Efficacy of GLP-1 Receptor Agonists and DPP-4 Inhibitors: Meta-Analysis and Systematic Review

Clinical Therapeutics/Volume 34, Number 6, 2012
Efficacy of GLP-1 Receptor Agonists and DPP-4 Inhibitors:

Meta-Analysis and Systematic Review
Vanita R. Aroda, MD1; Robert R. Henry, MD2; Jenny Han, MS3; Wenying Huang, PhD3;
Mary Beth DeYoung, PhD3; Tamara Darsow, PhD3; and Byron J. Hoogwerf, MD4
1
Department of Endocrinology, Diabetes, and Metabolism, MedStar Health Research Institute, Hyattsville,
Maryland; 2Division of Endocrinology and Metabolism, University of California, San Diego, California;
3
Amylin Pharmaceuticals, Inc, San Diego, California; and 4Eli Lilly and Company, Indianapolis, Indiana
ABSTRACT
Background: Considerable clinical data on the
treatment of type 2 diabetes with incretin-based therapies (glucagon-like peptide 1 receptor agonists [GLP1RAs] and dipeptidyl-peptidase IV [DPP-4] inhibitors)
are available.
Objective: This meta-analysis was performed to
support the understanding of the overall evidence by
summarizing the findings from studies of the incretinbased therapies.
Methods: The MEDLINE, EMBASE, BIOSIS, and
BIOSIS trial databases were searched for relevant literature published between January 1, 1990, and June 30,
2011. Search terms included GLP-1, DPP-4, the
names of drugs that have been approved by the US
Food and Drug Administration for the treatment of
diabetes, and the names of drugs that have not been
approved but are in late-stage research. Studies were
included if they were randomized controlled trials of
12 to 52 weeks duration and having change from
baseline in hemoglobin (Hb) A1c as the primary end
point. The random effects meta-analyses models examined HbA1c, fasting plasma glucose (FPG), and body
weight for individual therapies, but did not compare
effects between therapies.
Results: The reviewers identified 362 unique clinical
studies, of which 80 were eligible for inclusion in the
present meta-analysis. Mean baseline HbA1c values
ranged from 7.4% to 10.3% (GLP-1RA studies) and
7.2% to 9.3% (DPP-4 inhibitor studies). The highest
maintenance doses of the GLP-1RAs and the DPP-4
inhibitors were associated with changes from baseline
in mean HbA1c of 1.1% to 1.6% and 0.6% to
1.1%, respectively. Mean reductions in FPG with
exenatide once weekly (QW) or liraglutide once daily
were apparently greater than those with exenatide
twice daily (BID) and the DPP-4 inhibitors, with the
June 2012
exception of vildagliptin. Mean weight losses with the

GLP-1RAs and the DPP-4 inhibitors were 2.0 and
0.2 to 0.6 kg, respectively. The limitations of the
present analysis included a lack of adjustment for placebo use and interstudy heterogeneity associated with
differences in methodology (eg, management of concurrent medications, blinding, criteria for treatment
discontinuation).
Conclusions: All of the incretin-based therapies in the
present meta-analysis were associated with significant reductions from baseline in HbA1c and FPG. Further direct
comparative studies between the GLP-1RAs and the
DPP-4 inhibitors and within the GLP-1RA class are
justified. (Clin Ther. 2012;34:12471258) 2012 Elsevier
HS Journals, Inc. All rights reserved.
Key words: body weight, DPP-4 inhibitors, fasting
plasma glucose, GLP-1, hemoglobin A1c, incretin mimetics, meta-analysis, systematic review.
INTRODUCTION
Multiple. therapies that affect glucose homeostasis via
glucagon-like peptide (GLP)-1 signaling pathways are
now available for the treatment of type 2 diabetes mellitus.1 These incretin-based therapies are classified as
GLP-1 receptor agonists (GLP-1RAs) or dipeptidylpeptidase IV (DPP-4) inhibitors. GLP-1RAs adminisEarlier versions of the analyses in this article were presented in abstract
and poster format at the 46th Meeting of the European Association for
the Study of Diabetes; Stockholm, Sweden; September 20 24, 2010;
the abstract was published in Diabetologia. 2010;52(Suppl 1):S333. Abstract: 836.
*Current affiliation: American Diabetes Association, Alexandria,
Virginia.
Accepted for publication April 16, 2012.
http://dx.doi.org/10.1016/j.clinthera.2012.04.013
0149-2918/$ - see front matter
2012 Elsevier HS Journals, Inc. All rights reserved.
1247
Clinical Therapeutics
tered at therapeutic concentrations directly activate the
GLP-1R, whereas DPP-4 inhibitors slow GLP-1 degradation, thereby increasing endogenous GLP-1 concentrations that activate GLP-1R.2,3 The potential glycemic
benefits of these treatments include glucose-dependent
stimulation of insulin production and secretion; suppression of inappropriate glucagon secretion; slowing
of gastric emptying, which reduces the rate of glucose
appearance in the circulation; and satiety, which may
reduce food intake.3 The latter 2 mechanisms may be
responsible for the modest weight loss associated with
incretin-based therapies.3
There are key structural differences between various
GLP-1RAs and DPP-4 inhibitors. Amino acid backbones in GLP-1RAs resemble human GLP-1 (eg, liraglutide4) or exendin-4 (eg, exenatide), which shares
53% amino acid sequence identity with human GLP1.3,5 Liraglutide contains a fatty acid to improve in
vivo stability.4 The half-lives of GLP-1RAs range from
2.4 hours (exenatide6) to 13 hours (liraglutide).4 Continuous release of once-weekly exenatide was achieved
by embedding exenatide in biodegradable microspheres that release exenatide over 10 weeks.6 Among
the available GLP-1RAs, differences in structure, halflife, and plasma excursion of the active molecule may
affect potential efficacy. Small-molecule DPP-4 inhibitors also differ in molecular structure and mechanism:
alogliptin, linagliptin, and sitagliptin are in the xanthine class that forms a noncovalent bond with DPP-4,
whereas saxagliptin and vildagliptin are cyanopyrrolidines that form a covalent bond with the active site serine.7 Nonetheless, all DPP-4 inhibitors reduce the degradation of GLP-1.7 Because DPP-4 inhibitors have been
associated with increased endogenous concentrations of
human GLP-1, they may exert similar physiologic effects.
The comprehensive clinical trial programs for the
development of GLP-1RAs and DPP-4 inhibitors provide extensive data on efficacy to inform evidencebased clinical decisions. The volume of available data
is difficult to synthesize and makes differences in methodology difficult to assess, yet affords an opportunity
for a systematic review with the goal of providing physicians with an overview of the available clinical evidence. Evidence-based reviews are valuable for organizing large amounts of published data, identifying
trends in treatment efficacy, and assisting in the categorization of therapies (eg, as a metaclass [incretinbased therapies], as 2 classes [GLP-1RAs and DPP-4
inhibitors], or as 2 classes based on duration of ac-
1248
tion [short-acting GLP-1RAs, long-acting GLP-1RAs,

and DPP-4 inhibitors]).
METHODS
Data Sources and Literature Search
The MEDLINE, EMBASE, BIOSIS, and BIOSIS trials databases were searched for English-language articles published from January 1, 1990, to June 30, 2011.
Unpublished studies were excluded. Prespecified
search terms were dipeptidyl peptidase-4 inhibitor, dipeptidyl peptidase-IV inhibitor, DPP-4 inhibitor,
DPP4 inhibitor, DPP-IV inhibitor, exenatide, exendin,
byetta, liraglutide, victoza, taspoglutide, albiglutide,
GLP-1 receptor agonist, GLP-1 mimetic, GLP-1 analogue, GLP-1 analog, glucagon-like peptide-1 receptor
agonist, glucagon-like peptide-1 mimetic, glucagonlike peptide-1 analogue, glucagon-like peptide-1 analog, sitagliptin, januvia, alogliptin, linagliptin, tradjenta, vildagliptin, saxagliptin, incretin mimetic,
AVE0010 (now referred to as lixisenatide), and albugon. Titles, abstracts, study drug names, and subject
headings were searched. A search of these terms was
also conducted in the 2011 abstract databases from the
American Diabetes Association and the European Association for the Study of Diabetes. All of the records
identified were stored in an electronic database (Procite
version 5.0.3, ISI ResearchSoft, Carlsbad, California),
which identified duplicate citations and allowed citations to be searched by key word and grouped for
analyses.
Study Selection
Studies included in these analyses were randomized
controlled trials (RCTs) of 12 weeks duration that
included 10 patients with type 2 diabetes per treatment arm, reported change in hemoglobin (Hb) A1c as
the primary endpoint, and studied the effects of adding
a single drug (not multiple therapies) in a representative population. Therapies were included if Phase III
data were available as of June 30, 2011, and the highest effective maintenance dose was known. The analysis focused on therapies or therapeutic combinations
approved for use in the United States or the European
Union or that were in postPhase III development at
the time of the search.
Publications matching the prespecified criteria were
identified stepwise. After the exclusion of duplicate citations, the list was limited to RCTs in patients with
type 2 diabetes. The following article types were ex-
Volume 34 Number 6
V.R. Aroda et al.

cluded: reviews, letters, opinions, or treatment guidelines; abstracts published before January 2011 or duplicate abstracts; experimental non-human studies;
studies in type 1 diabetes; studies in obesity in the absence of diabetes; studies of drug mechanism of action,
pharmacokinetics, or pharmacodynamics properties.
Potential references were screened for key words indicating study design (eg, case, crossover, interim, extension, meta, pooled, model, simulation, real-world, retrospective, subgroup) by a reviewer (M.B.D.), with
verification of the sorted reference citations by another
reviewer (T.D.). Studies were eliminated if no Phase III
data were available, only a nonapproved administration method or dose was studied, the protocol was
unclear, or data were missing. Because of the dependence of efficacy on dose, data from treatment arms
were excluded if a dose was not approved or was not
included in Phase III trials. Doses selected are specified in Supplemental Appendix Table I in the online
version at http://dx.doi.org/10.1016/j.clinthera.
2012.04.013. The remaining studies were reviewed independently by 3 individuals (M.B.D., T.D., and Y.P.),
and ambiguous decisions were resolved by consultation with a fourth reviewer (J.R.; see Acknowledgments). The final study list was assessed against published reviews and meta-analyses to confirm
completeness.1,8 Analyses of therapies within each
class are reported in alphabetical order for consistency and fair balance.
tal Appendix Table III in the online version at http://

dx.doi.org/10.1016/j.clinthera.2012.04.013). No
adjustments were made for these parameters.
Data Extraction
RESULTS
Search Results and Study Characteristics
Mean baseline characteristics and demographic

data collected included duration of diabetes, age,
HbA1c, fasting plasma glucose (FPG), and body
weight (see Supplemental Appendix Table II in the
online version at http://dx.doi.org/10.1016/j.clinthera.
2012.04.013). In addition, mean and/or least-squares
(LS) mean (95% CI) changes from baseline to study
endpoint for HbA1c, FPG, and weight were extracted
from each treatment arm but were not imputed if
data were missing. Rates of adverse events were not
extracted. Data entered into the statistical model
were checked for accuracy against the original references by 4 individuals (M.B.D., T.D., Y.P., and
C.V.R. [see Acknowledgements]).
To assess study quality, studies were reviewed for
parameters such as discontinuation rate, medication
changes prior to baseline assessment, baseline differences, blinding, and analysis methods (see Supplemen-
June 2012
Statistical Analysis
Because the studies included varying dosing regimens, 2 types of analysis were conducted. One included the highest maintenance dose approved or
tested, and 1 included all doses approved for use or
tested in a Phase III trial (including titration doses with
lesser efficacy). In both analyses, weighted mean (SE)
differences from baseline were derived from multiple
treatment arms per study and then per drug. Missing
SDs or SEs were imputed from the pooled, nonmissing
SDs or SEs within studies of the same drug.9 Models
tested included fixed-effects, random-effects, and
Bayesian random-effects meta-regression models. The
extent of heterogeneity determined whether a fixedeffects or random-effects model was used in the analysis. Fixed-effects and random-effects models were used
with low and high levels of heterogeneity, respectively.
The Bayesian meta-regression model was used for determining whether differences in baseline HbA1c significantly affected the magnitude of calculated changes in
HbA1c. Mean baseline HbA1c was assessed as a modulator in meta-regression models. Study heterogeneity
(I2) was determined. Publication bias was assessed by
funnel plots, with the precision (1/SE) plotted against
the effect size.
From the initial database search, 362 potentially relevant publications were identified (Figure 1). For individual characteristics of the 80 publications selected
for final analysis, see Supplemental Appendix Table I
in the online version at http://dx.doi.org/10.1016/
j.clinthera.2012.04.013.
The majority of studies in these analyses (73/80)
were conducted during Phase III development, and
98% (78/80) were published in 2005 or later. Sixtyfive percent (52/80) were 24 to 30 weeks in duration,
9% (7/80) were 6 months in duration, and 85%
(68/80) included 90 patients per arm. In the majority of the studies (76% [61/80]), oral glucoselowering therapies were administered in combination with the agent of interest. Mean baseline HbA1c
values generally ranged between 8.0% and 8.6%,
and rates of discontinuation of active therapies were
1249
Relevant papers
identified (n = 362)
Excluded based on study design (n = 137)

Meta- or pooled analysis (n = 41)
Clinical practice or retrospective studies (n = 31)
Extension studies (n = 24)
Case studies (n = 12)
Subgroup analysis (n = 14)
Model or simulation (n = 7)
Interim analysis (n = 5)
Crossover studies (n = 3)
Studies identified for

detailed review (n = 225)
Studies identified for

qualititative analysis (n = 80)
GLP-1RAs (n = 33)
Exenatide BID (n = 15)
Exenatide QW (n = 7)
Liraglutide
once daily (n = 11)
Excluded based on details of study* (n = 145)

Primary endpoint not HbA1c (n = 50)
Nonapproved administration method or dose (n = 31)
Studied in a special patient population (n = 19)
Multiple therapies added simultaneously (n = 11)
Duration <12 wks (n = 8)
Incomplete information for analysis (n = 8)
No Phase III data available (n = 8)
Post-hoc analysis (n = 8)
Single arm study (n = 2)
DPP-4 inhibitors (n = 47)

Alogliptin (n = 5)
Linagliptin (n = 9)
Saxagliptin (n = 7)
Sitagliptin (n = 20)
Vildagliptin (n = 6)
Excluded arms from trials (n = 22)
Arms with nonapproved doses (n = 18)
GLP-1RA (n = 4)
DPP-4 inhibitors (n = 14)
Arms with nonapproved combination therapy (n = 4)
Arms from 80 studies identified
for analysis of all doses
Excluded because they did not test the
highest maintenance dose of a therapy (n = 3)
Arms from 77 studies identified
for analysis of highest maintenance dose
Figure 1. Study flow. HbA1c hemoglobin A1c. *Studies may have been excluded for 1 reason.
similar (see Supplemental Appendix Table II in the online version at http://dx.doi.org/10.1016/j.clinthera.

