Introduction

Rheumatoid arthritis (RA) is a chronic inflammatory disease which affects the synovial membrane causing joint damage and physical dysfunction [1]. Patients with RA suffer cardiovascular (CV) complications 1.5–2 fold than the general population [2]. Atherosclerosis constitutes an abnormal thickening of arterial walls as a result of endothelial layer dysfunction. Nowadays, it is considered a chronic inflammatory disease, as inflammation molecules and pathways play a significant role in all stages of the atherosclerotic process [3].

Oxidized LDL (oxLDL) plays a central role in the pathogenesis of atherosclerosis, promoting the activation of endothelial cells [4, 5]. Increased lipid oxidation has been described during inflammation, and oxLDL has been detected in the synovial membrane and synovial fluid of patients with RA [6,7,8,9]. Also, it has been reported that patients with RA and active disease had significantly increased serum oxLDL levels compared to patients without active disease [10]. Some studies have conflicting results questioning the pathogenic role of oxLDL in the development of CV disease in these patients, but most have confirmed that oxLDL is increased in RA [11, 12]. Many studies have tried to fill gaps in knowledge concerning the role of autoantigens and their associated autoantibodies in the development of atherosclerosis. In the case of autoantibodies against oxidized LDL (anti-oxLDL), data are ambiguous. Several studies have shown an atheroprotective effect of anti-oxLDL antibodies, while others are not in accordance with these results [13]. Indeed, various investigators assessed the association of IgM and IgG anti-oxLDL antibodies and cardiovascular events in subjects without CV disease. They found that IgM anti-oxLDL antibodies appear to have a protective effect for several CV events [14, 15], while the relationship of IgG anti-oxLDL is more complex and needs further investigation [13, 16].

In patients with RA, the association between lipid levels and CV events is less proportional than in the general population (lipid paradox in RA) [17]. In the present study, we investigated the impact of biological (b) disease-modifying anti-rheumatic drugs (DMARDs) on subclinical atherosclerosis in RA. Changes in the titer of anti-oxLDL antibodies after bDMARD initiation were also evaluated, and we tried to provide evidence regarding their general implication in the atherosclerotic process. An additional aim of our study was to determine possible relations among anti-oxLDL antibodies, inflammation, and atherosclerosis.

Methods

Study design and patients

This is a prospective controlled study conducted in a single University Centre between June 2016 and November 2020. During this time period, 753 RA patients visited the Rheumatology Department of the University Hospital of Ioannina, Greece. Among them, 70 patients met the inclusion criteria and were enrolled in the study.

Patients with no previous history of CV disease, eligible for the study, were those aged 18–65 years old, fulfilling the 2010 American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) classification criteria for RA [18]. They were receiving conventional synthetic (cs) DMARDs, such as methotrexate (MTX) or leflunomide (LFN). Biologic agents were administered in these patients after being carefully evaluated and screened appropriately [19, 20]. Exclusion criteria were the following: age above 65 years, established or familiar history of coronary artery disease, history of smoking, valvular heart defects, chronic infectious diseases, thyroid gland disorders, diabetes mellitus, neoplasms, chronic renal failure, hypertension, dyslipidemia of any type and metabolic syndrome, and patients that received any agent that can influence the lipid profile such as statins, diuretics, thyroid drugs, and others.

Thirty-one apparently healthy, non-smoking volunteers, with no previous history of CV disease, participated in the study and were used as a control group. These subjects were randomly selected during their routine medical check-up tests, the last 3 months of the patients’ selection period, and fulfilled the same exclusion criteria reported for the patient group. A written informed consent form was signed by all patients, and the study protocol was approved by the institutional ethics committee.

Clinical assessment

RA patients underwent a complete physical examination at baseline and after 6 months of bDMARD initiation. Age, gender, body mass index (BMI), disease activity score for 28-joint count indices based on C-reactive protein values (DAS28-CRP), and extra-articular manifestations of RA were recorded.

