Association of Genetically Enhanced Lipoprotein Lipase-Mediated Lipolysis and Low-Density Lipoprotein Cholesterol-Lowering Alleles With Risk of Coronary Disease and Type 2 Diabetes

Abstract

Importance: Pharmacological enhancers of lipoprotein lipase (LPL) are in preclinical or early clinical development for cardiovascular prevention. Studying whether these agents will reduce cardiovascular events or diabetes risk when added to existing lipid-lowering drugs would require large outcome trials. Human genetics studies can help prioritize or deprioritize these resource-demanding endeavors.

Objective: To investigate the independent and combined associations of genetically determined differences in LPL-mediated lipolysis and low-density lipoprotein cholesterol (LDL-C) metabolism with risk of coronary disease and diabetes.

Design, setting, and participants: In this genetic association study, individual-level genetic data from 392 220 participants from 2 population-based cohort studies and 1 case-cohort study conducted in Europe were included. Data were collected from January 1991 to July 2018, and data were analyzed from July 2014 to July 2018.

Exposures: Six conditionally independent triglyceride-lowering alleles in LPL, the p.Glu40Lys variant in ANGPTL4, rare loss-of-function variants in ANGPTL3, and LDL-C-lowering polymorphisms at 58 independent genomic regions, including HMGCR, NPC1L1, and PCSK9.

Main outcomes and measures: Odds ratio for coronary artery disease and type 2 diabetes.

Results: Of the 392 220 participants included, 211 915 (54.0%) were female, and the mean (SD) age was 57 (8) years. Triglyceride-lowering alleles in LPL were associated with protection from coronary disease (approximately 40% lower odds per SD of genetically lower triglycerides) and type 2 diabetes (approximately 30% lower odds) in people above or below the median of the population distribution of LDL-C-lowering alleles at 58 independent genomic regions, HMGCR, NPC1L1, or PCSK9. Associations with lower risk were consistent in quintiles of the distribution of LDL-C-lowering alleles and 2 × 2 factorial genetic analyses. The 40Lys variant in ANGPTL4 was associated with protection from coronary disease and type 2 diabetes in groups with genetically higher or lower LDL-C. For a genetic difference of 0.23 SDs in LDL-C, ANGPTL3 loss-of-function variants, which also have beneficial associations with LPL lipolysis, were associated with greater protection against coronary disease than other LDL-C-lowering genetic mechanisms (ANGPTL3 loss-of-function variants: odds ratio, 0.66; 95% CI, 0.52-0.83; 58 LDL-C-lowering variants: odds ratio, 0.90; 95% CI, 0.89-0.91; P for heterogeneity = .009).

Conclusions and relevance: Triglyceride-lowering alleles in the LPL pathway are associated with lower risk of coronary disease and type 2 diabetes independently of LDL-C-lowering genetic mechanisms. These findings provide human genetics evidence to support the development of agents that enhance LPL-mediated lipolysis for further clinical benefit in addition to LDL-C-lowering therapy.

Figure 1.. Associations of Genotype Category With Cardiometabolic Disease Outcomes in 2 × 2 Factorial Genetic Analyses

Associations of each genetic score group with risk of coronary artery disease and type 2 diabetes compared with the reference group. The reference group includes those with a low-density lipoprotein cholesterol (LDL-C)–lowering score and a triglyceride-lowering LPL score less than or equal to the median score; the genetically lower triglyceride levels only group, those with a triglyceride-lowering LPL score greater than the median but an LDL-C–lowering score less than or equal to the median; the genetically lower LDL-C levels only group, those with an LDL-C–lowering score greater than the median but a triglyceride-lowering LPL score less than or equal to the median; and the group with both exposures, those with both scores greater than the median. Analyses include individual-level genetic data from 390 470 participants of the UK Biobank, EPIC-Norfolk, and EPIC-InterAct studies. Median values and interquartile ranges for lipid levels in a given genotype category are from the EPIC-Norfolk study. To convert LDL-C level to micromoles per liter, multiply by 0.0259. To convert triglyceride level to micromoles per liter, multiply by 0.0113. IQR indicates interquartile range; LPL, lipoprotein lipase; NA, not applicable; OR, odds ratio.

Figure 2.. Associations of Triglyceride-Lowering LPL Alleles With Cardiometabolic Disease Outcomes in Individuals Above or Below the Median of the Population Distribution of Low-Density Lipoprotein Cholesterol (LDL-C)–Lowering Genetic Variants

Analyses include individual-level genetic data from 390 470 participants of the UK Biobank, EPIC-Norfolk, and EPIC-InterAct studies.

Figure 3.. Associations of Triglyceride-Lowering LPL Alleles With Cardiometabolic Disease Outcomes Within Quintiles of the Population Distribution of Genetic Variants at 58 Low-Density Lipoprotein Cholesterol (LDL-C)–Associated Genetic Loci

Data are from the UK Biobank, EPIC-Norfolk, and EPIC-InterAct studies. Median values and interquartile ranges for lipid levels within each stratum are from the EPIC-Norfolk study. To convert LDL-C level to micromoles per liter, multiply by 0.0259. To convert triglyceride level to micromoles per liter, multiply by 0.0113. IQR, interquartile range; OR, odds ratio.

Figure 4.. Associations of Loss-of-Function Alleles With Cardiometabolic Disease Outcomes in ANGPTL4 and ANGPTL3

A, Associations of the ANGPTL4 p.Glu40Lys loss-of-function allele with cardiometabolic disease outcomes. Groups with genetically higher or lower low-density lipoprotein cholesterol (LDL-C) levels were defined on the basis of the median value of the 58-variant LDL-C–lowering genetic score. Associations are scaled to represent the odds ratio (OR) per SD of genetically lower triglyceride levels. Data are from the UK Biobank, EPIC-Norfolk, and EPIC-InterAct studies. B, Associations of different genetic exposures associated with lower LDL-C levels with protection against coronary disease. A clear log-linear relationship between genetic difference in LDL-C level and lower risk is observed for several mechanisms, while ANGPTL3 loss-of-function variants are outliers in this relationship. For individual variants, the estimates represent per-allele differences; for quintiles of the LDL-C score, the difference is compared with the bottom quintile; for the overall genetic score, the difference is per SD of genetically lower LDL-C level; and for ANGPTL3 variants, the difference is in carriers compared with noncarriers.