2012.04.013). Key differences among the clinical
trial programs for individual drugs included number
of studies, number of patients per trial, number of
doses tested, proportion of double-blinded trials,
and proportion of trials preceded by discontinuation
1250
of prior oral glucose-lowering therapies (run-in versus add-on design) (Table). A high level of study
heterogeneity was identified among studies for each
incretin-based treatment assessed (see Supplemental
Appendix Table IV in the online version at http://
dx.doi.org/10.1016/j.clinthera.2012.04.013), suggesting
that a random-effects model would accurately describe
Volume 34 Number 6
June 2012
Table. Selected characteristics of the studies for individual GLP-1R agonists and DPP-4 inhibitors included in the meta-analysis.
Class/Drug
GLP-1 receptor
agonists
Exenatide BID
Exenatide QW
Liraglutide
once daily
No. of Total ITT Active-Treatment

Studiesa Population ITT Populationb
Doses Studied
Doses Not Studied
Discontinuations
of Active
Prestudy
Background
Treatment,
Medication
Medication,c
Range, % of
Change,d
% of Patients
Patients
% of Patients
19
7
11
7328
2210
6533
4038
842
4178
5 and 10 ge BID
2 mg QWe
0.6, 0.9, 1.2, and
1.8 mge once
daily
2.5 g BID
0.8 mg QW
0.045, 0.1, 0.225, 0.3,
0.45, and 0.75 mg
once daily
94.7
85.7
72.7
9.729.8
021.0
4.535.5
0
0
54.5
DPP-4 inhibitors
Alogliptin
2503
1976
60.0
6.737.4
20.0
Linagliptin
Saxagliptin
9
7
5177
3187
3221
1566
66.7
57.1
5.815.2
9.435.8
77.8
14.3
Sitagliptin
23
10,893f
5274f
60.9
2.756.9
60.9
Vildagliptin
3976
2497
12.5 and 25 mge

once daily
5 mg once dailye 1 and 10 mg once daily
5 mg once dailye 2.5, 10, 20, 40, and
100 mg once daily
100 mg once
5, 12.5, 25, 50 mg
dailye
BID; 25, 50, 200 mg
once daily
50 mg once
10, 25 mg BID; 25,
daily,g 50 mg
100 mg once daily
BIDeh
83.3
11.221.5
V.R. Aroda et al.
1251
BID twice daily; DPP-4 dipeptidyl-peptidase IV; GLP-1 glucagon-like peptide 1; QW once weekly.
a
Some studies evaluated multiple drugs, and so were included in the overall for each drug studied; therefore, the total count of studies exceeds 80.
b
All GLP-1RA or DPP-4 inhibitor groups included in the analysis.
c
Concurrent glucose-lowering therapy.
d
Included discontinuation of oral glucose-lowering therapy and discontinuation with transition to a different glucose-lowering therapy or therapies (see Supplemental
Appendix Table III in the online version at http://dx.doi.org/10.1016/j.clinthera.2012.04.013).
e
Highest maintenance dose.
f
Data unavailable in 3 studies.
g
Not as monotherapy, only with a sulfonylurea.
h
Not as monotherapy, only with metformin or a thiazolidinedione.
the data. Funnel plots assessing the precision of the data
suggested an even distribution of mean values for the
parameters studied (data not shown), interpreted as minimal publication bias.
Efficacy Findings
On analysis of the pooled data, each incretin-based
therapy was associated with significant mean reductions from baseline in HbA1c. For the highest maintenance dose studied (see Supplemental Appendix Table
I in the online version at http://dx.doi.org/10.1016/
j.clinthera.2012.04.013), the mean changes from baseline in HbA1c were apparently greater with the longacting GLP-1RAs than with the DPP-4 inhibitors
(Figure 2A and see Supplemental Figures 1 to 4 in
the online version at http://dx.doi.org/10.1016/
j.clinthera.2012.04.013). This apparent difference was
also reported after adjustment for differences in baseline HbA1c (on Bayesian analysis, mean [95% CI]
changes in HbA1c: exenatide BID, 1.08 [1.22 to
0.94]; exenatide QW, 1.54 [1.73 to 1.36]; liraglutide once daily, 1.22 [1.39 to 1.05]; alogliptin,
0.70 [0.90 to 0.50]; linagliptin, 0.60 [0.80 to
0.40]; saxagliptin, 0.71 [0.89 to 0.54]; sitagliptin, 0.70 [0.78 to 0.63]; and vildagliptin, 0.98
[1.46 to 0.52]). A difference in mean HbA1c reductions between the GLP-1RAs and the DPP-4 inhibitors
was also apparent at all doses approved or in late-stage
development (see Supplemental Appendix Table I in
the online version at http://dx.doi.org/10.1016/
j.clinthera.2012.04.013) (exenatide BID, 1.05
[1.19 to 0.92]; exenatide QW, 1.59 [1.70 to
1.48]; and liraglutide once daily, 1.21 [1.35 to
1.06] versus alogliptin, 0.65 [0.80 to 0.50];
linagliptin, 0.61 [0.75 to 0.46]; saxagliptin,
0.68 [0.78 to 0.57]; sitagliptin, 0.67 [0.75 to
0.60]; and vildagliptin, 0.98 [1.37 to 0.59]).
On pooled analysis, FPG was significantly reduced
from baseline with all of the GLP-1RAs and DPP-4
inhibitors studied. Among the GLP-1RAs, the mean
changes in FPG were numerically greater with the longacting GLP-1RAs than with exenatide BID (Figure 2B).
The mean changes in FPG with the DPP-4 inhibitors
were 0.97, 1.04, 0.73, 0.87, and 1.57
(alogliptin, linagliptin, saxagliptin, sitagliptin, and
vildagliptin, respectively) with vildagliptin demonstrating a broad range of FPG reductions. Similar or
lesser changes in FPG were observed in the all-labeleddose analyses (mean [95% CI] changes in FPG [mmol/
1252
L]: exenatide BID, 1.12 [1.31 to 0.92]; exenatide

QW, 2.12 [2.28 to 1.96]; liraglutide once daily,
1.71 [2.00 to 1.43]; alogliptin, 0.92 [1.24 to
0.59]; linagliptin, 1.04 [1.59 to 0.49]; saxagliptin, 0.73 [0.95 to 0.50]; sitagliptin, 0.87
[0.98, 0.77]; and vildagliptin, 1.32 [2.01 to
0.62]).
Weight Effects
Effects on body weight differed among the GLP1RAs and the DPP-4 inhibitors in the pooled analysis
(Figure 2C). Short- and long-acting GLP-1RAs were
associated with significant reductions from baseline in
body weight, whereas DPP-4 inhibitors only were associated with a trend toward weight loss. Similar findings were obtained when the effects of each dose on
body weight were studied (mean [95% CI] changes in
weight [kg]: exenatide BID, 1.94 [2.35 to 1.53];
exenatide QW, 2.41 [2.83 to 1.99]; liraglutide
once daily, 1.66 [2.43 to 0.88]; alogliptin, 0.27
[0.87 to 0.34]; saxagliptin, 0.64 [1.11 to
0.16]; sitagliptin, 0.29 [0.61 to 0.03]; and
vildagliptin, 0.21 [0.84 to 0.42]).
DISCUSSION
The findings from the present meta-analysis suggest
potential differences between the GLP-1RAs and the
DPP-4 inhibitors in terms of clinical response for glycemia and weight. With the exception of vildagliptin,
all of the DPP-4 inhibitors appeared to have been associated with similar mean decreases in HbA1c, FPG,
and body weight, with overlapping 95% CIs, across all
of the clinical trials assessed. In contrast, the responses
in mean HbA1c within the GLP-1RA class appeared to
differ between therapies, with overlapping 95% CIs
between exenatide BID and liraglutide once daily only.
With respect to mean responses in FPG in the GLP1RA class, the 95% CIs overlapped only with exenatide QW and liraglutide once daily. In contrast,
the mean responses in body weight appeared to have
been similar with all of the GLP-1RAs across trials.
The pattern of response for glycemic end points appeared to differ between GLP-1RAs, and the DPP-4
inhibitors seem to have had greater uniformity of
response.
The reasons that the patterns of overall outcomes
for HbA1c, FPG, and weight appeared different between the 2 drug classes are a matter of speculation.
The responses across all DPP-4 inhibitors may have
Volume 34 Number 6
V.R. Aroda et al.
Mean HbA1c Difference (95% CI)

Exenatide BID
1.10 [1.22 to 0.99]
Exenatide QW
1.59 [1.70 to 1.48]
Liraglutide
1.27 [1.41 to 1.13]
Alogliptin
0.69 [0.85 to 0.54]
Linagliptin
0.60 [0.75 to 0.46]
Saxagliptin
0.68 [0.78 to 0.57]
Sitagliptin
0.67 [0.75 to 0.60]
Vildagliptin
1.06 [1.48 to 0.64]
2.0
1.5
1.0
HbA1c Change (%)
0.5
Mean FPG Difference (95% CI)

Exenatide BID
1.16 [1.35 to 0.97]
Exenatide QW
2.12 [2.28 to 1.96]
Liraglutide
1.82 [2.07 to 1.57]
Alogliptin
0.97 [1.27 to 0.67]
Linagliptin
1.04 [1.59 to 0.49]
Saxagliptin
0.73 [0.95 to 0.50]
Sitagliptin
0.87 [0.98 to 0.77]
Vildagliptin
1.57 [2.23 to 0.90]
2.5
2.0
1.5
1.0
FPG Change (mmol/L)
0.5
Mean Weight Difference (95% CI)

Exenatide BID
2.03 [2.46 to 1.60]
Exenatide QW
2.41 [2.83 to 1.99]
Liraglutide
2.29 [2.99 to 1.59]
Alogliptin
0.30 [0.90 to +0.30]
Saxagliptin
0.64 [1.11 to 0.16]
Sitagliptin
0.29 [0.61 to +0.03]
Vildagliptin
0.16 [0.92 to +0.60]
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0.5
Weight Change (kg)
Figure 2. Overall mean changes from baseline in (A) hemoglobin A1c (HbA1c), (B) fasting plasma glucose (FPG),
and (C) weight with the use of glucagon-like peptide 1 receptor agonists (GLP-1RAs) or dipeptidylpeptidase IV (DPP-4) inhibitors at the highest maintenance doses evaluated in this meta-analysis and
systemic review of the efficacy of incretin-based therapies in type 2 diabetes mellitus.
been uniform because these agents have similar effects

on endogenous GLP-1 concentrations. In contrast, the
variability in mean HbA1c and FPG between the GLP1RAs may have been related to known differences in
the structure, dose, and pharmacokinetic properties
June 2012
between each of the GLP-1RAs. Additional study is

needed to support or refute these hypotheses.
Direct comparative studies have evaluated the potential differences in clinical outcomes among agents
between the 2 classes or among agents within the same
1253
class and are notable. In a Phase IIIb, comparative,
head-to-head, 18-week study in 800 patients treated
with either saxagliptin or sitagliptin, similar reductions
in HbA1c (treatment difference, 0.09%) and FPG
(treatment difference, 0.30 mmol/L) were observed.10
No significant differences in proinsulin, C-peptide, or
HOMA-B measurements were noted between these 2
agents. Additional comparative studies of sitagliptin or
saxagliptin versus vildagliptin may be warranted.
Greater reductions in HbA1c and FPG were found with
vildagliptin compared with the other DPP-4 inhibitors,
with marked variability introduced primarily by a single trial.11
Several RCTs directly compared the efficacy of a
GLP-1RA and an DPP-4 inhibitor using endpoints related and not related to glycemia. A randomized, parallel-group trial compared treatment with liraglutide
1.2 or 1.8 mg once daily to sitagliptin 100 mg once
daily over 26 weeks and reported findings on differences in HbA1c, FPG, and weight consistent with those
estimated in the present analysis. In that direct comparison of liraglutide and sitagliptin, significant differences in HbA1c (0.6%), FPG (1.3%), and weight
(2.4 kg) between highest maintenance doses were reported.12 Furthermore, differences in -cell function
between a GLP-1RA and an DPP-4 inhibitor were reported in that study, in a 4-week crossover study, and
in an 8-week crossover study.1214 The first study reported significantly greater fasting C-peptide concentration, proinsulin:insulin ratio, and HOMA-B (but
not HOMA-IR) with liraglutide than with sitagliptin.12 The 4-week crossover comparison between exenatide BID and sitagliptin (N 61) reported a significant 1.5-fold higher index of insulin secretion and 0.9fold lower secretion of postprandial glucagon in
patients treated for 2 weeks with exenatide BID compared with sitagliptin.14 A difference in gastric emptying between the 2 therapies was noted, with exenatide
slowing gastric emptying by 44% compared with sitagliptin. Similar findings were reported in the 8-week
crossover study, in which exenatide was associated
with greater effects on mean 24-hour glucose, postprandial glucose, postprandial glucagon, HOMA-B,
and caloric intake compared with sitagliptin.13 The
4-week crossover study reported a 4-fold difference in
the molar plasma concentration of GLP-1 or GLP-1RA
with exenatide BID versus sitagliptin (63.8 vs 15.1
pmol/L, respectively), which is a potential explanation
for the differences in efficacy.14
1254
Differences within the GLP-1RA class appear to be

related to the timing, duration, and half-life of therapy.
The decreases in HbA1c and FPG were greater with
exenatide 2.0 mg QW and liraglutide 1.8 mg once daily
compared with those with exenatide BID in direct comparative studies,15,16 presumably reflecting the continuous steady-state level of GLP-1 activity achieved with
these agents, whereas short-acting exenatide BID was
more efficacious in regulating postprandial glucose
compared with either exenatide QW or liraglutide
once daily, reflecting its premeal dosing, short half-life,
and more rapid peak effect. Differentiation between
exenatide 2.0 mg QW and liraglutide 1.8 mg once daily
was examined in a 26-week, randomized, open-label,
parallel-group noninferiority clinical study in 911 patients with type 2 diabetes. That trial, published in
abstract format after June 30, 2011, reported greater
HbA1c lowering (0.2%) and weight loss with once
daily liraglutide injections than with weekly exenatide
injections.17
A network meta-analysis estimated the comparative
efficacy of exenatide QW with that of liraglutide 1.2
mg once daily, the dosage recommended by National
Institute for Health and Clinical Excellence in the
United Kingdom.18 In that analysis, the differences in
HbA1c between exenatide QW and liraglutide once
daily were calculated based on differences in HbA1c
with the common comparators insulin glargine and
exenatide BID. The estimated mean differences in
HbA1c between exenatide QW and liraglutide 1.2 and
1.8 mg once daily were 0.14% and 0.03%, respectively. The investigators of that analysis concluded that
exenatide QW and both dosages of liraglutide were
similarly efficacious in lowering HbA1c.
Other published meta-analyses, including analyses
by Amori et al,1 Fakhoury et al,19 Shyangdan et al,20
and McIntosh et al,21 have evaluated GLP-1 based
therapies. However, some meta-analyses have been
limited to GLP-1RAs only,8,22 DPP-4 inhibitors
only,23,24 or oral therapies only25; others have focused
on specific subgroups (eg, exenatide added to oral therapies [Pinelli et al26] and maximal doses [Pinelli and
Hurren27]). Because of the substantive increase in data
on GLP-1 based therapies, an updated analysis that
included newly developed therapies and expanded
data was warranted. The present analysis is timely
given the approval of 2 GLP-1RA and 4 DPP-4 inhibitors in recent years. Compared with previously published meta-analyses of GLP-1 based therapies,1,19 21
Volume 34 Number 6
V.R. Aroda et al.

the present analysis is more inclusivein part due to
the accumulating data in this fieldwith evaluation of
8 therapies from 80 studies (through June 2011) versus
4 therapies from 29 studies (to May 20, 2007) in the
meta-analysis by Amori et al, 4 therapies from 38 studies (to July 2009) in the meta-analysis by Fakhoury et
al, 6 therapies from 28 studies (to July 2010) in the
meta-analysis by Shyangdan et al, and 5 therapies from
40 studies (to May 2009) in the meta-analysis by
McIntosh et al. Inclusion/exclusion criteria in the present analysis were similar to those in the meta-analyses
by Amori et al and McIntosh et al, but the meta-analysis by Fakhoury et al included only blinded, placebocontrolled trials and required 100 participants per
trial. Recently conducted studies were more likely to
compare 2 active treatments, to be more culturally and
nationally diverse than previous studies, and to be
Phase III rather than Phase II. In addition, in the present analysis, study data were screened by dose for maximum concordance with clinical practice.
An important concern in assessing the collected
study data is heterogeneity in the standard procedures
for studying different drugs. Systematic bias may have
been introduced into studies of individual therapies as
a result of differences in sample size, study design,
blinding, and populations analyzed; treatment prior to
the study; background therapies; duration of treatment
with stable background therapy; previous treatment
failure; treatment duration; treatment dose; comparators; and dropout rates. The use of a run-in design
associated with a recent medication change or an
add-in design has also been identified as a key methodologic difference. Similarly, differences in the protocol regarding changes in other medications may have
affected the study findings. However, studies of a given
drug were likely conducted using similar processes.
The primary processes affecting comparability of glycemic changes with different therapies were likely differences in the methods of measuring HbA1c,28 the placebo effect associated with enrollment in a clinical
study program, general medical care provided during
the trial, and patient management prior to measurement of baseline HbA1c. Although documentation of
the former issues is difficult, discontinuation of previous therapies (with or without the addition of a new
therapy) before the measurement of baseline HbA1c
was reported in the majority of studies of liraglutide,
linagliptin, or sitagliptin. Changes in glucose-lowering
medications in the period immediately before random-
June 2012
ization may have affected the magnitude of subsequent

