`
`Diabetes, Obesity and Metabolism 13: 394–407, 2011.
`© 2011 Blackwell Publishing Ltd
`
`An overview of once-weekly glucagon-like peptide-1
`receptor agonists—available efficacy and safety data
`and perspectives for the future
`S. Madsbad1, U. Kielgast1, M. Asmar2, C. F. Deacon2, S. S. Torekov2 & J. J. Holst2
`1DepartmentofEndocrinology,CopenhagenUniversity,HvidovreHospital,Hvidovre,Denmark
`2DepartmentofBiomedicalSciences,ThePanumInstitute,UniversityofCopenhagen,Copenhagen,Denmark
`
`Incretin-based therapies, such as the injectable glucagon-like peptide-1 (GLP-1) receptor agonists and orally administered dipeptidyl peptidase-4
`(DPP-4) inhibitors, have recently been introduced into clinical practice. At present, the GLP-1 receptor agonists need to be administered once
`or twice daily. Several once-weekly GLP-1 receptor agonists are in phase 3 development. This review examines the efficacy, safety and
`perspective for the future of the once-weekly GLP-1 receptor agonists: exenatide once weekly, taspoglutide, albiglutide, LY2189265 and
`CJC-1134-PC, and compared them to the currently available agonists, exenatide BID and liraglutide QD. A greater reduction in haemoglobin A1c
`(HbA1c) and fasting plasma glucose was found with the once-weekly GLP-1 receptor agonists compared with exenatide BID, while the effect
`on postprandial hyperglycaemia was modest with the once-weekly GLP-1 receptor agonist. The reduction in HbA1c was in most studies greater
`compared to oral antidiabetic drugs and insulin glargine. The reduction in weight did not differ between the short- and long-acting agonists. The
`gastrointestinal side effects were less with the once-weekly agonists compared with exenatide BID, except for taspoglutide. Antibodies seem
`to be most frequent with exenatide once weekly, while hypersensitivity has been described in few patients treated with taspoglutide. Injection
`site reactions differ among the long-acting GLP-1 receptor agonists and are observed more frequently than with exenatide BID and liraglutide.
`In humans, no signal has been found indicating an association between the once-weekly agonists and C-cell cancer. The cardiovascular safety,
`durability of glucose control and effect on weight will emerge from several ongoing major long-term trials. The once-weekly GLP-1 receptor
`analogues are promising candidates for the treatment of type 2 diabetes, although their efficacy may not be superior to once-daily analogue
`liraglutide.
`Keywords: albiglutide, antidiabetic drug, CJC-1134-PC, exenatide once weekly, GLP-1 analogue, GLP-1 receptor agonists, incretin therapy,
`LY2189265, taspoglutide, type 2 diabetes mellitus
`
`Date submitted 16 September 2010; date of first decision 24 September 2010; date of final acceptance 21 December 2010
`
`Introduction
`The incidence and prevalence of type 2 diabetes are rising
`steadily worldwide, primarily as a consequence of the increasing
`prevalence of obesity [1]. Type 2 diabetes mellitus is a complex
`disease that
`involves genetic susceptibility for abnormal
`β-cell function resulting in relative insulin deficiency and
`insulin resistance in liver, muscle and fat cells as well
`as excessive glucagon secretion [2]. The defective β-cell
`function also involves an impaired responsiveness to the
`two incretin hormones glucagon-like peptide-1 (GLP-1) and
`glucose-dependent insulinotropic polypeptide (GIP) [2–4].
`The increased morbidity and mortality in type 2 diabetes are
`both consequences of microvascular (renal disease, neuropathy,
`retinopathy) and macrovascular complications [5].
`A variety of therapeutic options for the treatment of
`hyperglycaemia in people with type 2 diabetes are available.
`It is generally accepted that the initial therapy should consist
`
`Correspondenceto: Prof. Sten Madsbad, Department of Endocrinology, Copenhagen
`University, Hvidovre Hospital, Hvidovre 2650, Denmark.
