Causal relationship between circulating inflammatory proteins and atherosclerosis: a bidirectional Mendelian randomization study and meta-analysis

Background
Atherosclerosis (AS) is a chronic inflammatory disease that significantly contributes to cardiovascular morbidity and mortality. Despite extensive research efforts, the connections between circulating inflammatory proteins (CIPs) and different subtypes of AS remain poorly understood. This study aims to clarify these relationships through Mendelian randomization (MR) analysis.
Methods
We utilized summary statistics from genome-wide association studies (GWAS) that included 14,824 European participants to analyze inflammatory protein levels, alongside data from the IEU GWAS database for AS phenotypes. Our primary approach for MR analysis was the inverse variance weighted method. To ensure the validity and robustness of the causal relationships, we conducted tests for pleiotropy and heterogeneity, as well as reverse MR analysis to assess the possibility of reverse causality. Finally, we performed a meta-analysis to consolidate and interpret our findings comprehensively.
Results
Our MR analysis identified several significant associations: elevated artemin [odds ratio (OR) = 1.195], glial cell line-derived neurotrophic factor (hGDNF) (OR = 1.173), and tumor necrosis factor (TNF) (OR = 1.179) levels increased peripheral atherosclerosis (PA) risk; higher CUB domain-containing protein 1 (OR = 0.534), interleukin (IL)-8 (OR = 0.274), monocyte chemoattractant protein-3 (OR = 0.373), transforming growth factor-alpha (OR = 0.306), and tumor necrosis factor receptor superfamily member 9 (OR = 0.423) levels decreased cerebral artery atherosclerosis risk; fibroblast growth factor 21 (FGF-21) (OR = 1.122), hGDNF (OR = 1.108), and IL-22 receptor subunit alpha-1 (IL-22RA1) (OR = 1.235) levels were positively associated with coronary artery atherosclerosis (COA) risk; while IL-13 (OR = 0.909) and TNF-beta levels (OR = 0.954) were negatively associated with COA risk. C-X-C motif chemokine 6 levels (CXCL6) (OR = 1.353) and hGDNF (OR = 1.161) were identified as risk factors for atherosclerosis, excluding cerebral, coronary, and peripheral arterial disease (AECCP). In contrast, IL-2 receptor subunit beta levels (OR = 0.801) and IL-6 levels (OR = 0.788) were found to be protective factors for AECCP. Additionally, CXCL6 (OR = 1.261), FGF-21 (OR = 1.090), IL-22RA1 (OR = 1.127), and hGDNF (OR = 1.134) exhibited a risk effect against overall AS risk, while IL-6 (OR = 0.834) exhibited a protective effect against overall AS risk.
Conclusions
This study identifies specific CIPs that have significant causal effects on various forms of AS through MR analysis. The findings suggest potential biomarkers and treatment targets for preventing and managing different manifestations of AS in clinical practice.