Abstract Category: Clinical Research
Title of Abstract: Serum Amyloid A Levels, in Early Aggressive Rheumatoid Arthritis, are Differentially Lowered by Treatment with TNF Antagonists Compared to Oral DMARDs
Authors: Aarat M. Patel, Douglas W. Chew, Danielle Goudeau, Donald M. Jones, S. Louis Bridges, Jr., Stacey S. Cofield, Stephen R. Wisniewski, Larry W. Moreland and Marc C. Levesque
Background/Purpose: Amyloidosis in rheumatoid arthritis (RA) is due to abnormal tissue deposition of serum amyloid A (SAA). Therapies aimed at the underlying inflammation can ameliorate RA, reduce serum SAA levels and reduce amyloid deposits. TNF antagonists have yielded remarkable benefits in case studies of RA patients with amyloidosis, and in vitro studies suggest that blocking TNF reduces SAA production. Therefore, our objectives were to determine 1) whether RA patients treated with TNF antagonists compared to oral disease modifying anti-rheumatic drugs (DMARDs) have greater decreases in serum SAA levels and 2) whether there were associated changes in SAA and CRP levels during treatment.
Methods: Serum SAA levels were measured by ELISA (Invitrogen, Inc.) and high-sensitivity C-reactive protein (CRP) assays were performed in a clinical lab by nephelometry from a cross-sectional sampling of RA subjects (n = 115) seen in local clinics who met the 1987 and/or 2010 ACR RA diagnostic criteria. A DAS28-CRP was determined for each subject. Analyses of SAA and CRP levels were also performed on samples from the Treatment of Early Aggressive RA (TEAR) study. TEAR subjects were treated initially with either methotrexate (MTX) plus etanercept (ETN), triple oral DMARDs (MTX, sulfasalazine and hydroxychloroquine), or MTX only. The latter group was stepped up to MTX plus ETN or triple oral DMARDs at 24 weeks or remained on MTX only if the DAS28-ESR was < 3.2. Blood samples and clinical disease activity measures (DAS28-ESR) were collected at baseline and 24, 48, and 102 wks after baseline. Serum SAA and hsCRP levels were measured by nephelometry from TEAR samples. In the cross-sectional analysis, SAA and CRP levels were compared between subjects grouped on the basis of treatment with TNF antagonists or oral DMARDs using log-normalized data and unpaired t tests. For continuous variables, Spearman’s correlation was determined, and Fisher’s exact test was used for contingency analyses. In the longitudinal analysis of TEAR samples, we used a mixed effects regression model.
Results: In the cross-sectional analysis, RA subjects treated with TNF antagonists had lower mean SAA levels compared to subjects treated with oral DMARDs (107 vs. 204 µg/mL, respectively; p = 0.0824). SAA levels were positively but weakly correlated with CRP levels (r2 = 0.2671; p < 0.0001). In the longitudinal analysis of TEAR subjects (n = 198), there were equal reductions in disease activity in all 5 treatment groups. SAA and CRP levels declined significantly from baseline in all groups (p<0.0001). In contrast to disease activity, there were significant differences in SAA and CRP levels between treatment groups (p=0.0207 and 0.0118, respectively) with greater reductions in SAA and CRP in the groups treated with MTX plus ETN. There was a positive, moderate correlation between SAA and CRP levels in the TEAR subjects (r2 = 0.4346; p < 0.0001).
Conclusions: Our results indicated that RA patients treated with TNF antagonists have lower SAA levels than patients treated with oral DMARDs. The lack of a strong correlation between CRP and SAA levels suggests that oral DMARDs and TNF antagonists differentially regulate SAA levels. This may have important implications for the treatment of patients with amyloidosis due to RA.