`Experimental Diabetes Research
`Volume 2012, Article ID 672658, 8 pages
`doi:10.1155/2012/672658
`
`Research Article
`Effects of Glucagon-Like Peptide-1 Receptor Agonists on
`Body Weight: A Meta-Analysis
`
`Matteo Monami,1 Ilaria Dicembrini,2 Niccol `o Marchionni,1
`Carlo M. Rotella,2 and Edoardo Mannucci3
`
`1 Geriatric Cardiology, Careggi Teaching Hospital and University of Florence, 50141 Florence, Italy
`2 Obesity Agency, Careggi Teaching Hospital and University of Florence, 50141 Florence, Italy
`3 Diabetes Agency, Careggi Teaching Hospital and University of Florence, 50141 Florence, Italy
`
`Correspondence should be addressed to Edoardo Mannucci, edoardo.mannucci@unifi.it
`
`Received 22 January 2012; Accepted 19 March 2012
`
`Academic Editor: Giovanni Di Pasquale
`
`Copyright © 2012 Matteo Monami et al. This is an open access article distributed under the Creative Commons Attribution
`License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
`cited.
`
`Glucagon-Like Peptide-1 receptor agonists (GLP-1RAs), approved as glucose-lowering drugs for the treatment of type 2 diabetes,
`have also been shown to reduce body weight. An extensive Medline, Cochrane database, and Embase search for “exenatide,”
`“liraglutide,” “albiglutide,” “semaglutide,” and “lixisenatide” was performed, collecting all randomized clinical trials on humans up
`to December 15, 2011, with a duration of at least 24 weeks, comparing GLP-1 receptor agonists with either placebo or active drugs.
`Twenty two (7,859 patients) and 7 (2,416 patients) trials with available results on body weight at 6 and 12 months, respectively,
`were included. When compared with placebo, GLP-1RAs determine a reduction of BMI at 6 months of −1.0 [−1.3; −0.6] kg/m2.
`Considering the average BMI at baseline (32.4 kg/m2) these data means a weight reduction of about 3% at 6 months. This result
`could seem modest from a clinical standpoint; however, it could be affected by many factors contributing to an underestimation of
`the effect of GLP-1RA on body weight, such as non adequate doses, inclusion criteria, efficacy of GLP-1RA on reducing glycosuria,
`and association to non-pharmacological interventions not specifically aimed to weight reduction.
`
`1. Introduction
`
`Most drugs developed for the therapy of obesity have failed to
`show a sufficient efficacy and safety for long-term treatment.
`In particular, agents which stimulate energy expenditure
`(e.g.,
`thyroid hormones, sympathoadrenergic drugs, or
`sibutramine) do not have an adequate cardiovascular safety,
`whereas centrally acting anorexants either are ineffective in
`the long term (e.g., serotonin reuptake inhibitors) or show
`neuropsychiatric adverse effects (e.g., amphetamine deriva-
`tives or cannabinoid receptor antagonists) [1]. As a result,
`orlistat, which inhibits lipid absorption, is the only available
`drug for obesity in many countries. Even for drugs which do
`not show relevant problems of long-term safety, such as orli-
`stat, the unsatisfactory tolerability profile limits clinical use.
`Glucagon-like peptide-1 (GLP-1) is a gastrointestinal
`hormone, produced mainly in the postprandial phase, which
`stimulates insulin secretion and inhibits glucagon release in
`
`a dose-dependent fashion [2]. Due to this properties, the
`hormone reduces hyperglycemia without inducing hypo-
`glycemia in patients with type 2 diabetes [3]. The rapid inac-
`tivation of GLP-1 in vivo and the consequent short half-life (a
`few minutes after subcutaneous administration) prevents its
`therapeutic use. Long-acting GLP-1 receptor agonists, which
`can be administered via subcutaneous injection once or twice
`a day or once a week, have been developed as glucose-
`lowering drugs for the treatment of type 2 diabetes [4], but
`they have also been shown to reduce body weight [5, 6]. The
`effects of GLP-1 and its agonists on body weight appears to
`be due to a reduction in food intake, mainly determined by a
`direct central (hypothalamic) effect of the hormone [7]. The
`stimulation of GLP-1 receptor also retards gastric emptying;
`this latter effect is again due, at least partly, to a central action,
`mediated via the autonomous nervous system [8]. One of
`the side effect of GLP-1 receptor agonists, nausea (sometimes
`associated with vomiting), could contribute to the weight
`
`MPI EXHIBIT 1064 PAGE 1
`
`MPI EXHIBIT 1064 PAGE 1
`
`
`
`2
`
`Experimental Diabetes Research
`
`reducing effect; however, weight loss has also been observed
`when analyzing separately patients who do not report nausea
`[8].
