LOWERING LDL-CHOLESTEROL IS PROTECTIVE FOR THE FRAILTY INDEX: A MENDELIAN RANDOMIZATION STUDY IN THE UK BIOBANK

摘要

Background Low density lipoprotein (LDL) cholesterol lowering drugs are effective treatments for cardiovascular disease. Studies using genetic variants with LDL-lowering allele scores have shown protective effects similar to clinical trials. The aim of our study was to investigate the association between genetically determined LDL-lowering effects and the frailty index (FI). Methods Genetic risk scores (GRS) for LDL-cholesterol levels were composed from genome-wide significant variants (GRS_large, n=274), independent variants (GRS_small, n=50), and for specific lipid-lowering drug targets (HMGCR, PCSK9, NPC1L1, APOB, APOC3, LDLR) using data from the UK biobank (n=383,901 unrelated individuals, 40–69 years). The FI, using Rockwood deficit accumulation model, was used as the outcome and the inverse-variance weighted method, with Egger regression and weighted medians as sensitivity analyses, was applied in a Mendelian Randomization design. Results In general, protective alleles in LDL variants were associated with a decrease in the FI. In the large and small GRS, effects were 0.226 (95% CI 0.182, 0.270) and 0.231 (0.144, 0.318), respectively, per added item on the FI-scale per standard deviation increase in LDL levels (both p<0.0005). Sensitivity analyses, including the FI without cardiovascular items and in individuals over 60 years only, were confirmatory. For drug-target specific loci, HMGCR, APOC3 and LDLR showed protective effects for the FI, and sensitivity analyses were, for the most part, confirmatory. Interestingly, APOB variants demonstrated a consistent null association with the FI throughout analyses. Conclusions Life-long effects of lipid lowering are protective for the FI. Working drug targets include HMGCR, APOC3 and LDLR, but not APOB.