Apolipoprotein B outperforms low density lipoprotein particle number as a marker of cardiovascular risk in the UK biobank

Abstract Background When Apolipoprotein B (ApoB) and LDL cholesterol (LDL-C) or non-high density lipoprotein cholesterol (non-HDL-C) are discordant, ApoB is a stronger predictor of atherosclerotic cardiovascular disease (ASCVD)(1-2). It is unclear whether ApoB also provides better risk stratification when ApoB and LDL particle number (LDL-P) are discordant. Purpose This study examines the relationship between ApoB and LDL-P in the UK Biobank to determine which biomarker provides more accurate risk prediction when ApoB and LDL-P are discordant. Methods The UK Biobank is a prospective observational study of 0.5 million adults. Analyses were restricted to 41,099 participants with at least 10 years of data following enrollment, three or more recorded ICD codes, plasma lipoprotein and apolipoprotein measurements, and available baseline characteristics. Major adverse cardiovascular events (MACE) and coronary artery disease (CAD) events were plotted against LDL-P to ApoB for all participants. Concordance was defined as the linear regression line with y-intercept forced to zero. Discordant subpopulations were defined as populations 2, 4, 6, 8, 10, 20, and 30% of the mean above or below the regression line. The hazard ratio (HR) of cases to controls was determined for the discordant subpopulations and the concordant control group. A HR1 means that the risk is greater in the discordant group than in the reference group, whereas a HR1 suggests that the cases are less common in the discordant group than in the reference group. Results For the discordant subpopulations with ApoB less than LDL-P, there was no significant difference in HR for MACE or CAD events compared to the concordant group. In contrast, among discordant subpopulations where ApoB was greater than LDL-P, the HR for both MACE and CAD events increased as discordance increased and was statistically significant at all percentage discordance cutoffs. At 2% discordance, the HR for MACE was 1.1 (P0.0001), and the HR for CAD events was also 1.1 (P0.0001). At 30% discordance, the HR for MACE increased to 1.4 (P=0.002), and the HR for CAD events rose to 2.5 (P0.0001). Conclusions This study suggests that when discordant, ApoB is a more accurate marker for cardiovascular risk than LDL-P. In keeping with prior data examining discordance between ApoB and LDL-C or non-HDL-C, this data reinforces the utility of ApoB in guiding lipid-lowering strategies and cardiovascular risk assessment.