Laboratory monitoring

Blood samples were obtained at baseline and after 6 months of bDMARD initiation. Total cholesterol (TC), triglycerides (TG), and high-density lipoprotein cholesterol (HDL-C) were determined by a clinical chemistry analyzer. Low-density lipoprotein cholesterol (LDL-C) levels were calculated using the Friedewald formula, while CRP was measured by nephelometry. Erythrocyte sedimentation rate (ESR) was measured by an automatic analyzer. IgM rheumatoid factor (IgM RF) was measured by latex test. Concentrations of anti-citrullinated protein/peptide antibodies (ACPAs) were determined by immune enzyme analysis (ELISA).

Determination of antibody titer against oxLDL

ELISA sandwich

Serum of patients was incubated on ELISA plates containing immobilized antigen which in this case was oxidized LDL (oxLDL) after ultracentrifugation from pooled fresh plasma of normolipidemic volunteers. The unbound antibody was removed by multiple rinsing. Then, incubation was performed with HRP-conjugated rabbit anti-human IgG monoclonal antibody (Dako Cytomation) (diluted 1:1000, v/v). Excess of the secondary antibody was removed by successive rinsing, and the antigen complex with the antibodies remained immobilized on the ELISA plate—color development followed by the addition of hydrogen peroxide and chromophore (o-phenylene diamine) as substrates [21]. The previous reaction was stopped by the addition of hydrochloric acid. The results of anti-oxLDL titers are expressed as optical density (OD) values. All measurements were performed at 492 nm, in a multifunctional microplate reader (InfiniteM 200 Pro, Tecan, Italy), before and after 6 months of biological treatment initiation.

Subclinical atherosclerosis evaluation

For the evaluation of the carotid intima-media thickness (cIMT), we used a GE Healthcare LOGIQ E9 ultrasound machine, equipped with a GE 9L Linear Array Probe with a maximum frequency of 10 MHz. All the measurements were performed by a single observer unaware of the participants’ clinical or laboratory characteristics. Patients were examined in a supine position with the neck extended. The distance between the lumen-intima interface and the leading edge of the media-adventitia interface of the far wall corresponds with cIMT. The measurement of cIMT was always performed at the same location in the same common carotid artery, proximal to the carotid bifurcation, both at baseline and at 6 months. Values were expressed in millimeters [19].

Statistical analysis

Statistical analysis was performed using SPSS Statistics, version 21.0. Data are expressed as mean (SD) or median (IQR) as appropriate. Variable distribution was tested for normality. Comparisons between groups were made with the chi-square test for categorical variables and the independent sample t-test or the Mann–Whitney U test for continuous variables. Changes over time were analyzed with the paired sample t-test or the non-parametric alternative Wilcoxon signed-ranks test. All tests were 2-sided and the significance level was set at 0.05.

Linear regression analyses adjusted for age and sex were performed to estimate the association between cIMT and anti-oxLDL, HDL-C, ESR, and CRP before and 6 months after the administration of biological agents and to calculate effect estimates (betas, 95% confidence intervals). Results were considered statistically significant if the p-value was < 0.05. Linear regression analyses were performed using Stata (version 13.1; StataCorp, College Station, TX, USA).

Results

Among the 70 RA patients who initially fulfilled the enrollment criteria, 11 were excluded. Five refused to receive any biological treatment, 3 had positive hepatitis B surface antigen, 2 had an abnormal chest x-ray, and one was positive for hepatitis C virus. Patient demographic data, baseline clinical characteristics, immunological profile, and their comparison with controls are presented in Table 1. The majority of patients were women, and mean age was 53.9 years with a mean disease duration of 2.8 years. The extra-articular manifestations included rheumatoid nodules in 6 patients, anemia of chronic disease in 5 patients, secondary Sjogren syndrome in 2 patients, and Raynaud phenomenon in 2 patients. All patients received csDMARDs, MTX 50 (84.6%), and LFN 9 (15.4%), while 35 received tumor necrosis factor a (TNFa) inhibitors, and 24 received non-anti-TNFa therapy. More specifically, among TNFa inhibitor users, 12 patients received adalimumab, 10 infliximab, 7 etanercept, and 6 golimumab. On the other hand, among non-anti-TNFa users, 15 patients received tocilizumab, an interleukin-6 receptor antagonist, while 9 were treated with abatacept, a T-cell modulatory agent. The choice of bDMARD therapy was based on a shared decision between patients and doctors, following the EULAR recommendations for RA management [22]. Patients presented with statistically significant lower TC and HDL-C levels and a statistically significant elevation of cIMT levels at baseline, compared with controls (Table 1).

