6 page paper on diabetes.
budsimpson
researchpaper2.pdf
Using Personalized Medicine in the Management of
Diabetes Mellitus
Nina Elk 1* and Otito F. Iwuchukwu
2
1 Division of Pharmacy Practice, Fairleigh Dickinson University School of Pharmacy, Florham Park, New Jersey; 2 Division of Pharmaceutical Sciences, Fairleigh Dickinson University School of Pharmacy, Florham Park, New
Jersey
Diabetes mellitus is a worldwide problem with an immense pharmacoeconomic burden. The multifac- torial and complex nature of the disease lends itself to personalized pharmacotherapeutic approaches to treatment. Variability in individual risk and subsequent development of diabetes has been reported in addition to differences in response to the many oral glucose lowering therapies currently available for diabetes pharmacotherapy. Pharmacogenomic studies have attempted to uncover the heritable com- ponents of individual variability in risk susceptibility and response to pharmacotherapy. We review the current pharmacogenomics evidence as it relates to common oral glucose lowering therapies and how they can be utilized in the management of polygenic and monogenic forms of diabetes. Evidence supports the use of genetic testing and personalized approaches to the treatment of monogenic dia- betes of the young. The data are not as robust for the current application of pharmacogenetic approaches to the treatment of polygenic type 2 diabetes mellitus, but there are suggestions as to future applications in this regard. We reviewed pertinent primary literature sources as well as current evidence-based guidelines on diabetes management. KEY WORDS pharmacogenomics, diabetes, monogenic diabetes syndrome, maturity-onset diabetes of the young, oral glucose lowering agents. (Pharmacotherapy 2017;37(9):1131–1149) doi: 10.1002/phar.1976
Diabetes mellitus (DM) presents a significant public health problem worldwide and is recog- nized as a cause of premature death and disabil- ity.
1 According to the World Health Organization’s
Global Report, the number of people diagnosed with DM worldwide has been increasing, reaching 422 million adults in 2014 and causing 1.5 million deaths.
2 In the United States, although newly diag-
nosed cases of DM are declining, DM remains at an all-time high with 29 million diagnoses reported for 2016 (https://www.cdc.gov/chronicdis ease/resources/publications/aag/diabetes.htm). DM was the seventh leading cause of death in 2013, and data indicate that patients with diabetes are twice as likely to develop cardiac complications at a younger age compared with people who do not have the disease.
3 The economic burden of DM
remains high. Direct and indirect costs stand at $245 billion and cost of medical expenditures for individuals with DM at $17,000 annually, which is reportedly 2.3 times greater than for those without DM.
4
The authors, employees of Fairleigh Dickinson Univer- sity School of Pharmacy, certify that they have no affilia- tions with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or nonfinancial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manu- script.
*Address for correspondence: Nina Elk, School of Phar- macy, Fairleigh Dickinson University, 230 Park Avenue, M-SP1-01, Florham Park, NJ 07932; e-mail: [email protected]. � 2017 Pharmacotherapy Publications, Inc.
R E V I E W O F T H E R A P E U T I C S
Prevention of DM, early diagnosis, and delay of disease progression are critical measures to avoid life-threatening complications. Current guidelines emphasize the importance of personal- ized care for patients with DM and recommend use of patient-specific pharmacotherapy approaches to treatment.
5 An algorithmic
approach to the management of DM incorporates steps such as identification of diabetes type, gly- cemic target selection, use of medications based on proven efficacy, and side-effect profiles.
5
In addition, addressing comorbidities attributing to cardiovascular diseases and counseling patients on a healthy lifestyle are components of personalized care for diabetes management. Genetic findings in DM risk assessment, patho- physiology, disposition, and efficacy of drugs used in DM allow expanding the definition of patient-specific therapy (Figure 1). This article reviews the genetic findings in pathophysiology and risk association of DM and pharmacogenet- ics of common oral glucose lowering agents, fol- lowed by a discussion of a pharmacogenomic contribution to personalized therapy.
Pharmacogenomic Landscape of Diabetes
Mellitus
Although DM is classified into four general categories based on pathophysiology, further differentiation based on genetic components is possible (Table 1). For the remainder of this section, we focus on the polygenic type 2 DM (T2DM) with some discussion on the monogenic forms of diabetes that include maturity-onset diabetes of the young (MODY) and neonatal dia- betes mellitus (NDM).
Type 2 Diabetes Mellitus: A Polygenic, Multifactorial Disease
T2DM is the most prevalent type of diabetes, and its polygenicity is supported by many lines of evidence showing both a distinct familial aggregation and a lack of Mendelian segrega- tion.
6 Pathophysiology of T2DM involves multi-
ple abnormalities in the regulation of glucose homeostasis (Figure 2). The search for suscepti- bility risk genes involved in the pathophysiology of T2DM and its observed phenotypes has been ongoing for decades. Several strategies that have been used include linkage, candidate gene, gen- ome, and exome-wide association studies.
9
Newer approaches, such as the reverse genetics phenome-wide approach of surveying clinically validated phenotypes with variants in known susceptibility genes, have also been used to vali- date diabetes-linked genes and pathways.
10
Candidate gene studies (CGS) focus on genes related to the disease in question and a priori knowledge of any mechanistic roles in the dis- ease pathway. Genes involved in glucose home- ostasis, specifically those encoding pathways of insulin production, secretion, and response (such as the insulin receptor [INSR], glycogen synthase 1 [GYS1], glucose transporter 4 [GLUT4], and insulin receptor site-1 [IRS-1]) were the main focus of T2DM CGS.
11, 12 Overall,
CGS contribution to valid genetic associations in T2DM was limited and sometimes not repro- ducible due to the pleiotropic nature of the dis- ease, small sample sizes, high genotyping costs, and low genetic signal throughput. Despite these shortcomings, CGS led to the identification of common variants in confirmed susceptibility genes such as peroxisome proliferator-activated receptor gamma (PPAR-c) and KCNJ11 encoding the inwardly rectifying Kir6.2 component of the b-cell adenosine triphosphate (ATP)-sensitive potassium channel (KATP channel).
13
Genome-wide association studies (GWAS) are comprehensive studies that use an unbiased and hypothesis-free approach to interrogate the whole genome for associations between genetic variants and observed phenotypes.
7 Because
GWAS are a special type of case-control studies where individual genomes of subjects with dis- ease (cases) are scanned for genetic markers dif- fering in frequency from control subjects, there is potential power to discover novel genetic loci underlying predisposition to and development of T2DM.
7 Some limitations (small sample size,
low throughput) of diabetes CGS were overcome Figure 1. Factors to consider when designing diabetes mellitus (DM) treatment regimen.
1132 PHARMACOTHERAPY Volume 37, Number 9, 2017
with GWAS, and due to the high-powered nat- ure of this approach, many genetic loci were both discovered and/or validated.
11, 14 The initial
set of GWAS studies on diabetes was published a decade ago, and about a dozen loci were strongly associated with T2DM risk across these studies.
15 Indeed, the strongest genetic signal(s)
associated to date with diabetes susceptibility was found with single nucleotide polymorphisms (SNPs) in the transcription factor 7-like 2 (TCF7L2) gene. This loci was discovered by whole-genome microsatellite marker analysis in a cohort of ~1000 Icelandic subjects16 and sub- sequently validated and replicated in T2DM GWAS.
14, 17 The protein encoded by TCF7L2 is
a high-mobility group box containing transcrip- tion factor postulated to affect diabetes risk through pancreatic b-cell dysfunction via the Wnt signaling pathway.
16 Other validated signals
reported include SNPs in the KCNJ11, PPAR-c, Fat Mass and Obesity (FTO) associated genes
15
(Table 2). At the time of last access, the publicly available NHGRI-EBI Catalog of published GWAS
18 lists over 157 studies on diabetes with
more than 1500 associations at genome-wide significance levels (p=5 9 10�8). One limiting assumption of the GWAS approach is that genetic variants conferring disease susceptibility are common (more than 5% frequency) in the studied populations,
14 an assumption that has
not held true with T2DM. It was postulated that
only a fraction of heritability was attributable to common variants and that combinations of rare alleles of small effect size were more likely to explain a greater fraction of T2DM heritability.
19
This rare variant hypothesis is still unproven. A very recent study by two of the largest diabetes GWAS consortia concluded that large-scale sequencing did not support the idea of a major role of lower frequency rare variants in predis- position to T2DM.
17
Monogenic Diabetes Syndrome
The monogenic diabetes syndrome covers a heterogeneous group of single-gene autosomally inherited diabetes that cannot be classified as type 1 or type 2 DM. The syndrome can be fur- ther divided into two main forms, MODY and NDM.
Maturity-Onset Diabetes of the Young
MODY (unlike polygenic T2DM) arises from monogenic defects in pancreatic b-cell function with little or no defect in insulin action. The clini- cal features of MODY include autosomal domi- nant inheritance, early onset (before age 25), absence of obesity, and evidence of b-cell function as seen with endogenous insulin secretion.
20 Cur-
rently ~13 different types of MODY (1–14) exist, associated with abnormalities in more than six
Table 1. Differentiation of Diabetes Mellitus Types 5–8
DM type Prevalence Pathophysiology Genetic component
Type 1 5–10% Autoantibodies to pancreatic islet cells and insulin are present that cause pancreatic b-cell destruction leading to insulin deficiency.
Mostly associated with genetic variation of HLA association genes responsible for immune system response and pancreatic b-cell function. Other genes are linked to type 1 DM.
a
Type 2 90–95% Increase in glucose production by the liver, insulin resistance in target tissues, and dysfunction of pancreatic b cells in the pancreas causing impairment of insulin secretion.
Mutations in specific genes associated with disorders of glucose homeostasis.
Monogenic diabetes syndrome: NDM
<5% Insulin is produced but not secreted through KATP channels in pancreatic b cells.
Mutations in a gene that encodes KATP channels in pancreatic b cells.
Monogenic diabetes syndrome: MODY
<5% In MODY 1 and MODY 3, hyperglycemia is present due to partial pancreatic b-cell dysfunction.
Mutations in the gene that encodes HNF4a in MODY 1 and HNF1a in MODY 3
In MODY 2, hyperglycemia is due to dysfunction of glucose-level monitoring by pancreatic b cells.
Mutations in the gene that encodes GCK and several transcription factors relevant to pancreatic b-cell development.
