Identification of genetically supported candidate therapeutic targets for angina pectoris using mendelian randomization.

Abstract

Angina pectoris is a heterogeneous clinical syndrome that may reflect obstructive coronary atherosclerosis, coronary vasomotor dysfunction, microvascular dysfunction, or myocardial oxygen supply-demand imbalance. We used Mendelian randomization (MR) to prioritize circulating proteins that may influence the broad risk of angina while distinguishing atherosclerosis-related pathways from other potential mechanisms. This study aimed to identify genetically supported candidate therapeutic targets for angina pectoris and characterize the relative strength and potential mechanisms of the evidence supporting each candidate. The primary analysis combined European cis-protein quantitative trait locus instruments with a broad genome-wide association study -defined angina phenotype. Steiger testing, Bayesian co-localization, phenotypic scanning, external validation, supplementary European trans- protein quantitative trait locus and Asian-ancestry analyses, phenotype-stratified MR, two-step mediation MR, enrichment analysis, bulk RNA-seq, protein-protein interaction analysis, and exploratory single-cell transcriptomic analyses were used to evaluate robustness and biological context. Seven protein-angina associations met the primary Bonferroni threshold; 6 proteins (LPA, TMEM106B, APOA5, TIRAP, PCSK9, and SWAP70) were retained after TGFB1 failed Steiger filtering. Independent external MR supported LPA in UK Biobank and LPA, PCSK9, and APOA5 in FinnGen. Phenotype-stratified analyses generally supported associations across stable angina, unstable angina, coronary artery disease, and myocardial infarction, although instrument availability and the strength of evidence differed among proteins. Two-step MR supported indirect pathways through coronary artery calcification for LPA, through LDL cholesterol for PCSK9, and through triglycerides for APOA5, but not through C-reactive protein for TIRAP. Exploratory single-cell analyses localized TMEM106B, TIRAP, and SWAP70 to relevant vascular and immune cell populations and provided limited, context-dependent functional support. LPA, PCSK9, and APOA5 have the strongest convergent genetic and external-replication evidence and appear to influence angina risk predominantly through atherosclerotic lipid-related pathways. TMEM106B, TIRAP, and SWAP70 remain potentially novel, genetically supported candidates with exploratory cellular evidence but without direct experimental validation. These candidates should not be considered established therapeutic targets until proteomic, functional, rescue, and clinical studies are completed.