`
`In fact, some drugs of this class (i.e., liraglutide and
`long-acting exenatide) are currently under development for
`the treatment of obesity [9–12]. A phase II, 20-week trial
`enrolling patients without diabetes showed that liraglutide
`has a higher efficacy than orlistat in promoting weight
`loss [13]. Another longer-term (52 weeks) trial with same
`molecule, the results of which have not been published in
`full but partly disclosed [14], confirms that liraglutide is
`an interesting option for the treatment of obesity. Another
`molecule of the same class, exenatide, has been reported
`to induce a significant weight loss in a 24-week placebo-
`controlled trial [15]. Most of what is known on the effect
`of GLP-1 receptor agonists on body weight comes from
`clinical trials performed on patients with type 2 diabetes,
`with glucose control as the principal endpoint. Currently
`ongoing trials enrolling subjects with obesity and without
`diabetes will provide, in due time, further information. In
`the meanwhile, a systematic evaluation of data collected in
`studies on type 2 diabetes can provide a more defined picture
`of what we can realistically expect from GLP-1 receptor
`agonists as weight-reducing agents.
`A recent meta-analysis has shown a weight loss of
`approximately 3% at endpoint in available published trials,
`with a duration ranging from 20 to 52 weeks [6]. This
`analysis does not provide information on the time-course of
`weight loss with GLP-1 receptor agonists. Furthermore, no
`distinction is made between placebo- and active comparator-
`controlled trials, with some of the comparators (i.e., insulin,
`thiazolidinediones, and sulfonylureas) possibly inducing
`weight gain. Aim of the present meta-analysis is to assess the
`effects of GLP-1 receptor agonists on body weight at 6 and
`12 months of treatment, separating placebo-controlled trials
`from comparisons with active drugs. Furthermore, a meta-
`regression analysis will be performed to explore predictors of
`weight change during treatment.
`
`2. Methods
`
`The meta-analysis was reported following the PRISMA
`checklist [16].
`
`2.1. Data Sources, Searches, and Extraction. An extensive
`Medline, Cochrane database, and Embase search for all ar-
`ticles in English using the keywords “exenatide”, “lirag-
`lutide”, “albiglutide”, “semaglutide”, and “lixisenatide” was
`performed collecting all randomized clinical
`trials on
`humans up to December 15, 2011. Completed but still
`unpublished trials were identified through a search of http://
`www.clinicaltrials.gov/ website. FDA (http://www.fda.gov/)
`and European Medicines Agency (EMA, http://www.ema
`.europa.eu/) reviews of approved drugs, as well as published
`infor-mation provided to FDA in response to queries during
`the approval process, were also searched for retrieval of un-
`published trials. Results of those trials were retrieved, if avail-
`able, on http://www.novonordisk-trials.com/ or http://www
`.clinicaltrials.org/. For unpublished and published trials
`
`which were not exhaustively disclosed, an attempt was made
`(through e-mail) to contact principal investigators in order
`to retrieve missing data. For all published trials, results
`reported in papers were used as the primary source of infor-
`mation, when available.
`The identification of relevant abstracts, the selection of
`studies based on the criteria described previously, and the
`subsequent data extraction were performed independently
`by two of the authors (E. Mannucci, M. Monami), and
`conflicts these resolved by the third investigator (N. Mas-
`chionni).
`
`2.2. Study Selection. A meta-analysis was performed includ-
`ing all randomized clinical trials, with a duration of at least
`24 weeks, comparing full therapeutic doses Glucagon-like
`Peptide-1 (GLP-1) receptor agonists (i.e., at least 1.8 mg/day
`liraglutide, 20 μg/day for exenatide b.i.d., 2 mg/day for
`exenatide once weekly) and with placebo or other active
`drugs. Trials with a shorter duration were excluded, due to
`the fact that they could not yield relevant information on
`body weight reduction. No review protocol was published
`elsewhere. Trials without any information on body mass
`index (BMI) at 6 or 12 months were also excluded.
`
`2.3. Quality Assessment. The quality of trials was assessed
`using some of the parameters proposed by Jadad et al. [17].
`The score was not used as a criterion for the selection of trials,
`whereas some items were used only for descriptive purposes.
`
`2.4. Data Synthesis and Analysis. The principal outcome was
`the effect of full therapeutic doses of GLP-1 receptor agonists,
`compared with other hypoglycemic agents or placebo, on
`BMI at 6 months and 12 months (when available). Between-
`group differences in endpoint BMI were assessed as a
`measure of treatment effect, without considering differ-
`ences from baseline. Secondary outcomes included glycated
`hemoglobin (HbA1c) at 6 and 12 months. Separate analyses
`were performed for trials with different GLP-1 receptor
`agonists and with different comparators, whenever possible.
`Furthermore, separate analyses were performed for trials
`with different principal endpoints. Metaregression analysis
`was performed on placebo-controlled trials, in order to
`identify possible predictors of weight loss.
`Heterogeneity was calculated using the I2 statistics.
