Glucagon-like peptide-1 receptor analogs linked to fewer blood clots than dipeptidyl peptidase-4 inhibitors in people with rheumatoid arthritis

Glucagon-like peptide-1 (GLP-1) analogs, widely used in type 2 diabetes mellitus (T2DM), have shown reduced risks of cardiovascular events, stroke, thrombotic events and all-cause mortality in populations with diabetes and obesity in previous studies [1, 2]. In contrast, DPP4i have not shown cardioprotective benefits, while they are effective in control of T2DM [1]. It is thought that this is likely via anti-inflammatory activity [1], and by attenuating platelet aggregation [2].

Rheumatoid arthritis (RA) is associated with higher risk of cardiovascular events [3] and thromboembolic events [4]. Whether the same benefit is achievable in persons with RA has been less clear. A recent population-based cohort study showed GLP-1 analog use reduced risk of stroke, major adverse cardiovascular events (MACE), and all-cause mortality [5]. Notably, in the RA subgroup, which comprised 38% of the total study population, GLP-1 analog use was also associated with a significant reduction in MACE and all-cause mortality. However, the impact on thromboembolic events, particularly deep vein thrombosis (DVT) and pulmonary embolism (PE), remains unclear. Here we evaluated a real-world cohort of RA and T2DM, focusing on arterial and venous thrombosis outcomes.

RA is one of the most prevalent chronic inflammatory autoimmune diseases with joints being predominately involved. It carries increased risk for cardiovascular disease [3], as well as venous thrombotic events [4]. Though the exact etiology of RA associated thrombosis is not completely clear, systemic inflammation leading to endothelial dysfunction and hypercoagulation are thought to play a role [6]. In RA, early endothelial dysfunction is driven by inflammatory cytokines including tumor necrosis factor-α (TNF-α), interleukin (IL)−1, and is reflected by increased C-reactive protein (CRP). This leads to increased expression of adhesion molecules and enhanced leukocyte and platelet adhesion to the endothelium, as well as increased permeability, and plasminogen activator inhibitor-1 (PAI-1), predisposing to thrombosis and disease progression [7]. Furthermore, RA can upregulate a hypercoagulability state by increasing inflammatory markers like fibrinogen, D-dimer, cytokines (IL-6, IL-8, TNF), and CRP, which drive coagulation at both extravascular and intravascular sites [6]. Activated platelets and microparticles (MP), influenced by oxidative stress and autoimmunity, can also enhance thrombin generation and platelet activation [8].

The cardiovascular protective and anti-thrombotic effects of GLP-1 analogs can be attributed to multiple mechanisms. Research has shown that GLP-1 analogs reduce inflammation through various pathways, including decreasing the production of reactive oxygen species [9], and lowering levels of inflammatory cytokines such as TNF-α, IL-1β, and IL-6 in mononuclear cells [10]. Additionally, they inhibit nuclear factor (NF)-κB binding in mononuclear blood cells, which may contribute to their anti-atherogenic effects [11]. GLP-1 analogs also show efficacy in reducing Thromboxane A2 (TXA2), a pro-inflammatory mediator involved in platelet activation [2]. Moreover, as obesity is linked to chronic inflammation originating from adipose tissue [12] and GLP-1 analogs are effective in promoting weight loss [13] thus reduce inflammation. DPP4i, which also act on incretin-based pathways like GLP-1 analogs, have not been shown to provide significant cardiovascular protection or anti-thrombotic effects [1]. While they share a similar pathway and demonstrate some anti-inflammatory properties [14], they lack evidence of effects on inhibiting endothelial function [15] and weight loss. Further research is needed to determine whether the anti-inflammatory effects of GLP-1 analogs are more pronounced compared to DPP4i, though the identified effects on TNF, IL1B and IL6 do provide overlap with those pathways engaged in RA, and provide an area of focus for future studies.

While a recent meta-analysis by Liu et al. [16] reported that long-term use of GLP-1 analogs was associated with increased risk of DVT in general population with T2DM or other metabolic syndromes, contrasting with our findings, our study focused on a distinct population with RA who carry chronic systemic inflammation, and have unique pathophysiological mechanisms including endothelial dysfunction and upregulated inflammatory cytokines which are known to increase thrombotic risk. In this context, the anti-inflammatory effects of GLP-1 analogs may confer protective benefits in RA populations and reduce thrombotic risk.

This study has several limitations. First, as a retrospective analysis using the TriNetX database, patient identification relied on ICD-10 codes rather than classification criteria, introducing risks of misdiagnosis and overdiagnosis. Second, despite propensity score matching, unmeasured confounders, such as duration and dosage of glucocorticoid use, the length of tobacco use, family history, and medication compliance may remain. Third, even after matching, the BMI in the GLP-1 analog group was significantly higher than in the DPP4i group, likely because patients with higher BMI are more inclined to use GLP-1 analogs. Nonetheless, the GLP-1 analog cohort exhibited a lower risk of thrombotic events despite starting with a higher BMI. Fourth, some factors could not be quantified using the TriNetX database, such as atherosclerotic cardiovascular disease (ASCVD) score. Fifth, due to the limitations of the database, we were unable to include RA-specific information such as disease duration, disease activity including the Disease Activity Score in 28 joints (DAS28) and the Clinical Disease Activity Index (CDAI), and seropositivity defined by RF and anti-cyclic citrullinated peptide antibodies (anti-CCP). Whether serologic status influences the reduced thrombotic risk associated with GLP-1 analog use warrants further investigation. Sixth, DMARD exposure was identified but RA duration was unavailable. Seventh, although TriNetX collects laboratory data such as CRP and erythrocyte sedimentation rate (ESR), we were unable to access individual patient charts to verify the etiologies of abnormal results and variability precluded their use in matching. Eighth, we did not compare the changes in HbA1c between the two groups because these levels were measured multiple and differing times throughout the study period, preventing standardized comparisons between groups.

This study demonstrated that GLP-1 analog use is associated with decreased risk of all types of thrombotic events including arterial as well as venous thrombosis, with decreased all-cause mortality compared with DPP4i. These findings offer valuable insights into the potential role of GLP-1 analogs in managing RA patients with co-existing diabetes and provide perspective on inflammatory pathways that may be particularly relevant in persons with RA and T2DM.

Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 22 KB)