DM = diabetes mellitus; GCK = glucokinase; HLA = human leukocyte antigen; HNF = hepatocyte nuclear factor; KATP channel = ATP-sensi- tive K
+ channel; MODY = maturity-onset diabetes of the young; NDM = neonatal diabetes mellitus.
a For a complete list of genes associated with type 1 DM, visit https://ghr.nlm.nih.gov/condition/type-1-diabetes#.
PERSONALIZED MEDICINE IN DIABETES MANAGEMENT Elk and Iwuchukwu 1133
different loci on different chromosomes and sub- classified by the deficient gene causing the observable phenotype.
21 MODY makes up ~1–4%
of all pediatric DM cases, and because presenting characteristics include those associated with either T1DM (early onset, lean body mass) or T2DM (preserved b-cell function, familial aggre- gation), misdiagnosis is common.
22, 23
MODY 1, 2, and 3 are the most well character- ized of all MODY subtypes. MODY 1 makes up ~10% of cases in characterized cohorts of MODY subjects.
21 It is caused by mutations in HNF4a,
encoding the transcription factor hepatocyte nuclear factor 4-a.22 A defect in a similar modi- fier gene, HNF1a, is responsible for the most common subtype MODY 3 that together with MODY 2 makes up ~50% of all cases.21 These hepatic transcription factors promote the tran- scription of multiple genes related to glucose metabolism and insulin production and secretion. MODY 2 is due to mutations in the glucoki-
nase (GCK) gene, causing decreased function of glucokinase, the enzyme responsible for blood glucose level homeostasis in pancreatic b cells.20
The resultant impaired hepatic cell glucose phosphorylation is responsible for the very mild hyperglycemia observed in patients with MODY 2
8, 22 (Table 3).
Neonatal Diabetes Mellitus Syndrome
NDM syndrome is a rare form of diabetes pre- senting within the first 6 months of life. Infants with NDM do not produce enough insulin, lead- ing to increased blood glucose levels. In about half of presenting patients, the condition persists and is termed permanent neonatal diabetes mel- litus (PNDM). In the other half of cases, the condition disappears during infancy and is ter- med transient neonatal diabetes mellitus (TNDM). NDM has a number of genetic causes attributed to it including variations in genes such as KCNJ11, ABCC8, GCK, INS, and ZFP57.
23–26 The most common causes are muta-
tions in KCNJ11 and ABCC8, the two genes encoding different subunits of the ATP-depen- dent potassium channel in pancreatic islet b cells. These mutations serve to decrease insulin
Figure 2. Schematic presentation of type 2 diabetes mellitus (T2DM) pathophysiology. Pathophysiology of T2DM involves deficiency of insulin action that leads to hyperglycemia and inability to maintain glucose homeostasis. Deficiency of insulin action may be due to decrease in amount of insulin secreted by islet cells of the pancreas, insulin resistance with skeletal muscles’ inability to uptake glucose for utilization, or a combination of both. Decreased insulin secretion results from b-cell dysfunction, decrease in b-cell mass, and reduced action of glucagon-like peptide 1 hormone that is involved in insulin secretion. Insulin resistance linked to obesity is due to a decreased number of insulin receptors, increased fatty acids, and triglycerides in adipose tissue interfering with insulin action, as well as causing an increase in production of glucose by the liver. Disease progresses with unresolved hyperglycemia, increase in gluconeogenesis, and loss of b-cell function.
1134 PHARMACOTHERAPY Volume 37, Number 9, 2017
secretion by preventing membrane depolariza- tion (Table 3).
Pharmacogenetic-Based Associations for
Common Oral Glucose Lowering Therapies
Various agents are available for the treatment of DM. They differ in their mechanisms of action and efficacy, and ability to attain DM treatment goals such as prevention of macrovascular and microvascular complications. In addition, antihy- perglycemic agents are classified by their propen- sity to cause side effects that have a negative impact on DM disease state treatment
5, 27
(Table 4).
Metformin
Metformin belongs to the biguanide class of oral glucose lowering therapies (OGLTs). Although many aspects of metformin’s action are still not well established, it has been postulated to exert its glucose-lowering effects by many pleiotropic mechanisms.
28 One major hypothesis
is metformin-induced activation of AMP-acti- vated protein kinase (AMPK).
29 In addition,
metformin prevents hyperglycemia by decreasing intestinal glucose absorption and increasing peripheral glucose uptake and utilization.
30 The
glucose-lowering effect of metformin is highly variable, with 35% of patients failing to achieve initial decreases in hemoglobin HbA1c.
31 About
Table 2. Selected Genes and Single Nucleotide Polymorphisms Associated with Insulin Action or b-Cell Function in T2DM Risk/Susceptibility GWAS
Gene Chromosome rs number Overall
odds ratio Function Postulated mechanism Year
ADAMTS9 11, 14
3 rs4607103 1.09 Secreted metalloproteinase
Insulin action 2008
ADCY5 14
3 rs11708067 1.12 Adenylyl cyclase Insulin action 2010 BCAR1
14, 17 16 rs7202877 1.12 Docking protein Regulation of
b-cell function 2012
BCL11A 14, 17
2 rs243021 1.08 Zinc finger protein Regulation of b-cell function
2010
CCND2 17
12 rs76895963 0.53 Cyclin D2 Cell cycle regulator/Enhanced insulin secretion
2014
CDKAL1 9, 11, 14, 17
6 rs7754840 1.15 Methylthiotransferase Regulation of b-cell function
2007
CDKN2A/B 9, 11, 14, 17
9 rs7754840 1.20 Cyclin-dependent kinase inhibitor
Regulation of b-cell function
2007
FTO 9, 11, 14, 17
16 rs8050136 1.27 Fat Mass and Obesity Associated/Nucleic acid demethylase
Insulin action 2007
GCKR 11, 17
2 rs780094 1.08 Glucokinase regulator protein
Insulin action 2007
HHEX/IDE 9, 11, 14, 17
10 rs1111875 1.15 Transcriptional repressor Intracellular insulin degradation
2007
IGF2BP2 9, 11, 14, 17
3 rs4402960 1.17 Insulinlike growth factor II mRNA-binding protein
Regulation of b-cell function
2007
IRS1 11, 14, 17
2 rs2943640 1.12 Docking/binding protein Insulin action 2009 JAZF1
9, 11, 14, 17 7 rs864745 1.10 Zinc finger protein Regulation of
b-cell function 2008
KCNJ11 11, 17
11 rs5219 1.14 Inwardly rectifying potassium channel Kir6.2
Regulation of insulin secretion
2003
KCNQ1 9, 11, 14, 17
11 rs2237892 1.23 Potassium channel b-cell function 2008 NOTCH2
9, 11, 14, 17 1 rs10923931 1.13 Transmembrane receptor Pancreatic
cell development 2008
PPAR-c 9, 11, 14, 15, 17 3 rs1801282 1.11 Peroxisome proliferator-activated receptor
Regulation of insulin action
2000
SLC30A8 9, 11, 14, 17
8 rs13266634 1.15 b-cell zinc efflux transporter Insulin storage and secretion
2007
TCF7L2 14, 15
10 rs7903146 1.4 T-cell transcription factor b-cell function 2006 TSPAN8
9, 11, 14, 17 12 rs7961581 1.09 Cell surface glycoprotein b-cell function 2008
GWAS = genome-wide association study; rs = reference sequence. Note: The rs numbers indicate the reference single nucleotide polymor- phism (SNP) cluster ID used for identification within the SNP public archive database (https://www.ncbi.nlm.nih.gov/projects/SNP/). All GWAS are documented in the Catalog of Published Genome-Wide Association Studies (http://www.ebi.ac.uk/gwas).
PERSONALIZED MEDICINE IN DIABETES MANAGEMENT Elk and Iwuchukwu 1135
T a b le
3 . C o m m o n M O D Y S u b ty p e s , P h e n o ty p e s , a n d S e le c t A s s o c ia te d G e n e ti c M u ta ti o n s
S u b ty p e a
Im p li c a te d g e n e
(a ff e c te d p ro te in )
M u ta ti o n sb
C h ro m o so m e
F re q u e n c y in
M O D Y p a ti e n ts
A g e a t d ia g n o si s
P a th o p h y si o lo g y
P h e n o ty p e
M O D Y 1 2 0 – 2 3
H N F -4 a
(H e p a to c y te
N u c le a r F a c to r 4 a )
N a tu ra ll y o c c u rr in g ,
H e te ro z y g o u s (Q
2 6 8 X ,
R 1 5 4 X , R 1 2 7 W )
2 0
5 %
A d o le sc e n c e /E a rl y
a d u lt h o o d
T ra n sc ri p ti o n
fa c to r (b -c e ll
fu n c ti o n in
p a n c re a s)
N e o n a ta l
h y p e ri n su li n is m ,
d ia b e te s
M O D Y 2 8 , 2 0 – 2 3
G C K (G
lu c o k in a se )
H e te ro z y g o u s (A
3 7 8 T ,
E 3 3 9 K )
7 1 0 – 6 0 %
B ir th
to e a rl y
c h il d h o o d
H e x o k in a se
(g lu c o se
se n so r in
p a n c re a s
a n d li v e r)
M il d h y p e rg ly c e m ia
M O D Y 3 2 0 – 2 3
T C F 1 o r H N F -1 a
(H e p a to c y te
N u c le a r F a c to r 1 a )
H e te ro z y g o u s (2 9 1 + C )
1 2
3 0 – 6 0 %
A d o le sc e n c e /E a rl y
a d u lt h o o d
T ra n sc ri p ti o n fa c to r
(b -c e ll fu n c ti o n in
p a n c re a s a n d k id n e y )
D ia b e te s
M O D Y 4 2 0 – 2 3
IP F 1 o r P D X 1
(I n su li n P ro m o te r
F a c to r 1 )
H e te ro z y g o u s
(P ro 6 3 fs d e lC )
1 3
< 1 %
E a rl y a d u lt h o o d
T ra n sc ri p ti o n fa c to r
(b -c e ll fu n c ti o n
in p a n c re a s)
D ia b e te s
M O D Y 5 2 0
T C F 1 o r H N F -1 b
(H e p a to c y te
N u c le a r F a c to r 1 b )
H e te ro z y g o u s
(E 1 0 1 X , d e lT )
1 7
3 – 1 0 %
A d o le sc e n c e /E a rl y
a d u lt h o o d
T ra n sc ri p ti o n fa c to r
(b -c e ll fu n c ti o n
in p a n c re a s)
D ia b e te s
M O D Y 6 2 0
N E U R O D 1
(N e u ro g e n ic
D if fe re n ti a ti o n fa c to r 1 )
H e te ro z y g o u s
(2 0 6 + C )
2 V e ry
ra re
L a te r in
li fe
T ra n sc ri p ti o n fa c to r
(b c e ll fu n c ti o n in
p a n c re a s a n d k id n e y )
D ia b e te s
T N D M
2 0 , 2 3 – 2 6
Z A C , A B C C 8 ,
K C N J1 1 , H N F -1 b
H e te ro z y g o u s
R a re
B ir th
to 6 m o
T ra n si e n t
d ia b e te s
P N D M
2 0 , 2 3 – 2 6
K C N J1 1 , A B C C 8 ,
G C K * , IP F 1 * ,
H e te ro z y g o u s,
* H o m o z y g o u s
R a re
B ir th
to 6 m o
P e rm
a n e n t
d ia b e te s
M O D Y = m a tu ri ty -o n se t d ia b e te s o f th e y o u n g .
a M O D Y su b ty p e s 1 – 1 4 h a v e b e e n c h a ra c te ri z e d a t th e g e n e ti c le v e l, b u t o n ly
su b ty p e s 1 – 6 h a v e c li n ic a ll y d e fi n e d p h e n o ty p e s.
