Background/Purpose: Complement activation, a key component of innate immunity and activator of adaptive immunity has been linked to RA pathogenesis. Anti-citrullinated peptide antibody (ACPA) and rheumatoid factor (RF) may cause complement activation.  We examined the association of the terminal complement component C5 and its breakdown product C5a with ACPA in a preclinical Indigenous North American (INA) population at high risk for severe RA.

Methods:  C5, C5a, and ACPA were examined by enzyme-linked immunosorbent assay ELISA, RF by nephlometry, in 267 sera from INA with RA (n=43), first degree relatives (FDR) (ACPA positive n=16; ACPA negative n=64) and FDR who later developed synovitis (FDR-S; n=10).  Samples were spun and stored at -20C.  ACPA positive FDRs samples were matched to one or two ACPA negative FDR samples based on date of sample collection.  Levels were correlated with sample storage duration to assess for in vitro complement activation.  Associations between C5, C5a, ACPA, and subject group were determined using Chi squared, Mann Whitney U tests and Wilcoxon signed ranks tests for paired samples.  Correlations were tested by Spearman correlation.  Multivariable models controlling for time stored, ACPA, RF, visit sequence and sex were generated to determine the independent influence of ACPA on C5 and C5a serum levels.  Data are reported as median (interquartile range (IQR)) and B with 95% confidence limits (CL).

Results: C5a levels increased with increasing time stored (rho=0.69; p<0.0001); C5 levels decreased (rho= -0.33;p=0.002); median (IQR) time stored 6.3(5.5) years).  This presumed in vitro activation was most important for samples stored over 5 years. In FDR samples matched for time stored, C5a and C5 levels were similar between FDR persistently ACPA positive vs  ACPA negative (C5a 17.8 (6.3) vs 17.0 (6.7) ng/ml; p=ns; C5 135.2 (81.7) vs  130.8 (87.5) µg/ml; p=ns).  Comparing all samples from FDRs, C5 levels were higher in ACPA positive vs ACPA negative samples (135.2 (46.6) vs  71.2 (75) µg/ml; p<0.001). Using linear regression to determine predictors of C5 levels in 153 samples from 80 FDR, the presence of ACPA independently predicted higher C5 level (B (CL) 42 (17-71)) after correcting for duration  of time sample stored, visit sequence, sex and RF (titre > 40 U/ml being positive). This association with ACPA was not seen for C5a. In FDR-S (n=10), ACPA titers increased prior to synovitis.  FDR-S had higher C5a levels than FDR (238 (225) vs 133 (362) ng/ml; p<0.0001) an association that was significant after correcting for duration of storage, visit sequence, ACPA, RF and sex (B=95(16-174); p=0.02).

Conclusion: In vitro complement activation may occur with prolonged sample storage and must be considered in any analysis of stored samples. C5 level is independently predicted by ACPA in FDR in an INA prospective cohort suggesting decreased activity of the terminal complement cascade in the presence of ACPA. Those who transitioned to synovitis showed evidence of increased complement activity and prior to onset of synovitis showed expansion of ACPA. ACPA related complement cascade activation in imminent synovitis cannot be confirmed based on this data.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/in-indigenous-north-americans-at-high-risk-for-ra-complement-c5-level-is-associated-with-acpa-positivity-and-c5a-with-transition-to-synovitis-even-after-correcting-for-in-vitro-complement-activation-f/