changes in HbA1c. Glycemic responses in trials with
adjustment in background therapies prior to baseline
HbA1c measurement must be interpreted with caution.
Although safety profile tolerability endpoints were
not addressed in the present meta-analysis, differences
in adverse events may have influenced clinical decisions. Most of the adverse events reported with the
GLP-1RAs (primarily gastrointestinal events, eg, nausea, vomiting) were mild or moderate, and the rates of
hypoglycemia were low when treatment with GLP1RAs was not combined with a sulfonylurea.29 33 In a
direct-comparative trial between the DPP-4 inhibitors
saxagliptin and sitagliptin, no differences in tolerability or serious adverse events were reported,10 and the
most common AEs were infections and headache.
However, differences in tolerability between a GLP1RA and a DPP-4 inhibitor were reported in 3 directcomparator trials.12,14,34 Discontinuation rates over
26 weeks with liraglutide 1.8 mg once daily were 2fold those with liraglutide 1.2 mg once daily and sitagliptin 100 mg.12 Nausea was transient in most cases
and was reported in 4-fold more patients with liraglutide 1.2 or 1.8 mg once daily than with sitagliptin.
Vomiting was twice as common with liraglutide, but
the rates of serious adverse events were similar. DeFronzo et al14 reported similar findings in a 4-week
crossover study that compared exenatide BID to sitagliptin; transient nausea occurred 3-fold more frequently with exenatide BID than with sitagliptin. Similar results were obtained in an 8-week crossover study
that compared exenatide BID to sitagliptin: nausea and
vomiting were 2- to 3-fold more common with exenatide BID treatment, but headache and diarrhea
were observed at similar frequencies.13 In a study that
compared clinical outcomes with exenatide QW and
sitagliptin, nausea occurred more than twice as often in
patients treated with exenatide as with sitagliptin, but
a similar rate of serious adverse events was reported.34
Differences in tolerability between different GLP-1RAs
have also been reported. In DURATION-1 (Diabetes
Therapy Utilization: Researching Changes in A1C,
Weight and Other Factors Through Intervention With
Exenatide Once Weekly),16 which compared the efficacy and safety profiles of exenatide BID and QW, the
prevalence of transient nausea appeared to have been
lower with exenatide QW than with exenatide BID
(26% vs 35%), although the prevalence of serious adverse events was similar (5.4% vs 3.4%). In the
1255
LEAD-6 (Liraglutide Effect Action in Diabetes) trial,15
which compared the efficacy and safety profiles of liraglutide 1.8 mg once daily with those of exenatide BID
10 g, the proportion of patients who experienced
nausea was similar (26% vs 28%). However, the incidences of serious and severe adverse events appeared
be greater in liraglutide-treated patients (serious, 5.1%
vs 2.6%; severe, 7.2% vs 4.7% with liraglutide once
daily vs exenatide BID, respectively). In the DURATION-6 study,17 which compared the efficacy and
safety profiles of exenatide QW with liraglutide 1.8 mg
once daily, both therapies were reported as having a
favorable safety profile.
Differences in the practical aspects of care were not
considered in the present analysis. The DPP-4 inhibitors are orally administered 1 time per day, whereas
the GLP-1RAs are administered by subcutaneous injection twice daily, once daily, or once every 7 days.
Exenatide BID must be administered within 60 minutes before the 2 largest meals of the day,35 whereas the
timing of the administration of liraglutide once daily or
exenatide QW is flexible.15,16
Limitations of the present meta-analysis included
difficulty in capturing all relevant papers due to the
rapidly evolving evidence base, which led to selection
of a specific end-date. An additional limitation was
inadequate information to adjust the data for non
study-drug related improvements in HbA1c, FPG, or
weight. As in clinical practice, the contributions to efficacy of improved lifestyle and/or positive treatment
expectations cannot be quantified. Because of increasing interest in active comparators among the therapies
for type 2 diabetes, the ability to adjust data for a
placebo effect is diminishing.25 Only trial-level published data (not subject-level data) were available for
the analyses in the present report.
ACKNOWLEDGMENTS
The authors thank James Ruggles for discussions on
the inclusion/exclusion of articles, Yvette Peters and
Carmelle V. Remillard for reviewing the accuracy of
the data, and Loretta Nielsen for editorial and medical
writing assistance.
Dr. Aroda was responsible for the study design, data
interpretation, writing and editing of the paper. Dr.
Henry was responsible for the study design, data interpretation, critical review and editing of the paper. Ms.
Han and Dr. Huang were responsible for the study
design, data collection, data analysis, critical review of
the paper. Dr. DeYoung was responsible for the study
design, literature search, data analysis, creation of figures and tables, writing of the paper. Drs. Darsow and
Hoogwerf were responsible for the study design, literature search, data analysis, critical review and editing
of the paper.
CONFLICTS OF INTEREST
All of the incretin-based therapies analyzed in the present

study were associated with reductions from baseline in
HbA1c and FPG. The findings emphasize a continued
need for direct-comparative trials between GLP-1 based
therapies to further inform treatment decisions.
Statistical and writing support for the manuscript was

provided or funded by Amylin Pharmaceuticals, Inc
and Eli Lilly and Company, the developers of
exenatide.
Dr. Aroda has received research grant support from
Amylin Pharmaceuticals, Inc, Aegerion Pharmaceuticals, Inc; Bayhill Therapeutics (2008 2009); InteKrin
Therapeutics, Inc; Merck & Co, Inc; Novo Nordisk;
Roche Pharmaceutical Corporation; The sanofi-aventis Group; Takeda Pharmaceuticals North America,
Inc; and VeraLight, Inc. Dr. Aroda has received consultants fees from Roche Pharmaceutical Corporation
(2008), Tethys (2009), and VeraLight (2009). Dr.
Henry has received consultants fees from, and is an
advisory board member at, Amylin Pharmaceuticals,
Inc, and receives research support through a Veterans
Administration medical research grant from Amylin
Pharmaceuticals Inc. Ms. Han, Dr. Huang, and Dr.
DeYoung are employees of, and hold stock options in,
Amylin Pharmaceuticals, Inc. Dr. Darsow was an employee of Amylin Pharmaceuticals, Inc at the time that
these analyses were begun. Dr. Hoogwerf is an employee of and may hold stock options in Eli Lilly and
Company.
SUPPLEMENTAL MATERIAL
REFERENCES
CONCLUSIONS
Supplemental appendixes and figures accompanying

this article can be found in the online version at
http://dx.doi.org/10.1016/j.clinthera.2012.04.013.
1256
1. Amori RE, Lau J, Pittas AG. Efficacy and safety of incretin

therapy in type 2 diabetes: systematic review and metaanalysis. JAMA. 2007;298:194 206.
Volume 34 Number 6
V.R. Aroda et al.

2. Mentlein R. Dipeptidyl-peptidase
IV (CD26)role in the inactivation
of regulatory peptides. Regul Pept.
1999;85:9 24.
3. Drucker DJ, Nauck MA. The incretin system: glucagon-like peptide-1
receptor agonists and dipeptidyl
peptidase-4 inhibitors in type 2
diabetes. Lancet. 2006;368:1696
1705.
4. MontanyaE,SestiG.Areviewofefficacy
and safety data regarding the use of
liraglutide, a once-daily human glucagon-likepeptide1analogue,inthetreatment of type 2 diabetes mellitus. Clin
Ther. 2009;31:24722488.
5. Bunck MC, Diamant M, Cornr A, et al.
One-year treatment with exenatide improves -cell function, compared with
insulin glargine, in metformin-treated
type 2 diabetic patients. Diabetes Care.
2009;32:762768.
6. Fineman M, Flanagan S, Taylor K, et al.
Pharmacokinetics and pharmacodynamics of exenatide extended-release
after single and multiple dosing. Clin
Pharmacokinet. 2011;50:6574.
7. Gwaltney SL 2nd. Medicinal chemistry approaches to the inhibition of
dipeptidyl peptidase IV. Curr Top
Med Chem. 2008;8:15451552.
8. Monami M, Marchionni N, Mannucci
E.Glucagon-likepeptide-1receptoragonists in type 2 diabetes: a meta-analysis
ofrandomizedclinicaltrials.EurJEndocrinol. 2009;160:909917.
9. Ma J, Liu W, Hunter A, Zhang W.
Performing meta-analysis with incomplete statistical information in clinical
trials. BMC Med Res Methodol. 2008;8:
56.
10. Scheen AJ, Charpentier G, stgren
CJ, et al. Efficacy and safety of
saxagliptin in combination with metformin compared with sitagliptin in
combination with metformin in
adult patients with type 2 diabetes
mellitus. Diabetes Metab Res Rev.
2010;26:540 549.
11. Bosi E, Dotta F, Jia Y, Goodman M.
Vildagliptin plus metformin combination therapy provides superior glycaemic control to individual mono-
June 2012
12.
13.
14.
15.
16.
17.
18.
therapy in treatment-naive patients

with type 2 diabetes mellitus.
Diabetes Obes Metab. 2009;11:506
515.
Pratley RE, Nauck M, Bailey T, et al, for
the 1860-LIRA-DPP-4 Study Group. Liraglutide versus sitagliptin for patients
with type 2 diabetes who did not have
adequate glycaemic control with metformin: a 26-week, randomised, parallel-group, open-label trial. Lancet. 2010;
375:14471456.
Berg JK, Shenouda SK, Heilmann
CR, et al. Effects of exenatide twice
daily versus sitagliptin on 24-h glucose, glucoregulatory and hormonal
measures: a randomized, doubleblind, crossover study. Diabetes Obes
Metab. 2011;13:982989.
DeFronzo RA, Okerson T, Viswanathan
P, et al. Effects of exenatide versus sitagliptin on postprandial glucose, insulin
and glucagon secretion, gastric emptying, and caloric intake: a randomized,
cross-over study. Curr Med Res Opin.
2008;
24:29432952.
Buse JB, Rosenstock J, Sesti G, et al,
for the LEAD-6 Study Group. Liraglutide once a day versus exenatide twice
a day for type 2 diabetes: a 26-week
randomised, parallel-group, multinational, open-label trial (LEAD-6).
Lancet. 2009;374:39 47.
Drucker DJ, Buse JB, Taylor K, et al,
for the DURATION-1 Study Group.
Exenatide once weekly versus twice
daily for the treatment of type 2
diabetes: a randomised, open-label,
non-inferiority study. Lancet. 2008;
372:1240 1250.
Buse JB, Nauck MA, Forst T, et al.
Efficacy and safety of exenatide once
weekly versus liraglutide in subjects
with type 2 diabetes (DURATION6): a randomised, open-label study.
Diabetologia. 2011;54(Suppl 1):S38.
Scott D, Boye KS, Timlin L, et al. A
network meta analysis to compare
glycaemic control in patients with
type 2 diabetes treated with exenatide once weekly or liraglutide.
Value Health. 2011;14:A472A473.
19. Fakhoury WK, Lereun C, Wright D.

A meta-analysis of placebo-controlled
clinical trials assessing the efficacy
and safety of incretin-based medications in patients with type 2 diabetes.
Pharmacology. 2010;86:44 57.
20. Shyangdan DS, Royle PL, Clar C, et
al. Glucagon-like peptide analogues
for type 2 diabetes mellitus: systematic review and meta-analysis. BMC
Endocr Disord. 2010;10:20.
21. McIntosh B, Cameron C, Singh SR, et
al. Second-line therapy in patients with
type 2 diabetes inadequately controlled
withmetforminmonotherapy:asystematic review and mixed-treatment comparison meta-analysis. Open Med. 2011;
5:e35e48.
22. Esposito K, Mosca C, Brancario C,
et al. GLP-1 receptor agonists and
HBA1c target of 7% in type 2
diabetes: meta-analysis of randomized controlled trials. Curr Med Res
Opin. 2011;27:1519 1528.
23. Monami M, Iacomelli I, Marchionni
N, Mannucci E. Dipeptydil peptidase-4 inhibitors in type 2 diabetes:
a meta-analysis of randomized clinical trials. Nutr Metab Cardiovasc Dis.
2010;20:224 235.
24. Esposito K, Cozzolino D, Bellastella
G, et al. Dipeptidyl peptidase-4 inhibitors and HbA1c target of 7%
in type 2 diabetes: meta-analysis of
randomized controlled trials. Diabetes Obes Metab. 2011;13:594 603.
25. Sherifali D, Nerenberg K, Pullenayegum
E, et al. The effect of oral antidiabetic
agents on A1C levels: a systematic review and meta-analysis. Diabetes Care.
2010;33:18591864.
26. Pinelli NR, Cha R, Brown MB, Jaber
LA. Addition of thiazolidinedione or
exenatide to oral agents in type 2
diabetes: a meta-analysis. Ann Pharmacother. 2008;42:15411551.
27. Pinelli NR, Hurren KM. Efficacy and
safety of long-acting glucagon-like peptide-1 receptor agonists compared with
exenatide twice daily and sitagliptin in
type 2 diabetes mellitus: a systematic
reviewandmeta-analysis.AnnPharmacother. 2011;45:850860.
1257
28. Hanas R, John G. 2010 consensus
statement on the worldwide standardization of the hemoglobin A1C
measurement. Diabetes Care. 2010;
33:19031904.
29. Moretto TJ, Milton DR, Ridge TD, et
al. Efficacy and tolerability of exenatide monotherapy over 24 weeks
in antidiabetic drug-naive patients
with type 2 diabetes: a randomized,
double-blind, placebo-controlled,
parallel-group study. Clin Ther. 2008;
30:1448 1460.
30. Garber AJ, Foley JE, Banerji MA, et
al. Effects of vildagliptin on glucose
control in patients with type 2 diabetes inadequately controlled with a
sulphonylurea. Diabetes Obes Metab.
2008;10:10471056.
31. Fleck P, Inzucchi S, Seufert J, et al.
Initial therapy with combination
alogliptin plus pioglitazone in type 2
diabetes patients inadequately controlled with diet and exercise. Diabetologia. 2009;52(Suppl 1):S294.
32. Rosenstock J, Aguilar-Salinas C, Klein E,
etal,fortheCV181-011StudyInvestigators. Effect of saxagliptin monotherapy
in treatment-naive patients with type 2
diabetes. Curr Med Res Opin. 2009;25:
24012411.
33. Scott R, Wu M, Sanchez M, Stein P.
Efficacy and tolerability of the dipeptidyl peptidase-4 inhibitor sitagliptin as monotherapy over 12 weeks in
patients with type 2 diabetes. Int
J Clin Pract. 2007;61:171180.
34. Bergenstal RM, Wysham C, Macconell L, et al. Efficacy and safety of
exenatide once weekly versus sitagliptin or pioglitazone as an adjunct
to metformin for treatment of type 2
diabetes (DURATION-2): a randomised trial. Lancet. 2010;376:
431 439.
35. Forti A, Garcia EG, Yu MB, et al.
Efficacy and safety of exenatide administered before the two largest
daily meals of Latin American patients with type 2 diabetes. Curr Med
Res Opin. 2008;24:24372447.
1258
Address correspondence to: Vanita R. Aroda, MD, Department of Endocrinology, Diabetes, and Metabolism, MedStar Health Research Institute, 6525 Belcrest Road, Suite 700, Hyattsville, MD 20782. E-mail:
[email protected]
Volume 34 Number 6
V.R. Aroda et al.
Supplemental Appendix
REFERENCES
1. Bergenstal R, Lewin A, Bailey T, et al, for the NovoLog
Mix-vs.-Exenatide Study Group. Efficacy and safety of
biphasic insulin aspart 70/30 versus exenatide in subjects
with type 2 diabetes failing to achieve glycemic control with
metformin and a sulfonylurea. Curr Med Res Opin. 2009;25:
6575.
2. Bolli GB, Rosenstock J, Balas B, et al. Superior glycaemic
control with taspoglutide, a once-weekly human GLP-1
analogue, compared with twice daily exenatide in type 2
diabetes: the T-emerge 2 Trial. Diabetologia. 2010;
53(Suppl):S38. Abstract.
3. Buse JB, Henry RR, Han J, et al, for the Exenatide-113
Clinical Study Group. Effects of exenatide (exendin-4) on
glycemic control over 30 weeks in sulfonylurea-treated
patients with type 2 diabetes. Diabetes Care. 2004;27:2628
2635.
4. DeFronzo RA, Ratner RE, Han J, et al. Effects of exenatide
(exendin-4) on glycemic control and weight over 30 weeks
in metformin-treated patients with type 2 diabetes. Diabetes
Care. 2005;28:10921100.
5. Forti A, Garcia EG, Yu MB, et al. Efficacy and safety of
exenatide administered before the two largest daily meals
of Latin American patients with type 2 diabetes. Curr Med
Res Opin. 2008;24:24372447.
6. Gallwitz B, Bhmer M, Segiet T, et al. Exenatide twice daily
versus premixed insulin aspart 70/30 in metformin-treated
patients with type 2 diabetes. Diabetes Care. 2011a;34:604
606.
7. Gao Y, Yoon KH, Chuang LM, et al, for the GWAA Study
Group. Efficacy and safety of exenatide in patients of Asian
descent with type 2 diabetes inadequately controlled with
metformin or metformin and a sulphonylurea. Diabetes Res
Clin Pract. 2009;83:69 76.
8. Heine RJ, Van Gaal LF, Johns D, et al. Exenatide versus
insulin glargine in patients with suboptimally controlled
type 2 diabetes: a randomized trial. Ann Intern Med.
2005;143:559 569.
9. Kadowaki T, Namba M, Yamamura A, et al. Exenatide
exhibits dose-dependent effects on glycemic control over
12 weeks in Japanese patients with suboptimally controlled
type 2 diabetes. Endocr J. 2009;56:415 424.
10. Kadowaki T, Namba M, Imaoka T, et al. Improved glycemic
control and reduced bodyweight with exenatide: a doubleblind, randomized, phase 3 study in Japanese patients with
suboptimally controlled type 2 diabetes over 24 weeks. J
Diabetes Invest. 2011;2:210 217.
11. Kendall DM, Riddle MC, Rosenstock J, et al. Effects of
exenatide (exendin-4) on glycemic control over 30 weeks in
patients with type 2 diabetes treated with metformin and a
sulfonylurea. Diabetes Care. 2005;28:10831091.
June 2012
12. Liutkus J, Rosas Guzman J, Norwood P, et al. A placebocontrolled trial of exenatide twice-daily added to thiazolidinediones alone or in combination with metformin.
Diabetes Obes Metab. 2010;12:1058 1065.
13. Moretto TJ, Milton DR, Ridge TD, et al. Efficacy and
tolerability of exenatide monotherapy over 24 weeks in
antidiabetic drug-naive patients with type 2 diabetes: a
randomized, double-blind, placebo-controlled, parallelgroup study. Clin Ther. 2008;30:1448 1460.
14. Nauck MA, Duran S, Kim D, et al. A comparison of
twice-daily exenatide and biphasic insulin aspart in patients with type 2 diabetes who were suboptimally controlled with sulfonylurea and metformin: a non-inferiority
study. Diabetologia. 2007a;50:259 267.
15. Sowa H, Kadowaki T, Namba M, et al. Efficacy and safety
of exenatide in Japanese patients with type 2 diabetes
suboptimally controlled despite oral therapy. Endocr J.
2010;57(Suppl 2):S362. Abstract.
16. Zinman B, Hoogwerf BJ, Duran Garcia S, et al. The effect of
adding exenatide to a thiazolidinedione in suboptimally
controlled type 2 diabetes: a randomized trial. Ann Intern
Med. 2007;146:477 485.
17. Bergenstal RM, Wysham C, MacConell L, et al. Efficacy and
safety of exenatide once weekly versus sitagliptin or pioglitazone as an adjunct to metformin for treatment of type 2
diabetes (DURATION-2): a randomised trial. Lancet. 2010a;
376:431 439.
18. Blevins T, Pullman J, Malloy J, et al. DURATION-5:
Exenatide once weekly resulted in greater improvements in
glycemic control compared with exenatide twice daily in
patients with type 2 diabetes. J Clin Endocrinol Metab.
2011;96:13011310.
19. Cuddihy RM, Russell-Jones D, Hanefeld M, et al. Efficacy
and safety of exenatide once weekly versus metformin,
pioglitazone, and sitagliptin used as monotherapy in
drug-nave patients with type 2 diabetes. Diabetes. 2011;60
(Suppl. 1):A77. Abstract.
20. Diamant M, Van Gaal L, Stranks S, et al. Once weekly
exenatide compared with insulin glargine titrated to target
in patients with type 2 diabetes (DURATION-3): an
open-label randomised trial. Lancet. 2010;375:2234
2243.
21. Drucker DJ, Buse JB, Taylor K, et al, for the DURATION-1
Study Group. Exenatide once weekly versus twice daily for
the treatment of type 2 diabetes: a randomised, openlabel, non-inferiority study. Lancet. 2008;372:1240 1250.
22. Kim D, MacConell L, Zhuang D, et al. Effects of onceweekly dosing of a long-acting release formulation of
exenatide on glucose control and body weight in subjects
with type 2 diabetes. Diabetes Care. 2007;30:14871493.
23. MacConell L, Malloy J, Huang W, et al. Safety and efficacy
of once-monthly exenatide over 20 weeks in patients with
1258.e1
24.
25.
26.
27.
28.
29.
30.
31.
32.
type 2 diabetes. Diabetes. 2011;60 (Suppl 2):LB13LB14.