M O D Y 7 – 1 4 a re
v e ry
ra re
h e te ro z y g o u s si n g le -p o in t m u ta ti o n s
ty p ic a ll y o c c u rr in g in
e a rl y c h il d h o o d /a d o le sc e n c e .
b M u ta ti o n s a re
se le c t re p re se n ta ti v e v a ri a n ts
li st e d fo r st u d ie s o f a ff e c te d p a ti e n ts
in th e O n li n e M e n d e li a n In h e ri ta n c e in
M a n d a ta b a se
a t h tt p s: // w w w .o m im
.o rg . B e c a u se
M O D Y is
a re la ti v e ly
ra re
ty p e o f d ia b e te s, n o t a ll m u ta ti o n s li st e d in
th e ta b le
a re
re p re se n te d in
a ll M O D Y p a ti e n ts .
1136 PHARMACOTHERAPY Volume 37, Number 9, 2017
a third of patients fail to achieve adequate glyce- mic control on initial metformin monotherapy, and over time many patients become less sensi- tive (responsive) to its glucose-lowering effects.
32 The genetic contribution to this vari-
ability in response has been studied with a focus on the pharmacokinetics (PK) and pharmacody- namics (PD) of metformin.
Metformin: Pharmacokinetics/Pharmacodynamics
Metformin is not metabolized upon ingestion; rather, it is actively transported into various tis- sues and cleared by renal excretion. Metformin is taken up into the intestinal cells by plasma monoamine transporters, or PMAT (SLC29A4),
taken up into the bloodstream by the organic cation transporter OCT1 (SLC22A1) that is also responsible for its active hepatic uptake.
28 Sub-
sequent secretion from the blood into the renal tubular cells is mediated by another organic cation transporter, OCT2 (SLC22A2). Final excretion into the urine is mediated by two transporters in the multidrug and toxin extru- sion family, MATE1 and MATE2 (SLC47A1 and SLC47A2) (Figure 3).
28
Metformin: Evidence of Pharmacogenetic Associa- tions
Although PD targets of metformin are not well established, the pharmacogenetics of the
Table 4. Classification of Oral Medications Used in Diabetes Mellitus Treatment 5, 27
Drug class Drugs Mechanism of action: cellular
Mechanism of action: physiologic
HbA1c % decrease
Hypoglycemia risk
Weight effect
Biguanides *
Metformin Activate AMPK, which then increase intracellular AMP
Cause decrease in hepatic glucose production and intestinal absorption of glucose; improve insulin sensitivity
1.0–1.5 Neutral Slight loss
Sulfonylureas* Glyburide Close KATP channels on pancreatic b-cell plasma membranes
Stimulate insulin release from b cells; reduce glucose output from the liver, increase insulin sensitivity
0.8 Moderate to severe
Gain Glipizide Glimepiride
Meglitinides Repaglinide Close KATP channels on pancreatic b-cell plasma membranes
Stimulate insulin release from pancreatic b cells
0.7 Mild to moderate
Gain Nateglinide
Thiazolidinedione* Pioglitazone Activate the nuclear transcription factor PPAR-c
Improve target cell response to insulin
0.8 Neutral Gain Rosiglitazone
a-Glucosidase inhibitors
Acarbose Inhibit intestinal alpha-glucosidase enzyme
Slow intestinal carbohydrate digestion/Absorption
0.6 Neutral Neutral Miglitol
DPP-4 inhibitors Sitagliptin Inhibit DPP-4 activity, increasing postprandial active incretin (GLP-1, GIP) concentrations
Glucose-dependent increase in insulin secretion and decrease in glucose production
0.7 Neutral Neutral Saxagliptin Linagliptin Alogliptin
SGLT2 inhibitors Canagliflozin Inhibit SGLT2 in the proximal nephron
Block glucose reabsorption in kidneys
0.7–1.0 Neutral Loss Empagliflozin
GLP-1 receptor agonists
Exenatide Activate GLP-1 receptors
Glucose-dependent increase in insulin secretion and decrease in glucagon secretion; slow gastric emptying and increase satiety
1 Neutral Loss Liraglutide Albiglutide Dulaglutide
Amylin mimetics Pramlintide Activate amylin receptors
Decrease in glucagon secretion; slow down gastric emptying and increase satiety
0.3 Severe Loss
AMPK = AMP-activated protein kinase; DPP-4 = dipeptidyl peptidase-4; GIP = gastric inhibitory polypeptide; GLP-1 = glucagon-like peptide 1; KATP = ATP-sensitive K
+ channel; PPAR-c = peroxisome proliferator-activated receptor gamma; SGLT2 = sodium-glucose cotransporter-2.
*Medications marked with an asterisk have evidence of a pharmacogenetic component.
PERSONALIZED MEDICINE IN DIABETES MANAGEMENT Elk and Iwuchukwu 1137
metformin PD-based response have been explored using both CGS and GWAS (Table 5).
33 One of the most promising results
was seen in the Genetics of DARTS (GoDARTS) study. GoDARTs was a large-scale study with 1024 T2DM patients on metformin, a subset from 3200 metformin-treated patients from the parent U.K. Diabetes Audit and Research in Tayside Scotland (DARTS) study.
53 DARTS
aimed to identify all diabetes patients in the Tayside community using electronic record linkage of multiple data sources and compare ascertainment of cases with that of primary care registers.
53 Subjects in the GoDARTs substudy
consented to have their genomes interrogated. 39
In this GWAS, the strongest association with metformin response was found with an intronic variant (rs11212617) near a candidate gene, ATM, or the ataxia telangiectasia mutated gene that encodes a serine/threonine kinase.
39 The
minor C allele of rs11212617 was associated with successful treatment outcomes on met- formin, defined as a reduction in HbA1c and an ability to reach treatment target HbA1c values of 7% or lower.
39 It has been postulated that
mutations in the ATM gene may affect upstream regulation of AMPK and alter glyce- mic response to metformin.
39
The rs11212617 when studied in three other different cohorts supported the positive
association between the minor C allele and met- formin response,
37 an association further con-
firmed in meta-analyses of the three cohorts alone and in combination with two previously published studies.
37 The Diabetes Prevention
Program (DPP), another large T2DM cohort study, failed to show an association between this top ATM variant and metformin effect in delay- ing progression to diabetes from impaired glu- cose tolerance.
40 These differences are most
likely due to the different phenotypes (predia- betes and overt diabetes) and study outcomes (impact on diabetes incidence and ability to reach treatment goals) in DPP and the other studies. PK studies seeking to explain differences in
metformin exposure and response mainly focused on variations in genes coding for OCT1, OCT2, and MATE1.
30, 33, 38, 54 Multiple genetic variants
in SLC22A1 have been associated with decreased metformin efficacy and increased renal clear- ance.
34, 54 The GoDARTS study did not replicate
associations between two common SLC22A1 vari- ants (rs12208357, rs72552763) and glycemic response.
35 The DPP study discovered a novel
SLC22A1 variant (rs683369) associated with a 31% decrease in diabetes incidence risk in met- formin-treated patients and also validated SLC47A1 variants that were previously associated with increased response to metformin (Table 5).
33
Figure 3. Schematic presentation of pharmacogenomics of metformin. Metformin is absorbed in the small intestine via plasma monoamine transporters (PMATs) encoded by SLC29A4 and is taken up into the bloodstream by an organic cation transporter (OCT1) encoded by the SLC22A1 gene that is also responsible for active hepatic uptake. Metformin is secreted from the blood to tubular cells by the organic cation transporter (OCT2) encoded by the SLC22A2 gene. Finally, metformin is excreted into the urine via multidrug and toxin extrusion family transporters MATE1 and MATE2 that are encoded by genes SLC47A1 and SLC47A2, respectively.
