APOB to estimated APOB ratio for screening for the APOE2 genotype
Auger, C.; Sampson, M.; Zubiran, R.; Cole, J.; Wolska, A.; Otvos, J. D.; Sniderman, A. D.; Remaley, A. T.
Show abstract
BackgroundFamilial dysbetalipoproteinemia (FDB) is a genetic lipoprotein disorder that can develop in patients homozygous for the APOE2 genotype ({varepsilon}2/{varepsilon}2). It is associated with decreased clearance of remnant lipoproteins and increased atherosclerotic cardiovascular disease (ASCVD) risk disproportionate to their level of LDL-C. A goal of this study was to develop a screening test for the {varepsilon}2/{varepsilon}2 genotype based on routinely available lipid tests and to determine those at most risk for ASCVD. MethodsAfter assembly of a primary prevention cohort from the UK Biobank (n= 269,895), gene array and exome data was utilized to classify patients as being {varepsilon}2/{varepsilon}2 genotype positive or negative. Lipid profiles and APOB levels were extracted and the number of ASCVD events was tabulated during a 15-year follow-up period. ResultsUsing a newly developed equation for estimating APOB (eAPOB) with lipid panel test results, the ratio of measured APOB to eAPOB was better than any other individual lipid test or ratio for identifying patients with the {varepsilon}2/{varepsilon}2 genotype (AUC: APOB/eAPOB: 0.990 (0.986-0.994), nonHDL-C/APOB: 0.961 (0.952-0.970), APOB: 0.955 (0.949-0.961), VLDL/TG: 0.788 (0.771-0.804)). The majority of {varepsilon}2/{varepsilon}2 patients could be identified with the APOB/eAPOB ratio even before they expressed the FDB phenotype with elevated TG and nonHDL-C. The PCE or PREVENT risk equations were the most accurate method for identifying higher risk patients (AUC: PREVENT: 0.690 (0.637-0.742), PCE: 0.697 (0.645-0.749)). ConclusionThe APOB/eAPOB ratio can be used to accurately identify the {varepsilon}2/{varepsilon}2 genotype and conventional risk equations are the best method for determining those at risk for ASCVD.
Matching journals
The top 7 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Baseline Cardiometabolic Profiles and SARS-CoV-2 Infection in the UK Biobank 95%
- Actionable absolute risk prediction of atherosclerotic cardiovascular disease: a behavior-management approach based on data from 464,547 UK Biobank participants 94%
- Mendelian Randomization Study of Whole Blood Viscosity and Cardiovascular Diseases 93%
Similar papers in this journal
- Transgelin: A New Gene Involved in LDL Endocytosis Identified by a Genome-wide CRISPR-Cas9 Screen 93%
- Aggressive Cholesterol Lowering Normalizes Atherosclerosis Regression in Jak2V617F Mice 91%
- Structure dynamics of ApoA-I amyloidogenic variants in small HDL increase their ability to mediate cholesterol efflux 91%
Similar papers in this journal
- Identification of endothelial-derived proteins in plasma associated with cardiovascular risk factors 94%
- A functional variant of the SIDT2 gene involved in cholesterol transport is associated with HDL-C levels and premature coronary artery disease. 93%
- Trimethylamine-N-oxide affects cell type-specific pathways and networks in mouse aorta to promote atherosclerotic plaque vulnerability 93%
Similar papers in this journal
- A Mendelian randomization study of the role of lipoprotein subfractions in coronary artery disease 96%
- Associations of genetic and infectious risk factors with coronary heart disease 93%
- EPHX1 mutations cause a lipoatrophic diabetes syndrome due to impaired epoxide hydrolysis and increased cellular senescence 92%
Similar papers in this journal
- Sex Differences in Middle Cerebral Artery Reactivity and Hemodynamics Independent from Changes in Systemic Arterial Stiffness in Adult Sprague-Dawley Rats 91%
- Intermuscular Adipose Tissue And Muscle Function In Patients On Maintenance Hemodialysis 91%
- Medin Induces Pro-Inflammatory Activation of Human Brain Vascular Smooth Muscle Cells 89%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.