The American Diabetes Association's 70th Scientific Sessions took place in Orlando, Fla., from June 25 to June 29. At the meeting, researchers presented numerous abstracts exploring the genetic underpinnings of diabetes susceptibility, investigating SNPs showing variable patient responses to commonly prescribed medications; and analyzing genetic risk for conditions that can heighten people's chances of developing diabetes. Below are a handful of PGx abstracts presented at the meeting.
PPARG, Insulin Resistance and Metformin Pharmacogenetics — A GoDARTS Study
Kaixin Zhou of the Dundee Diabetes Genetics Group and others assessed the role of Pro12Ala (rs1801282) in PPARg on insulin sensitivity, adipokine concentrations, and glycemic response to metformin.
The researchers genotyped Rs1801282 in 5,993 cases and 6,713 controls from the GoDARTS (Genetics of Diabetes Audit and Research in Tayside, Scotland) cohort. Logistic and linear regression analyses conducted in 2,552 patients was used to model whether treatment target was reached. In 2,016 patients, Cox regression was applied to model metformin failure. The models considered covariates such as dose, adherence, pre-metformin HbA1c, creatinine clearance, and assumed additive genetic effect.
The models showed that "the PPAR Pro12Ala allele was protective for diabetes and increases insulin sensitivity in 1,843 controls," the researchers reported. "This Ala allele is associated with lower odds of achieving the treatment target."
Additionally, researchers reported that the Ala allele carriers were also likely to fail metformin monotherapy earlier. The study found no association between the Pro12Ala variant and adiponectin and leptin concentrations in 2,468 healthy controls.
Touting this as the "largest diabetes pharmacogenetics study to date," researchers concluded that the Ala allele of rs1801282 reduces efficacy of metformin. "This may reflect alteration by PPARg of insulin sensitivity by altering adipocyte differentiation or by directly influencing phosphorylation of AMPK," the researchers reported.
Molecular Dissection of Sulfonylurea Pharmacogenomics in the ATP-Sensitive Potassium Channel Type 2 Diabetes Risk K23/A1369 Haplotype
Noting that previous studies have linked E23K and S1369A genes to higher susceptibility to type 2 diabetes and more responsive to sulfonylurea gliclazide, Yiqiao Lang and Peter Light from the Diabetes Institute at the University of Alberta attempted to determine the structural aspects of the gliclazide molecule that allow the treatment to inhibit the K23/A1369 variant KATP channel. This analysis was conducted to shed light on why other structurally related slufonylureas "may paradoxically display a decreased inhibitory potency on the K23/A1369 variant KATP channel."
"The molecular structure of gliclazide can be dissected into two parts: the sulfonylurea and ring-fused pyrrole motifs," the researchers reported. Then they picked structurally similar sulfonylureas, such as sulfonylurea tolbutamide and glinide mitiglinide, and measured their inhibitory profiles on recombinant human KATP channels containing genetic variants K23/A1369 or E23/S1369, using the excised inside-out patch-clamp technique. The effort revealed tolbutamide as having a decreased inhibitory effect on K23/A1369 variants, and mitiglinide was similar to gliclazide, having an increased inhibitory effect on K23/A1369 variants.
"In conclusion, the ring-fused pyrrole motif present in gliclazide and mitiglinide but not tolbutamide is the key structural feature that confers the increased potency on K23/A1369 variant KATP channel," Lang and Light reported. "These results confirm that a molecular pharmacogenomic approach may be used to rationally design optimized drug therapies in T2D patients carrying K23/A1369 haplotype."
Therapeutic Response to Sulfonylureas in Patients with Type 2 Diabetes Is Altered by TCF7L2 Variants
Andreas Holstein of Klinikum Lippe-Detmold in Germany and others investigated whether the diabetes risk allele rs7903146 in the TCF7L2 gene might alter type 2 diabetes patients' hypoglycemic response to sulfonylureas.
The researchers recruited 189 patients with type 2 diabetes treated with sulfonylurea agents and genetically tested them to determine if they harbored the TCF7L2 rs7903146 allele, which also has been shown in past studies to confer an increased risk for diabetes. If patients failed sulfonylurea treatment, researchers gave them additional treatment with insulin.
According to the abstract, "TCF7L2 genotype and diabetes duration were the main predictors of sulfonylurea treatment failure." The rs7903146 T-allele was significantly more frequent in the group of patients additionally treated with insulin (40 percent) compared to the group treated only with sulfonylureas (28 percent). The researchers stated that the genotype effect seen in the study was independent of age, sex, and diabetes duration.
"Our data suggest that patients with diabetes risk alleles in TCF7L2 have an altered hypoglycemic response to sulfonylureas, thus supporting previously reported findings and indicating the potential of pharmacogenomics in the therapy of type 2 diabetes," the researchers concluded.