1138 PHARMACOTHERAPY Volume 37, Number 9, 2017
T a b le
5 . S e le c t P K -P D
B a s e d G e n o ty p e -P h e n o ty p e A s s o c ia ti o n s fo r M o s t C o m m o n ly
S tu d ie d O ra l G lu c o s e L o w e ri n g T h e ra p ie s in
T 2 D M
P a ti e n ts
D ru g
G e n e
V a ri a n t
S tu d y o u tc o m e s
E ff e c t
B ig u a n id e s
(m e tf o rm
in )
S L C 2 2 A 1 (O
C T 1 )3
3 – 3 6
rs 6 2 2 3 4 2
H b A 1 c
0 .5 8 %
re d u c ti o n in
H b A 1 c fo r C C
h o m o z y g o te s
rs 6 2 8 0 3 1 (M
4 0 8 V )
H b A 1 c
N o si g n if ic a n t e ff e c t
rs 3 5 1 6 7 5 1 4 (M
4 2 0 d e l)
H b A 1 c
N o e ff e c t
rs 1 2 2 0 8 3 5 7 (R 6 1 C )
H b A 1 c
N o e ff e c t
m e tf o rm
in st e a d y -s ta te
le v e ls
a n d d e c re a se
in H b A 1 c
G e n e d o se
e ff e c t: D e c re a si n g A U C
a n d
p h a rm
a c o d y n a m ic
e ff e c ts
a t 6 a n d
2 4 m o c o rr e sp o n d in g
to n u m b e r o f O C T 1 re d u c e d fu n c ti o n
a ll e le s
rs 1 2 2 0 8 3 5 7 (R 6 1 C ),
rs 3 5 1 6 7 5 1 4 (M
4 2 0 d e l)
G a st ro in te st in a l si d e
e ff e c ts
2 -f o ld
h ig h e r o d d s o f g a st ro in te st in a l
e ff e c ts
in in d iv id u a ls
w it h O C T 1
re d u c e d fu n c ti o n a ll e le s
rs 6 8 3 3 6 9
T 2 D M
p re v e n ti o n
N o m in a l a ss o c ia ti o n w it h d ia b e te s
p re v e n ti o n
rs 6 2 2 3 4 2
H b A 1 c
N o e ff e c t
S L C 2 2 A 2 (O
C T 2 )3
3 , 3 7 , 3 8
rs 3 1 6 0 1 9 (G
8 0 8 T )
M e tf o rm
in st e a d y -s ta te
le v e ls
a n d d e c re a se
in H b A 1 c
N o e ff e c t
rs 6 6 2 3 0 1
T 2 D M
p re v e n ti o n
N o m in a l a ss o c ia ti o n w it h d ia b e te s
p re v e n ti o n
S L C 4 7 A 1 (M
A T E 1 )3
3 , 3 6 , 3 7
rs 2 2 8 9 6 6 9
H b A 1 c
In c re a se d re d u c ti o n in
H b A 1 c le v e ls
rs 2 2 5 2 2 8 1
H b A 1 c
In c re a se d re d u c ti o n in H b A 1 c le v e ls
fo r
h o m o z y g o u s C C
p a ti e n ts
rs 1 2 9 4 3 5 9 0
H b A 1 c
D e c re a se d in
H b A 1 c re d u c ti o n fo r
A A h o m o z y g o te s
rs 3 4 3 9 9 0 3 5
M e tf o rm
in st e a d y -s ta te
le v e ls
a n d d e c re a se
in H b A 1 c
N o e ff e c t
A T M
3 7 , 3 9 , 4 0
rs 1 1 2 1 2 6 1 7
H b A 1 c
In c re a se d re d u c ti o n in
H b A 1 c a n d
in c re a se d tr e a tm
e n t su c c e ss
T 2 D M
p re v e n ti o n
N o e ff e c t o n ti m e to
p ro g re ss io n to
d ia b e te s
(c o n ti n u e d )
PERSONALIZED MEDICINE IN DIABETES MANAGEMENT Elk and Iwuchukwu 1139
T a b le
5 (c o n ti n u e d )
D ru g
G e n e
V a ri a n t
S tu d y o u tc o m e s
E ff e c t
S u lf o n y lu re a s
C Y P 2 C 9 4 1 – 4 4
rs 1 0 5 7 9 1 0
H y p o g ly c e m ia
In c re a se d ri sk
o f h y p o g ly c e m ia
w it h
v a ri a n t a ll e le
rs 1 0 5 7 9 1 0 , rs 1 7 9 9 8 5 3
H y p o g ly c e m ia
H e te ro z y g o u s a n d h o m o z y g o u s v a ri a n ts
m o re
c o m m o n in
S U -t re a te d p a ti e n ts
w it h h y p o g ly c e m ia
rs 1 0 5 7 9 1 0 , rs 1 7 9 9 8 5 3
F a st in g b lo o d g lu c o se ,
S U
d o se
re q u ir e m e n ts
In c re a se d lo w e ri n g o f fa st in g p la sm
a g lu c o se , in c re a se d to lb u ta m id e d o se
re d u c ti o n b e tw
e e n d o se
1 a n d 1 0 fo r
v a ri a n t a ll e le s
rs 1 0 5 7 9 1 0 , rs 1 7 9 9 8 5 3
H b A 1 c
In c re a se d re d u c ti o n in
a b so lu te
H b A 1 c
le v e ls
rs 1 0 5 7 9 1 0 , rs 1 7 9 9 8 5 3
H y p o g ly c e m ia
N o o v e rr e p re se n ta ti o n o f v a ri a n t a ll e le s
in S U -t re a te d p a ti e n ts
w it h
h y p o g ly c e m ia
rs 1 0 5 7 9 1 0 , rs 1 7 9 9 8 5 3
T im
e to
st a b le
d o se
N o si g n if ic a n t e ff e c ts
w it h v a ri a n t a ll e le s
A B C C 8 3 6 , 3 7 , 4 5 – 4 7
rs 7 5 7 1 1 0
H b A 1 c , fa st in g p la sm
a g lu c o se
G a ll e le
c a rr ie rs
h a d g re a te r H b A 1 c
re d u c ti o n s c o m p a re d to
th e T T
h o m o z y g o te s
H b A 1 c , fa st in g p la sm
a g lu c o se
G re a te r re d u c ti o n s in
fa st in g p la sm
a g lu c o se
a n d H b A 1 c w it h h e te ro z y g o u s
a n d h o m o z y g o u s v a ri a n t a ll e le s
S e v e re
h y p o g ly c e m ia
N o e ff e c t
K C N J1 1 4 1 – 4 4
rs 5 2 1 9
H b A 1 c , fa st in g p la sm
a g lu c o se
S ig n if ic a n tl y a ss o c ia te d w it h g re a te r
d e c re a se s in
fa st in g p la sm
a g lu c o se
H b A 1 c
G re a te r H b A 1 c re d u c ti o n in
v a ri a n t
a ll e le
c a rr ie rs
v s w il d -t y p e
h o m o z y g o te s
S e c o n d a ry
fa il u re
H b A 1 c , se v e re
h y p o g ly c e m ia
V a ri a n t a ll e le
a ss o c ia te d w it h in c re a se d
H b A 1 c le v e ls
a n d w a s le ss
fr e q u e n t in
S U -t re a te d p a ti e n ts
w it h se v e re
h y p o g ly c e m ia
v s c o n tr o ls
S U
re sp o n se
N o e ff e c t
T C F 7 L 2 4 8 , 4 9
rs 1 2 2 5 5 3 7 2 , rs 7 9 0 3 1 4 6
H b A 1 c < 7 %
In c re a se d ri sk
o f tr e a tm
e n t fa il u re
w it h
v a ri a n ts
a ll e le s
rs 7 9 0 3 1 4 6
H b A 1 c < 7 %
In c re a se d ri sk
o f tr e a tm
e n t fa il u re
w it h
v a ri a n t a ll e le
H b A 1 c a n d fa st in g
p la sm
a g lu c o se
re d u c ti o n
L e ss
H b A 1 c re d u c ti o n a n d fa st in g p la sm
a g lu c o se
le v e ls
w it h v a ri a n t a ll e le s
c o m p a re d w it h w il d -t y p e a ll e le s
(c o n ti n u e d )
1140 PHARMACOTHERAPY Volume 37, Number 9, 2017
Sulfonylureas
Sulfonylureas (SUs) are insulin secretagogues shown to be efficacious in the treatment of T2DM by producing favorable lowering effect of HbA1c and are recommended to be used in com- bination with other agents for better blood glu- cose control.
55 SUs bind to their target site on
pancreatic b islet cells, the sulfonylurea receptor type 1 (SUR1), encoded by ABCC8, a member of the ATP binding cassette gene superfamily.
56, 57
When SUs bind to SUR1, it results in closing of the pancreatic b-cell KATP channel, causing membrane depolarization and triggering calcium influx with subsequent exocytosis of insulin- containing vesicles and increased insulin secretion.
57
Sulfonylureas: Pharmacokinetics/Pharmacodynamics
The PK and PD responses of SUs are well characterized. SUs undergo extensive and vari- able hepatic metabolism by cytochrome P450 2C9 (CYP2C9). Polymorphisms in CYP2C9 were shown to be one of the reasons behind the vari- ability in elimination rates for most SUs.
58 Two
CYP2C9 reduced function variants of interest, CYP2C9*2 (rs1799853) and CYP29*3 (rs1057910), are associated with decreased clear- ance and increased area under the curve (AUC) of SUs in healthy volunteers.
36
Response to SU therapy depends on b-cell function, PK differences among patients, and current T2DM status. Interpatient variability in response to SU treatment is observed with fail- ure of therapy and the incidence of hypo- glycemia. Primary failure with SUs is reported in 10–20% of patients and secondary failure in 5– 7% of patients.
56 Evidence from trials shows that
hypoglycemia develops in 31% of patients within the first year of starting SU, progressing to 38% of patients at 5 years of therapy.
31, 59 Accord-
ing to one meta-analysis, the incidence of severe hypoglycemia in SU-treated patients was 0.8%.
60
Sulfonylureas: Evidence of Pharmacogenetic Associ- ations
The largest pharmacogenetics-based clinical study to date was the GoDARTS study with 1073 patients, half of whom were on SU monotherapy and half on dual therapy (SU and metformin).
41 In this study, patients with two
copies of the CYP2C9*2 or *3 alleles were 3.4 times more likely to achieve target treatmentT
a b le
5 (c o n ti n u e d )
D ru g
G e n e
V a ri a n t
S tu d y o u tc o m e s
E ff e c t
T h ia z o li d in e d io n e s
P P A R -c
5 0
rs 1 8 0 1 2 8 2
F a st in g p la sm
a g lu c o se ,
H b A 1 c re d u c ti o n
V a ri a n t a ll e le
c a rr ie rs
h a d g re a te r
d e c re a se s in
fa st in g p la sm
a g lu c o se
a n d
H b A 1 c
C Y P 2 C 8 5 1
rs 1 1 5 7 2 0 8 0 , rs 1 0 5 0 9 6 8 1
R o si g li ta z o n e st e a d y -
st a te
le v e ls , H b A 1 c ,
e d e m a ri sk
V a ri a n t a ll e le s a ss o c ia te d w it h lo w e r
p la sm
a le v e ls
o f ro si g li ta z o n e , re d u c e d
th e ra p e u ti c re sp o n se , a n d a lo w e r ri sk
o f d e v e lo p in g e d e m a
rs 1 1 5 7 2 0 8 0
H b A 1 c , w e ig h t g a in
L e ss
re d u c ti o n in
H b A 1 c , le ss
w e ig h t
g a in
a ft e r ro si g li ta z o n e . N o a ss o c ia ti o n
w it h p io g li ta z o n e
S L C O 1 B 1 5 1
rs 4 1 4 9 0 5 6
H b A 1 c , w e ig h t g a in
G re a te r H b A 1 c re d u c ti o n b u t n o t w e ig h t
g a in
a ft e r ro si g li ta z o n e . N o a ss o c ia ti o n
w it h p io g li ta z o n e
A D IP O Q 5 2
rs 2 2 4 1 7 6 6 , rs 1 5 0 1 2 9 9
H b A 1 c , fa st in g p la sm
a g lu c o se
D e c re a se
in fa st in g p la sm
a g lu c o se
fo r
b o th
S N P s, L e ss
re d u c ti o n in
H b A 1 c fo r
h o m o z y g o te
v a ri a n ts
o f rs 2 2 4 1 7 6 6
A U C
= a re a u n d e r th e c u rv e ; H b = h e m o g lo b in ; rs
= re fe re n c e se q u e n c e ; S N P = si n g le
n u c le o ti d e p o ly m o rp h is m ; S U
= su lf o n y lu re a ; T 2 D M
= ty p e 2 d ia b e te s m e ll it u s.