`Weighted mean differences were calculated for BMI and
`HbA1c at 6 and 12 months, and a random effects model was
`used for the meta-analysis. Publication/disclosure bias was
`estimated separately for placebo-controlled trials and studies
`versus active comparators, using Kendall’s tau without
`continuity correction, and one-sided P, were calculated,
`together with the fail-safe N, and Funnel plot analysis. All
`those analyses were performed using Comprehensive Meta-
`analysis Version 2, Biostat, (Englewood, NJ, USA).
`
`3. Results
`
`The trial flow summary is reported in Figure 1. Trials with
`available results on body mass index at 6 months were 21
`
`MPI EXHIBIT 1064 PAGE 2
`
`MPI EXHIBIT 1064 PAGE 2
`
`
`
`Experimental Diabetes Research
`
`3
`
`a significant (P = 0.002) BMI reduction of 1.6 (0.6–2.5)
`kg/m2 in comparison with placebo). One further trial, which
`enrolled patients with type 2 diabetes, had been designed
`for the assessment of weight reduction with exenatide as the
`principal outcome [19], with similar results.
`In the 19 trials performed in patients with diabetes,
`GLP-1 receptor agonists were associated with a significantly
`lower BMI at 6 months in comparison with placebo and
`with any active glucose-lowering agent, with the exception
`of the only 2 available head-to-head comparisons with
`thiazolidinediones. No differences in the weight-reducing
`effects were observed between exenatide and liraglutide
`(Figure 3(a)). A subgroup analysis of placebo-controlled
`trials was performed on the basis of the minimum BMI
`chosen as inclusion criterion; in trials excluding (N = 4)
`or including (N = 5) nonoverweight (BMI < 25 kg/m2),
`the difference in 6-month BMI between active treatment
`and control groups was −1.0 [−1.6; −0.4] and −0.8 [−1.3;
`−0.3] kg/m2, respectively (bothP < 0.001).
`For 18 out of 19 of those trials, the principal endpoint was
`HbA1c, which was significantly reduced by GLP-1 receptor
`agonists in comparison with placebo, DPP4 inhibitors,
`and thiazolidinediones, whereas differences with respect to
`sulfonylureas and insulin were not statistically significant
`(Figure 3(b)).
`Metaregression analysis was performed on all placebo-
`controlled trials, including those on nondiabetic individuals,
`irrespective of the principal endpoint of the study. In the
`11 available trials, mean baseline BMI, age, and duration of
`diabetes (in the 9 trials on patients with diabetes) were not
`significantly correlated with treatment effect on BMI.
`
`3.2. Results at 12 Months. Results on BMI at 12 months were
`available in 7 trials, 6 of which were performed in patients
`with diabetes. The only one trial [14] enrolling subjects
`without diabetes, which had weight loss as its principal
`endpoint,
`liraglutide,
`induced a significant reduction of
`weight in comparison with placebo (−1.2 [−2.3; −0.1] kg/m2
`in 1-year BMI; P = 0.04). The results of the other 6 trials, all
`with active comparators, are summarized in Table 2. In these
`studies, a further reduction of body weight was observed
`after the first six months of treatment. Similar results were
`obtained when the only trial which did not report 6-month
`BMI [14] was excluded from the analysis (data not shown).
`
`Medline/unpublished
`n = 127/157
`
`Not randomized trials
`n = 19/22
`
`Non humans
`n = 1/0
`
`No external comparison
`n = 4/7
`
`Short duration
`n = 55/77
`
`Duplicate
`n = 15/3
`
`Already published on medline
`n = 0/25
`
`BMI at 6 months not available
`n = 12/23
`
`Fulfilling all inclusion criteria
`n = 21/0
`
`Included
`N = 21
`
`Figure 1: Trial flow summary.
`
`Funnel plot of standard error by difference in means
`
`0
`
`0.2
`
`0.4
`
`0.6
`
`0.8
`
`Standarderror
`
`1−3
`
`−2
`
`−1
`1
`0
`Difference in means
`
`2
`
`3
`
`Figure 2: Funnel plot for bias/disclosure publication.
`
`(19 of which in patients with diabetes), whereas those with
`data at one year were 7 (6 of which on diabetes); the
`characteristics of those studies are summarized in Table 1.
`Funnel plot analysis on 6-month trials on diabetes (Figure 2)
`did not reveal any major publication/disclosure bias for BMI,
`as confirmed by Kendall’s tau (t = 0.14, P = 0.36) and fail-
`safe N (number of missing studies that would bring P > 0.05:
`733). I2 for BMI at 6 months was 83.6 (P < 0.001).