Abstract.
Buse JB, Rosenstock J, Sesti G, et al, for the LEAD-6
Study Group. Liraglutide once a day versus exenatide
twice a day for type 2 diabetes: a 26-week randomised,
parallel-group, multinational, open-label trial (LEAD6). Lancet. 2009;374:39 47.
Garber A, Henry R, Ratner R, et al, for the LEAD-3 (Mono)
Study Group. Liraglutide versus glimepiride monotherapy
for type 2 diabetes (LEAD-3 Mono): a randomised, 52week, phase III, double-blind, parallel-treatment trial.
Lancet. 2009;373:473 481.
Kaku K, Rasmussen MF, Clauson P, Seino Y. Improved
glycaemic control with minimal hypoglycaemia and no
weight change with the once-daily human glucagon-like
peptide-1 analogue liraglutide as add-on to sulphonylurea
in Japanese patients with type 2 diabetes. Diabetes Obes
Metab. 2010;12:341347.
Madsbad S, Schmitz O, Ranstam J, et al, for the NN22111310 International Study Group. Improved glycemic control with no weight increase in patients with type 2 diabetes
after once-daily treatment with the long-acting glucagonlike peptide 1 analog liraglutide (NN2211): a 12-week,
double-blind, randomized, controlled trial. Diabetes Care.
2004;27:13351342.
Marre M, Shaw J, Brandle M, et al, for the LEAD-1 SU Study
Group. Liraglutide, a once-daily human GLP-1 analogue,
added to a sulphonylurea over 26 weeks produces greater
improvements in glycaemic and weight control compared
with adding rosiglitazone or placebo in subjects with type 2
diabetes (LEAD-1 SU). Diabet Med. 2009;26:268 278.
Nauck M, Frid A, Hermansen K, et al, for the LEAD-2
Metformin Study Group. Efficacy and safety comparison of
liraglutide, glimepiride, and placebo, all in combination
with metformin, in type 2 diabetes: The LEAD (Liraglutide
Effect and Action in Diabetes)-2 study. Diabetes Care.
2009a;32:84 90.
Pratley RE, Nauck M, Bailey T, et al, for the 1860-LIRADPP-4 Study Group Liraglutide versus sitagliptin for patients with type 2 diabetes who did not have adequate
glycaemic control with metformin: a 26-week, randomised,
parallel-group, open-label trial. Lancet. 2010;375:1447
1456.
Russell-Jones D, Vaag A, Schmitz O, et al. Liraglutide vs
insulin glargine and placebo in combination with metformin and sulfonylurea therapy in type 2 diabetes mellitus
(LEAD-5 metSU): a randomised controlled trial. Diabetologia. 2009;52:2046 2055.
Yang W, Chen L, Ji Q, et al. Liraglutide provides similar
glycaemic control as glimepiride (both in combination
with metformin) and reduces body weight and systolic
blood pressure in Asian population with type 2 diabetes
from China, South Korea and India: a 16-week, random-
1258.e2
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
ized, double-blind, active control trial. Diabetes Obes Metab.

2011a;13:81 88.
Zinman B, Gerich J, Buse JB, et al, for the LEAD-4 Study
Investigators. Efficacy and safety of the human GLP-1
analog liraglutide in combination with metformin and TZD
in patients with type 2 diabetes mellitus (LEAD-4
MetTZD). Diabetes Care. 2009;32:1224 1230.
DeFronzo RA, Fleck PR, Wilson CA, Mekki Q, for the
Alopliptin Study 010 Group. Efficacy and safety of the
dipeptidyl peptidase-4 inhibitor alogliptin in patients with
type 2 diabetes and inadequate glycemic control: a randomized, double-blind, placebo-controlled study. Diabetes Care.
2008a;31:23152317.
Fleck P, Inzucchi S, Seufert J, et al. Initial therapy with
combination alogliptin plus pioglitazone in type 2 diabetes
patients inadequately controlled with diet and exercise.
Diabetologia. 2009;52(Suppl 1):S294. Abstract.
Nauck MA, Ellis GC, Fleck PR, et al, for the Alogliptin Study
008 Group. Efficacy and safety of adding the dipeptidyl
peptidase-4 inhibitor alogliptin to metformin therapy in
patients with type 2 diabetes inadequately controlled with
metformin monotherapy: a multicentre, randomised,
double-blind, placebo-controlled study. Int J Clin Pract.
2009a;63:46 55.
Pratley RE, Kipnes MS, Fleck PR, et al, for the Alogliptin
Study 007 Group. Efficacy and safety of the dipeptidyl
peptidase-4 inhibitor alogliptin in patients with type 2
diabetes inadequately controlled by glyburide monotherapy. Diabetes Obes Metab. 2009a;11:167176.
Pratley RE, Reusch JE, Fleck PR, et al, for the Alogliptin
peptidase-4 inhibitor alogliptin added to pioglitazone in
patients with type 2 diabetes: a randomized, double-blind,
placebo-controlled study. Curr Med Res Opin. 2009b;25:
23612371.
Del Prato S, Barnett AH, Huisman H, et al. Effect of
linagliptin monotherapy on glycaemic control and markers
of beta-cell function in patients with inadequately controlled type 2 diabetes: a randomized controlled trial.
Forst T, Uhlig-Laske B, Ring A, et al. Linagliptin (BI 1356), a
potent and selective DPP-4 inhibitor, is safe and efficacious
in combination with metformin in patients with inadequately controlled Type 2 diabetes. Diabet Med. 2010;27:
1409 1419.
Gallwitz B, Uhlig-Laske B, Bhattacharaya S, et al. Linagliptin has similar efficacy to glimepiride but improved cardiovascular safety over 2 years in patients with type 2 diabetes
inadequately controlled on metformin. Diabetes. 2011b;
60(Suppl 2):LB11LB12. Abstract.
Gomis R, Espadero RM, Jones R, et al. Efficacy and safety of
initial combination therapy with linagliptin and pioglitazone in patients with inadequately controlled type 2
Volume 34 Number 6
V.R. Aroda et al.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
diabetes: a randomized, double-blind, placebo-controlled

study. Diabetes Obes Metab. 2011;13:653 661.
Kawamori R, Inagaki N, Araki E, et al. Linagliptin
monotherapy improves glycemic control in Japanese
patients with T2DM over 12 weeks. Diabetes. 2010a;59
(Suppl 1):A190. Abstract.
Kawamori R, Inagaki N, Araki E, et al. Linagliptin provides
superior glycemic control compared to voglibose as monotherapy in Japanese patients with type 2 diabetes. Diabetes.
2010b;59(Suppl 1):A172A173. Abstract.
Lewin AJ, Arvay L, Liu D, Patel S, Woerle HJ. Safety and
efficacy of linagliptin as add-on therapy to a sulphonylurea
in inadequately controlled type 2 diabetes. Diabetologia.
2010;53(Suppl 1):S326. Abstract.
Owens DR, Swallow R, Jones P, et al. Linagliptin improves
glycemic control in type 2 diabetes patients inadequately
controlled by metformin and sulfonylurea without weight
gain or hypoglycemia. Diabetes. 2010;59(Suppl 1):A149.
Abstract.
Taskinen MR, Rosenstock J, Tamminen I, et al. Safety
and efficacy of linagliptin as add-on therapy to metformin in patients with type 2 diabetes: a randomized,
double-blind, placebo-controlled study. Diabetes Obes
Metab. 2011;13:6574.
Chacra AR, Tan GH, Apanovitch A, et al, for the CV181040 Investigators. Saxagliptin added to a submaximal dose
of sulphonylurea improves glycaemic control compared
with uptitration of sulphonylurea in patients with type 2
diabetes: a randomised controlled trial. Int J Clin Pract.
2009;63:13951406.
DeFronzo RA, Hissa MN, Garber AJ, et al, for the Saxagliptin 014 Study Group. The efficacy and safety of saxagliptin
when added to metformin therapy in patients with inadequately controlled type 2 diabetes with metformin alone.
Diabetes Care. 2009;32:1649 1655.
Fonseca V, Zhu T, Karyekar C, Hirshberg B. Adding
saxagliptin (SAXA) 5 mg is superior to uptitrating metformin extended release (MET XR) in type 2 diabetes
mellitus (T2DM) patients with inadequate glycemic control on a stable dose of MET XR 1500 mg. Diabetes.
2011a;60 (Suppl 1):A279. Abstract.
Gke B, Gallwitz B, Eriksson J, et al, for the D.C. Investigators. Saxagliptin is non-inferior to glipizide in patients with
type 2 diabetes mellitus inadequately controlled on metformin alone: a 52-week randomised controlled trial. Int
J Clin Pract. 2010;64:1619 1631.
Rosenstock J, Sankoh S, List JF. Glucose-lowering activity of
the dipeptidyl peptidase-4 inhibitor saxagliptin in drugnaive patients with type 2 diabetes. Diabetes Obes Metab.
2008;10:376 386.
Rosenstock J, Aguilar-Salinas C, Klein E, et al, for the
CV181-011 Study Investigators. Effect of saxagliptin mono-
June 2012
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
therapy in treatment-naive patients with type 2 diabetes.

Curr Med Res Opin. 2009;25:24012411.
Scheen AJ, Charpentier G, stgren CJ, et al. Efficacy and
safety of saxagliptin in combination with metformin compared with sitagliptin in combination with metformin in
adult patients with type 2 diabetes mellitus. Diabetes Metab
Res Rev. 2010;26:540 549.
Arechavaleta R, Seck T, Chen Y, et al. Efficacy and safety of
treatment with sitagliptin or glimepiride in patients with
type 2 diabetes inadequately controlled on metformin
monotherapy: a randomized, double-blind, non-inferiority
trial. Diabetes Obes Metab. 2011;13:160 168.
Aschner P, Kipnes MS, Lunceford JK, et al, for the
Sitagliptin Study 021 Group. Effect of the dipeptidyl
peptidase-4 inhibitor sitagliptin as monotherapy on glycemic control in patients with type 2 diabetes. Diabetes Care.
2006;29:26322637.
Aschner P, Katzeff HL, Guo H, et al. Efficacy and safety of
monotherapy of sitagliptin compared with metformin in
patients with type 2 diabetes. Diabetes Obes Metab. 2010;12:
252261.
Bergenstal R, Forti A, Chiasson J-L, et al. Once weekly
taspoglutide, a human GLP-1 analog, is superior to
sitagliptin in improving glycemic control and weight loss
in patients with type 2 diabetes (T2D): results from the
T-emerge 4 trial. Diabetes. 2010b;59(Suppl 1):A16.
Abstract.
Charbonnel B, Karasik A, Liu J, Wu M, Meininger G, for the
Sitagliptin Study 020 Group. Efficacy and safety of the
dipeptidyl peptidase-4 inhibitor sitagliptin added to ongoing metformin therapy in patients with type 2 diabetes
inadequately controlled with metformin alone. Diabetes
Care. 2006;29:2638 2643.
Fonseca V, Staels B, Morgan JDI, et al, for the Sitagliptin
036 Study Group. The addition of sitagliptin to metformin and pioglitazone therapy enhances glycemic
control in patients with type 2 diabetes. Diabetes.
2011b;60(Suppl 1):A308. Abstract.
Goldstein B, Feinglos M, Lunceford J, et al, for the
Sitagliptin 036 Study Group. Effect of initial combination
therapy with sitagliptin, a dipeptidyl peptidase-4 inhibitor,
and metformin on glycemic control in patients with type 2
diabetes. Diabetes Care. 2007;30:1979 1987.
Hanefeld M, Herman GA, Wu M, et al, for the Sitagliptin
Study 014 Investigators. Once-daily sitagliptin, a dipeptidyl peptidase-4 inhibitor, for the treatment of patients with
type 2 diabetes. Curr Med Res Opin. 2007;23:1329 1339.
Hermansen K, Kipnes M, Luo E, et al, for the Sitagliptin
peptidase-4 inhibitor, sitagliptin, in patients with type 2
diabetes mellitus inadequately controlled on glimepiride
alone or on glimepiride and metformin. Diabetes Obes
Metab. 2007;9:733745.
1258.e3
64. Iwamoto Y, Taniguchi T, Nonaka K, et al. Dose-ranging
efficacy of sitagliptin, a dipeptidyl peptidase-4 inhibitor, in
Japanese patients with type 2 diabetes mellitus. Endocr J.
2010;57:383394.
65. Mohan V, Yang W, Son HY, et al. Efficacy and safety of
sitagliptin in the treatment of patients with type 2 diabetes
in China, India, and Korea. Diabetes Res Clin Pract. 2009;83:
106 116.
66. Nauck MA, Meininger G, Sheng D, Terranella L, for the
Sitagliptin Study 024 Group. Efficacy and safety of the
dipeptidyl peptidase-4 inhibitor, sitagliptin, compared
with the sulfonylurea, glipizide, in patients with type 2
diabetes inadequately controlled on metformin alone: a
randomized, double-blind, non-inferiority trial. Diabetes
Obes Metab. 2007b;9:194 205.
67. Nonaka K, Kakikawa T, Sato A, et al. Efficacy and safety of
sitagliptin monotherapy in Japanese patients with type 2
diabetes. Diabetes Res Clin Pract. 2008;79:291298.
68. Raz I, Hanefeld M, Xu L, et al, for the Sitagliptin Study 023
Group. Efficacy and safety of the dipeptidyl peptidase-4
inhibitor sitagliptin as monotherapy in patients with type 2
diabetes mellitus. Diabetologia. 2006;49:2564 2571.
69. Raz I, Chen Y, Wu M, et al. Efficacy and safety of sitagliptin
added to ongoing metformin therapy in patients with type
2 diabetes. Curr Med Res Opin. 2008;24:537550.
70. Rosenstock J, Brazg R, Andryuk PJ, et al, for the Sitagliptin
peptidase-4 inhibitor sitagliptin added to ongoing pioglitazone therapy in patients with type 2 diabetes: a 24-week,
multicenter, randomized, double-blind, placebo-controlled, parallel-group study. Clin Ther. 2006;28:1556
1568.
71. Rosenstock J, Jelaska A, Seman L, et al. Efficacy and safety
of BI10773, a new sodium glucose cotransporter-2
(SGLT-2) inhibitor, in type 2 diabetes inadequately controlled on metformin. Diabetes. 2011;60(Suppl 1):A271.
Abstract.
72. Scott R, Loeys T, Davies MJ, Engel SS, for the Sitagliptin
Study 801 Group. Efficacy and safety of sitagliptin when
1258.e4
73.
74.
75.
76.
77.
78.
79.
80.
added to ongoing metformin therapy in patients with type