PERSONALIZED MEDICINE IN DIABETES MANAGEMENT Elk and Iwuchukwu 1141
HbA1c levels of less than 7% compared with patients with two copies of the wild-type CYP2C9*1 allele. Variant allele carriers were less likely to experience treatment failure with SU monotherapy.
41 Reports were published on asso-
ciations between CYP2C9 genotypes and the risk of hypoglycemia.
42, 43 An initial study showed
an odds ratio of 5.2 for the *2/*3 and the *3/*3 genotypes for risk of SU-induced severe hypo- glycemia,
43 but a subsequent 10-year follow-up
study did not corroborate these findings. 44
Although confirming the role of CYP2C9 in SU metabolism, the GoDARTS study and other stud- ies
41, 43, 44 also emphasize that determination of
CYP2C9 genotype may not be sufficient itself to predict SU response. Translating these PK gene- based studies to clinical outcomes, such as fast- ing serum glucose levels and hypoglycemia, showed only modest effects due most likely to limited sample size and lack of statistical power. For the PD of SUs, initial CGS studies focused
on the ABCC8 and KCNJ11 genes encoding SUR1 and Kir6.2, the different subunits of the pancreatic b-cell KATP channels based on a priori knowledge of their association with dia- betes (Figure 4).
45, 46, 61 Results were consistent
for variants previously associated with T2DM: E23K (rs5219) in KCNJ11 and S1369A
(rs757110) in ABCC8, two tightly linked vari- ants occurring together as a haplotype due to linkage.
62 This haplotype was shown to be less
sensitive to SU treatment in vitro. 62
Clinical studies probing associations of these variants with either therapeutic response or occurrence of hypoglycemia have proven equivocal with some studies showing a positive association and others showing no association.
45, 47, 61, 63, 64 Of
note is the fact that studies differed with respect to SUs used, dose, study outcomes, ethnicities, cohort characteristics, and sample size. With the advent of whole-genome sequencing
technologies, novel genetic loci associated with T2DM and SU response have been discovered; the most replicated is the TCF7L2 loci.
16, 48, 49
In 901 SU-treated T2DM patients identified from GoDARTS, an intronic SNP, rs12255732, in TCF7L2 was strongly associated with SU response.
48 TCF7L2 is expressed in pancreatic b
cells, and reduced function alleles have been associated with lower insulin secretion sugges- tive of a direct or indirect role on b-cell func- tion.
48 Individuals with two copies of the variant
T allele were two times less likely to reach target HbA1c levels below 7% within 1 year of treat- ment compared with those with two copies of the G variant of rs12255732.
48 These findings
Figure 4. Schematic diagram of pharmacogenomics of sulfonylureas. Sulfonylureas bind to sulfonylurea receptor type 1 (SUR1) located on the pancreatic b islet cell that is encoded by the ABCC8, a member of the ATP-binding cassette gene superfamily. When SUs bind to SUR1, it results in closing of the pancreatic b-cell potassium channel (KATP), made up of Kir6.2 subunits encoded by KCNJ11. KATP channel closing causes membrane depolarization and triggering calcium influx with subsequent exocytosis of insulin-containing vesicles and increased insulin secretion, a process postulated to be regulated by TCF7L2. Insulin binds to insulin receptor site (IRS-1) encoded by IRS-1 and located on the surface of insulin-sensitive tissue. IRS-1 gets phosphorylated and regulates the function of insulin inside the cell.
1142 PHARMACOTHERAPY Volume 37, Number 9, 2017
were confirmed by other independent stud- ies.
49, 65 Another intron variant in TCF7L2,
rs7903146, was associated with lower response and increased risk of treatment failure for carri- ers of the T allele compared with the CC geno- types.
48, 49 Other PD targets such as Insulin
receptor substrate-1 encoded by the IRS-1 gene have been associated with T2DM and SU response. The most common variant in IRS-1, rs1801278, was associated with an increased risk of secondary failure to SU therapy in two sepa- rate cohorts of T2DM patients.
66, 67
Thiazolidinediones
Although their use is limited due to risk of severe adverse events such as heart failure, myocardial infarction, and bladder cancer, thia- zolidinediones (TZDs) still have a role in dia- betes pharmacotherapy. Decision to use a TZD in a patient is made after careful risk versus ben- efit evaluation. TZDs are insulin sensitizers that increase insulin-dependent glucose elimination, reduce hepatic glucose production, and decrease circulating free fatty acids. TZDs act by activat- ing PPAR-c; the activated receptor then binds to DNA response elements and controls the expres- sion of genes involved in maintenance of meta- bolic homeostasis.
68 PPAR-c activation in
mature adipose cells induces expression of genes involved in the insulin signaling cascade and as a result improves insulin sensitivity in diabetic patients. In addition, these genes were demon- strated to reduce insulin resistance.
69
As seen with the other OGLTs, there is pro- nounced interindividual variability in response to TZD treatment. Clinical studies of pioglita- zone and rosiglitazone show that between 12% and 45% of T2DM patients failed to achieve suf- ficient reduction in either fasting plasma glucose and/or HbA1c concentrations.
70
Thiazolidinediones: Pharmacokinetics/Pharmacody- namics and Evidence of Pharmacogenetic Associa- tions
TZDs and the impact of variation in PK genes on response has been evaluated. TZDs are lar- gely metabolized by the polymorphic cyto- chrome P450 2C8 (CYP2C8) and to a much lesser extent by CYP2C9 and CYP3A4.
71 Genetic
variation in CYP2C8 was reported to influence the PK of rosiglitazone with significantly lower AUC and higher oral clearance for *1/*3 sub- jects compared with *1/*1 subjects.71 Although
it was hypothesized that the lower metabolic activity CYP2C8*2 and *3 variants would result in decreased metabolism and increased drug levels, clinical studies did not corroborate these assumptions. Instead, they showed higher clear- ance and lower AUC for subjects with two copies of the *3 alleles compared with the wild type.
72 The CYP2C8*3 allele is relatively rare in
whites, and adjustments for baseline variables may have reduced the impact of genetic variabil- ity on rosiglitazone PK. Furthermore, TZDs are substrates of OATP1B (SLCO1B1), the main organic anion transporter protein responsible for their active shuttling into hepatocytes.
71 A joint
investigation on the effects of CYP2C8 and SLCO1B1 polymorphisms was recently con- ducted in 833 Scottish patients with T2DM trea- ted with TZDs.
51 Results showed that although
the CYP2C8*3 variant was associated with reduced glycemic response to rosiglitazone, the common SCLO1B1 521T>C variant, in contrast, was associated with an enhanced glycemic response.
51 A subset of patients known as super-
responders based on combined CYP2C8 (*1/*1) and SCLO1B1 (CC) genotypes had a 0.39% greater HbA1c reduction with rosiglitazone com- pared with poor responders.
51
Studies on TZDs and PD targets have shown some progress. Many studies to date have focused on polymorphisms in the PPAR-c gene, a logical first choice because it encodes the binding site for TZD ligands (Figure 5). The PPAR-c gene contains a common nonsynony- mous SNP, rs1801282, that gives rise to a codon 12 proline to alanine substitution (Pro12Ala) implicated in more than 50 studies on diabetes susceptibility.
73 Meta-analyses of these associa-
tions show an almost 20% reduction in T2DM risk for Ala12 allele carriers compared with car- riers of the Pro12 allele.
73, 74 Clinical transla-
tional studies probed for associations with the Ala12 allele and response to TZD therapy in at- risk patients and patients with T2DM. Reports on the Ala12 allele and TZD response in dia- betes patients are conflicting. Whereas one study showed no difference in frequency of subjects with one copy of the Ala12 allele,
75 another
study reported a higher frequency of the Ala12 allele in responders to pioglitazone.
76 One study
reported a significantly higher response to rosiglitazone at 12 weeks in Ala12 subjects com- pared with homozygous Pro12 subjects (43.72% vs 86.67%).
50 Again, key differences among
studies that may have contributed to inconsis- tent results include differences in TZD type,
PERSONALIZED MEDICINE IN DIABETES MANAGEMENT Elk and Iwuchukwu 1143
treatment duration, inclusion criteria, baseline metabolic conditions, and ethnicity. Another non-PPAR-c TZD target gene of inter-
est is the ADIPOQ gene that encodes adiponectin (Figure 5). Low levels of adiponectin have been associated with insulin resistance, and TZDs are known to increase adiponectin levels that may contribute in part to their insulin-sensitizing effects.
74 Common exonic (45T>G; rs16861194)
and intronic (276G>T; rs1501299) SNPs in ADI- POQ have been studied. Carriers of the 45G variant and the 276G wild-type alleles, whether studied alone or in haplotype combinations, had smaller decreases in fasting plasma glucose and HbA1c compared with patients carrying the other alleles.
52 In addition to studies probing
drug response, occurrence of the many troubling and limiting adverse effects with TZDs also led to genetic explorations of any associated mecha- nisms. One study reported an association between weight gain and a SNP in the perilipin gene.
77 A genetic substudy of the Diabetes
Reduction Assessment with Ramipril and Rosiglitazone Medication (DREAM) study showed a significant association with rosiglita- zone-induced edema and rs6123045 in the nuclear factor of the activated T-cells cytoplas- mic calcineurin-dependent 2 (NFATC2) gene in subjects of European descent.
78 Neither of these
studies have yet to be replicated. Other PD genes linked to TZDs have been studied without any significant replication reports.
Gliptins
Gliptins, also known as dipeptidyl peptidase IV (DPP4) inhibitors, are classified as incretin mimetics. They are newer agents that exert their effects by inhibiting a key enzyme in the incretin signaling pathway, preventing the degradation of endogenous glucagon-like peptide-1 (GLP-1) receptor agonist and gastric inhibitory peptide (GIP).