`E-mail: sten.madsbad@hvh.regionh.dk
`
`of lifestyle changes plus metformin [6–9]. Some years after
`diagnosis most patients require combination therapy to
`achieve effective glycaemic control, but the lack of consensus
`regarding which agent to add to metformin has provoked
`debate among physicians [6–10]. Sulphonylurea (SU) and
`metformin represent together with insulin the ‘old agents’,
`while thiazolidinediones (TZD) have been used for the last
`decade. The incretin-based therapies, such as GLP-1 receptor
`agonists and dipeptidyl peptidase-4 (DPP-4) inhibitors, were
`recently introduced into clinical practice [11–13]. In addition,
`all of the agents, used alone or in combination, are associated
`with different adverse events including hypoglycaemia (SU and
`insulin), weight gain (SU, insulin and TZD), gastrointestinal
`side effects (metformin and GLP-1 receptor agonist) and
`increased risk of fractures (TZD) [2,6–10,12]. The DPP-4
`inhibitors are weight neutral [11,12,14,15].
`The incretin-based therapies have been the focus of
`much attention during the last years because of
`their
`unique mechanisms of action [3,16–18]. The GLP-1 receptor
`agonists potentiate insulin secretion, inhibit glucagon release,
`delay gastric emptying and reduce appetite and thereby
`
`article
`review
`
`
`
`MPI EXHIBIT 1052 PAGE 1
`
`MPI EXHIBIT 1052 PAGE 1
`
`
`
`DIABETES, OBESITY AND METABOLISM
`
`GLP-1 receptor agonists
`SC administered peptides
`
`Human GLP-1 backbone
`
`Exendin-4 backbone
`
`Weekly
`
`Once-daily
`
`Weekly
`
`QD BID
`
`Taspoglutide
`
`Liraglutide
`
`Albiglutide
`
`LY2189265
`
`CJC-1131
`
`Semaglutide
`
`Exenatide QW
`CJC-1134-PC
`
`Exenatide
`
`Lixisenatide
`
`Figure 1. Shows the glucagon-like peptide-1 (GLP-1) receptor agonists,
`which have already been approved (exenatide BID and liraglutide). The
`agonists are subdivided in relation to whether the backbone of the
`compound is human GLP-1 or exenatide, and in relation to the frequency
`of administration (once weekly or once daily or twice daily). The once-
`weekly GLP-1 receptor agonists illustrated with yellow are discussed in
`details in the present review.
`
`induce weight loss [3,16]. The DPP-4 inhibitors primarily
`improve insulin secretion and inhibit glucagon release, but
`have no clinically relevant effects on gastric emptying and
`appetite [3,11,16,17,19]. The potentiation of
`insulin and
`glucagon release is glucose dependent and therefore associated
`with a low risk of hypoglycaemia [3,11,16,17,19–21]. The
`combined improvement of glycaemic control and weight
`loss has promoted a particular interest in GLP-1 receptor
`agonists [11–13].
`The short half-life of native GLP-1 (1–2 min) has
`necessitated the development of long-acting GLP-1 receptor
`agonists for the management of type 2 diabetes [3,12,13,16,
`22]. The short half-life is due to inactivation by cleavage by
`the enzyme DPP-4 at the alanine residue at position 2 of the
`molecule [3,16,17,23, 24].
`The present review provides an update on currently available
`clinical trials that have assessed the efficacy and safety of
`exenatide twice daily (BID) and liraglutide as well as the
`long-acting once-weekly GLP-1 receptor agonists: exenatide
`once weekly, taspoglutide and albiglutide (figure 1). The once-
`weekly LY219265 and CJC-1134-PC will also be presented
`briefly (figure 1). Two other long-acting GLP-1 receptor
`agonists in development, CJC-1131 and semaglutide, will not be
`discussed, as information about the compounds is very sparse.
