NEW YORK (GenomeWeb) – Genetic variants linked to lower levels of low-density lipoprotein cholesterol are also associated with an increased risk of type 2 diabetes, a new study has found.
A University of Cambridge-led team of researchers conducted meta-analyses of genome-wide association studies that drew on some 51,000 people with type 2 diabetes and 270,000 controls to examine any links between LDL-C-lowering alleles like those near NPC1L1 and risk of developing type 2 diabetes.
As they reported in the Journal of the American Medical Association today, the investigators found LDL-C-lowering variants near NPC1L1 and other genes were associated with a lower risk of coronary artery disease, but a higher risk of type 2 diabetes, data that provide "insights into potential adverse effects of LDL-C-lowering therapy," senior author Nicholas Wareham from Cambridge and his colleagues wrote in their paper.
Wareham and his colleagues pulled together data from the European Prospective Investigation into Cancer and Nutrition (EPIC)-InterAct study, the UK Biobank study, and the Diabetes Genetics Replication And Meta-analysis (DIAGRAM) study. They also folded in an additional 11 studies that examined the association between type 2 diabetes and rs12916 in the HMGCR gene. The study participants were all from Europe or the US.
The researchers focused on the effect of variants near NPC1L1because it is inhibited by ezetimibe, an approved lipid-lowering drug, but they also included variants in or near the targets of other lipid-lowering therapies or prospective therapies like HMGCR, PCSK9, ABCG5/G8, and LDLR. For each gene, they used Mendelian randomization statistical methods to estimate associations of LDL-C-lowering genetic variants and outcomes, and they estimated absolute risk differences.
Alleles at the NPC1L1 locus that lower LDL-C levels were inversely associated with coronary heart disease and directly associated with type 2 diabetes, Wareham and his colleagues reported. This association with type 2 diabetes held, they added, even when they accounted for a known, nearby association signal near the GCK gene with type 2 diabetes and glycemic traits.
Exome-sequencing association results, meanwhile, indicated that there wasn't enrichment of NPC1L1 protein-truncating alleles in cases with type 2 diabetes, but there was an overrepresentation of missense NPC1L1 variants in people with type 2 diabetes as compared to controls, the researchers added.
At the same time, Wareham and his colleagues also reported that LDL-C-lowering alleles in HMGCR were inversely associated with coronary heart disease and directly associated with type 2 diabetes, and that a loss-of-function variant in PCSK9 was associated with a higher risk of type 2 diabetes. A deeper analysis of the signals at PCSK9 suggested that there could be nine distinct association signals there.
The link between the LDL-C-lowering alleles at ABCG5/G8 and LDLR and type 2 diabetes did not reach statistical significance, the researchers noted.
Wareham and his colleagues also found that though each of these LDL-C lowering alleles each gave similar reductions in coronary artery disease risk, their effect on type 2 diabetes risk wasn't as uniform. This suggests that there is a gene-specific mechanism at play that underlies diabetes risk, they said.
And this could mean that LDL-C-lowering drugs with different targets could have varying influences on type 2 diabetes risk. "This may have clinical implications for the future of lipid-lowering therapy in the context of the increasing number of approved drugs acting on different molecular targets," the researchers wrote.