03, p=0.75).

Correlation analysis of SAA and CRP levels with several conventional assessments showed (SAA and CRP, respectively): SJC - 0.17 vs 0.08, TJC - 0.18 vs 0.09, DAS28-4v - 0.32 vs 0.41, SDAI - 0.23 vs 0.18, CDAI - 0.18 vs 0.11, ESR – 0.32 vs 0.55; all p values <0.05 except SJC, TJC and SDAI for CRP values. There was no statistically significant correlation of these acute-phase protein with HAQ and VAS-P. ESR levels showed very weak correlation with all parameters (p>0.05), except with DAS28-4v (r=0.58, p<0.001).

We also noted that SAA levels were raised (>6.4mg/L) in 37.6% (n=65) of patients in which the CRP concentration was normal (<0.3mg/dL). However, only 6% (n=11) with normal SAA levels had raised concentrations of CRP. When evaluated SAA and CRP levels in patients with active disease (DAS28-4v ³2.6, SDAI>3.3 and CDAI>2.8) we found more patients with normal CRP values than SAA (62 vs 26, 82 vs 32, 84 vs 33, respectively).

Interestingly, when comparing the effects of biological therapy on APP we noted that in TCZ group the CRP, ESR and SAA levels were lower than in group under TNF antagonists and rituximab (p<0.001 for all, Mann-Whitney test).

Conclusions The lack of a strong correlation between SAA and CRP or ESR levels coupled with their better correlation with markers of RA activity suggests that changes in their levels may provide a more sensitive indicator of disease activity, especially during treatment with biologic therapy.

Disclosure of Interest None declared