Genetic Architecture And Clinical Outcomes Of The Fredrickson-Levy-Lees Dyslipoproteinemias

Background and AimsThe genetic basis and clinical relevance of the classical Fredrickson-Levy-Lees (FLL) dyslipoproteinemia classifications has not been studied in general population-based cohorts. We aimed to evaluate the phenotypic and genetic characteristics of FLL disorders. MethodsAmong UK Biobank participants free of prevalent coronary artery disease (CAD), we used blood lipids and apolipoprotein B concentrations to infer FLL classes (Types I, IIa, IIb, III, IV, and V). For each FLL class, Cox proportional hazards regression estimated risk of incident CAD. Phenome-wide association testing was performed. GWAS were performed, followed by in silico causal gene prioritization and heritability analyses. Prevalence of disruptive Mendelian lipid variants was assessed from whole exome sequencing. ResultsOf 450,636 individuals, 259,289 (57.5%) met criteria for a FLL dyslipoproteinemia: 63 (0.01%) type I; 40,005 (8.9%) type IIa; 94,785 (21.0%) type IIb; 13,998 (3.1%) type III; 110,389 (24.5%) type IV; and 49 (0.01%) type V. Over median 11.1 years follow-up, compared to normolipidemics the type IIb pattern conferred the highest hazard of incident CAD overall (HR 1.92, 95% CI 1.84-2.01, P<0.001) and in meta-analysis across matched non-HDL-C strata (HR 1.45, 95% CI 1.30-1.60). GWAS revealed 250 loci associated with FLL, of which 13 were shared across all classes; compared to GWAS of isolated lipid traits, 72 additional loci were detected. Mendelian lipid variants were rare (2%), but polygenic heritability was high, ranging from 23% (type III) to 54% (type IIb). ConclusionsFLL classes have distinct genetic architectures yielding new insights for cardiometabolic disease beyond single lipid analyses.