1649-P: Effect of Fasting Insulin-Derived Gene Variant Clusters on Cardiometabolic Traits

• Published in: Diabetes (New York, N.Y.) Vol. 69; no. Supplement_1
• Main Authors: SEVILLA, MAGDALENA, UDLER, MIRIAM, KIM, HYUNKYUNG, HSU, SARAH, MEIGS, JAMES B., MANNING, ALISA
• Format: Journal Article
• Language: English
• Published: New York American Diabetes Association 01.06.2020
• Subjects: Blood pressure; Cardiovascular disease; Coronary artery; Diabetes; Epidermal growth factor receptors; Fasting; Genetic diversity; Genome-wide association studies; Genomes; Glomerular filtration rate; Heart diseases; High density lipoprotein; Insulin; Insulin resistance; Low density lipoprotein
• ISSN: 0012-1797; 1939-327X
• DOI: 10.2337/db20-1649-P

Background: Fasting insulin concentrations reflect insulin resistance, which increase is associated with subsequent risk of heterogeneous cardiometabolic traits, including coronary artery disease (CAD), blood pressure (BP), estimated glomerular filtration rate (eGFR), BMI and lipid traits (TG, HDL, and LDL). We clustered insulin-associated genetic variants to define heterogeneous insulin resistance domains and tested whether these were associated with different cardiometabolic traits. Methods: We studied subjects of European Ancestry with genetic and clinical data available (N=18,127) in the Partners Biobank. We applied a soft clustering to 56 genome-wide association study (GWAS) fasting insulin genetic variants and 25 fasting insulin-related traits. We generated weighted genetic risk scores (GRS) for each cluster (alleles were aligned in the direction of increased fasting insulin effect). Then we aimed to identify clinical consequences between the clusters and cardiometabolic outcomes including CAD, SBP, DBP, eGFR, BMI, TG, HDL, and LDL. We tested the hypothesis that different genetic insulin domains will be cross-sectionally associated with some or all of these cardiometabolic traits. Results: We obtained three novel clusters of genetic variants. Each cluster was represented by a set of traits and a set of loci. The first represented VF and SAT, with loci near DLG2, FAM169B and, RGS12 genes, and its GRS was positive associated with CAD (p=0.004). The second cluster represents insulin sensitivity and lipid traits with loci near IRS1, ARRDC4 and, SNTG1, and its GRS was associated with lower TG (p=0.00001), and higher HDL(p=0.007). The last cluster represents proinsulin with loci near FABP2 and, PCAT5. No associations were found with this cluster. Conclusion: These results reveal different genetic insulin domains indicating different mechanisms for cardiometabolic traits associated with fasting insulin.