`
`3.1. Results at 6 Months. Only two trials [15, 18] in
`subjects with obesity not associated with diabetes reported
`outcomes on body weight of exenatide at 6 months (with
`
`4. Discussion
`
`The few available trials designed with weight loss as the
`principal endpoint and enrolling nondiabetic patients with
`obesity have shown that GLP-1 receptor agonists have a
`potential use as drugs for the treatment of overweight
`[14, 15]. Similar results were obtained in a trial on over-
`weight patients with polycystic ovary syndrome, in which
`restoration of menstrual cycles was the principal endpoint
`[20]. The much wider evidence collected in subjects with
`type 2 diabetes confirms this effect, as previously reported
`[6, 18]. This action is consistent across trials, and it cannot
`be attributed to selective reporting as shown by Funnel
`
`MPI EXHIBIT 1064 PAGE 3
`
`MPI EXHIBIT 1064 PAGE 3
`
`
`
`4
`
`Experimental Diabetes Research
`
`6.7/6.8
`7.2/7.1
`7.5/7.6
`6.9/6.8
`7.9/8.1
`8.1/8.0
`7.4/8.3
`7.7/8.6
`6.6/7.5
`7.8/8.7
`6.9/7.6
`
`NR/NR
`5.6/5.7
`
`7.2/7.7
`7.2/7.4
`6.8/7.0
`
`7.0/7.2
`6.9/7.3
`7.5/8.1
`7.1/7.8
`7.3/7.4
`7.0/8.1
`7.0/7.9
`7.3/8.5
`7.5/8.7
`
`(%,ID/C)
`6-month
`HbA1cat
`
`#Studieswithmultiplecomparators;NR:notreported;ID:interventionaldrug;C:comparator;DM:diabetesmellitus;wks:weeks.
`
`7.5
`8.2
`8.6
`7.9
`8.7
`8.8
`8.2
`8.5
`8.4
`8.6
`7.8
`
`NR
`5.6
`
`8.5
`8.5
`8.3
`
`8.3
`8.4
`8.4
`8.3
`8.3
`8.3
`8.5
`8.3
`8.5
`
`29.9/30.4
`30.6/32.0
`29.9/30.5
`32.0/33.2
`27.5/28.5
`27.3/28.9
`33.0/33.9
`32.9/33.7
`33.2/33.5
`32.4/33.8
`30.4/31.4
`
`39.1/40.8
`37.8/38.8
`
`31.2/31.7
`31.2/33.0
`31.1/32.5
`
`29.9/30.9
`29.9/31.5
`29.9/30.2
`32.3/33.6
`30.1/31.6
`29.9/31.2
`33.0/33.7
`30.1/31.1
`29.9/30.3
`
`30.5
`31.3
`30.4
`33.1
`28.4
`28.6
`34.0
`34.0
`33.5
`33.3
`31.5
`
`40.4
`39.5
`
`32.0
`32.0
`32.0
`
`30.3
`32.8
`29.8
`33.0
`31.1
`30.9
`33.5
`31.3
`30.0
`
`5.0
`9.5
`9.9
`5.0
`NR
`NR
`5.8
`9.0
`12.0
`6.3
`2.0
`
`0.0
`0.0
`
`6.0
`6.0
`8.0
`
`9.0
`6.0
`7.0
`5.0
`8.0
`9.0
`9.0
`8.0
`7.0
`
`58
`59
`59
`57
`55
`56
`53
`55
`59
`55
`54
`
`29
`46
`
`52
`52
`58
`
`57
`55
`56
`53
`57
`57
`55
`57
`56
`
`52
`26
`52
`26
`52
`52
`30
`30
`30
`30
`24
`
`24
`24
`
`26
`26
`26
`
`26
`52
`26
`52
`26
`26
`26
`26
`26
`
`Glargine
`Glargine
`
`BiAsp
`BiAsp
`
`Glimepiride
`Glibenclamide
`
`Placebo
`Placebo
`Placebo
`Placebo
`Placebo
`
`Placebo
`Placebo
`
`Sitagliptin
`Pioglitazone
`
`Glargine
`
`Glargine
`Sitagliptin
`
`Rosiglitazone
`Glimepiride
`Glimepiride
`
`Placebo
`Placebo
`Placebo
`Placebo
`
`36/33
`282/267
`253/248
`248/246
`57/54
`61/62
`113/113
`241/247
`138/123
`129/123
`77/78
`
`20/20
`73/79
`
`160/166
`160/165
`233/233
`
`230/232
`225/219
`234/232
`498/248
`242/242
`230/114
`355/175
`242/121
`234/114
`
`Buncketal.[38]
`Heineetal.[37]
`Naucketal.[36]
`Gallwitzetal.[35]
`Derosaetal.[34]
`Derosaetal.[33]
`DeFronzoetal.[32]
`Kendalletal.[19]
`Buseetal.[31]
`Buseetal.[30]
`Morettoetal.[29]
`
`Type2diabetes
`Elkind-Hirschetal.[20]
`Rosenstocketal.[15]
`Obesity
`Exenatide
`Bergenstaletal.[28]#
`Bergenstaletal.[28]
`Diamantetal.[27]
`ExenatideLAR
`Russell-Jonesetal.[27]#
`Pratleyetal.[26]
`Marreetal.[21]#
`Garberetal.[25]
`Naucketal.[22]#
`Russell-Jonesetal.[24]
`Zinmanetal.[23]
`Naucketal.[22]
`Marreetal.[21]
`Liraglutide
`
`baseline(%)
`
`(Kg/m2)
`
`(Kg/m2)
`
`(ys)
`
`HbA1c
`
`BMIat6-month
`
`BMIBaseline
`
`DurationofDM
`
`Age(ys)
`
`(wks)
`
`Trialduration
`
`Comparator
`
`patients(ID/C)
`
`Numberof
`
`Study(Reference)
`
`Table1
`
`MPI EXHIBIT 1064 PAGE 4
`
`MPI EXHIBIT 1064 PAGE 4
`
`
`
`Experimental Diabetes Research
`
`5
`
`Table 2: Weighted mean differences in 6- and 12-month BMI between GLP-1 receptor agonists and different active comparators.