Table 1 Clinical and laboratory characteristics of RA patients at baseline
Full size table

After a six-month follow-up, the re-evaluation of carotids revealed a statistically important decrease of cIMT values. This observation was accompanied by a statistically significant elevation of HDL-C levels and a reduction of the titer of anti-oxLDL antibodies regardless of the bDMARD that was administered (Table 2). There was also a clinical improvement, and the reduction of the DAS28-CRP score reflects a low disease activity (Table 2). A comparative analysis of parameters and change of parameters was also performed between the anti-TNF group and the group of patients who did not receive anti-TNF at baseline and after 6 months of treatment, respectively. Based on the results in Table 3, the two groups did not show a statistically significant difference regarding the parameters at baseline and their change over a 6-month follow-up.

Table 2 Changes over time in disease activity indexes, lipid profile, cIMT, and anti-oxLDL in RA patients treated with anti-TNF and non-anti-TNF for 6 months
Full size table
Table 3 Comparison analysis of parameters between the anti-TNF and non-anti-TNF groups at baseline and their change over a six-month period, after bDMARD administration
Full size table

No statistically significant association was found between the cIMT and anti-oxLDL, HDL-C, CRP, or DAS28-CRP neither before nor 6 months after the administration of biological agents. Moreover, the change of cIMT within 6 months of the bDMARD administration was not correlated with the changes in anti-oxLDL, HDL-C, or CRP respectively (Table 4).

Table 4 Linear regression analyses in order to investigate the association between carotid atherosclerosis, lipid parameters, and inflammation markers in RA patients treated with bDMARDs
Full size table

Discussion

In our study, RA patients who presented with significant subclinical atherosclerosis and high titer of anti-oxLDL antibodies at baseline ended up with favorable outcomes concerning their CV risk as there was a reduction of anti-oxLDL antibody titer and cIMT values after 6 months of bDMARD initiation. No statistically significant correlation was revealed between the cIMT and anti-oxLDL, HDL-C, or CRP neither before nor 6 months after the administration of biological agents.

Elevated HDL-C levels were also observed, even from the first 6 months of therapy. Several studies have focused on examining other CV risk factors and in particular HDL-C function. The increase in TC and LDL-C after suppression of inflammation was accompanied by a HDL-C increase, which protects from CV disease [23, 24]. In addition, there is evidence that anti-atherogenic properties of HDL-C appeared to improve when inflammation is reduced [25, 26]. Quantitative as well as qualitative lipid changes are important in the assessment of the atherosclerotic process. HDL-C particles contain, among others, apolipoprotein A1, whose main role is to prevent the oxidation of LDL-C and as a consequence to inhibit one of the first steps of atherosclerosis [25].

Anti-oxLDL antibodies are associated with atherosclerosis in the general population and have been reported in several autoimmune diseases. In a previous study, their concentrations were measured in the serum of 140 patients with RA. The analysis revealed a positive correlation with CRP and a negative correlation with HDL-C. An independent positive correlation was also found with cIMT [7]. In another study, their titer was found to be high in patients with early RA, and a reduction was observed after csDMARD administration and subsequent control of inflammation [27]. In the present study, no correlation was found between cIMT and anti-oxLDL. These contradictory findings suggest that the role of the humoral immune response to oxLDL is unclear, and the antibody formation may serve a different purpose through the various stages of atherosclerosis.