`
`Exenatide BID (Byetta) and Liraglutide
`(Victoza)
`Currently, two GLP-1 receptor agonists with extended half-lives
`are available for the treatment of type 2 diabetes [13,25–29].
`The first GLP-1 receptor agonist to reach the market (2005),
`(cid:2)
`exenatide (synthetic exendin-4; Byetta
`, Amylin Pharmaceu-
`ticals, Inc., San Diego, CA; Eli Lilly Company, Indianapolis,
`IN, USA), shares 53% amino acid homology with human GLP-
`(cid:2)
`1 [26,28,29]. Liraglutide (Victoza
`, Novo Nordisk, Bagsværd,
`Denmark) is 97% identical to the native hormone and has a
`fatty acid side chain promoting binding to human albumin
`
`review article
`
`after administration [25,27]. The half-life of exenatide after sc
`administration is about 2.4 h, and exenatide is therefore given
`twice daily, whereas the half-life of liraglutide is about 13 h
`and is administered as once-daily sc injection [13,25–29]. Exe-
`natide is given in micrograms (mcg), starting with 5 mcg BID
`and 10 mcg BID after 4 weeks if tolerated [13,26]. Liraglutide
`treatment is initiated with 0.6 mg once daily, increasing to
`1.2 mg after 1 week and in some patients up to 1.8 mg [13].
`The concentration of liraglutide is much higher than that
`of exenatide, but the fraction of the hormones that is not
`bound to albumin is very low, so that the concentration of
`free hormone is probably similar to that of the peak con-
`centration of exenatide [13,25,27]. Because of the large depot
`of bound liraglutide, its concentration varies little through-
`out the day (which also means that the timing of injection is
`uncritical), whereas the concentration of exenatide given twice
`daily (BID) varies from very low to therapeutic values [26,29].
`The uptitration is employed to reduce gastrointestinal side
`effects [13,25–27,29]. As nausea probably occurs at high peak
`plasma concentrations of GLP-1 [30], the lower incidence of
`nausea with liraglutide compared with exenatide BID may be
`explained by its sustained release formulation and tachyphy-
`laxia resulting from the sustained plasma level [25,27,31,32].
`Exenatide BID is currently approved (2005) for use as an
`adjunct twice-daily (BID) formulation injected before break-
`fast and dinner [13,26,29]. Once-daily liraglutide was approved
`in 2009 in Europe and in 2010 in USA and Japan.
`The reduction in haemoglobin A1c (HbA1c) with exenatide
`BID is about 0.5–1.0% in patients with a baseline HbA1c
`of 7.9–8.4%, whereas open-label comparator studies showed
`HbA1c reduction of 1.1–1.5% from baseline HbA1c values
`of 8.2–9.0% [13,26]. The reduction in HbA1c with liraglutide
`in clinical controlled trials of the LEAD (Liraglutide Effect
`and Action in Diabetes) programme was 0.8–1.5% in patients
`with an average baseline HbA1c of 8.2–8.5%. The reduction
`was in most cases greater or at least similar to oral comparator
`antidiabetic drugs [13]. In patients with a mean baseline HbA1c
`of about 8.5 and 9.8%, the reduction was 1.4 and 2.3%,
`respectively [13,33]. In a head-to-head comparison (LEAD
`6), the reduction in HbA1c was 0.33% greater (−1.12 vs.
`−0.79%) with liraglutide compared with exenatide [31,32].
`Also reduction in fasting plasma glucose was greater (−1.6 vs.
`−0.6 mmol/l), while weight loss did not differ significantly
`(−3.24 vs. −2.87 kg) [31,32]. In most phase 3 studies with
`exenatide and liraglutide, the weight loss was in the range of
`2–3 kg after 26 weeks of treatment compared with placebo
`and greatest when added to metformin [13]. In LEAD 6
`pancreatic β-cell function was improved, and triglycerides
`and free fatty acids were reduced to a greater extent with
`liraglutide than exenatide. However, the ability to reduce blood
`pressure (−2.5/1.1 vs. −2.0/1.9 mm Hg) was similar [31,32].