`
`Overall
`DPP-4 inhibitors
`Sulphonylureas
`Insulin
`
`Number
`of trials
`6
`1
`3
`2
`
`6-month BMI
`−1.2 [−1.5; −0.8]
`−1.6 [−2.6; −0.8]
`−1.4 [−2.4; −0.7]
`−0.7 [−1.4; 0.0]
`
`P
`
`<0.001
`<0.001
`0.001
`0.048
`
`12-month BMI
`−1.9 [−3.0; −0.8]
`−1.7 [−2.7; −0.7]
`−2.3 [−4.2; −0.5]
`−1.5 [−2.1; −0.8]
`
`P
`
`<0.001
`0.001
`0.012
`<0.001
`
`P
`
`Number of
`trials
`
`−3
`
`−2.5
`
`−2
`
`−1.5
`
`−1
`
`−0.5
`
`0
`
`0.5
`
`1
`
`2
`
`2
`
`4 2 2
`
`6 4 2
`
`9 5 4
`
`(a)
`
`DPP-4 inhibitors
`
`MD LL,95% CI UL,95% CI
`−1.08 −2.15
`
`0.001
`
`0.05
`
`Thiazolidinediones
`
`−1.02 −2.49
`
`0.45
`
`0.17
`
`Sulfonylureas
`Exenatide
`Liraglutide
`
`−1.31 −1.62
`−1.29 −1.88
`−1.41 −2.03
`
`Insulin
`Exenatide
`Liragl./Exen. LAR
`
`−1.1 −1.45
`−1.05 −1.47
`−1.2 −2.15
`
`−1
`−0.7
`−0.79
`
`−0.75
`−0.64
`−0.25
`
`Placebo
`Exenatide
`Liraglutide
`
`−0.967 −1.322
`−1.058 −1.563
`−0.85 −1.375
`
`−0.591
`−0.554
`−0.315
`
`<0.001
`<0.001
`<0.001
`
`<0.001
`<0.001
`0.021
`
`<0.001
`<0.001
`0.002
`
`MD LL,95% CI UL,95% CI
`
`P
`
`Number of
`trials
`
`−1.6 −1.4 −1.2 −1 −0.8 −0.6 −0.4 −0.2
`
`0
`
`0.2
`
`0.4
`
`2
`
`2
`
`4 2 2
`
`6 4 2
`
`9 54
`
`(b)
`
`DPP-4 inhibitors
`
`Thiazolidinediones
`
`Sulfonylureas
`Exenatide
`Liraglutide
`
`Insulin
`Exenatide
`Liragl./Exen. LAR
`
`Placebo
`Exenatide
`Liraglutide
`
`−0.42
`
`−0.41
`
`−0.23
`−0.04
`−0.4
`
`−0.06
`0.02
`−0.2
`
`−0.86
`−0.87
`−0.85
`
`−0.54
`
`−0.81
`
`−0.58
`−0.34
`−0.99
`
`−0.18
`−0.09
`−0.36
`
`−1.25
`−1.46
`−1.37
`
`−0.3
`
`−0.02
`
`0.12
`0.25
`0.19
`
`0.06
`0.13
`−0.04
`
`−0.46
`−0.28
`−0.31
`
`<0.001
`
`0.037
`
`0.2
`0.76
`0.18
`
`0.3
`0.67
`0.017
`
`<0.001
`0.004
`0.002
`
`Figure 3: Weighted mean differences in 6-month BMI (a) and HbA1c (b) between GLP-1 receptor agonists and different active comparators
`or placebo, in trials performed in type 2 diabetic patients. MD: weighted mean differences; LL: lower limits; UL: upper limits.
`
`MPI EXHIBIT 1064 PAGE 5
`
`MPI EXHIBIT 1064 PAGE 5
`
`
`
`6
`
`Experimental Diabetes Research
`
`plot analysis and Kendall’s tau calculation. GLP-1 receptor
`agonists have a beneficial effect on body weight not only in
`comparisons with drugs that induce weight gain (such as
`insulin or sulfonylureas) but also with respect to placebo.