2 diabetes. Diabetes Obes Metab. 2008;10:959 969.
Seck T, Nauck M, Sheng D, et al, for the Sitagliptin Study
024 Group. Safety and efficacy of treatment with sitagliptin
or glipizide in patients with type 2 diabetes inadequately
controlled on metformin: a 2-year study. Int J Clin Pract.
2010;64:562576.
Yang W, Guan Y, Shentu Y, et al. Sitagliptin added to
ongoing metforming therapy in Chinesse patients with type
2 diabetes significantly enhances glycemic control. Diabetes. 2011b;60:A306. Abstract.
Bolli G, Dotta F, Rochotte E, Cohen SE. Efficacy and
tolerability of vildagliptin vs. pioglitazone when added to
metformin: a 24-week, randomized, double-blind study.
Diabetes Obes Metab. 2008;10:8290.
Bosi E, Camisasca RP, Collober C, et al. Effects of
vildagliptin on glucose control over 24 weeks in patients
with type 2 diabetes inadequately controlled with metformin. Diabetes Care. 2007;30:890 895.
Bosi E, Dotta F, Jia Y, Goodman M. Vildagliptin plus
metformin combination therapy provides superior glycaemic control to individual monotherapy in treatment-naive
patients with type 2 diabetes mellitus. Diabetes Obes Metab.
2009;11:506 515.
Filozof C, Gautier J-F. A comparison of efficacy and safety
of vildagliptin and gliclazide in combination with metformin in patients with Type 2 diabetes inadequately
controlled with metformin alone: a 52-week, randomized
study. Diabet Med. 2010;27:318 326.
Garber AJ, Schweizer A, Baron MA, et al. Vildagliptin in
combination with pioglitazone improves glycaemic control
in patients with type 2 diabetes failing thiazolidinedione
monotherapy: a randomized, placebo-controlled study.
Garber AJ, Foley JE, Banerji MA, et al. Effects of vildagliptin
on glucose control in patients with type 2 diabetes
inadequately controlled with a sulphonylurea. Diabetes
Obes Metab. 2008;10:10471056.
Volume 34 Number 6
V.R. Aroda et al.
Supplemental Table I. Drug doses included in the present meta-analysis of the efficacy of the glucagon-like
peptide 1 receptor agonists (GLP-1RAs) and dipeptidyl-peptidase IV (DPP-4) inhibitors
in type 2 diabetes mellitus.
Drug
Highest Maintenance Dose
All Doses
GLP-1RAs
Exenatide BID
Exenatide QW
Liraglutide
10 g BID
2 mg QW
1.8 mg once daily
5 and 10 g BID
2 mg QW
0.6, 0.9, 1.2, and 1.8 mg once daily
DPP-4 inhibitors
Alogliptin
Linagliptin
Saxagliptin
Sitagliptin
Vildagliptin
25 mg once daily
5 mg once daily
5 mg once daily
100 mg once daily
50 mg BID*
12.5 and 25 mg once daily

5 mg once daily
5 mg once daily
100 mg once daily
50 mg BID,* 50 mg once daily
*Not as monotherapy, only with a sulfonylurea.
Not as monotherapy, only with metformin or a thiazolidinedione.
June 2012
1258.e5
Study
Study
ITT
Phase Duration, wk Population
GLP-1RA: Exenatide BID

Bergenstal et al IIIb
(2009)1
24
Bolli et al
(2010)2
III
24
Buse et al
(2004)3
III
30
DeFronzo et al
(2005)4
III
30
Forti et al
(2008)5
III
12
Volume 34 Number 6
Gallwitz et al
(2011)6
Gao et al
(2009)7
Heine et al
(2005)8
Kadowaki et al
(2009)9
III
26
III
16
III
26
III
12
Kadowaki et al
(2011)10
III
24
Kendall et al
(2005)11
III
30
Liutkus et al
(2010)12
Moretto et al
(2008)13
III
26
III
24
Nauck et al
(2007)14
Sowa et al
(2010)15
III
52
III
24
Zinman et al
(2007)16
III
16
Diabetes Baseline Baseline

Baseline
Duration, y HbA1c, % Weight, kg FPG, mmol/L
Female, %
White, %
124
124
124
373
384
392
129
125
123
113
110
113
190
51.6
51.6
52.4
NG
NG
NG
42.6
40.8
37.4
39.8
48.2
40.7
58.4
NG
NG
NG
0
0
0
59.7
61.6
66.7
79.6
77.3
72.6
41.0
8.6
8.4
9.9
NG
NG
NG
6.6
6.3
5.7
4.9
6.2
6.6
8.5
10.2
10.1
10.3
8.05
8.08
8.08
8.6
8.5
8.7
8.18
8.26
8.2
8.4
96.6
96.9
93.8
94.5
93.2
95.5
95
95
99
101
100
100
82.7
10.9
11.2
9.9
9.87
9.91
9.83
10.0
10.8
9.3
9.8
9.4
8.9
10.9
187
51.3
45.0
8.3
8.5
81.8
9.0
181
173
234
232
282
267
40
38*
37
38
72
72
35
241
245
247
111
54
78
77
77
253
248
72
72
35
121
112
NG
NG
52.0
59.0
45.0
43.4
25.0
29.7
32.4
37.8
31.9
31.9
31.4
40.7
40.8
44.1
40
43
38
48.0
45.0
47.0
51.0
NG
NG
NG
46.3
42.9
NG
NG
0
0
79.8
80.5
0
0
0
0
NG
NG
NG
66.4
69.0
68.4
57
61
72.0
65.0
66.0
NG
NG
NG
NG
NG
85.1
82.1
5
5
8
8
9.9
9.2
11.9
14.8
11.3
9.6
11.6
12.2
12.4
8.7
8.7
9.4
6.3
6.4
2
2
1
9.8
10
NG
NG
NG
7.3
8.2
7.9
7.9
8.3
8.3
8.2
8.3
8.1
8.0
7.9
7.9
8.2
8.3
8.1
8.5
8.5
8.5
8.2
8.3
7.8
7.9
7.8
8.6
8.6
NG
NG
NG
7.9
7.9
NG
NG
69.6
67.9
87.5
88.3
71.1
64.9
65.6
70.3
69.1
67.0
70.3
98.0
97.0
99.0
94.5
92.6
86
85
86
85.5
83.4
NG
NG
NG
97.5
96.9
NG
NG
9.3
9.3
10.1
10.4
8.9
9.4
9.1
9.3
9.1
9.1
8.9
9.9
10.1
10.0
9.2
9.0
8.5
9.2
8.9
11.1
11.3
NG
NG
NG
9.1
8.8
Therapy
Studied
Background
Therapy
Dropout
Rate, %
Control
10 g EBID
MET SFU
Biphasic INS Asp once daily

EBID
Biphasic INS Asp BID
EBID
10 g EBID
MET TZD
EBID
10 mg/wk TASPO
20 mg/wk TASPO
10 g EBID
SFU
PBO
5 g EBID
PBO
PBO
10 g EBID
MET
PBO
5 g EBID
PBO
PBO
10 g EBID at breakfast, MET SFU
dinner
TZD
10 g EBID at lunch, dinner
EBID at breakfast,
dinner
10 g EBID
MET
INS Aspart
10 g EBID
MET SFU
PBO
PBO
10 g EBID
MET SFU
INS
INS glargine
EBID
PBO
SFU BG or
2.5 g EBID
PBO
SFU TZD
5.0 g EBID
PBO
10 g EBID
PBO
10 g EBID
SFU, SFU
PBO
PBO
5 g EBID
BG, or SFU
PBO
TZD
10 g EBID
MET SFU
PBO
5 g EBID
PBO
PBO
10 g EBID
TZD MET
PBO
PBO
10 g EBID
D/E
PBO
5 g EBID
PBO
PBO
10 g EBID
MET SFU
INS
Biphasic INS Asp
BID
10 g EBID
SFU MET
PBO
PBO
5 g EBID
TZD
PBO
10 g EBID
TZD MET
PBO
PBO
29.8
16.1
19.4
NG
NG
NG
29.5
24.0
39.8
17.7
18.2
21.2
11.1
16.0
25.4
20.8
17.5
10.3
19.1
9.4
2.5
8.1
10.8
16.2
27.4
9.7
5.6
17.8
15.9
23.9
14.0
7.0
13.0
14.0
12.0
21.3
10.1
NG
NG
NG
28.9
14.3
(continued)
1258.e6
Supplemental Table II. Studies included in the present meta-analysis of the efficacy of the glucagon-like peptide 1 receptor agonists (GLP-1RAs) and
dipeptidyl-peptidase IV (DPP-4) inhibitors in type 2 diabetes mellitus.
June 2012
Supplemental Table II (continued).

Study
Study
ITT
GLP-1RA: Exenatide QW
III
Bergenstal et al
(2010)17
26
160
166
165
123
129
248
246
163
163
113
100
147
148

Baseline
Female, %
White, %
44
48
52
45
40
NG
NG
NG
NG
48
45
49.0
45.0
33
30
39
55
63
NG
NG
NG
NG
82
85
73.0
83.0
6
5
6
7
7
NG
NG
NG
NG
8.0
8.3
6
7
8.6
8.5
8.5
8.4
8.5
NG
NG
NG
NG
8.3
8.3
8.3
8.3
89
87
88
94.3
97.0
NG
NG
NG
NG
91.2
90.6
102
102
Therapy
Studied
Background
Therapy
9.2
9.1
9.1
9.3
9.6
NG
NG
NG
NG
9.9
9.7
9.2
9.6
2 mg EQW
100 mg SITA once daily
45 mg PIO
10 g EBID
2 mg EQW
2 mg EQW
2.5 mg MET once daily
45 mg PIO once daily
2 mg EQW
INS glargine
10 g EBID
2 mg EQW
MET
D/E or MET
SFU TZD
D/E
Control
Dropout
Rate, %
EBID
INS Glargine
EBID
21
13
21
22.8
15.5
NG
NG
NG
NG
10.3
6.3
11.6
13.5
PBO
PBO
0
0
14.3
III
24
III
26
Diamant et al
(2010)20
Drucker et al
(2008)21
III
26
III
30
Kim et al
(2007)22
II
15
15
16*
14
33.0
25.0
64.0
60.0
56.0
64.0
4
5
4
8.3
8.6
8.6
110
107
101
9.3
10.3
10.2
2 mg EQW
0.8 mg EQW
PBO
20
26
28
27
29
43
23
43
33
80
94
77
93
4.7
6.5
6.8
5.9
8.6
8.4
8.6
8.6
92
101
94
101
10.4
10.1
10.0
10.4
5 mg EQM
8 mg EQM
11 mg EQM
2 mg EQW
D/E or MET or
PIO or MET
PIO
EQW
EQW
EQW
10
7
7
0
26
233
231
251
247
248
88
88
88
30
28*
29
26*
26*
25*
27*
234
228
233
114
231
51.0
45.0
53.0
51.0
46.0
40
33
35
33.3
41.4
31.0
37.0
34.6
36.0
33.3
47.0
55.0
46.0
53.0
53.0
93.0
91.0
80.0
75.0
77.0
0
0
0
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
8.5
7.9
5.2
5.3
5.6
9.3
11.6
10.1
4.6
6.1
3.4
3.4
NG
NG
NG
6.5
6.7
6.5
6.5
6.6
8.2
8.1
8.3
8.3
8.4
8.6
8.21
8.45
7.4
7.4
7.4
7.4
NG
NG
NG
8.5
8.5
8.4
8.4
8.4
9.8
9.5
9.3
9.5
9.5
9.85
9.16
9.48
10.8
9.9
9.7
NG
NG
NG
NG
9.7
9.8
10.8
9.5
9.9
1.8 mg LIRA once daily

10 g EBID
8 mg GLIM once daily
PBO
PBO
1-4 mg/d GLIM
PBO
4 mg ROSI
METSFU
EBID
LIRA
GLIM
GLIM
PBO
PBO
PBO
PBO
PBO
PBO
PBO
PBO
PBO
PBO
PBO
LIRA
14.2
19.5
35.5
30.1
38.7
5.7
4.5
15.9
6.7
7.1
17.2
0
11.5
12
25.9
9
14
10.7
27.2
16
GLP-1RA: Exenatide QM
MacConell et al II
(2011)23
GLP-1RA: Liraglutide
III
Buse et al
(2009)24
III
Garber et al
(2009)25
52
Kaku et al
(2010)26
III
24
Madsbad et al
(2004)27
II
12
Marre et al
(2009)28
III
26
1258.e7
93.1
93
92.5
92.8
93.4
66.1
64.5
66.7
NG
NG
NG
NG
NG
NG
NG
81.6
81.6
81.6
81.6
81.6
MET SFU
D/E, MET,
SFU, TZD or 2
OADs
D/E MET
D/E
SFU
D/E
SFU
(continued)
V.R. Aroda et al.
Blevins et al
(2011)18
Cuddihy et al
(2011)19
Study
Study
ITT
Nauck et al
(2009)29
III
26
Pratley et al
(2010)30
III
26
Russell-Jones
et al (2009)31
III
26
Yang et al
(2011)32
III
16
Zinman et al
(2009)33
III
26
DPP-4 inhibitor: Alogliptin

III
DeFronzo et al
(2008)34
Fleck et al
(2009)35
III
26
26
Nauck et al
(2009)36
III
26
Pratley et al
(2009)37
III
26
Pratley et al
(2009)38
III
26
Volume 34 Number 6
DPP-4 inhibitor: Linagliptin

III
Del Prato et al
(2011)39
II
Forst et al
(2010)40
24
12
242
241
242
244
122
225
221
219
232
115
234
231
233
234
231
178
178
177
133
131
64
164
163
164*
Female, %
White, %
41.0
46.0
38.0
43.0
40.0
48
48
45
43.0
51.0
40.0
45.9
45.1
46.2
41.6
49.0
43.0
38.0
88.0
88.0
84.0
89.0
88.0
82
87
91
NG
NG
NG
0
0
0
0
83.0
81.0
84.0
46.8 (overall) 66.9 (overall)

NG
NG
NG
NG
NG
NG
NG
NG
NG
NG

Baseline
8
7
7
8
8
6.0
6.4
6.3
9.2
9.4
9.7
7.4
7.5
7.2
7.8
9
9
9
8.4
8.3
8.4
8.4
8.4
8.4
8.4
8.5
8.3
8.3
8.2
8.5
8.6
8.6
8.5
8.6
8.5
8.4
88.6
88.6
88.6
88.6
88.6
93.7
94.6
93.1
85.5
85.7
85
68.6
67.4
68.2
68.2
NG
NG
NG
10.1
9.9
10.2
10.0
10.0
10.1
9.9
10.0
9.1
9.4
9.1
9.8
9.5
9.9
9.6
10.3
10.1
10.0
NG
NG
NG
3.2P
3.2P
3.2P
NG
NG
NG
8.8P
8.8P
8.8P
NG
NG
NG
NG
NG
NG
9.6
9.5
9.7
10.6P
10.6P
10.6P
164*
NG
NG
3.2P
8.8P
NG
10.6P
213
210
104
203
198
99
97
197
199
52.6
45.7
52.2
45.3
50.0
48.5
45.4
44.7
37.2
80.0
76.0
76.0
69.5
71.2
72.7
73.2
72.6
76.4
6
6
6
7.8
7.6
7.7
7.8
7.7
7.4
7.9
7.9
8.0
8.1
8.1
NG
8
8.1
8.0
NG
NG
NG
NG
NG
NG
NG
NG
NG
9.4
9.5
9.9
NG
NG
NG
9.5
9.5
9.5
336
167
65*
66
66*
65*
71
51.2
52.7
44.6
43.9
47.0
36.9
38.0
53.6
53.9
99
100
99
99
97
NG
NG
6.9
7.3
8.2
6.7
6.2
8.0
8.0
8.2
8.5
8.4
8.2
8.4
78.5
79.2
92.5
90.7
89.9
90.5
93.1
9.1
9.2
10.1
10.5
10.5
10.0
10.3
Therapy
Studied
Background
Therapy

PBO
PBO
INS glargine
PBO
MET
MET
MET SFU
MET
MET TZD
12.5 mg ALO once daily

D/E
25 mg ALO once daily
PBO
D/E
PIO 30 mg once daily
30 mg PIO/d
25 mg ALO once daily 30
mg PIO/d
MET
PBO
SFU
PBO
PBO PIO
SFU or MET
12.5 mg ALO PIO
25 mg ALO PIO
5 mg LIN once daily
PBO
1 mg LIN once daily
5 mg LIN once daily
10 mg LIN once daily
1-3 mg GLIM once daily
PBO
D/E
MET
Control
Dropout
Rate, %
PBO
PBO
PBO
LIRA
SITA
SITA
PBO or INS
LIRA
GLIM
GLIM
GLIM
PBO
PBO
21.1
18.3
14.0
13.9
39.3
24.9
13.6
11.4
10.0
16.0
6.0
12.6
23.2
26.1
8.7
14.0
7.8
18.0
PBO
PBO
PIO
21.7 (overall)
NG
NG
NG
NG
NG
NG
PBO
PBO
PBO
PBO
PBO PIO
PBO PIO
8.0
13.3
6.7
24.6
25.3
37.4
14.4
12.7
10.6
PBO
PBO
PBO
PBO
10.1
17.4
20.0
15.2
9.1
6.2
19.7
(continued)
1258.e8
June 2012

Study
Study
ITT
White, %
764
755
259
130
159
160
80
159
160
162
161
84
793
265
524
177
NG
NG
41.3
34.6
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
47
43
0
0
74.5
74.6
0
0
0
0
0
0
NG
NG
NG
NG
75
79
24
248
253
267
54.4
56.5
53.9
59.7
59.7
56.9
192
191
181*
179
138
144
428
430
67
55
47
63*
41*
44*
54*
52*
102
106
95
98*
403
398
56.8
46.1
47.5
46.4
NG
NG
50.5
46.0
37.0
60.0
47.0
37.0
44.0
39.0
30.0
42.0
43.1
49.1
50.5
54.1
52.9
49.2
79.7
83.2
79.6
83.8
NG
NG
82.2
84.2
87.0
85.0
87.0
84.0
80.0
77.0
87.0
92.0
87.3
87.7
83.2
81.6
67.7
65.1
Gallwitz et al
(2011)41
Gomis et al
(2011)42
Kawamori et al
(2010)43
III
104
III
24
III
12
Kawamori et al
(2010)44
III
26
Lewin et al
(2010)45
Owens et al
(2010)46
Taskinen et al
(2011)47
III
18
III
24
III
24
DPP-4 inhibitor: Saxagliptin

