Background Chronic inflammation and immune dysregulation may have a role in the increased cardiovascular (CV) risk in Rheumatoid Arthritis (RA) patients, but the actual mechanisms remain unclear. Although the presence of antibodies against HDL lipoproteins has been reported in other autoimmune conditions, whether they are present in RA patients is still unknown.
Objectives To evaluate the presence of IgG anti-HDL antibodies in RA patients and its potential associations with clinical features, traditional CV risk factors and proinflammatory cytokines.
Methods IgG anti-HDL antibodies and total IgG levels were quantified by ELISA techniques in serum samples from 212 RA patients, 131 sex- and age-matched healthy controls (HC) and 52 subjects with marked traditional CV risk factors (CVR: 22 (42.3%) diabetes, 32 (61.5%) hypertension, 32 (61.5%) dyslipidemia, 19 (36.5) smoking and 17 (32.6%) obesity). A subgroup of 13 RA patients, naïve to TNFα-blockers was prospectively followed upon TNFα-blockade. The cut-off for anti-HDL/IgG positivity was considered the 90th percentile of the HC group. Lipid blood analyses were performed by conventional methods. TNFα, IFNα, MIP1α, IFNγ, IL-8, VEGF, GM-CSF, IL-17, MCP1, SDF-1α, resistin and leptin serum levels were quantified by immunoassays.
Results IgG anti-HDL levels were higher in RA patients compared to both HC (p<0.0001) and CVR subjects (p=0.015). Differences with HC remain after correction by total IgG levels (p<0.003). Anti-HDL/IgG were correlated with HDL-cholesterol levels in patients at onset (n=47, r=-0.502, p=0.002), which exhibit lower HDL levels than their long-standing counterparts (p<0.001). Similarly, multiple regression analysis revealed that anti-HDL/IgG levels were negatively associated to HDL levels in the whole RA population (-1.182 [-1.823, -0.541], p=0.0003) after adjusting for demographical, clinical, inflammatory parameters and treatments. RA patients classified as anti-HDL/IgG-positive (n=40, 18.8%) were more likely to have experienced a CV event than their negative counterparts (40.0 vs 12.7%, p<0.0001), but no differences were registered in traditional CV risk factors, except for dyslipidemia. Similarly, these patients exhibited higher levels of several proinflammatory mediators (CRP p<0.001, IFNα p=0.006, MIP1α p<0.001, IFNγ p=0.004, IL-8 p=0.006, GM-CSF p=0.033, IL-17 p=0.002 and MCP1 p=0.024). Finally, DAS28 improvement upon TNFα-blockade was associated with decreasing anti-HDL/IgG (r=-0.654, p=0.015) and increasing HDL levels (r=0.699, p=0.011), thus being the change in anti-HDL/IgG antibodies parallel to that of HDL levels (r=-0.664, p=0.018). However, TNFα blockade did not significantly change HDL or anti-HDL/IgG levels (p=0.170 and p=0.101, respectively).
Conclusions Anti-HDL antibodies are increased in RA patients independently of commorbidities and associated with a proinflammatory milieu and impaired lipid blood profile which may contribute to the increased rate of CV events in these patients.
Acknowledgements This work was supported by European Union FEDER funds and the Fondo de Investigaciόn Sanitaria (FIS, PI12/00523). J.R.-C. is a recipient of a FPU fellowship from the Ministerio de Educaciόn (Spain).
Disclosure of Interest None declared