`The only exception is represented by direct comparisons with
`thiazolidinediones: in this case, despite a mean difference
`in 6-month BMI similar to that observed for other active
`comparators, the statistical significance is not reached, due
`to the small number of available trials.
`In order to evaluate the weight reducing effect of
`GLP-1 receptor agonists, the most interesting results are
`those obtained in placebo-controlled trials, which allow to
`discriminate the beneficial action of these drugs from the
`adverse effects on body weight of other glucose-lowering
`agents. In these studies, the mean weight loss at 6 months
`is 1.0 kg/m2; considering that the average BMI at baseline
`is about 33.9 kg/m2, this means that the actual ponderal
`reduction is in the 3% range. The estimated weight loss seems
`to be larger than that reported in a previous meta-analysis
`[6]; this result could be due to the exclusion of patients
`treated with submaximal doses of GLP-1 receptor agonists.
`This result could seem modest from a clinical standpoint;
`however, several factors should be considered. In all trials on
`patients with diabetes except one the principal endpoint was
`the improvement in HbA1c, and not weight loss. This means
`that patients were selected on the basis of unsatisfactory
`glucose control, and not for their overweight; the minimum
`BMI for inclusion was not specified in some studies, and
`ranged from 25 to 45 kg/m2 in the others, meaning that,
`in all trials, part of the patients enrolled were not actually
`obese. Notably, those trials that excluded normal-weight
`subjects showed a greater effect of GLP-1 receptor agonists on
`weight loss. Furthermore, patients with diabetes could have
`greater difficulties in losing weight than similarly overweight
`subjects with normal glucose tolerance. In those who had
`elevated HbA1c at baseline, the reduction of glycosuria
`determined by drug treatment could have been an obstacle
`to weight loss. Finally, the nonpharmacological interventions
`associated to drugs in trials for glycemic control in type 2 dia-
`betes are not specifically aimed at weight reduction. All these
`factors could have contributed to an underestimation of the
`effect of GLP-1 receptor agonists on body weight. It should
`also be recognized that weight loss in clinical trials could
`be quite different from that obtained in real-life conditions.
`The selection of patients with greater compliance and the
`more accurate follow-up produces a greater weight loss from
`baseline in randomized clinical trials. On the other hand, for
`the same reasons, as long as the between-group differences
`are assessed, as in the present study, the lifestyle/dietary
`intervention associated with drug treatment in randomized
`trials can partly mask the actual effect of the drug.
`It should also be considered that treatment with GLP-1
`receptor agonists could have some further beneficial effects
`on other metabolic alterations of obese patients (e.g., insulin
`resistance, risk of diabetes, blod pressure, etc.), beyond
`weight loss. The assessment of those effects was not among
`the aims of the present meta-analysis.
`The effect of GLP-1 and its receptor agonists on food
`intake and body weight is dose dependent [13]. For this
`
`reason, it is possible that doses needed for the treatment of
`obesity are higher than those indicated for type 2 diabetes.
`For example, liraglutide 3.0 mg/day induces a greater weight
`loss than 1.8 mg/day, whereas no additional effect on blood
`glucose is expected over 1.8 mg/day [13]. Obviously, at least
`some of the adverse effects of these drugs (e.g., nausea
`and vomiting) are also dose-dependent and they could be
`amplified by the increase in daily doses. In the case that re-
`commended doses for obesity exceed in a relevant manner
`those for diabetes, the safety profile of GLP-1 receptor ago-
`nists, which is satisfactory when they are used in the treat-
`ment of type 2 diabetes, should be verified on a sufficiently
`wide amount of data.
`Some interesting information can be obtained from the
`analysis of data collected in trials on type 2 diabetes with a 1-
`year follow-up; in fact, the effect of GLP-1 receptor agonists
`at 1 year seems to be larger than that observed, in the same
`trials, after 6 months of treatment. The number of studies
`is limited, and none of them includes a comparison with
`placebo; in fact, a longer-term treatment without any active
`drug would be unethical in patients with unsatisfactory
`control of diabetes. Active comparisons can be misleading, as
`the comparators often induce weight gain (e.g., insulin, sul-
`fonylureas, thiazolidinediones); this means that the increased
`difference between GLP-1 receptor agonists and control
`groups at 1 year could be partly due to weight gain induced
`by comparators. Despite these limitations, the possibility
`that the maximum effect of GLP-1 receptor agonists on
`body weight is reached after 6 months should be considered
`and taken into account in the design of future clinical
`trials.
