A potentially underestimated predictor of coronary artery disease risk: The ApoB/ApoA1 ratio

Abstract

Background: Cardiovascular disease (CVD) is the leading cause of death worldwide, and statin therapy is the cornerstone of atherosclerotic cardiovascular disease. However, clinical practice is unsatisfactory, and there is significant interest in the risk of residual cardiovascular events. Traditional study methods make it difficult to exclude the crosstalk of confounding factors, and we investigated the impact of the ApoB/ApoA1 ratio on CVD using two-sample Mendelian randomization (MR) and multivariate Mendelian randomization (MVMR) methods.

Methods: Two-sample MR and MVMR analyses were performed using pooled statistics from genome-wide association studies (GWAS) of ApoB/ApoA1 ratio (BAR), lipoprotein (a) (Lp(a)), and triglyceride (TG) in Europeans to assess the causal relationship between BAR, Lp(a), and TG with coronary artery disease (CAD).

Results: The genetic prediction of BAR was significantly correlated with CAD (Inverse variance weighted (IVW) beta = 0.255; OR = 1.291; 95 % CI = 1.061-1.571; P = 0.011) in a two-sample MR analysis. MVMR studies showed that BAR (beta = 0.373; OR = 1.452; 95 % CI = 1.305-1.615; P = 7.217e-12), Lp (a) (beta = 0.238; OR = 1.269; 95 % CI = 1.216-1.323; P = 2.990e-28), and TG (beta = 0.155; OR = 1.168; 95 % CI = 1.074-1.270; P = 2.829e-04) were significantly associated with CAD. After further colinearity analyses of LASSO regressions, the results of multivariate analyses were similar for IVW, MR-Egger, MR-Lasso, and median methods.

Conclusion: BAR is causally related to coronary artery disease. BAR is an independent predictor of CAD risk, independent of routine lipid measurements and other risk factors. TG and Lp(a) may be causally related to CAD, subject to verification in clinical practice.

Keywords: ApoB/ApoA1 ratio; Coronary artery disease; GWAS; Multivariable Mendelian randomization; Two-sample MR.

Fig. 1

Forest plot of causal effects of ApoB/ApoA1 ratio associated single nucleotide polymorphisms on coronary artery disease. The red dots represent the joint causal estimation used in a single tool using all SNPs, using two different methods (inverse variance weighted [IVW] random effects and MR-Egger).

Fig. 2

Scatter plot of genetic association of ApoB/ApoA1 ratio with coronary heart disease. The slope of each line represents the causal association for each method.

Fig. 3

Funnel plot assessing heterogeneity. The blue line represents the IVW estimate, and the dark blue line represents the Egger Mendelian randomization estimate.

Fig. 4

Sensitivity analysis to investigate the possibility that causal association was driven by a unique single nucleotide polymorphism in coronary heart disease. Each black dot represents the IVW MR method used to estimate the causal effect of BAR on CAD after excluding this particular variant from the analysis. The red dots represent IVW estimates using all SNPs.