III
Chacra et al
(2009)48
DeFronzo et al
(2009)49
III
24
Fonseca et al
(2011)50
Gke et al
(2010)51
Rosenstock
et al (2008)52
III
18
IIIb
52
III
12
6
12
1258.e9
Rosenstock
et al (2009)53
III
24
Scheen et al
(2010)54
IIIb
18

Baseline
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
Therapy
Studied
Background
Therapy
MET
NG
NG
8.6
8.58
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
8.09
8.02
NG
NG
78.3
82.7
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
82.2
83.3
NG
NG
10.5
10.6
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
9.4
9.2
5 mg LIN once daily

1-4 mg D GLIM
5 mg LIN once daily
PBO
5 mg LIN once daily
PBO
5 mg LIN once daily
0.2 mg TID VOGL
5 mg LIN once daily
PBO
5 mg LIN once daily
PBO
5 mg LIN once daily
PBO
7.1
6.8
6.8
8.4
8.5
8.4
75.2
76.2
75.6
9.4
9.7
9.7
6.7
6.4
6.3
6.7
8.1
8.1
8
8.1
NG
NG
7.7
7.7
8
7.7
7.9
8
7.5
7.8
7.9
7.8
7.9
8
7.9
7.8
7.7
7.7
86
87.3
87.8
87.1
NG
NG
88.7
88.6
93.1
86.6
89.8
92.4
91.2
92.2
88.9
86.8
92.1
90.87
86.56
89.3
NG
NG
9.7
10.0
9.8
9.7
NG
NG
9.0
8.9
9.1
8.6
9.4
9.4
8.0
8.5
9.6
8.8
9.9
9.5
9.5
9.9
8.86
8.89
2.5 mg SAXA once daily

5 mg SAXA once daily
PBO 7.5 mg GLY once
daily
PBO
MET
5-20 mg GLIP once daily
PBO for low dose
PBO for high dose
PBO
NG
NG
5.5
5.4
1.8median
1.0median
0.8median
0.7median
0.3median
0.5median
1.7median
1.3median
3.1
2.5
2.3
2.3
6.3
6.3
Control
Dropout
Rate, %
GLIM
PBO
PBO
PBO
VOGL
VOGL
PBO
PBO
PBO
NG
NG
5.8
14.6
NG
NG
NG
NG
NG
NG
NG
NG
NG
NG
7.5
7.9
GLY
PBO GLY
PBO GLY
22.6
22.9
34.1
MET
MET
PBO
PBO
PBO
MET
GLIP
PBO
PBO
PBO
PBO
PBO
PBO
PBO
PBO
PBO
22.9
25.1
22.7
37.4
NG
NG
27.1
25.3
17.9
12.7
21.3
14.3
14.6
0
20.4
13.5
28.4
35.8
42.1
29.6
9.4
6.0
PIO
D/E
D/E
SFU
MET SFU
MET
D/E
MET
D/E
D/E
D/E
MET
(continued)
V.R. Aroda et al.
Female, %
Study
Study
ITT
DPP-4 inhibitor: Sitagliptin

III
Arechavaleta
et al (2011)55
III
Aschner et al
(2006)56
30
24
Aschner et al
(2010)57
Bergenstal et al
(2010)58
III
24
III
24
Charbonnel
et al (2006)59
Fonseca et al
(2011)60
Goldstein et al
(2007)61
III
24
III
26
III
24
Volume 34 Number 6
Hanefeld et al
(2007)62
II
24
Hermansen
et al (2007)63
III
24
Iwamoto et al
(2010)64
III
12

Baseline
Female, %
White, %
516
519
238
250*
253
528
522
90
177
182*
187*
464
237
NG
NG
182*
182*
190*
45.0
46.2
42.9
53.2
48.6
52.0
56.0
NG
NG
NG
NG
44.2
40.5
NG
NG
51.1
54.9
44.7
57.6
57.4
51.3
52.8
50.2
NG
NG
NG
NG
NG
NG
63.1
67.1
NG
NG
47.8
58.2
53.7
6.8
6.7
4.4P
4.4P
4.4P
2.6
2.1
NG
NG
NG
NG
6.2P
6.2P
NG
NG
4.5
4.5
4.5
7.5
7.5
8.01
8.08
8.03
7.2
7.2
8.03
7.94
7.55
7.57
7.96
8.03
NG
NG
8.9
8.68
8.79
80.6
82.0
NG
NG
NG
NG
NG
91.1
92.5
93.6
91.8
NG
NG
NG
NG
NG
NG
NG
8.0
8.1
9.5
9.7
9.8
7.9
7.9
NG
NG
NG
NG
9.4
9.6
NG
NG
11.4
10.9
11.3
182*
57.7
52.2
4.5
8.76
NG
10.9
176
179
111
110
110
111
111
222
219
106*
116*
106*
113*
80
72
70
68*
73
47.2
48.0
36.9
54.5
44.5
55.9
48.6
47.3
46.6
47.2
47.4
45.3
47.8
31.5
36.3
34.7
48.6
41.2
46.0
52.0
78.4
85.7
88.2
81.1
88.3
61.3
63.9
57.5
64.7
55.7
71.7
0
0
0
0
0
4.5
4.5
3.3
3.3
3.6
4.5
3.6
8.3
9.3
7.2
9.3
8
10.6
6.4
4.7
5.6
5.4
5.1
8.68
8.87
7.59
7.6
7.78
7.79
7.71
8.34
8.34
8.42
8.27
8.43
8.26
7.74
7.49
7.57
7.56
7.65
NG
NG
NG
NG
NG
NG
10.9
11.2
9.7
9.4
9.8
9.5
9.7
10.0
10.1
10.1
10.0
10.3
9.9
8.7
8.1
8.0
7.9
8.2
86.5
85.9
85.8
87.2
85.1
86.7
NG
NG
NG
NG
NG
Therapy
Studied
Background
Therapy

1 mg once daily GLIM
PBO
1g BID MET
PBO
10 mg/wk TASPO
20 mg/wk TASPO
PBO
PBO
500 mg BID MET
1 g BID MET
50 mg BID SITA 500 mg
BID MET
50 mg BID SITA 1 g BID
MET
PBO
PBO
50 mg BID SITA
PBO
PBO
PBO
PBO
MET
D/E
D/E
MET
MET
MET PIO
D/E
D/E
SFU MET
SFU MET
SFU
SFUMET
SFU
SFU MET
D/E
Control
Dropout
Rate, %
GLIM
PBO
PBO
MET
PBO
PBO, SITA
PBO, SITA
PBO
PBO
PBO
PBO
PBO
9.3
9.8
12.2
PBO
9.9
PBO
PBO
PBO
PBO
PBO
PBO
PBO
PBO
PBO
PBO
PBO
PBO
27.8
20.7
27.0
5.5
16.4
9.9
13.6
16.7
18.3
21.7
12.1
17.9
18.6
3.8
1.4
2.9
1.5
6.8
14.7
11.6
14.4
NG
NG
NG
NG
10.4
19.0
NG
NG
15.9
14.3
13.7
(continued)
1258.e10
June 2012

Study
Study
ITT
Mohan et al
(2009)65
Nauck et al
(2007)66
Nonaka et al
(2008)67
Raz et al
(2006)68
III
18
III
52
II
12
III
18
Raz et al
(2008)69
Rosenstock
et al (2006)70
Rosenstock
et al (2011)71
III
30
III
24
II
12
Scott et al
(2008)72
III
18
Seck et al
(2010)73
Yang et al
(2011)74
III
104
III
24
DPP-4 inhibitor: Vildagliptin

III
Bolli et al
(2008)75
III
Bosi et al
(2007)76
Bosi et al
(2009)77
II
24
24
24

Baseline
Background
Therapy
10.5
10.5
8.8
8.8
9.1
9.1
10.2
10.0
10.2
11.0
11.2
9.3
9.2
NG
NG
NG
NG
NG
NG
NG
NG
8.7
8.7
8.9
8.4
8.5
NG
NG

PBO
520 mg/d GLIP
100 mg SITA
PBO
PBO
PBO
PBO
PBO1
1 mg BI once daily
5 mg BI once daily
10 mg BI once daily
25 mg BI once daily
50 mg BI once daily
PBO2
100 mg ROSI once daily
PBO
5-20mg GLIP once daily
PBO
D/E
91.6
92
92.5
95.3
94.8
90
11.0
11.0
9.7
9.9
10.1
10.8
8.6
87
10.2
8.7
8.6
88
88
10.3
10.5
50 mg BID VILDA
30 mg/d PIO
50 mg VILDA once daily
50 mg BID VILDA
PBO
50 mg BID VILDA 1000
mg BID MET
50 mg BID VILDA 500 mg
BID MET
50 mg BID VILDA
1000 mg MET once daily
White, %
352
178
588
584
75
76
103
193
199*
94
96
175
178
NG*
NG*
NG*
NG*
NG*
NG*
NG
NG
94
87
92
248
256
NG
NG
43.0
40.0
42.9
38.7
40.0
34.0
37.3
46.3
49.5
58.5
49.0
46.9
42.1
NG
NG
NG
NG
NG
NG
NG
NG
45
37.0
41.0
42.7
37.1
NG
NG
0
0
73.5
74.3
0.0
0.0
61.8
69.3
70.9
47.0
42.0
72.6
72.5
NG
NG
NG
NG
NG
NG
NG
NG
61.0
59.0
61.0
77.4
78.5
0
0
2.1
1.9
6.5
6.2
4
4.1
4.5P
4.5P
4.5P
7.3
8.4
6.1
6.1
NG
NG
NG
NG
NG
NG
NG
NG
4.9
4.6
5.4
5.8
5.7
NG
NG
8.7
8.8
7.48
7.52
7.54
7.69
8.05
8.04
8.14
9.1
9.3
8.05
8
NG
NG
NG
NG
NG
NG
NG
NG
7.75
7.73
7.68
7.3
7.3
NG
NG
66.8
66.6
89.5
89.7
NG
NG
NG
NG
NG
81.2
81.5
90.9
86.4
NG
NG
NG
NG
NG
NG
NG
NG
83.1
84.9
84.6
88.5
90.3
NG
NG
295
281
143
143
130
292*
35.9
38.3
42.7
38.5
46.9
42.0
81.9
82.4
74.1
74.1
73.1
73.9
6.4
6.4
6.8
5.8
6.2
1.9
8.6
8.5
8.4
8.4
8.3
8.7
290*
44.1
73.1
1.9
297
292
40.0
41.8
75.0
72.1
2.1
2.2
MET
D/E
D/E
MET
PIO
MET
MET
MET
MET
MET
MET
D/E
Control
Dropout
Rate, %
PBO
GLIP
SITA
PBO
PBO
PBO
PBO
PBO
PBO1
PBO1
PBO1
PBO1
PBO1
PBO2
PBO
PBO
GLIP
PBO
13.0
25.0
34.4
29.5
2.7
11.8
17.3
8.3
10.7
14.9
17.7
14.9
11.2
NG
NG
NG
NG
NG
NG
NG
NG
10.0
2.0
9.0
56.9
54.2
NG
NG
PIO
VILDA
PBO
PBO
MET
11.2
13.2
13.6
15.1
16.5
11.9
MET
15.5
MET
18.3
17.2
(continued)
V.R. Aroda et al.
1258.e11
Therapy
Studied
Female, %
1258.e12
Study
Study
ITT
Female, %
White, %

Baseline
Therapy
Studied
Background
Therapy
Control
Dropout
Rate, %
Filozof and
Gautler
(2010)78
Garber et al
(2007)79
III
52
513
494
47.8
48.2
78.9
77.5
6.4
6.8
8.5
8.5
85.7
84.2
10.8
10.6
50 mg BID VILDA
320 mg GLIC once daily
MET
GLIC
20.7
16.6
III
24
III
24
45.2
55.1
49.3
40.9
40.2
41.7
83.9
79.4
78.3
68.9
70.5
67.4
4.7
4.6
4.8
6.9
6.7
7.8
8.6
8.7
8.7
8.5
8.6
8.5
NG
NG
NG
91.5
87.3
89.4
10.3
10.0
10.1
10.5
10.5
10.3

50 mg BID VILDA
PBO
50 mg BID VILDA
PBO
TZD
Garber et al
(2008)80
124
136
138
132
132
144
PBO
PBO
PBO
PBO
15.6
21.5
19.0
14.4
15.9
25.0
SFU
SFU
SFU
ALO alogliptin; Asp aspart; BG biguanide; BI SGLT-2 inhibitor BI-10773; D/E diet and exercise; EBID exenatide BID; EQM exenatide once monthly; EQW exenatide once weekly; GLIC gliclazide;
GLIM glimepiride; GLIP glipizide; GLY glyburide; INS insulin; LIN linagliptin; LIRA liraglutide; MET metforming; NG not given; OAD oral antidiabetes drug; P pooled; PBO, placebo; PIO
pioglitazone; PBO placebo; QM once monthly; QW once weekly; ROSI rosiglitazone; SAXA saxagliptin; SFU any sulfonylurea; SITA sitagliptin; TASPO taspoglutide; TZD any thiazolidinedione;
VILDA vildagliptin; VOGL voglibose.
*Data row excluded from analysis.
Study excluded from analysis of highest maintenance dose.
Volume 34 Number 6
June 2012
Supplemental Table III. Comparison of designs of the studies included in the present meta-analysis of the efficacy of the glucagon-like peptide 1
receptor agonists (GLP-1RAs) and dipeptidyl-peptidase IV (DPP-4) inhibitors in type 2 diabetes mellitus.
Study
GLP-1RA: Exenatide BID
Bergenstal et al
(2009)1
Bolli et al (2010)2
Buse et al (2004)3
DeFronzo et al
(2005)4
Forti et al (2008)5
Gallwitz et al
(2011)6
Gao et al (2009)7
Heine et al (2005)8
Kadowaki et al
(2009)9
Kadowaki et al
(2011)10
Kendall et al
(2005)11
Liutkus et al
(2010)12
Moretto et al
(2008)13
Nauck et al
(2007)14
Sowa et al (2010)15
Zinman et al
(2007)16
Differences in
Baseline
Data?
Missing
Baseline Data
Double-Blind?
N
N
N
N
N
N
Y
N
N
N
N
N
N
N
N
N
Multi-National?
(Country or No.
of Countries)
Superiority or
Noninferiority Trial?
Missing
Weight Data
Analysis Population
Data Type
N (US)
Both
ITT & PP
Mean
N
Y
Y
Y (NG)
N (US)
N (US)
NI
S
S
N
N
N
ITT
ITT & evaluable
ITT & evaluable
LS mean
Mean
Mean
N
Y
Y
N
Y (2)
N (Germany)
NI
NI
N
N
PP
ITT
LS mean
LS mean
N
N
N
N
N
N
Y
N
N
Y (4)
Y (13)
N (Japan)
S
NI
S
N
N
N
ITT & PP
ITT & PP
ITT
LS mean
LS mean
Mean
N (Japan)
ITT
LS mean
N (US)
ITT & evaluable
Mean
Y (5)
ITT
LS mean
Y (23)
ITT
LS mean
Y (race)
Y (13)
NI
ITT & PP
Mean
N
N
N
N
Y
N
Y
Y
N (Japan)
Y (3)
S
S
Y
N
ITT
ITT & PP
LS mean
Mean
Y (3)
ITT
LS mean
N (US)
NI
ITT
LS mean
Y (NG)
ITT
LS mean
Y (7)
ITT
LS mean
NG
NI
ITT & evaluable
LS mean
Y (sex)
N (US)
ITT
Mean
(continued)
V.R. Aroda et al.
1258.e13
GLP-1RA: Exenatide QW
Bergenstal et al
(2010)17
Blevins et al
(2011)18
Cuddihy et al
(2011)19
Diamant et al
(2010)20
Drucker et al
(2008)21
Kim et al (2007)22
Washout/Rx
Change Prior to
Randomization?
Washout/Rx
Change Prior to
Randomization?
Differences in
Baseline
Data?
Missing
Baseline Data
Double-Blind?
Multi-National?
(Country or No.
of Countries)
NG
N
N
N
N
N
N
N
Y
Y (15)
Y (2)
N
Y (4-wk OAD washout)
N
N
N
Y (race)
Y
Y
N (Japan)
Y (2)
Y (24-wk transition to
GLIM)
MET)
N
Y (race)
MET & GLIM)
Y
Y (69-wk PIO to ROSI,
MET)
Missing
Weight Data
Analysis Population
Data Type
ITT & evaluable
LS mean
Both
Both
N
N
ITT & PP
ITT
Mean
Mean
S
S
N
N
ITT & PP
ITT
Mean
Adj mean
N (US)
Both
ITT
LS mean
Y (21)
Both
ITT
Mean
Y (13)
Both
ITT & PP
Mean
Y (race)
Y (17)
Both
ITT
Mean
N
N
N
Y (weight)
Y
Y
Y (3)
Y (2)
Both
S
Y
N
ITT
ITT
Mean
Mean
DPP-4 inhibitor: Alogliptin