[ pagebreak ]
A Genetic Risk Score To Improve the Prediction of Coronary Heart Disease in Type 2 Diabetes
Lu Qi from Harvard University and colleagues evaluated whether previously identified genetic markers associated with coronary heart disease in the general population could predict CHD in diabetes patients. Researchers considered three studies of US white individuals with type 2 diabetes, including the prospective Nurses' Health Study involving 309 CHD cases and 544 controls; the Health Professional Follow-up Study with 345 CHD cases and 451 controls; and the cross-sectional Joslin Heart Study of 422 CHD cases and 436 controls. All study subjects were genotyped for fifteen risk variants identified by genome-wide association studies.
Qi et al. showed that five SNPs (rs4977574 [CDKN2A/2B], rs12526453 [PHACTR1], rs646776 [CELSR2-PSRC1-SORT1], rs2259816 [HNF1A], and rs11206510 [PCSK9] were consistently associated with CHD throughout the three studies, with adjusted odds ratios and in the combined analyses, respectively. Out of the number of risk alleles carried at these five genetic loci, researchers created a genotype risk score, or GRS.
"After adjusting for clinical risk factors such as age, gender, [myocardial infarction], smoking status, HbA1c, and history of hypertension and hypercholesterolemia, the [odds ratio] of CHD per GRS unit was 1.19" with a 95 percent confidence interval of 1.13-1.26. Individuals with a GRS greater than 8, which included 44 percent of diabetic subjects, had a two-fold increase in CHD risk as compared to individuals with GRS lower than or equal to 5, reported Qi et al. Additionally, researchers noted that their prediction of CHD was "significantly improved" when the GRS was added to a model including clinical predictors.
"The incremental predictive value of GRS, measured as the increase in the area under the receiver-operating characteristic curve (AUC) after adding GRS, was 0.0136," the researchers stated. "In conclusion, our results indicate that a genetic score based on five SNPs significantly improves CHD prediction among individuals with type 2 diabetes." Qi et al. added that cost-effectiveness analyses are needed in order to determine if these tests should be implemented into clinical practice.
A Comprehensive Assessment of 40 Candidate Genes in the Diabetes Prevention Program (DPP)
Jose Florez from Harvard Medical School, researchers from the Diabetes Prevention Program, and others examined how genetic variants identified through linkage studies, candidate gene studies, and genome-wide association scans impact diabetes incidence in an at-risk population, and whether these variants influence response to preventive interventions.
In this study, funded by the American Diabetes Association, Florez et al. selected 1,590 SNPs from previously conducted GWAS for type 2 diabetes and related traits for capturing common variations in European and African HapMap populations in 40 candidate genes already associated with type 2 diabetes. These candidate genes have been implicated in previous studies in monogenic forms of diabetes, encoding T2D drug targets, or drug-metabolizing enzymes, or have shown activity in cellular metabolism, hormonal regulation, or response to exercise.
Additionally, the researchers analyzed SNPs for association with diabetes incidence and their interaction with response to metformin or lifestyle intervention in 2,994 participants in the Diabetes Prevention Program, using an additive genetic model. The researchers estimated ancestry proportions, adjusted for multiple hypothesis testing by permutation, and assessed false discovery rates.
The team found "nominal associations" with diabetes incidence at 85 loci, and "nominal interactions" with metformin and lifestyle intervention at 91 and 69 loci, respectively. The loci associated with metformin response and lifestyle interventions were "mostly non-overlapping loci."
"The most significant associations with diabetes incidence occurred at several SNPs in the AMP kinase (AMPK) gene PRKAG2, with five directionally consistent associations for SNPs originally identified in 100K SNP GWAS," the abstract notes. Additionally, researchers replicated the association of a variant in the metformin transporter MATE1 with metformin response and detected nominal interactions at SNPs in the AMPK kinase gene LKB1, the AMPK genes PRKAA1 and PRKAA2, and a missense SNP in the gene encoding the metformin transporter OCT1.
"With regard to interactions with the lifestyle intervention, several of our top findings clustered around the peroxisome proliferator-associated receptor γ coactivators 1α and 1β (PPARGC1A and PPARGC1B, respectively)," researchers noted, adding that their findings "merit confirmation in independent samples."
Patient Interest and Response to Genetic Testing for Risk of Type 2 Diabetes in a Primary Care Setting
Scott Joy, medical director of Duke Health Center, and colleagues garnered consent from 144 patients getting tested for their blood glucose at a primary care clinic in an effort to gauge their interest and behavior related to genetic testing.
"The potential of personal genomics to influence primary prevention of type 2 diabetes necessitates that patients both express interest in genetic testing, and if at genetic risk, engage in practices that reduce their risk," Joy et al. state in the abstract.
