Novel insights into the association between genetically proxied inhibition of proprotein convertase subtilisin/kexin type 9 and risk of sarcopenia

Abstract

Background: The effects of lipid-lowering drugs [including statins, ezetimibe, and proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors] on hyperlipidaemia have been established. Some may have treatment effects beyond their reported properties, offering potential opportunities for drug repurposing. Epidemiological studies have reported conflicting findings on the relationship between lipid-lowering medication use and sarcopenia risk.

Methods: We performed a two-sample Mendelian randomization (MR) study to investigate the causal association between the use of genetically proxied lipid-lowering drugs (including statins, ezetimibe, and PCSK9 inhibitors, which use low-density lipoprotein as a biomarker), and sarcopenia risk. The inverse-variance weighting method was used with pleiotropy-robust methods (MR-Egger regression and weighted median) and colocalization as sensitivity analyses.

Results: According to the positive control analysis, genetically proxied inhibition in lipid-lowering drug targets was associated with a lower risk of coronary heart disease [PCSK9 (OR, 0.67; 95% CI, 0.61 to 0.72; P = 7.7E-21); 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR; OR, 0.68; 95% CI, 0.57 to 0.82; P = 4.6E-05), and Niemann-Pick C1-like 1 (NPC1L1; OR, 0.53; 95% CI, 0.40 to 0.69; P = 3.3E-06)], consistent with drug mechanistic actions and previous trial evidence. Genetically proxied inhibition of PCSK9 (beta, -0.040; 95% CI, -0.068 to -0.012; P = 0.005) and circulating PCSK9 levels (beta, -0.019; 95% CI, -0.033 to -0.005; P = 0.006) were associated with reduced appendicular lean mass (ALM) with concordant estimates in terms of direction and magnitude. Validation analyses using a second instrument for PCSK9 yielded consistent results in terms of direction and magnitude [(PCSK9 to ALM; beta, -0.052; 95% CI, -0.074 to -0.032; P = 7.1E-7); (PCSK9 protein to ALM; beta, -0.060; 95% CI, -0.106 to -0.014; P = 0.010)]. Genetically proxied inhibition of PCSK9 gene expression in the liver may be associated with reduced ALM (beta, -0.013; 95% CI, -0.035 to 0.009; P = 0.25), consistent with the results of PCSK9 drug-target and PCSK9 protein MR analyses, but the magnitude was less precise. No robust association was found between HMGCR inhibition (beta, 0.048; 95% CI, -0.015 to 0.110; P = 0.14) or NPC1L1 (beta, 0.035; 95% CI, -0.074 to 0.144; P = 0.53) inhibition and ALM, and validation and sensitivity MR analyses showed consistent estimates.

Conclusions: This MR study suggested that PCSK9 is involved in sarcopenia pathogenesis and that its inhibition is associated with reduced ALM. These findings potentially pave the way for future studies that may allow personalized selection of lipid-lowering drugs for those at risk of sarcopenia.

Keywords: Mendelian randomization; Older people; Proprotein convertase subtilisin/kexin type 9; Sarcopenia; Therapeutic target prediction.

Figure 1

Study design overview. (A) MR model. (B) Genetic instrument construction and analysis plan in the present study. For each drug target or LDL, genetic instruments were constructed by obtaining summary genetic association data on SNPs associated with LDL located within or near the gene encoding the drug target ([HMGCR; chr5:74,632,154‐74,657,929], [NPC1L1; chr7:44,552,134‐44,580,914], [PCSK9; chr1:55,505,221‐55,530,525]) or independent of genomic position (LDL) from the Global Lipids Genetics Consortium (GLGC). We constructed instruments for circulating PCSK9 levels from pQTLs in participants in the deCODE cohort (n = 35,559). We constructed a second PCSK9 protein instrument derived from a separate study providing PCSK9 pQTLs in 12,271 participants. Liver PCSK9 expression data were derived from genotype‐tissue expression version 8 liver tissue (n = 178). Cis‐located variants (+/− 100 kb) extracted from respective pQTL and eQTL GWASs; summary genetic association data for these SNPS were then extracted from genome‐wide association studies of CHD, ALM and LHGS. MR analyses were performed using inverse‐variance weighted random‐effects models as primary analyses and various approaches as sensitivity analyses. ALM, appendicular lean mass; CHD, coronary heart disease; HMGCR, 3‐hydroxy‐3‐methylglutaryl‐CoA reductase; LDL, low‐density lipoprotein; LHGS, low handgrip strength; MR, Mendelian randomization; NPC1LI, Niemann–Pick C1‐like 1; PCSK9, proprotein convertase subtilisin/kexin type 9.

Figure 2

Associations between genetically proxied lipid‐lowering drugs, LDL and risk of coronary heart disease as the positive control in the primary analysis. CHD, coronary heart disease; CI, confidence interval; HMGCR, 3‐hydroxy‐3‐methylglutaryl CoA reductase; LDL, Iow‐density lipoprotein; NPCIL1, Niemann–Pick C1‐like 1; OR, odds ratio; PCSK9, proprotein convertase subtilisin/kexin type 9.

Figure 3

Drug‐target MR results of genetically proxied PCSK9 inhibition on ALM using genetic variants and cis‐QTLs instruments. ALM, appendicular lean mass; CI, confidence interval; PCSK9, proprotein convertase subtilisin/kexin type 9.