79 GLP-1 is secreted by the intestines after
intake of food and acts to induce insulin secre- tion from pancreatic b cells, suppresses the release of glucagon, inhibits gastric emptying, and reduces appetite. Inhibiting the degradation of GLP-1 leads to prolonged secretion of insulin and inhibition of glucagon.
79
Gliptins: Evidence of Pharmacogenetic Associations
In terms of their metabolic profile, DPP-4 inhibitors are not typical substrates for either cytochrome P450 (except saxagliptin, metabo- lized by CYP3A4/A5) or any major transporter families because their main mode of clearance is by renal excretion.
7 Due to the relatively recent
Figure 5. Schematic presentation of pharmacogenomics of thiazolidinediones. (A) Thiazolidinediones (TZDs) are substrates of OATP1B, the organic anion transporter protein encoded by SLCO1B1. OATP1B1 is the main hepatic transporter responsible for the active shuttling of organic anion compounds like TZDs into hepatocytes. (B) TZDs act by activating the PPAR-c receptor encoded by the PPAR-c gene; the activated receptor then binds to DNA response elements and controls the expression of genes involved in maintenance of metabolic homeostasis. (C) Adiponectin encoded by the ADIPOQ gene is a protein hormone expressed abundantly in adipose tissue where it promotes glucose uptake and fatty acid oxidation. Adiponectin actions are controlled by PPAR-c receptor activation.
1144 PHARMACOTHERAPY Volume 37, Number 9, 2017
entry of these drugs into the market, very few pharmacogenetic studies have been conducted.
80
Studies on genetic influence on insulin secretion after administration of human GLP-1 to healthy individuals may be extrapolated to explain observed differences in response to GLP-1 recep- tor agonists and DPP-4 inhibitors. A recent study showed a SNP, rs7202877, near the CTRB1 and CTRB2 loci influenced the response to DPP-4 inhibitor treatment in patients with T2DM.
81 The CRTB locus encodes for the diges-
tive enzyme chymotrypsin, an important regula- tor of the incretin pathway.
79 The G allele of
rs7202877 was previously associated with an increased risk for T1DM and a decreased risk for T2DM.
14 In one study, the rs7202877G allele
was associated with enhanced GLP-1 secretion in healthy individuals and reduced response to DPP-4 inhibitor treatment in patients with T2DM.
Personalized Treatment Strategy for DM
Managing patients with DM presents a chal- lenge. Treatment failure, risk of hypoglycemia, and selection of appropriate agents based on pathogenesis of the disease creates a need for using precision medicine in DM. With so many different classes of medications available, the guidelines recommend use of patient-specific approaches when designing DM treatment plans.
55 Clinicians should select a treatment
strategy and goals of therapy that provide the most benefit with the least harm. To address the question of efficacy among commonly used dia- betes medications in various patient populations, a national randomized interventional study was proposed.
82 The Glycemia Reduction Approaches
in Diabetes: A Comparative Effectiveness Study (GRADE) is currently enrolling adult patients with a history of DM for less than 10 years and currently on metformin monotherapy. Patients are divided into four groups to receive met- formin with either glimepiride, sitagliptin, liraglutide, or basal insulin glargine. The primary outcome measure is time to primary failure of treatment defined by a HbA1c higher than 7% over a period of 4–7 years. Investigators will compare personalized response to treatment, looking at adverse events, tolerability, microvas- cular and macrovascular complications, quality of life, and cost-effectiveness. Participants’ phe- notypes will be studied to help guide selection of optimal diabetic agent(s).
82 The results of
GRADE should provide additional guidance
when selecting a diabetic agent and using a per- sonalized approach in DM management. Despite the multiple factors that clinicians need to con- sider when selecting a treatment strategy for DM today, there is still room for incorporating phar- macogenomics for a more personalized manage- ment of DM (Figure 1).
Using Genetics for the Prediction of DM
Primary studies and meta-analyses of genetics studies on diabetes risk have yielded many genetic associations that increase our under- standing of T2DM pathophysiology. Although these explorations only explain a small fraction of genetic contribution to phenotype, the associ- ations can be used to improve disease classifica- tion. One example of this is with the use of genetic risk scores (GRS) to predict the risk of diabetes incidence. GRS scores are typically weighted by participants, and the number of risk alleles per SNP are summed over the total num- ber of representative SNPs associated with the risk of diabetes. The Framingham Heart Study (FHS) and the Diabetes Prevention Program (DPP) are representative cohorts that have suc- cessfully used this approach to predict diabetes risk.
83, 84 In the FHS, a GRS based on 18 risk
alleles predicted new cases of diabetes with a slightly better prediction of risk than common risk factors (such as body mass index, hyperten- sion, family history of diabetes, ethnicity, and age) alone.
84 In the DPP, a high GRS (based on
34 risk alleles) was associated with an increased risk of developing diabetes but was subsequently attenuated by lifestyle interventions in the ana- lytical model.
83 Although these examples pro-
vide an illustration of how pharmacogenomics can be incorporated into the risk prevention paradigm, the challenge of selecting the most relevant and clinically validated risk variants is one that is yet to be overcome.
Applications of Pharmacogenomics in Monogenic Diabetes Syndrome
For disease management, the most extensive application of pharmacogenetics in DM to date has been with the monogenic forms of the dis- ease. Identification of the correct MODY subtype has a tremendous impact on treatment. It is rec- ommended that patients presenting with diabetes in the first 6 months of life be tested for muta- tions in KCNJ11 and ABCC8 to rule out TNDM.
23 The identification of the NDM subtype
PERSONALIZED MEDICINE IN DIABETES MANAGEMENT Elk and Iwuchukwu 1145
is significant due to associated mutations occur- ring in the pancreatic b-cell KATP channels and their role in guiding treatment choice. SU action on these channels causes increased insulin secre- tion, making them the most ideal agents for NDM management. TNDM patients compared with those with PNDM are likely to have signifi- cantly higher endogenous insulin production, making them much more responsive to SUs.
23
They can thus be successfully treated with low- dose SUs in place of standard insulin therapy versus higher SU doses for PNDM. The switch from insulin to SU should only be made after a molecular genetic diagnosis. There are estab- lished translational research diagnosis and treat- ment approach guidance/protocols for MODY and NDM as well as a MODY probability calcu- lator for clinicians’ reference.
85
Patients with MODY 1 and 3 are prime targets for personalized therapy as embodied by the reported relatively safe transition from insulin to SU monotherapy in these patients. The finding that these patients are generally hypersensitive to SUs makes such a transition possible. This hypersensitivity to SUs has been attributed to downregulation of HNF1-a and HNF4-a target genes in the liver, leading to decreased hepatic uptake of SUs and increased levels in the general circulation.
80
Personalized treatment strategy selection may not always include pharmacotherapy, a paradigm seen with the appropriate identification of patients with MODY 2, a frequently misdiag- nosed subset.
8 These patients present with mild
hyperglycemia, which improves on its own, and generally do not progress to overt diabetes with associated macrovascular and microvascular complications. Consequently, patients with MODY 2 typically have no need for pharma- cotherapy and can be managed by diet alone.
8
Ideally, clinicians ought to be aware that patients with an onset of diabetes before age of 25 should be referred for genetic testing to rule out MODY and prevent misdiagnosis. The Inter- national Society for Pediatric and Adolescent Diabetes (ISPAD) guidelines provide very detailed recommendations about when to sus- pect the presence of MODY.
23 The American
Diabetes Association guidelines also provide rec- ommendations for the screening of MODY in children and adolescents initially diagnosed with DM within the first 6 months of life.
5 Genetic
tests for MODY-related genes and CLIA- approved labs for such testing are available on the Genetic Testing Registry.
86 The National
Monogenic Diabetes Registry of MODY patients is useful for further guidance on current research and treatment strategies.
87
Pharmacogenomics in Predicting Response to Therapy
With respect to the pharmacogenetics of OGLTs, metformin is still the guideline-based first-line monotherapy for diabetes. The variabil- ity in response and treatment failure with met- formin is clearly indicative of a differential effect that may be genetically induced. Variants in metformin PK and transporter genes have been the most widely studied and have greatly advanced our knowledge of metformin pharma- cogenomics. Replicated SLC transporter variants could play a role in guideline development for metformin use in select patient populations based on combined GRS. Pharmacogenomics- guided therapy based on a greater understanding of metformin genes and pathways could help clinicians if outcomes (patient response and probable treatment failures) could be determined a priori. Alternative guideline-based recommen- dations can then be used. This type of pharma- cogenomics-guided approach could lead to long-term dollar cost savings compared with a “wait until treatment fails then switch” approach. Response to SU therapy depends on b-cell
function and its insulin production ability. Results with clinical studies of the CYP2C9 gene and SU response have been promising as seen in the GoDARTs study. Variations in KCNJ11 and ABCC8 genes were also identified as associated with a high response to SU treatment in certain patients with hyperglycemia.
46 These critical
findings may potentially lead to testing/screening patients for these polymorphisms. Advance knowledge of these and other validated strong genetic signals, such as those in TCF7L2, may also assist in careful dose titration for SU-sensi- tive versus SU-tolerant patients. On a pharmaco- genetic basis, SU therapy can therefore be initiated as the most effective choice in “high- response” patients and avoided in “low/no- response” patients similar to the current success with some MODY patients. With the TZDs, although results from many
of the pharmacogenomic studies conducted were clinically significant, the current guideli- nes stipulate consideration of risk versus benefit before initiating use. It is doubtful there will be any additional benefit to incorporation of
1146 PHARMACOTHERAPY Volume 37, Number 9, 2017
pharmacogenomics based on their current role in treatment. It is our opinion that although there are cer-
tainly advances in genetic involvement and asso- ciations with OGLTs, there are as yet no strong clinical guidelines or recommendations for genetic testing and application in clinical practice.
Future Prospects of Using Genetics for Therapy of DM
Common risk factors for T2DM include obe- sity, dyslipidemia, family history, age, low physi- cal activity, Hispanic and African-American ethnicity, smoking, and unhealthy diet.
5 Compos-
ite GRS may better predict diabetes risk in combi- nation with other risk factors but would not be expected to be an efficient discriminator due to low effect sizes of single common variants in models. However, GRS may be useful in younger patients who have not developed obvious risk factors and may also prove useful for newborn diabetes susceptibility genetic screening and fol- low-up.
83, 84 Another utility for GRS would be in
ethnically focused screening and targeted treat- ment based on the presence or absence of known susceptibility genes because reports of popula- tion-specific T2DM susceptibility variants con- tinue to emerge from large consortia studies.