N
DeFronzo et al
(2008)34
Y (16)
ITT
LS mean
Y (weight,
FPG,
diabetes
duration)
N
Y (NG)
ITT
LS mean
Y (15)
ITT
LS mean
N (US)
ITT
LS mean
Y (transition from ROSI

to PIO)
Y (4)
ITT
LS mean
Study
GLP-1RA: Exenatide QM
MacConell et al
(2011)23
GLP-1RA: Liraglutide
Buse et al (2009)24
Garber et al
(2009)25
Kaku et al (2010)26
Madsbad et al
(2004)27
Marre et al
(2009)28
Nauck et al
(2008)29
Pratley et al
(2010)30
Russell-Jones et al
(2008)31
Yang et al (2011)32
Zinman et al
(2009)33
Volume 34 Number 6
Fleck et al
(2009)35*
Nauck et al
(2009)36
Pratley et al
(2009)37
Pratley et al
(2009)38
Superiority or
(continued)
1258.e14
Supplemental Table III (continued).
June 2012
Study
DPP-4 inhibitor: Linagliptin
Del Prato et al
(2011)39
Forst et al
(2010)40
Gallwitz et al
(2011)41
Gomis et al
(2011)42
Kawamori et al
(2010)43
Kawamori et al
(2010)44
Lewin et al
(2010)45
Owens et al
(2010)46
Taskinen et al
(2011)47
Washout/Rx
Change Prior to
Randomization?
Differences in
Baseline
Data?
Missing
Baseline Data
Double-Blind?
Multi-National?
(Country or No.
of Countries)
Superiority or
Missing
Weight Data
Analysis Population
Data Type
Y (11)
ITT & PP
Adj mean
Y (6)
ITT
Mean
Y (NG)
NI
ITT & PP
Adj mean
Y (7)
ITT
Adj mean
N (Japan)
ITT
Adj mean
N (Japan)
ITT
Adj mean
NG
NI
ITT
Adj mean
NG
ITT
Adj mean
Y (10)
ITT
Adj mean
Y (NG)
ITT
Adj mean
Y (9)
PP
NG
ITT
LS mean & Adj

mean
Adj mean
N
N
N
N
Y
Y
Y (11)
N (US)
NI
S
N
N
PP
Modified ITT
Adj mean
Adj mean
Y (weight,
BMI, sex)
N
NG
ITT
Adj mean
Y (9)
NI
ITT & PP
Adj mean
(continued)
1258.e15
V.R. Aroda et al.
DPP-4 inhibitor: Saxagliptin

Chacra et al
GLY)
(2009)48
DeFronzo et al
N
(2009)49
N
Fonseca et al
(2011)50
Gke et al (2010)51
N
N
Rosenstock et al
(2008)52
Rosenstock et al
N
(2009)53
Scheen et al
N
(2010)54
Study
Washout/Rx
Change Prior to
Randomization?
Volume 34 Number 6
DPP-4 inhibitor: Sitagliptin

N
Arechavaleta et al
(2011)55
Aschner et al
Y (212-wk OAD
(2006)56
washout)
Aschner et al
N
57
(2010)
Bergenstal et al
N
(2010)58
Charbonnel et al
Y (219-wk OAD
(2006)59
washout)
Fonseca et al
N
(2011)60
Goldstein et al
Y (212-wk OAD
(2007)61
washout)
Hanefeld et al
Y (26-wk OAD
(2007)62
washout)
Hermansen et al
Y (210 wk OAD
63
(2007)
washout)
Iwamoto et al
Y
(2010)64
Mohan et al
Y (36-wk OAD
(2009)65
washout)
Nauck et al
Y (8 wk transition to
(2007)66
MET)
Nonaka et al
Y (28-wk OAD
67
(2008)
washout)
Raz et al (2006)68
Y (212-wk OAD
washout)
Raz et al (2008)69
MET)
Rosenstock et al
Y (214-wk OAD
(2006)70
washout)
Rosenstock et al
N
71
(2011)
Scott et al (2008)72
N
Y
Seck et al (2010)73
Y
Yang et al (2011)74
Differences in
Baseline
Data?
Missing
Baseline Data
Double-Blind?
Multi-National?
(Country or No.
of Countries)
Superiority or
Missing
Weight Data
Analysis Population
Data Type
NG
NI
ITT & PP
LS mean
Y (18)
ITT (all patients treated)
LS mean
Y (23)
NI
PP
LS mean
Y (NG)
ITT
Mean
Y (25)
LS mean
NG
ITT
Mean
Y (17)
LS mean
Y (sex)
Y (NG)
Mean
Y (NG)

& PP
LS mean
N (Japan)
ITT
LS mean
Y (3)
Full analysis set
LS mean
Y (31)
NI
PP
LS mean
N (Japan)
LS mean
Y (7)
LS mean
Y (NG)
Full analysis set
LS mean
Y (15)
LS mean
Y (NG)
ITT
Mean
N
N
Y
N
N
Y
Y
Y
Y
Y (6)
NG
N (China)
S
NI
S
N
N
Y

PP
ITT
LS mean
LS mean
LS mean
(continued)
1258.e16
June 2012
Study
DPP-4 inhibitor: Vildagliptin
Bolli et al (2008)75
Bosi et al (2007)76
Bosi et al (2009)77
Filozof and Gautier
(2010)78
Garber et al
(2007)79
Garber et al
(2008)80
Washout/Rx
Change Prior to
Randomization?
Differences in
Baseline
Data?
Missing
Baseline Data
N
N
N
N
N
N
N
N
N
N
Double-Blind?
Multi-National?
(Country or No.
of Countries)
Superiority or
Missing
Weight Data
Analysis Population
Data Type
N
N
N
N
Y
Y
Y
Y
Y (9)
Y (4)
Y (5)
NG
NI
S
S
NI
N
N
N
N
PP
ITT
ITT
ITT & PP
Adj mean
Adj mean
LS mean
Mean
Y (2)
ITT
Adj mean
Y (5)
ITT
Adj mean
Adj adjusted; BMI body mass index; FPG fasting plasma glucose; GLIM glimepiride; GLY glyburide; ITT intent to treat; LS least squares; MET metformin; NG not given; NI noninferiority; N
No; OAD oral antidiabetes drug; P pooled; PIO pioglitazone; PP per protocol; QM once monthly; QW once weekly; ROSI rosiglitazone; S, superiority; Y yes.
*Abstract; no article published to date.
V.R. Aroda et al.
1258.e17
Supplemental Table IV. Analysis of heterogeneity (%) of the studies included in the present meta-analysis of the
glucagon-like peptide 1 receptor agonists (GLP-1RAs) and dipeptidyl-peptidase IV
(DPP-4) inhibitors in type 2 diabetes mellitus (based on I2 statistics).
HbA1c Heterogeneity
Drug
FPG Heterogeneity
Body Weight Heterogeneity
Highest
Maintenance doses
All
Labeled Doses
Highest
Maintenance Doses
All
Labeled Doses
Highest
Maintenance Doses
All
Labeled Doses
GLP-1RAs
Exenatide QW
Exenatide BID
Liraglutide
once daily
57.12
88.54
57.12
93.14
0
81.17
0
86.71
61.39
92.46
61.39
93.69
81.74
91.51
63.51
91.02
92.27
97.21
DPP-4 inhibitors
Alogliptin
Linagliptin
Saxagliptin
Sitagliptin
Vildagliptin
63.63
93.99
83.85
89.10
98.30
76.58
93.99
83.85
89.10
98.16
70.48
96.24
80.90
68.55
97.04
86.15
96.24
80.90
68.55
97.46
90.59
NG
84.08
92.42
93.75
94.73
NG
84.08
92.42
91.70
FPG fasting plasma glucose; HbA1c hemoglobin A1c; NG not given.
1258.e18
Volume 34 Number 6
V.R. Aroda et al.
GLP-1RAs
DPP-4 Inhibitors
Exenatide BID
Alogliptin
1.75 [2.08 to 1.42]
0.90 [1.10 to 0.70]
0.98 [1.14 to 0.83]
0.86 [1.08 to 0.64]
0.79 [0.95 to 0.63]
0.78 [0.98 to 0.58]
1.50 [1.70 to 1.30]
1.15 [1.29 to 1.01]
1.00 [1.19 to 0.81]
1.20 [1.30 to 1.10]
1.11 [1.25 to 0.97]
1.40 [1.00 to 1.20]
1.62 [1.84 to 1.40]
0.77 [0.93 to 0.61]
0.84 [1.23 to 0.45]
0.90 [1.10 to 0.70]
1.04 [1.18 to 0.90]
1.60 [1.80 to 1.40]
0.89 [1.07 to 0.71]
1.10 [1.22 to 0.99]
Bergenstal 2009
Blevins 2011
Bolli 2010
Buse 2004
Buse 2009
DeFronzo 2005
Drucker 2008
Forti 2008
Gallwitz 2011a
Gao 2009
Heine 2005
Kadowaki 2009
Kadowaki 2011
Kendall 2005
Liutkus 2010
Moretto 2008
Nauck 2007a
Sowa 2010
Zinman 2007
All
3.0
2.5
2.0
1.5
1.0
0.5
0.0
HbA1c Change (%)

0.59 [0.84 to 0.34]
0.96 [1.18 to 0.74]
0.60 [0.80 to 0.40]
0.52 [0.72 to 0.32]
0.80 [1.00 to 0.60]
0.69 [0.85 to 0.54]
DeFronzo 2008
Fleck 2009
Nauck 2009
Pratley 2009a
Pratley 2009b
All
3.0
2.5
2.5
2.0
1.5
1.0
0.5
0.0
2.5
1.0
1.5
1.0
0.5
0.0

0.64 [0.74 to 0.54]
0.69 [0.83 to 0.55]
0.88 [1.01 to 0.75]
0.74 [0.81 to 0.67]
0.90 [1.17 to 0.63]
0.46 [0.62 to 0.30]
0.52 [0.60 to 0.44]
0.68 [0.78 to 0.57]
Chacra 2009
DeFronzo 2009
Fonseca 2011a
Gke 2010
Rosenstock 2008
Rosenstock 2009
Scheen 2010
All
2.5
2.0
1.5
1.0
0.5
0.0
HbA1c Change (%)
Sitagliptin
Buse 2009
Garber 2009
Marre 2009
Nauck 2009
Pratley 2010
RussellJones 2009
Yang 2011a
Zinman 2009
All
1.5
2.0
Saxagliptin
3.0

1.12 [1.28 to 0.96]
1.14 [1.30 to 0.99]
1.10 [1.27 to 0.93]
1.00 [1.20 to 0.80]
1.50 [1.63 to 1.37]
1.33 [1.51 to 1.15]
1.45 [1.64 to 1.26]
1.50 [1.70 to 1.30]
1.27 [1.41 to 1.13]
2.0
0.0

0.44 [0.54 to 0.34]
0.50 [0.70 to 0.30]
0.40 [0.48 to 0.32]
1.06 [1.18 to 0.94]
0.87 [1.02 to 0.72]
0.47 [0.61 to 0.33]
0.62 [0.69 to 0.55]
0.49 [0.57 to 0.41]
0.60 [0.75 to 0.46]
Liraglutide
2.5
0.5
HbA1c Change (%)
HbA1c Change (%)
3.0
1.0
Del Prato 2011

Forst 2010
Gallwitz 2011b
Gomis 2011
Kawamori 2010a
Lewin 2010
Owens 2010
Taskinen 2011
All
3.0

1.50 [1.70 to 1.30]
1.60 [1.80 to 1.40]
1.53 [1.67 to 1.39]
1.50 [1.60 to 1.40]
1.90 [2.10 to 1.70]
1.70 [2.29 to 1.11]
1.50 [1.89 to 1.11]
1.59 [1.70 to 1.48]
3.0
1.5
Linagliptin
Exenatide QW
Bergenstal 2010a
Blevins 2011
Cuddihy 2011
Diamant 2010
Drucker 2008
Kim 2007
MacConell 2011
All
2.0
HbA1c Change (%)
0.5
HbA1c Change (%)
0.0

0.47 [0.55 to 0.39]
0.61 [0.74 to 0.48]
0.43 [0.48 to 0.38]
0.90 [1.10 to 0.70]
0.89 [1.00 to 0.78]
0.67 [0.77 to 0.57]
1.15 [1.31 to 0.99]
0.70 [0.84 to 0.58]
0.66 [0.83 to 0.49]
0.44 [0.58 to 0.30]
0.45 [0.57 to 0.33]
0.69 [0.81 to 0.57]
0.70 [0.80 to 0.60]
0.67 [0.75 to 0.59]
0.65 [0.80 to 0.50]
0.90 [1.03 to 0.77]
0.48 [0.61 to 0.35]
1.00 [1.30 to 0.70]
0.85 [0.98 to 0.72]
0.45 [0.65 to 0.25]
0.62 [0.69 to 0.55]
0.73 [0.87 to 0.59]
0.54 [0.64 to 0.44]
0.82 [0.95 to 0.69]
0.67 [0.75 to 0.60]
Arechavaleta 2011
Aschner 2006
Aschner 2010
Bergenstal 2010a
Bergenstal 2010b
Charbonnel 2006
Cuddihy 2011
Fonseca 2011b
Goldstein 2007
Hanefeld 2007
Hermansen 2007
Iwamoto 2010
Mohan 2009
Nauck 2007b
Nonaka 2008
Pratley 2010
Raz 2006
Raz 2008
Rosenstock 2006
Rosenstock 2011
Scheen 2010
Scott 2008
Seck 2010
Yang 2011b
All
3.0
2.5
2.0
1.5
1.0
0.5
0.0
HbA1c Change (%)
Vildagliptin
0.88 [0.98 to 0.78]
0.90 [1.10 to 0.70]
1.70 [1.78 to 1.62]
0.81 [0.93 to 0.69]
1.00 [1.20 to 0.80]
1.06 [1.48 to 0.64]
Bolli 2008
Bosi 2007
Bosi 2009
Filozof 2010
Garber 2007
All
3.0 2.5 2.0 1.5 1.0 0.5
HbA1c Change (%)
0.0
Supplemental Figure 1. Changes from baseline in hemoglobin (Hb) A1c in the analysis of highest maintenance
dose in the present meta-analysis of the glucagon-like peptide 1 receptor agonists (GLP1RAs) and dipeptidyl-peptidase IV (DPP-4) inhibitors in type 2 diabetes mellitus.
June 2012
1258.e19
GLP-1RAs
DPP-4 Inhibitors
Exenatide BID
Alogliptin
Mean HbA1C Difference (95% CI)
1.75 [2.08 to 1.42]
0.90 [1.10 to 0.70]
0.98 [1.14 to 0.82]
0.66 [0.82 to 0.50]
0.79 [0.95 to 0.63]
0.61 [0.76 to 0.47]
1.50 [1.70 to 1.30]
1.15 [1.29 to 1.01]
1.00 [1.19 to 0.81]
1.20 [1.30 to 1.10]
1.11 [1.24 to 0.98]
1.30 [1.44 to 1.16]
1.47 [1.62 to 1.31]
0.66 [0.76 to 0.56]
0.84 [1.23 to 0.45]
0.80 [0.94 to 0.66]
1.04 [1.18 to 0.90]
1.45 [1.59 to 1.31]
0.89 [1.07 to 0.71]
1.05 [1.19 to 0.92]
Bergenstal 2009
Blevins 2011
Bolli 2010
Buse 2004
Buse 2009
DeFronzo 2005
Drucker 2008
Forti 2008
Gallwitz 2011a
Gao 2009
Heine 2005
Kadowaki 2009
Kadowaki 2011
Kendall 2005
Liutkus 2010
Moretto 2008
Nauck 2007a
Sowa 2010
Zinman 2007
All
3.0
2.5
2.0
1.5
1.0
0.5
0.0
HbA1C Change (%)
3.0
2.5
2.0
1.5
1.0
0.5
2.5
0.0
HbA1C Change (%)
1.5
1.0
0.5
0.0

0.44 [0.54 to 0.34]
0.50 [0.70 to 0.30]
1.06 [1.18 to 0.94]
0.40 [0.48 to 0.32]
0.87 [1.02 to 0.72]
0.47 [0.61 to 0.33]
0.62 [0.69 to 0.55]
0.49 [0.57 to 0.41]
0.60 [0.75 to 0.46]
DelPrato 2011
Forst 2010
Gomis 2011
Gallwitz 2011
Kawamori 2010a
Lewin 2010
Owens 2010
Taskinen 2011
All
2.5
2.0
1.5
1.0
0.5
0.0
Saxagliptin

0.64 [0.74 to 0.54]
0.69 [0.83 to 0.55]
0.88 [1.01 to 0.75]
0.74 [0.81 to 0.67]
0.90 [1.17 to 0.63]
0.46 [0.62 to 0.31]
0.52 [0.60 to 0.44]
0.68 [0.78 to 0.57]
Chacra 2009
DeFronzo 2009
Fonseca 2011a
Gke 2010
Rosenstock 2008
Rosenstock 2009
Scheen 2010
All
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Sitagliptin
Buse 2009
Garber 2009
Kaku 2010
Madsbad 2004
Marre 2009
Nauck 2009
Pratley 2010
RussellJones 2009
Yang 2011a
Zinman 2009
All
2.0
1.0
HbA1C Change (%)

1.12 [1.28 to 0.96]
0.99 [1.10 to 0.88]
1.51 [1.65 to 1.37]
0.70 [1.14 to 0.26]
1.10 [1.22 to 0.98]
0.90 [1.01 to 0.79]
1.37 [1.46 to 1.28]
1.33 [1.51 to 1.15]
1.31 [1.41 to 1.21]
1.50 [1.64 to 1.36]
1.21 [1.35 to 1.06]
2.5
1.5
Linagliptin
Liraglutide
3.0
2.0
HbA1C Change (%)
HbA1C Change (%)

1.50 [1.70 to 1.30]
1.60 [1.80 to 1.40]
1.53 [1.67 to 1.39]
1.50 [1.60 to 1.40]
1.90 [2.10 to 1.70]
1.70 [2.29 to 1.11]
1.50 [1.89 to 1.11]
1.59 [1.70 to 1.48]
Bergenstal 2010a
Blevins 2011
Cuddihy 2011
Diamant 2010
Drucker 2008
Kim 2007
MacConell 2011
All
3.0
3.0
Exenatide QW

