Genetic and non-genetic determinants of drug response in type 2 diabetes

Abstract

Background: Approximately 415 million people live with diabetes worldwide. Type 2 diabetes(T2D) accounts for 85%–95% of the cases. Despite the availability of severaldrugs, considerable interindividual variation in response to medicationsresults in unnecessary treatment failure. In addition to non-genetic factors, geneticfactors are thought to contribute to such variability. 

Aim: This thesisattempts to investigate genetic and non-genetic factors underlying thevariability in response to commonly used drugs in T2D.

Methods: We assessed drug response as efficacy and/or adverse effect following treatmentusing data gathered from medical records mainly from the Diabetes Research forPatient Stratification (DIRECT) consortium and the Genetics of Diabetes Auditand Research Tayside Study (GoDARTS). Genetic data was obtained usingchip-based arrays followed by imputation and TaqMan genotyping. Subsequentcandidate gene and genome wide association (GWAS) analyses were conducted usinglinear and logistic regressions followed by meta-analyses. In addition,downstream rare variant burden test and pathway analyses were performed.

Results: We showed robust association of metformin use with fasting glucagon likepeptide-1 (GLP-1) levels in diabetic and non-diabetic subjects. Gastrointestinal(GI) irritation is the most common side effect of metformin. Concomitantadministration of metformin with gut metformin transporter inhibiting drugssuch as tricyclic antidepressants, proton pump inhibitors and codeine increasedthe odds of GI intolerance. Moreover we showed association of the G allele atrs3889348-SLC29A4 (PMAT) withincreased odds of intolerance.  In theGoDARTS study performed on subjects who have been taking sulphonylureas (SUs)as an add-on therapy to metformin, carriers of the K allele at E23K-KCNJ11 had greater HbA1c reduction andthis was replicated in another study using data from the Diabetes Care SystemWest-Friesland (DCS) study. We also identified a novel locus, rs11535279-LHFPL3, associated with glycaemicresponse to SUs. Further analysis revealed enrichment of the insulin/IGFpathway-mitogen activated protein kinase/MAPK cascade in glycaemic response toSU treatment suggesting the role of the post insulin secretion pathway inglycaemic response to SUs. In a study investigating joint effect of variants intransporter (SLCO1B1 521T>C) andmetabolizing (CYP2C8*3) proteins withHbA1c reduction and weight gain to thiazolidinediones (TZDs), we showed a largeclinical impact on the therapeutic response to rosiglitazone.  Lastly, in a meta-analysis consisting of 1,235T2D subjects treated with GLP-1RAs, carriers of two or more variant allelesderived from GLP-1R variants(Gly168Ser and Pro7Leu) had significantly reduced efficacy compared tohomozygous carriers of the parent alleles. In addition, rare variant analysisrevealed suggestive evidence of association of the mutational load of variantsin genes previously implicated in GLP-1/glucose stimulated insulin secretionwith glycaemic response to GLP-1RAs.

Conclusions: In this thesis, we have identified clinical and novel genetic factorsunderlying treatment efficacy and adverse effects related to drugs used to treatin T2D.