17
Genetics-based association studies in OGLTs and their role in DM treatment identified herita- ble components of a patient’s response to therapy. Although at this time not enough evidence for clinical utility exists, future larger cohorts and combinations from academic, practice, and indus- try DM research groups may aid in identifying more genes implicated in OGLT response. The large evidence base generated from these high- powered consortia studies may be an impetus for development of guidelines for the use of genetic information in diabetes pharmacotherapy. Use of pharmacogenetics for the prediction
and prevention of side effects from OGLTs may be an idea whose time is around the corner. Metformin and SUs, although mainline agents for therapy, are more prone to cause intolerant gastrointestinal effects or hypoglycemia com- pared with other agents. As such, testing patients a priori for any implicated clinically val- idated metformin or SU intolerance or low- response susceptibility variants could be an adjunct tool to maximizing their use in therapy. The same extrapolations could also be made for restricted OGLTs such as rosiglitazone. In the
future, knowing which patients are likely to develop severe side effects can help with strati- fied medicine.
Conclusion
With so many different classes of medications available for managing diabetes, the recommenda- tion from the guidelines is to use a patient-speci- fic approach and to select treatment strategies providing the most benefit with the least harm. Using personalized medicine in the treatment of diabetes where patients’ diabetes types can be precisely identified with respect to pathology fits perfectly with the selection of agents that will best treat specific pathologies of such stratified dia- betes. To make it even more patient specific, pharmacogenomics may be used to identify variations in medication disposition. Based on the evidence, attaining glycemic goal(s) is not enough to prevent diabetes-related complications. Con- trolling comorbidities and most importantly pre- vention of diabetes onset and progression needs to be addressed. Overall, data from pharmacogenomics-based
studies may help present a clearer picture of any implicated molecular mechanisms en route to selecting pharmacoeconomically viable and rele- vant treatment strategies for the complex multi- factorial disease that is DM.
References
1. World Health Organization. Diabetes: Key Facts, November 2016. Available from www.who.int/mediacentre/factsheets/ fs312/en/. Accessed February 13, 2017.
2. World Health Organization. Diabetes: Global report on dia- betes. Available from www.who.int/diabetes/global-report/en/. Accessed February 13, 2017.
3. Centers for Disease Control and Prevention. Chronic Disease Prevention and Health Promotion: Diabetes, July 2016. Avail- able from www.cdc.gov/chronicdisease/resources/publications/ aag/diabetes.htm. Accessed February 13 2017.
4. American Diabetes Association. Economic costs of diabetes in the U.S. in 2012. Diabetes Care 2013;36:1033–46.
5. American Diabetes Association. Standards of medical care in diabetes. Diabetes Care 2016;39:S1–106.
6. Barroso I. Genetics of type 2 diabetes. Diabet Med 2005;22: 517–35.
7. Hupfeld CJ, Courtney H, Olefsky JM. Type 2 diabetes melli- tus: etiology, pathogenesis, and natural history. In: Jameson JL, De Groot LJ, eds. Endocrinology: adult and pediatric. Philadel- phia, PA: Elsevier Health Sciences, 2010:765–87.
8. Carmody D, Naylor RN, Bell CD, et al. GCK-MODY in the US National Monogenic Diabetes Registry: frequently misdiag- nosed and unnecessarily treated. Acta Diabetol 2016;53:703–8.
9. Tan JT, Chia KS, Ku CS. The molecular genetics of type 2 dia- betes: past, present and future. In: Encyclopedia of life sciences. Chichester, UK: John Wiley & Sons, 2009. doi: 10. 1002/9780470015902.a0021994.
10. Denny JC, Ritchie MD, Basford MA, et al. PheWAS: demon- strating the feasibility of a phenome-wide scan to discover gene-disease associations. Bioinformatics 2010;26:1205–10.
PERSONALIZED MEDICINE IN DIABETES MANAGEMENT Elk and Iwuchukwu 1147
11. Brunetti A, Chiefari E, FTI D. Recent advances in the molecu- lar genetics of type 2 diabetes mellitus. World J Diabetes 2014;5:128–40.
12. Owen KR, McCarthy MI. Genetics of type 2 diabetes. Curr Opin Genet Dev 2007;17:239–44.
13. Doria A, Patti M-E, Kahn CR. The emerging genetic architec- ture of type 2 diabetes. Cell Metab 2008;8:186–200.
14. Morris AP, Voight BF, Teslovich TM, et al. Large-scale associ- ation analysis provides insights into the genetic architecture and pathophysiology of type 2 diabetes. Nat Genet 2012;44: 981–90.
15. The Wellcome Trust Case Control Consortium. Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature 2007;447:661–78.
16. Grant SF, Thorleifsson G, Reynisdottir I, et al. Variant of transcription factor 7-like 2 (TCF7L2) gene confers risk of type 2 diabetes. Nat Genet 2006;38:320–3.
17. Fuchsberger C, Flannick J, Teslovich TM, et al. The genetic architecture of type 2 diabetes. Nature 2016;536:41–7.
18. GWAS Catalog. The NHGRI-EBI catalog of published gen- ome-wide association studies, February 2017. Available from www.ebi.ac.uk/gwas/. Accessed February 13, 2017.
19. Bonnefond A, Froguel P. Rare and common genetic events in type 2 diabetes: what should biologists know? Cell Metab 2015;21:357–68.
20. Vaxillaire M, Froguel P. Genetic basis of maturity-onset dia- betes of the young. Endocrinol Metab Clin North Am 2006;35: 371–84.
21. Chambers C, Fouts A, Dong F, et al. Characteristics of matu- rity onset diabetes of the young in a large diabetes center. Pediatr Diabetes 2016;17:360–7.
22. Pihoker C, Gilliam LK, Ellard S, et al. Prevalence, characteris- tics and clinical diagnosis of maturity onset diabetes of the young due to mutations in HNF1A, HNF4A, and glucokinase: results from the SEARCH for Diabetes in Youth. J Clin Endo- crinol Metab 2013;98:4055–62.
23. Rubio-Cabezas O, Hattersley AT, Njølstad PR, et al. ISPAD Clinical Practice Consensus Guidelines 2014. The diagnosis and management of monogenic diabetes in children and ado- lescents. Pediatr Diabetes 2014;15(Suppl 20):47–64.
24. Babenko AP, Polak M, Cave H, et al. Activating mutations in the ABCC8 gene in neonatal diabetes mellitus. N Engl J Med 2006;355:456–66.
25. Gloyn AL, Pearson ER, Antcliff JF, et al. Activating mutations in the gene encoding the ATP-sensitive potassium-channel subunit Kir6.2 and permanent neonatal diabetes. N Engl J Med 2004;350:1838–49.
26. Colombo C, Delvecchio M, Zecchins C, et al. Transient neonatal diabetes mellitus is associated with a recurrent (R201H) KCNJ11 (KIR6.2) mutation. Diabetologia 2005;48: 2439–41.
27. Canadian Diabetes Association Clinical Practice Guidelines Expert Committee. Pharmacologic management of type 2 dia- betes: 2016 interim update. Can J Diabetes 2016;40:193–5.
28. Gong L, Goswami S, Giacomini KM, Altman RB, Klein TE. Metformin pathways: pharmacokinetics and pharmacodynam- ics. Pharmacogenet Genomics 2012;22:820–7.
29. Zhou G, Myers R, Li Y, et al. Role of AMP-activated protein kinase in mechanism of metformin action. J Clin Invest 2001;108:1167–74.
30. Pawlyk AC, Giacomini KM, McKeon C, Shuldiner AR, Florez JC. Metformin pharmacogenomics: current status and future directions. Diabetes 2014;63:2590–9.
31. Kahn SE, Haffner SM, Heise MA, et al. Glycemic durability of rosiglitazone, metformin, or glyburide monotherapy. N Engl J Med 2006;355:2427–43.
32. Riedel AA, Heien H, Wogen J, Plauschinat CA. Loss of glyce- mic control in patients with type 2 diabetes mellitus who were receiving initial metformin, sulfonylurea, or thiazolidinedione monotherapy. Pharmacotherapy 2007;27:1102–10.
33. Jablonski KA, McAteer JB, De Bakker PIW, et al. Common variants in 40 genes assessed for diabetes incidence and
response to metformin and lifestyle intervention in the dia- betes prevention program. Diabetes 2010;59:2672–81.
34. Shu Y, Sheardown SA, Brown C, et al. Effect of genetic varia- tion in the organic cation transporter 1 (OCT1) on metformin action. J Clin Invest 2007;117:1422–31.
35. Zhou K, Donnelly LA, Kimber CH, et al. Reduced-function SLC22A1 polymorphisms encoding organic cation transporter 1 and glycemic response to metformin: a GoDARTS study. Diabetes 2009;58:1434–9.
36. Becker ML, Pearson ER, Tk�a�c I. Pharmacogenetics of oral antidiabetic drugs. Int J Endocrinol 2013;2013: Article ID 686315. doi: 10.1155/2013/686315.
37. van Leeuwen N, Nijpels G, Becker ML, et al. A gene variant near ATM is significantly associated with metformin treatment response in type 2 diabetes: a replication and meta-analysis of five cohorts. Diabetologia 2012;55:1971–7.
38. Chen Y, Li S, Brown C, et al. Effect of genetic variation in the organic cation transporter 2 on the renal elimination of met- formin. Pharmacogenet Genomics 2009;19:497–504.
39. Zhou K, Bellenguez C, Spencer CC, et al. Common variants near ATM are associated with glycemic response to metformin in type 2 diabetes. Nat Genet 2011;43:117–20.
40. Florez JC, Barrett-Connor E, Jablonski KA, et al. The C allele of ATM rs11212617 does not associate with metformin response in the diabetes prevention program. Diabetes Care 2012;35:1864–7.
41. Zhou K, Donnelly L, Burch L, et al. Loss-of-function CYP2C9 variants improve therapeutic response to sulfonylureas in type 2 diabetes: a Go-DARTS study. Clin Pharmacol Ther 2010;87: 52–6.
42. Ragia G, Petridis I, Tavridou A, Christakidis D, Manolopou- los VG. Presence of CYP2C9*3 allele increases risk for hypo- glycemia in type 2 diabetic patients treated with sulfonylureas. Pharmacogenomics 2009;10:1781–7.