0.57 [0.75 to 0.40]
0.96 [1.18 to 0.74]
0.60 [0.74 to 0.46]
0.45 [0.59 to 0.31]
0.73 [0.87 to 0.59]
0.65 [0.80 to 0.50]
DeFronzo 2008
Fleck 2009
Nauck 2009
Pratley 2009a
Pratley 2009b
All
0.5
HbA1C Change (%)
0.0

0.47 [0.55 to 0.39]
0.61 [0.74 to 0.48]
0.43 [0.48 to 0.38]
0.90 [1.10 to 0.70]
0.89 [1.00 to 0.78]
0.67 [0.77 to 0.57]
1.15 [1.31 to 0.99]
0.70 [0.84 to 0.56]
0.66 [0.83 to 0.49]
0.44 [0.58 to 0.30]
0.45 [0.57 to 0.33]
0.69 [0.81 to 0.57]
0.70 [0.80 to 0.60]
0.67 [0.75 to 0.59]
0.65 [0.80 to 0.50]
0.90 [1.03 to 0.77]
0.48 [0.61 to 0.35]
1.00 [1.30 to 0.70]
0.85 [0.98 to 0.72]
0.45 [0.65 to 0.25]
0.62 [0.69 to 0.55]
0.73 [0.87 to 0.59]
0.54 [0.64 to 0.44]
0.82 [0.95 to 0.69]
0.67 [0.75 to 0.60]
Arechavaleta 2011
Aschner 2006
Aschner 2010
Bergenstal 2010a
Bergenstal 2010b
Charbonnel 2006
Cuddihy 2011
Fonseca 2011b
Goldstein 2007
Hanefeld 2007
Hermansen 2007
Iwamoto 2010
Mohan 2009
Nauck 2007b
Nonaka 2008
Pratley 2010
Raz 2006
Raz 2008
Rosenstock 2006
Rosenstock 2011
Scheen 2010
Scott 2008
Seck 2010
Yang 2011b
All
3.0
2.5
2.0
1.5
1.0
0.5
0.0
HbA1C Change (%)
Vildagliptin
0.88 [0.98 to 0.78]
0.90 [1.10 to 0.70]
1.70 [1.78 to 1.62]
0.81 [0.93 to 0.69]
1.00 [1.20 to 0.80]
0.58 [0.78 to 0.38]
0.98 [1.37 to 0.59]
Bosi 2009
Filozof 2010
Garber 2007
Bolli 2008
Bosi 2007
Garber 2008
All
3.0 2.5 2.0 1.5 1.0 0.5
HbA1C Change (%)
0.0
Supplemental Figure 2. Reductions in hemoglobin (Hb) A1c for all doses in the present meta-analysis of the
glucagon-like peptide 1 receptor agonists (GLP-1RAs) and dipeptidyl-peptidase IV
(DPP-4) inhibitors in type 2 diabetes mellitus.
1258.e20
Volume 34 Number 6
V.R. Aroda et al.
GLP-1RAs
DPP-4 Inhibitors
Alogliptin
Exenatide BID
1.19 [1.26 to 1.12]
0.67 [1.21 to 0.12]
1.81 [2.18 to 1.44]
0.60 [1.19 to 0.01]
0.60 [0.99 to 0.21]
0.56 [1.04 to 0.08]
1.40 [1.79 to 1.01]
0.60 [0.90 to 0.29]
1.30 [1.60 to 1.00]
1.43 [1.72 to 1.13]
1.60 [2.25 to 0.96]
1.61 [2.02 to 1.20]
0.60 [0.99 to 0.21]
1.04 [1.47 to 0.60]
1.82 [2.21 to 1.42]
1.59 [2.02 to 1.16]
1.16 [1.35 to 0.97]
Bergenstal 2009
Blevins 2011
Bolli 2010
Buse 2004
Buse 2009
DeFronzo 2005
Drucker 2008
Forti 2008
Gao 2009
Heine 2005
Kadowaki 2009
Kadowaki 2011
Kendall 2005
Moretto 2008
Nauck 2007a
Zinman 2007
All
3.0 2.5 2.0 1.5 1.0 0.5
0.0

0.91 [1.31 to 0.51]
1.43 [1.81 to 1.05]
0.94 [1.27 to 0.62]
0.47 [0.84 to 0.10]
1.10 [1.45 to 0.76]
0.97 [1.27 to 0.67]
DeFronzo 2008
Fleck 2009
Nauck 2009b
Pratley 2009a
Pratley 2009b
All
3.0
2.5
3.0
2.0
1.5
1.0
0.5
0.5
0.0

0.50 [0.70 to 0.30]
1.22 [1.67 to 0.77]
1.80 [1.99 to 1.60]
1.09 [1.38 to 0.81]
0.60 [0.80 to 0.40]
1.04 [1.59 to 0.49]
2.5
2.0
1.5
1.0
0.5
0.0
FPG (mmol/L)
1.80 [2.20 to 1.40]
1.94 [2.49 to 1.40]
2.25 [2.52 to 1.98]
2.10 [2.49 to 1.71]
2.30 [2.69 to 1.91]
2.20 [3.18 to 1.22]
1.89 [2.87 to 0.91]
2.12 [2.28 to 1.96]
2.5
1.0
Del Prato 2011

Forst 2010
Gomis 2011
Kawamori 2010a
Taskinen 2011
All
Exenatide QW
3.0
1.5
Linagliptin
FPG (mmol/L)
Bergenstal 2010a
Blevins 2011
Cuddihy 2011
Diamant 2010
Drucker 2008
Kim 2007
MacConell 2011
All
2.0
FPG (mmol/L)
Saxagliptin
0.56 [0.78 to 0.33]
1.22 [1.49 to 0.95]
Chacra 2009
DeFronzo 2009
Gke 2010
Rosenstock 2008
Rosenstock 2009
Scheen 2010
All
0.50 [0.67 to 0.33]

1.20 [1.73 to 0.68]
0.50 [0.84 to 0.16]
0.60 [0.76 to 0.44]
0.72 [0.95 to 0.50]
0.0
3.0
FPG (mmol/L)
2.5
2.0
1.5
1.0
0.5
0.0
FPG (mmol/L)
Liraglutide
Sitagliptin
Buse 2009
Garber 2009
Marre 2009
Nauck 2009
Pratley 2010
RussellJones 2009
Yang 2011a
Zinman 2009
All
3.0 2.5 2.0 1.5 1.0 0.5 0.0
FPG (mmol/L)
1.61 [2.00 to 1.22]

1.42 [1.75 to 1.09]
1.59 [1.92 to 1.25]
1.70 [2.03 to 1.37]
2.14 [2.43 to 1.85]
1.55 [1.89 to 1.21]
2.12 [2.51 to 1.73]
2.44 [2.83 to 2.06]
1.82 [2.06 to 1.57]

0.83 [1.01 to 0.64]
0.70 [1.00 to 0.40]
0.64 [0.77 to 0.51]
0.90 [1.20 to 0.60]
1.13 [1.48 to 0.78]
0.97 [1.32 to 0.62]
0.97 [1.34 to 0.61]
0.94 [1.29 to 0.60]
0.24 [0.57 to +0.08]
0.81 [1.07 to 0.56]
1.40 [1.70 to 1.10]
0.56 [0.81 to 0.31]
1.25 [1.55 to 0.94]
0.83 [1.13 to 0.53]
0.70 [1.10 to 0.30]
1.60 [2.10 to 1.10]
0.93 [1.24 to 0.61]
0.68 [1.08 to 0.27]
0.90 [1.06 to 0.74]
0.65 [1.03 to 0.27]
1.10 [1.40 to 0.80]
0.80 [1.11 to 0.49]
0.87 [0.98 to 0.76]
Arechavaleta 2011
Aschner 2006
Aschner 2010
Charbonnel 2006
Fonseca 2011
Cuddihy 2011
Goldstein 2007
Hanefeld 2007
Hermansen 2007
Iwamoto 2010
Mohan 2009
Nauck 2007b
Nonaka 2008
Pratley 2010
Raz 2006
Raz 2008
Rosenstock 2006
Rosenstock 2011
Scheen 2010
Scott 2008
Seck 2010
Yang 2011
All
3.0
2.5
2.0
1.5
1.0
0.5
0.0
FPG (mmol/L)
Vildagliptin
Bolli 2008
Bosi 2009
Filozof 2010
Garber 2007
All
3.0 2.5 2.0 1.5 1.0 0.5
FPG (mmol/L)
1.40 [1.60 to 1.20]

2.42 [2.57 to 2.26]
1.31 [1.58 to 1.04]
1.10 [1.49 to 0.71]
1.57 [2.23 to 0.90]
0.0
Supplemental Figure 3. Changes from baseline in fasting plasma glucose (FPG) in the analysis of highest maintenance dose in the present meta-analysis of the glucagon-like peptide 1 receptor agonists (GLP-1RAs) and dipeptidyl-peptidase IV (DPP-4) inhibitors in type 2 diabetes
mellitus.
June 2012
1258.e21
GLP-1RAs
DPP-4 Inhibitors
Alogliptin
Exenatide BID

1.90 [2.57 to 1.23]
1.40 [2.05 to 0.75]
2.27 [2.98 to 1.56]
1.60 [2.19 to 1.01]
2.87 [3.52 to 2.22]
2.80 [3.78 to 1.82]
3.60 [4.58 to 2.62]
1.21 [1.48 to 0.93]
4.10 [4.53 to 3.67]
1.20 [1.50 to 0.90]
2.30 [2.73 to 1.87]
1.30 [1.89 to 0.71]
0.39 [0.94 to +0.16]
1.60 [1.99 to 1.21]
1.40 [2.58 to 0.22]
3.10 [3.69 to 2.51]
2.50 [2.89 to 2.11]
1.50 [2.09 to 0.91]
1.75 [2.41 to 1.09]
2.03 [2.46 to 1.60]
Bergenstal 2009
Blevins 2011
Boli 2010
Buse 2004
Buse 2009
DeFronzo 2005
Drucker 2008
Forti 2008
Gallwitz 2011a
Gao 2009
Heine 2005
Kadowaki 2009
Kadowaki 2011
Kendall 2005
Liutkus 2010
Moretto 2008
Nauck 2007
Sowa 2010
Zinman 2007
All
7 6 5 4 3 2 1 0
0.22 [0.73 to +0.29]

0.30 [0.73 to +0.13]
0.67 [1.05 to 0.29]
+0.68 [+0.31 to +1.05]
1.00 [1.39 to 0.61]
0.30 [0.90 to +0.30]
DeFronzo 2008
Fleck 2009
Nauck 2009b
Pratley 2009a
Pratley 2009b
All
7 6 5 4 3 2 1
Weight Change (kg)
Saxagliptin
0.87 [1.19 to 0.55]
1.10 [1.32 to 0.88]
0.23 [1.03 to 0.57]
0.10 [0.33 to 0.53]
0.64 [1.11 to 0.16]
DeFronzo 2009
Gke 2010
Rosenstock 2008
Rosenstock 2009
All
7 6 5 4 3 2 1 0 1
Weight Change (kg)
Weight Change (kg)
Sitagliptin
Exenatide QW

2.30 [2.90 to 1.70]
2.30 [3.01 to 1.59]
2.00 [2.39 to 1.61]
2.60 [2.99 to 2.21]
3.70 [4.68 to 2.72]
3.80 [6.54 to 1.06]
1.40 [2.58 to 0.22]
2.41 [2.83 to 1.99]
Bergenstal 2010a
Blevins 2011
Cuddihy 2011
Diamant 2010
Drucker 2008
Kim 2007
MacConell 2011
All
7 6 5 4 3 2 1
Weight Change (kg)
Liraglutide
3.24 [3.89 to 2.59]
2.45 [3.00 to 1.90]
0.20 [0.75 to +0.35]
2.80 [3.19 to 2.41]
3.38 [3.91 to 2.85]
1.80 [2.45 to 1.15]
2.44 [3.08 to 1.80]
2.00 [2.59 to 1.41]
2.29 [2.99 to 1.59]
Buse 2009
Garber 2009
Marre 2009
Nauck 2009
Pratley 2010
RussellJones 2009
Yang 2011a
Zinman 2009
All
7 6 5 4 3 2 1 0 1
Weight Change (kg)
0.80 [1.07 to 0.53]

0.20 [0.59 to 0.19]
0.60 [0.90 to 0.30]
0.89 [1.45 to 0.33]
0.80 [0.53 to 1.07]
0.60 [0.92 to 0.28]
0.80 [1.39 to 0.21]
+1.20 [+0.66 to +1.74]
+0.00 [0.51 to +0.51]
+0.80 [+0.40 to +1.20]
+0.10 [0.20 to +0.40]
+0.60 [+0.40 to +0.80]
1.50 [2.00 to 1.00]
0.10 [0.40 to +0.20]
0.96 [1.50 to 0.42]
0.60 [1.00 to 0.20]
0.50 [1.19 to +0.19]
+1.80 [+1.10 to +2.50]
0.84 [1.45 to 0.23]
0.40 [0.80 to +0.00]
1.60 [2.30 to 0.90]
0.29 [0.61 to +0.03]
Arechavaleta 2011
Aschner 2006
Aschner 2010
Bergenstal 2010a
Bergenstal 2010b
Charbonnel 2006
Cuddihy 2011
Fonseca 2011
Goldstein 2007
Hermansen 2007
Iwamoto 2010
Mohan 2009
Nauck 2007b
Nonaka 2008
Pratley 2010
Raz 2006
Raz 2008
Rosenstock 2006
Rosenstock 2011
Scott 2008
Seck 2010
All
7 6 5 4 3 2 1 0 1
Weight Change (kg)
Vildagliptin

0.30 [0.09 to +0.69]
+0.20 [0.39 to +0.79]
1.18 [1.49 to 0.87]
+0.08 [0.24 to +0.40]
0.16 [0.92 to +0.60]
Bolli 2008
Bosi 2007
Bosi 2009
Filozof 2010
All
7 6 5 4 3 2 1 0 1
Weight Change (kg)
Supplemental Figure 4. Changes from baseline in weight in the analysis of highest maintenance dose in the
present meta-analysis of the glucagon-like peptide 1 receptor agonists (GLP-1RAs) and
dipeptidyl-peptidase IV (DPP-4) inhibitors in type 2 diabetes mellitus.
1258.e22
Volume 34 Number 6

Efficacy of GLP-1 Receptor Agonists and DPP-4 Inhibitors: Meta-Analysis and Systematic Review

Uploaded by

Copyright:

Available Formats

Efficacy of GLP-1 Receptor Agonists and DPP-4 Inhibitors: Meta-Analysis and Systematic Review

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Efficacy of GLP-1 Receptor Agonists and DPP-4 Inhibitors: Meta-Analysis and Systematic Review

Uploaded by

Copyright:

Available Formats

Clinical Therapeutics/Volume 34, Number 6, 2012

Efficacy of GLP-1 Receptor Agonists and DPP-4 Inhibitors:

exception of vildagliptin. Mean weight losses with the

tion [short-acting GLP-1RAs, long-acting GLP-1RAs,

V.R. Aroda et al.

tal Appendix Table III in the online version at http://

Mean baseline characteristics and demographic

Excluded based on study design (n = 137)

Studies identified for

Studies identified for

Excluded based on details of study* (n = 145)

DPP-4 inhibitors (n = 47)

similar (see Supplemental Appendix Table II in the online version at http://dx.doi.org/10.1016/j.clinthera.

No. of Total ITT Active-Treatment

Doses Not Studied

12.5 and 25 mge

V.R. Aroda et al.

L]: exenatide BID, 1.12 [1.31 to 0.92]; exenatide

V.R. Aroda et al.

Mean HbA1c Difference (95% CI)

1.10 [1.22 to 0.99]

1.59 [1.70 to 1.48]

1.27 [1.41 to 1.13]

0.69 [0.85 to 0.54]

0.60 [0.75 to 0.46]

0.68 [0.78 to 0.57]

0.67 [0.75 to 0.60]

1.06 [1.48 to 0.64]

Mean FPG Difference (95% CI)

1.16 [1.35 to 0.97]

2.12 [2.28 to 1.96]

1.82 [2.07 to 1.57]

0.97 [1.27 to 0.67]

1.04 [1.59 to 0.49]

0.73 [0.95 to 0.50]

0.87 [0.98 to 0.77]

1.57 [2.23 to 0.90]

Mean Weight Difference (95% CI)

2.03 [2.46 to 1.60]

2.41 [2.83 to 1.99]

2.29 [2.99 to 1.59]

0.30 [0.90 to +0.30]

0.64 [1.11 to 0.16]

0.29 [0.61 to +0.03]

0.16 [0.92 to +0.60]

Weight Change (kg)

been uniform because these agents have similar effects

between each of the GLP-1RAs. Additional study is

Differences within the GLP-1RA class appear to be

V.R. Aroda et al.

ization may have affected the magnitude of subsequent

All of the incretin-based therapies analyzed in the present

Statistical and writing support for the manuscript was

Supplemental appendixes and figures accompanying

1. Amori RE, Lau J, Pittas AG. Efficacy and safety of incretin

V.R. Aroda et al.

therapy in treatment-naive patients

19. Fakhoury WK, Lereun C, Wright D.

V.R. Aroda et al.