The mechanism by which anti-oxLDL antibodies participate in the pathogenesis of atherosclerosis is not certain. A logical theory is that the reduction of their titer reflects the reduction of their substrate, oxLDL, which plays a central role in the whole atherogenic process. However, there is another theory that potentially places oxLDL among the most potent epitopes. The term oxLDL refers to a set of heterogeneous oxidative changes in LDL-C subunits and ApoB (apolipoprotein B). These changes include reorganization of the phospholipid shell upon exposure of phosphorylcholine (PC) and binding of aldehydes, such as malondialdehyde (MDA), to ApoB. Oxidized phospholipids derived from oxLDL and LDL-C can promote inflammatory activation of macrophages, vascular smooth muscle cells, and other cells, acting as damage-associated molecular patterns (DAMPs). DAMPs are endogenous molecules released by cells that are damaged or shed and activate the immune system [28, 29].

Measurement of titer of anti-oxLDL antibodies may also contribute to stratification of CV risk. Nevertheless, the role of anti-oxLDL antibodies in the atherosclerotic process remains controversial. A systematic review by van den Berg et al. has highlighted that IgM anti-oxLDL antibodies appear to be protective against severe coronary artery disease and cardiovascular events, while the relationship with IgG anti-oxLDL antibodies is complex enough and difficult to elucidate [13]. More specifically, a study by Tsimikas et al. found higher serum levels of IgM oxLDL antibodies which were associated with a lower risk of developing CV endpoints [14]. van den Berg et al. also found a strong protective effect of IgM anti-MDA-LDL antibodies on future CV disease development [15]. However, other studies did not discover a significant association between IgM anti-MDA-LDL antibodies and future CV events [30, 31]. Another study by Khamis et al. found a protective association between IgG anti-oxLDL antibodies and cardiovascular endpoints with cases having lower levels of IgG anti-oxLDL than controls [16]. However, the IgG class of autoantibodies is mostly pathogenic in many autoimmune rheumatic diseases like systemic lupus erythematosus, antiphospholipid syndrome, and others [32, 33]. Despite the paucity of studies that lack definite conclusive data as regards the role of oxLDL antibodies and CV events, it seems that IgM anti-oxLDL antibodies have a protective effect for severe CV events, while the relationship of IgG-oxLDL is more complex and needs further investigation [13].

The biological role of these autoantibodies requires further evaluation, as new evidence supports the notion that they constitute not only a simple biomarker of atherosclerosis but also a key component of the innate immune system. IgM anti-oxLDL antibodies perform homeostatic functions, maintaining the equilibrium of atherosclerosis development. Perhaps in the presence of overwhelming stimuli, such as traditional CV risk factors or other inflammatory triggers, a maladaptive immune response occurs with immunoglobulin class switching to IgG and leading to accelerated atheroma deposition or plaque rupture [28, 34, 35].

Our study provides evidence concerning the implication of IgG anti-oxLDL antibodies in atherosclerosis of RA patients treated with bDMARDs. The results of the present study suggest their potential atherogenic action as the reduction of their titer after treatment was accompanied by reduced cIMT values, even if no correlation was revealed between them, probably due to a type II error. Imaging modalities and especially non-invasive carotid ultrasound are very useful in depicting atherosclerotic plaques [36]. Carotid plaques and cIMT have been reported in several studies as markers of coronary artery disease [37,38,39,40,41].

Furthermore, no statistically important difference was found between anti-TNF and non-anti-TNF users as regards the parameters and their change over time. This observation emphasizes the necessity of suppression and control of inflammation, in an attempt of managing the CV risk in RA patients [16, 17]. Indeed, it is reported that the use of anti-TNF in these patients is associated with the reduction of insulin resistance [41].

A limitation of the present study is the relatively small number of patients included as a consequence of strict inclusion criteria and the short period of follow-up; nevertheless, a pure sample was selected. To our knowledge, there are a small number of studies investigating the role of these specific autoantibodies, and this number gets smaller when we focus on RA patients and bDMARDs. We conclude that biological therapy in RA patients is accompanied by a decrease of cIMT values, followed by the reduction of anti-oxLDL. Thus, there is growing evidence that these autoantibodies may constitute an independent biomarker of CV risk prediction, but further evaluation is needed to clarify their complicated mechanism of action.