43. Holstein A, Plaschke A, Ptak M, et al. Association between CYP2C9 slow metabolizer genotypes and severe hypogly- caemia on medication with sulphonylurea hypoglycaemic agents. Br J Clin Pharmacol 2005;60:103–6.
44. Holstein A, Hahn M, Patzer O, Seeringer A, Kovacs P, Stingl J. Impact of clinical factors and CYP2C9 variants for the risk of severe sulfonylurea-induced hypoglycemia. Eur J Clin Phar- macol 2011;67:471–6.
45. Feng Y, Mao G, Ren X, et al. Ser 1369Ala variant in sulfony- lurea receptor gene ABCC8 Is associated with antidiabetic effi- cacy of gliclazide in Chinese type 2 diabetic patients. Diabetes Care 2008;31:1939–44.
46. Florez JC, Jablonski KA, Kahn SE, et al. Type 2 diabetes-asso- ciated missense polymorphisms KCNJ11 E23K and ABCC8 A1369S influence progression to diabetes and response to interventions in the Diabetes Prevention Program. Diabetes 2007;56:531–6.
47. Holstein JD, Kovacs P, Patzer O, Stumvoll M, Holstein A. The Ser1369Ala variant of ABCC8 and the risk for severe sul- fonylurea-induced hypoglycemia in German patients with type 2 diabetes. Pharmacogenomics 2012;13:5–10.
48. Pearson ER, Donnelly LA, Kimber C, et al. Variation in TCF7L2 influences therapeutic response to sulfonylureas: a GoDARTs study. Diabetes 2007;56:2178–82.
49. Holstein A, Hahn M, Korner A, Stumvoll M, Kovacs P. TCF7L2 and therapeutic response to sulfonylureas in patients with type 2 diabetes. BMC Med Genet 2011;12:30.
50. Kang ES, Park SY, Kim HJ, et al. Effects of Pro12Ala polymor- phism of peroxisome proliferator-activated receptor gamma2 gene on rosiglitazone response in type 2 diabetes. Clin Phar- macol Ther 2005;78:202–8.
51. Dawed AY, Donnelly L, Tavendale R, et al. CYP2C8 and SLCO1B1 variants and therapeutic response to thiazolidine- diones in patients with type 2 diabetes. Diabetes Care 2016;39:1902–8.
52. Kang ES, Park SY, Kim HJ, et al. The influence of adiponectin gene polymorphism on the rosiglitazone response in patients with type 2 diabetes. Diabetes Care 2005;28:1139–44.
1148 PHARMACOTHERAPY Volume 37, Number 9, 2017
53. Morris AD, Boyle DI, MacAlpine R, et al. The diabetes audit and research in Tayside Scotland (DARTS) study: electronic record linkage to create a diabetes register. DARTS/MEMO Collaboration. BMJ 1997;315:524–8.
54. Shu Y, Brown C, Castro RA, et al. Effect of genetic variation in the organic cation transporter 1, OCT1, on metformin phar- macokinetics. Clin Pharmacol Ther 2008;83:273–80.
55. American Diabetes Association. 7. Approaches to glycemic treatment. Diabetes Care 2016;39(Suppl 1):S52–9.
56. DeFronzo RA. Pharmacologic therapy for type 2 diabetes mel- litus. Ann Intern Med 1999;131:281–303.
57. Aquilante CL. Sulfonylurea pharmacogenomics in type 2 dia- betes: the influence of drug target and diabetes risk polymor- phisms. Expert Rev Cardiovasc Ther 2010;8:359–72.
58. Kirchheiner J, Roots I, Goldammer M, Rosenkranz B, Brock- moller J. Effect of genetic polymorphisms in cytochrome p450 (CYP) 2C9 and CYP2C8 on the pharmacokinetics of oral antidiabetic drugs: clinical relevance. Clin Pharmacokinet 2005;44:1209–25.
59. Group UKPDS. Intensive blood-glucose control with sulpho- nylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes. Lancet 1998;352:837–53.
60. Schopman J, Simon A, Hoefnagel S, Hoekstra J, Scholten R, Holleman F. The incidence of mild and severe hypoglycaemia in patients with type 2 diabetes mellitus treated with sulfony- lureas: a systematic review and meta-analysis. Diabetes Metab Res Rev 2014;30:11–22.
61. Li Q, Chen M, Zhang R, et al. KCNJ11 E23K variant is associ- ated with the therapeutic effect of sulphonylureas in Chinese type 2 diabetic patients. Clin Exp Pharmacol Physiol 2014;41:748–54.
62. Hamming KSC, Soliman D, Matemisz LC, et al. Coexpression of the type 2 diabetes susceptibility gene variants KCNJ11 E23K and ABCC8 S1369A alter the ATP and sulfonylurea sen- sitivities of the ATP-sensitive K+ channel. Diabetes 2009;58:2419–24.
63. Javorsky M, Klimcakova L, Schroner Z, et al. KCNJ11 gene E23K variant and therapeutic response to sulfonylureas. Eur J Intern Med 2012;23:245–9.
64. Sesti G, Laratta E, Cardellini M, et al. The E23K Variant of KCNJ11 encoding the pancreatic b-cell adenosine 50-tripho- sphate-sensitive potassium channel subunit Kir6.2 is associated with an increased risk of secondary failure to sulfonylurea in patients with type 2 diabetes. J Clin Endocrinol Metab 2006;91:2334–9.
65. Kimber CH, Doney AS, Pearson ER, et al. TCF7L2 in the Go- DARTS study: evidence for a gene dose effect on both diabetes susceptibility and control of glucose levels. Diabetologia 2007;50:1186–91.
66. Sesti G, Marini MA, Cardellini M, et al. The Arg972 variant in insulin receptor substrate-1 is associated with an increased risk of secondary failure to sulfonylurea in patients with type 2 diabetes. Diabetes Care 2004;27:1394–8.
67. Prudente S, Morini E, Lucchesi D, et al. IRS1 G972R missense polymorphism is associated with failure to oral antidiabetes drugs in white patients with type 2 diabetes from Italy. Dia- betes 2014;63:3135–40.
68. van Leeuwen N, Swen JJ, Guchelaar HJ, ’t Hart LM. The role of pharmacogenetics in drug disposition and response of oral glucose-lowering drugs. Clin Pharmacokinet 2013;52:833–54.
69. Della-Morte D, Palmirotta R, Rehni AK, et al. Pharmacoge- nomics and pharmacogenetics of thiazolidinediones: role in diabetes and cardiovascular risk factors. Pharmacogenomics 2014;15:2063–82.
70. Aquilante CL, Zhang W, McCollum M. Race, ethnicity, and use of thiazolidinediones among US adults with diabetes. Curr Med Res Opin 2007;23:489–94.
71. Aquilante CL, Bushman LR, Knutsen SD, Burt LE, Rome LC, Kosmiski LA. Influence of SLCO1B1 and CYP2C8 gene poly- morphisms on rosiglitazone pharmacokinetics in healthy vol- unteers. Hum Genomics 2008;3:7–16.
72. Kirchheiner J, Thomas S, Bauer S, et al. Pharmacokinetics and pharmacodynamics of rosiglitazone in relation to CYP2C8 genotype. Clin Pharmacol Ther 2006;80:657–67.
73. Gouda HN, Sagoo GS, Harding A-H, Yates J, Sandhu MS, Hig- gins JPT. The association between the peroxisome prolifera- tor-activated receptor-gamma2 (PPARG2) Pro12Ala gene variant and type 2 diabetes mellitus: a HuGE review and meta- analysis. Am J Epidemiol 2010;171:645–55.
74. Aquilante CL. Pharmacogenetics of thiazolidinedione therapy. Pharmacogenomics 2007;8:917–31.
75. Bluher M, Lubben G, Paschke R. Analysis of the relationship between the Pro12Ala variant in the PPAR-gamma2 gene and the response rate to therapy with pioglitazone in patients with type 2 diabetes. Diabetes Care 2003;26:825–31.
76. Hsieh MC, Lin KD, Tien KJ, et al. Common polymorphisms of the peroxisome proliferator-activated receptor-gamma (Pro12Ala) and peroxisome proliferator-activated receptor- gamma coactivator-1 (Gly482Ser) and the response to pioglita- zone in Chinese patients with type 2 diabetes mellitus. Meta- bolism 2010;59:1139–44.
77. Kang ES, Cha BS, Kim HJ, et al. The 11482G>A polymor- phism in the perilipin gene is associated with weight gain with rosiglitazone treatment in type 2 diabetes. Diabetes Care 2006;29:1320–4.
78. Bailey SD, Xie C, Do R, et al. Variation at the NFATC2 locus increases the risk of thiazolidinedione-induced edema in the Diabetes REduction Assessment with ramipril and rosiglita- zone Medication (DREAM) study. Diabetes Care 2010;33: 2250–3.
79. Drucker DJ, Nauck MA. The incretin system: glucagon-like peptide-1 receptor agonists and dipeptidyl peptidase-4 inhibi- tors in type 2 diabetes. Lancet 2006;368:1696–705.
80. Kleinberger JW, Pollin TI. Personalized medicine in diabetes mellitus: current opportunities and future prospects. Ann N Y Acad Sci 2015;1346:45–56.
81. ’t Hart LM, Fritsche A, Nijpels G, et al. The CTRB1/2 locus affects diabetes susceptibility and treatment via the incretin pathway. Diabetes 2013;62:3275–81.
82. Nathan DM, Buse JB, Kahn SE, et al. Rationale and design of the glycemia reduction approaches in diabetes: a compara- tive effectiveness study (GRADE). Diabetes Care 2013;36: 2254–61.
83. Hivert M-F, Jablonski KA, Perreault L, et al. Updated genetic score based on 34 confirmed type 2 diabetes loci is associated with diabetes incidence and regression to normoglycemia in the diabetes prevention program. Diabetes 2011;60:1340–8.
84. Meigs JB, Shrader P, Sullivan LM, et al. Genotype score in addition to common risk factors for prediction of type 2 dia- betes. N Engl J Med 2008;359:2208–19.
85. Diabetes Genes. Genetic types of diabetes including maturity- onset diabetes of the young (MODY). Available from www.di abetesgenes.org. Accessed February 13, 2017.
86. Genetic Testing Registry. Available from www.ncbi.nlm.nih. gov/gtr/. Accessed February 13, 2017.
87. Monogenic Diabetes Registry. Available from http://monogenic diabetes.uchicago.edu/mody-registry/. Accessed February 13, 2017.
PERSONALIZED MEDICINE IN DIABETES MANAGEMENT Elk and Iwuchukwu 1149