In the study, participants first received a primer on genetics and how to give a biological sample. Then they were given a baseline survey to ascertain their interest in genetic testing and whether they would modify their diet and exercise patterns based on hypothetical genetic test results. Patients were also surveyed after receiving their blood glucose test results and three months after they completed the baseline survey.
The surveys revealed that study participants believed there was a strong association between genetics and T2DM risk. Additionally, they felt it was important to know even if one gene put them at higher risk. Approximately 64 percent of the participants expressed they would "probably not" or "definitely not" undergo genetic testing in the next six months.
Desire to be genetically tested was lowest in patients who expected their test to show lower genetic risk; had lower perceived risk for T2DM; felt it was less important to know one's genetic risk; and had less positive attitudes and feelings towards genetic testing.
When presented with hypothetical test results showing they had a higher genetic risk of developing type 2 diabetes, participants expressed stronger intentions to exercise more, follow a healthy diet of fruits and vegetables, and lower the amount of fat in their diet.
"Genetic testing for T2DM may not currently have widespread appeal to patients in a primary care setting, but genetics are felt to be strongly associated with T2DM," the researchers concluded. "Findings of higher genetic risk for developing type 2 diabetes may prompt changes in a patient's diet and exercise patterns."
[ pagebreak ]
Novel Type 2 Diabetes Susceptibility Loci Identified by Large-Scale Replication Using the "Metabochip"
Nigel Rayner of the University of Oxford's Center for Diabetes, members of the DIAGRAM (Diabetes Genetics Replication and Meta-analysis) Consortium, and others, used a custom Illumina microarray containing around 195,000 SNPs, called Metabochip, to identify novel loci conferring type 2 diabetes susceptibility.
Researchers analyzed 3,185 type 2 diabetes cases and 3,569 controls. Out of more than 185,000 SNPs, Rayner et al. ended up focusing on the top 5,000 independent autosomal signals from the DIAGRAM genome-wide association meta-analysis, which involved more than 12,000 type 2 diabetes cases and more than 56,000 controls of European descent.
"We observed directional consistency for all published T2D-loci including TCF7L2 (P
"Of these, 2,468 SNPs showed directionally consistent effects (binomial p
Although this follow-up study was of "modest size," joint analysis of DIAGRAM GWA and Tayside Metabochip data revealed several new signals that came near to or exceeded genome-wide significance. These included a locus near ARL15 (ADP-ribosylation factor-like 15) on chromosome 5p15.
"These data are consistent with a long tail of common variant association signals of modest effect contributing to T2D susceptibility," researchers concluded. They said they hoped that Metabochip-genotyping underway in more than 30,000 T2D cases and 50,000 controls will add to the list of validated T2D-susceptibility loci.
A Genome-Wide Association Analysis in over 177,000 Individuals Identifies 15 Loci Contributing to Variation in Central Obesity and Fat-Distribution
Cecilia Lindgren of the University of Oxford's Center for Diabetes, representing the work of the Genome-wide Investigation of Anthropometric measures (GIANT) consortium, presented data on a study that tried to detect common genetic variants for waist-to-hip ratio (WHR), a primary measure for obesity. The GIANT consortium joins study investigators and genetic epidemiologists around the world who want to pool genome-wide association results on anthropometric parameters such as body-mass-index and height.
"Obesity is an increasing public health issue, but not all forms of obesity carry the same risk," Lindgren reported in the abstract. Researchers noted that people with high WHR have increased morbidity, diabetes, hypertension, heart disease, stroke and certain cancers, but genetic variants that contribute to variation in WHR are not well characterized.
In order to detect common variants with modest effect sizes, researchers performed a meta-analysis of 35 genome-wide association studies comprising more than 77,000 individuals of European ancestry as part of the GIANT consortium. The investigation led to the discovery of 16 independent loci associated with WHR, which the researchers replicated in 30 additional cohorts comprising greater 100,000 individuals.
The combined analysis identified 15 loci that reached genome-wide significance. "We confirmed the known locus near LYPLAL1and identified 14 novel associations, including; RSPO3, TBX15, near VEGFA, and DNM3."
The researchers stated that "RSPO3 may regulate embryonic angiogenesis via the Wnt signaling pathway. TBX15 is differentially expressed between subcutaneous and visceral fat and the expression is correlated with WHR." VEGFA has been shown to play a role in diabetic nephropathy and retinopathy. The researchers also observed "a marked gender difference," since seven of the 15 loci appeared to have a stronger association in women than in men.
"Taken together, these results promise to enhance our knowledge of underlying biological pathways involved in central obesity," Lindgren concluded in the abstract. "Hopefully these advancements should support functional and translational advances in the management of obesity through development of novel diagnostic and therapeutic options, though this is an area for further study."