Background/Purpose:

Positive rheumatoid factor (RF) and/or anti-citrullinated protein antibodies (ACPA) identify a major rheumatoid arthritis (RA) subgroup, characterized by greater propensity to erosive joint damage and extra-articular disease manifestations as compared to seronegative patients. The purpose was to determine the associations between RF/ACPA and changes in functional immune response signatures over 5 years in patients with RA.

Methods:

A longitudinal analysis was performed of baseline and 5-year follow-up data from a prospective study of adult RA patients in a population-based incidence cohort. Data were available for 324 patients at baseline and for 155 patients at 5 years. Peripheral blood leukocytes (PBL) were freshly collected from patients and isolated by Ficoll density gradient centrifugation at both time-points. PBLs were stimulated at 37°C for 48 hrs under six separate conditions: anti-CD3/anti-CD28 monoclonal antibodies (CD3/CD28); phytohemagglutinin (PHA), staphylococcal enterotoxins A and B (SEA/SEB); phorbol myristate acetate and ionomycin (PMA); CpG oligonucleotides (CpG ODN2006); and medium alone. A panel of 17 cytokines and chemokines was analyzed using multiplexed immunoassays (Meso Scale Discovery). RF was measured using the nephelometry (CSL Behring), and ACPA was measured by Quanta Lite CCP3 IgG (Inova Diagnostics). Factor analysis of normalized data was performed using baseline data on 324 patients to make inferences about underlying cell types based on cytokine profiles. Among the 155 patients with follow-up, Spearman methods were used to assess correlations between RF and/or ACPA seropositivity and longitudinal changes in the factor scores for each stimulus, adjusting for age and sex.

Results:

A total of 109 (70%) of the 155 patients were seropositive at baseline (median age: 57.7 yrs; median disease duration: 9.7 yrs). RF/ACPA positively associated with CpG factor 1 (r = 0.28, p <.001), factor 2 (r = 0.21, p =.011), and factor 3 (r = 0.20, p =.015). Based on loadings >0.5, CpG factor 1 comprised IL-1β, G-CSF, TNF-α, MIP-1β, and IL-6; factor 2 comprised IFNγ, IL-2, and IL-13; and factor 3 comprised IL-12, and IL-5. Previous studies have shown that the type of CpG used in this study mainly activates B cells. Responses to medium alone paralleled those to CpG, as RF/ACPA positively correlated with medium alone factor 1 (r = 0.27, p <.001) and factor 3 (r = 0.29, p <.001) responses. Medium alone factor 1 comprised GM-CSF, MCP-1, IL-17A, IL-6, MIP-1β, TNF-α, and G-CSF; factor 3 comprised IL-1β and TNF-α. Based on this profile, multiple Toll-like receptor (TLR) ligands present in the medium fetal calf serum could have stimulated monocytes or other antigen presenting cells. In contrast, RF/ACPA negatively correlated with CD3/CD28-induced IL-5 (r = -0.21, p =.008), IL-17A (r = -0.22, p =.005), and GM-CSF (r = -0.20, p =.013).

Conclusion:

Our findings suggest that sustained stimulation of RF-containing immune complexes increases monocyte and B-cell activation states with increased expression of TLR. This could inform the development of functional assays for CpG and other TLR immune responses as predictors of response to B-cell directed therapy in RA.

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/longitudinal-associations-between-rheumatoid-factor-and-ex-vivo-cytokine-production-reveal-novel-mechanistic-insights-into-rheumatoid-arthritis/