Evaluating the Cardiometabolic Efficacy and Safety of Lipoprotein Lipase Pathway Targets in Combination With Approved Lipid-Lowering Targets: A Drug Target Mendelian Randomization Study

Abstract

Background: Therapies targeting the LPL (lipoprotein lipase) pathway are under development for cardiometabolic disease. Insights into their efficacy-both alone and in combination with existing lipid-lowering therapies-modes of action, and safety of these agents are essential to inform clinical development. Using Mendelian randomization, we aimed to (1) evaluate efficacy, (2) explore shared mechanisms, (3) assess additive effects with approved lipid-lowering drugs, and (4) identify secondary indications and potential adverse effects.

Methods: We selected triglyceride-lowering genetic variants located in the genes encoding ANGPTL3 (angiopoietin-like 3), ANGPTL4 (angiopoietin-like 4), APOC3 (apolipoprotein C3), and LPL and conducted drug target Mendelian randomization on primary outcomes including coronary artery disease and type 2 diabetes, and secondary outcomes, including apolipoprotein B, waist-to-hip ratio, body mass index, and 233 metabolic biomarkers. We conducted interaction Mendelian randomization analyses in 488 139 UK Biobank participants to test the effect of combination therapy targeting the LPL and LDLR (low-density lipoprotein receptor) pathways. Finally, we investigated potential secondary indications and adverse effects by leveraging genetic association data on 1204 disease end points.

Results: Genetically predicted triglyceride lowering through the perturbation of LPL pathway activation targets ANGPTL4, APOC3, and LPL was associated with a lower risk of coronary artery disease and type 2 diabetes and lower apolipoprotein B. Genetically predicted triglyceride lowering through ANGPTL4 was associated with a lower waist-to-hip ratio, suggestive of a favorable body fat distribution. There was no evidence of a multiplicative interaction between genetically proxied perturbation of ANGPTL4, APOC3, and LPL and that of HMGCR (HMG-CoA reductase) and PCSK9 (proprotein convertase subtilisin/kexin type 9) on coronary artery disease and type 2 diabetes, consistent with additive effects. Finally, associations of genetically predicted LPL pathway targeting were supportive of the broad safety of these targets.

Conclusions: Our findings provide genetic evidence supporting the efficacy and safety of LPL pathway activation therapies for the prevention of coronary artery disease and type 2 diabetes, alone or in combination with statins or PCSK9 inhibitors.

Keywords: Mendelian randomization analysis; PCSK9 inhibitors; biomarkers; cholesterol lowering medication; coronary heart disease; lipoprotein lipase.

Figure 1. Schematic overview of the study design.

ANGPTL3 = Angiopoietin-like 3; ANGPTL4 = Angiopoietin-like 4; APOC3 = Apolipoprotein CIII; LPL = Lipoprotein lipase; HMGCR = HMG-CoA reductase; PCSK9 = Proprotein convertase subtilisin/kexin type 9; LDLR = Low-density lipoprotein receptor. Created with BioRender.com.

Figure 2. Association between genetically predicted lipid-lowering targeting and cardiometabolic traits and diseases using Two sample Mendelian randomization.

A) Impact of LPL pathway activation on coronary artery disease and type 2 diabetes. B) Impact of PCSK9 and HMGCR inhibition on coronary artery disease and Type 2 diabetes. C) Impact of LPL pathway activation on ApoB and anthropometric traits. D) Impact of PCSK9 and HMGCR inhibition on ApoB and anthropometric traits. ANGPTL3 = Angiopoietin-like 3; ANGPTL4 = Angiopoietin-like 4; APOC3 = Apolipoprotein CIII; LPL = Lipoprotein lipase; HMGCR = HMG-CoA reductase; PCSK9 = Proprotein convertase subtilisin/kexin type 9.

Figure 3. Association between genetic enhancement of LPL and metabolites in relation with the association of genetic inhibition of ANGPTL4 and APOC3 on the same metabolites.

Each effect is scaled to a one standard deviation effect on TG levels. Panel A) ANGPTL4 inhibition vs. LPL enhancement. Panel B) APOC3 inhibition vs. LPL enhancement. Panel C) ANGPTL3 inhibition vs. LPL enhancement. Panel D) PCSK9 inhibition vs. LPL enhancement. Panel E) HMGCR inhibition vs. LPL enhancement. Panel F) PCSK9 inhibition vs. HMGCR inhibition. The dashed line represents the best fitting line scaled on the inverse of the variance of the estimates ANGPTL3 = Angiopoietin-like 3; ANGPTL4 = Angiopoietin-like 4; APOC3 = Apolipoprotein CIII; LPL = Lipoprotein lipase; HMGCR = HMG-CoA reductase; PCSK9 = Proprotein convertase subtilisin/kexin type 9.

Figure 4. Interaction between pair of genetic risk score of LPL pathway targets and LDLR pathway targets.

Display of the Mendelian randomization estimates in the genetic interaction analysis between LPL pathway targets and LDLR targets. Each Cox model was adjusted for sex and the ten first principal component of ancestry. ANGPTL3 = Angiopoietin-like 3; ANGPTL4 = Angiopoietin-like 4; APOC3 = Apolipoprotein CIII; LPL = Lipoprotein lipase; HMGCR = HMG-CoA reductase; PCSK9 = Proprotein convertase subtilisin/kexin type 9; LDLR = Low-density lipoprotein receptor.

Figure 5. Association between genetically predicted targeting of lipid-lowering targets and 1204 diseases in FinnGen.

A) LPL activation, B) ANGPTL4 inhibition. C) APOC3 inhibition. D) ANGPTL3 inhibition. The green line is the threshold for multiple testing significance. Triangles pointing down represent negative associations, whereas triangles pointing up represent positive associations. The ten most significant associations that pass multiple testing correction are annotated. ANGPTL3 = Angiopoietin-like 3; ANGPTL4 = Angiopoietin-like 4; APOC3 = Apolipoprotein CIII; LPL = Lipoprotein lipase.