`
`Conflict of Interests
`
`M. Monami has received speaking fees from Astra Zeneca,
`Bristol Myers Squibb, Eli-Lilly, Merck, Novo Nordisk, and
`Takeda. I. Dicembrini has received speaking fees from Bayer,
`Eli-Lilly, Novo Nordisk, and Takeda. C. M. Rotella has re-
`ceived consultancy fees from Eli Lilly, research grants from
`Eli Lilly, and speaking fees from Eli Lilly, Novo Nordisk, and
`Sanofi-Aventis. E. Mannucci has received consultancy fees
`from Merck and Novartis, speaking fees from Astra Zeneca,
`Bristol Myers Squibb, Merck, and Novartis, and research
`grants from Merck, Novartis, and Takeda. This work was
`performed as part of the institutional activity of the authors,
`without financial support from any third party.
`
`References
`
`[1] M. L. Vetter, L. F. Faulconbridge, V. L. Webb, and T. A. Wad-
`den, “Behavioral and pharmacologic therapies for obesity,”
`Nature Reviews Endocrinology, vol. 6, no. 10, pp. 578–588,
`2010.
`[2] M. E. Doyle and J. M. Egan, “Mechanisms of action of glu-
`cagon-like peptide 1 in the pancreas,” Pharmacology and Ther-
`apeutics, vol. 113, no. 3, pp. 546–593, 2007.
`[3] B. Ahr´en, “GLP-1 for type 2 diabetes,” Experimental Cell Re-
`search, vol. 317, no. 9, pp. 1239–1245, 2011.
`
`MPI EXHIBIT 1064 PAGE 6
`
`MPI EXHIBIT 1064 PAGE 6
`
`
`
`Experimental Diabetes Research
`
`7
`
`[4] A. J. Garber, “Long-acting glucagon-like peptide 1 receptor
`agonists: a review of their efficacy and tolerability,” Diabetes
`Care, vol. 34, pp. S279–S284, 2011.
`[5] B. Gallwitz, “Glucagon-like peptide-1 analogues for Type 2 di-
`abetes mellitus: current and emerging agents,” Drugs, vol. 71,
`no. 13, pp. 1675–1688, 2011.
`[6] T. Vilsbø, M Christensen, A. E. Junker, F. K. Knop, and L. L.
`Gluud, “Effects of glucagon-like peptide-1 receptor agonists
`on weight loss: systematic review and meta-analyses of ran-
`domised controlled trials,” British Medical Journal. In press.
`[7] M. R. Hayes, B. C. De Jonghe, and S. E. Kanoski, “Role of the
`glucagon-like-peptide-1 receptor in the control of energy ba-
`lance,” Physiology and Behavior, vol. 100, no. 5, pp. 503–510,
`2010.
`[8] S. E. Kanoski, L. E. Rupprecht, S. M. Fortin, B. C. De Jonghe,
`and M. R. Hayes, “The role of nausea in food intake and body
`weight suppression by peripheral GLP-1 receptor agonists,
`exendin-4 and liraglutide,” Neuropharmacology, vol. 62, no. 5-
`6, pp. 1916–1927, 2012.
`[9] “Effect of Liraglutide on Body Weight
`in Non-diabetic
`Obese Subjects or Overweight Subjects With Co-morbidities:
`SCALE—Obesity and Pre-diabetes,” http://www.clinicaltrials
`.gov/ct2/show/NCT01272219/.
`[10] “Effect of Liraglutide on Body Weight in Overweight or Obese
`Subjects With Type 2 Diabetes: SCALE—Diabetes,” http://
`www.clinicaltrials.gov/ct2/show/NCT01272232/.
`[11] “The Effects of Exenatide (Byetta(Trademark)) on Energy Ex-
`penditure and Weight Loss in Nondiabetic Obese Subjects,”
`http://www.clinicaltrials.gov/ct2/show/NCT00856609/.
`[12] “The Effect of Exenatide on Weight and Hunger in Obese,
`Healthy Women,” http://www.clinicaltrials.gov/ct2/show/NC-
`T00456885/.
`[13] A. Astrup, S. R¨ossner, L. Van Gaal et al., “Effects of liraglutide
`in the treatment of obesity: a randomised, double-blind, pla-
`cebo-controlled study,” The Lancet, vol. 374, no. 9701, pp.
`1606–1616, 2009.
`[14] “Comparison of Liraglutide Versus Placebo in Weight Loss
`Maintenance in Obese Subjects: SCALE Maintenance,” http://
`www.clinicaltrials.gov/ct2/show/NCT00781937?term=NCT0-
`0781937/.
`[15] J. Rosenstock, L. J. Klaff, S. Schwartz et al., “Effects of exe-
`natide and lifestyle modification on body weight and glucose
`tolerance in obese subjects with and without pre-diabetes,”
`Diabetes Care, vol. 33, no. 6, pp. 1173–1175, 2010.
`[16] D. Moher, A. Liberati, J. Tetzlaff et al., “Preferred reporting
`items for systematic reviews and meta-analyses: the PRISMA
`statement,” Annals of Internal Medicine, vol. 151, no. 4, pp.
`264–269, 2009.
`[17] A. R. Jadad, R. A. Moore, D. Carroll et al., “Assessing the
`quality of reports of randomized clinical trials: is blinding
`necessary?” Controlled Clinical Trials, vol. 17, no. 1, pp. 1–12,
`1996.
`[18] K. Elkind-Hirsch, O. Marrioneaux, M. Bhushan, D. Vernor,
`and R. Bhushan, “Comparison of single and combined treat-
`ment with exenatide and metformin on menstrual cyclicity in
`overweight women with polycystic ovary syndrome,” Journal
`of Clinical Endocrinology and Metabolism, vol. 93, no. 7, pp.
`2670–2678, 2008.
`[19] D. M. Kendall, M. C. Riddle, J. Rosenstock 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, vol. 28, no. 5, pp. 1083–1091,
`2005.
`
`[20] M. Monami, N. Marchionni, and E. Mannucci, “Glucagon-
`like peptide-1 receptor agonists in type 2 diabetes: a meta-
`analysis of randomized clinical trials,” European Journal of
`Endocrinology, vol. 160, no. 6, pp. 909–917, 2009.
`[21] M. Marre, J. Shaw, M. Br¨andle et al., “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),” Diabetic
`Medicine, vol. 26, no. 3, pp. 268–278, 2009.
`[22] M. Nauck, A. Frid, K. Hermansen et al., “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, vol. 32, no. 1, pp. 84–90, 2009.
`[23] B. Zinman, J. Gerich, J. B. Buse et al., “Efficacy and safety of the
`human glucagon-like peptide-1 analog liraglutide in combi-
`nation with metformin and thiazolidinedione in patients with
`type 2 diabetes (LEAD-4 Met+TZD),” Diabetes Care, vol. 32,
`no. 7, pp. 1224–1230, 2009.
`[24] D. Russell-Jones, A. Vaag, O. Schmitz et al., “Liraglutide Effect
`and Action in Diabetes 5 (LEAD-5) met+SU Study Group.
`Liraglutide vs insulin glargine and placebo in combination
`with metformin and sulfonylurea therapy in type 2 diabetes
`mellitus (LEAD-5 met+SU): a randomised controlled trial,”
`Diabetologia, vol. 52, no. 10, pp. 2046–2055, 2009.
`[25] A. Garber, R. Henry, R. Ratner et al., “Liraglutide ver-
`sus glimepiride monotherapy for type 2 diabetes (LEAD-3
`Mono): a randomised, 52-week, phase III, double-blind, para-
`llel-treatment trial,” The Lancet, vol. 373, no. 9662, pp. 473–
`481, 2009.
`[26] R. Pratley, M. Nauck, T. Bailey et al., “1860-LIRA-DPP-4 Study
`Group. One year of liraglutide treatment offers sustained and
`more effective glycaemic control and weight reduction com-
`pared with sitagliptin, both in combination with metformin,
`in patients with type 2 diabetes: a randomised, parallel-group,
`open-label trial,” International Journal of Clinical Practice, vol.
`65, no. 4, pp. 397–407, 2011.
`[27] M. Diamant, L. Van Gaal, S. Stranks 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,” The Lancet, vol. 375, no. 9733, pp. 2234–
`2243, 2010.
`[28] R. M. Bergenstal, C. Wysham, L. MacConell 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,” The Lancet,
`vol. 376, no. 9739, pp. 431–439, 2010.
`[29] T. J. Moretto, D. R. Milton, T. D. Ridge et al., “Efficacy and
`tolerability of exenatide monotherapy over 24 weeks in antidi-
`abetic drug-naive patients with type 2 diabetes: a randomized,
`double-blind, placebo-controlled, parallel-group study,” Clin-
`ical Therapeutics, vol. 30, no. 8, pp. 1448–1460, 2008.
`[30] J. B. Buse, R. R. Henry, J. Han, D. D. Kim, M. S. Fineman,
`and A. D. Baron, “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, vol. 27, no. 11, pp. 2628–2635, 2004.
`[31] J. B. Buse, R. M. Bergenstal, L. C. Glass et al., “Use of twice-
`daily exenatide in Basal insulin-treated patients with type 2
`diabetes: a randomized, controlled trial,” Annals of Internal
`Medicine, vol. 154, no. 2, pp. 103–112, 2011.
`[32] R. A. DeFronzo, R. E. Ratner, J. Han, D. D. Kim, M. S.
`Fineman, and A. D. Baron, “Effects of exenatide (exendin-4)
`
`MPI EXHIBIT 1064 PAGE 7
`
`MPI EXHIBIT 1064 PAGE 7
`
`
`
`8
`
`Experimental Diabetes Research
`
`on glycemic control and weight over 30 weeks in metformin-
`treated patients with type 2,” Diabetes Care, vol. 28, no. 5, pp.
`1092–1100, 2005.
`[33] G. Derosa, P. Maffioli, S. A. T. Salvadeo et