`The gastrointestinal side effects were most pronounced with
`exenatide BID, 28% having nausea and 9.9% vomiting
`compared with 25.5 and 6.0%, respectively, during treatment
`with liraglutide [31,32]. After 8–10 weeks the percentage of
`patients reporting nausea with liraglutide was below 10%,
`while in the exenatide group the level remained at about
`10% [31]. At week 26, only 2.5% of the liraglutide group
`
`Volume 13 No. 5 May 2011
`
`doi:10.1111/j.1463-1326.2011.01357.x 395
`
`
`
`MPI EXHIBIT 1052 PAGE 2
`
`MPI EXHIBIT 1052 PAGE 2
`
`
`
`review article
`
`had nausea compared with 8.6% in the exenatide group [31].
`Antibodies have been reported in approximately 50% during
`treatment with exenatide versus 4–13% in patients receiving
`liraglutide [13,34]. In most patients, the antibodies were of
`low titres and without apparent effect on efficacy [34,35]. In
`LEAD 6 liraglutide was less immunogenic than exenatide,
`and fewer than 10% of liraglutide-treated patients developed
`antibodies to liraglutide [34]. Overall, antiliraglutide antibodies
`were low, and did not impact the efficacy or safety of liraglutide
`treatment [34]. Overall,
`treatment
`satisfaction was rated
`slightly higher with liraglutide than exenatide BID [31,32].
`
`Exenatide Once Weekly (Bydureon®)
`An exenatide once-weekly (QW) formulation has been devel-
`oped using biodegradable polymeric microspheres that entrap
`exenatide (Amylin Pharmaceutical, Ely Lilly and Alkermes
`Incorporated, Cambridge, MA, USA) [36,37]. Exenatide is
`incorporated into a matrix of poly(d,l-lactide-co-glycolide)
`(PLG), which previously has been used as a biomaterial in
`sutures and in extended release preparations that allow gradual
`drug delivery at controlled rates [38]. Once released, exenatide
`is eliminated via the kidneys. After sc injection of 2 mg of exe-
`natide once weekly, a stable plasma exenatide level is obtained
`after 5–10 weeks, a level which is comparable to the peak
`concentrations observed with exenatide BID [36,37]. A plasma
`level of exenatide >50 pg/ml, which is known to reduce fasting
`plasma glucose concentration, is observed after about 2 weeks
`of treatment [36].
`
`Clinical Controlled Studies With Exenatide Once Weekly
`In a small clinical trial, patients with type 2 diabetes treated with
`diet and exercise or metformin monotherapy were randomized
`to placebo (n = 14), 0.8 mg exenatide once weekly (QW)
`(n = 16) or 2.0 mg exenatide QW (n = 15) [36]. The trial
`composed of 15 weeks of active treatment followed by a
`12-week follow-up. Average baseline HbA1c was 8.5%, and the
`reduction in HbA1c was −1.4 and −1.7% in the 0.8 and 2 mg
`groups, compared with an increase of +0.4% in the placebo
`group. The final HbA1c was 7.2 and 6.6% for the 0.8 and
`2.0 mg groups, respectively. More than 80% of patients treated
`with 2.0 mg reached HbA1c <7.0%. The study showed that the
`dose–response relationship for HbA1c reduction and weight
`control differs [36]. Thus, the 2.0 mg dose reduced weight
`significantly (−3.4 kg) compared with placebo treatment, while
`the 0.8 mg dose was without any effect on weight [36].
`
`DURATION-1
`In the DURATION-1 (Diabetes therapy Utilisation: Research-
`ing changes in HbA1c, weight and other factors Though
`Intervention with exenatide ONce weekly) trial, 10 mcg
`exenatide BID and 2 mg exenatide QW were compared in
`a 30-week trial including 295 type 2 patients [35]. Average
`baseline HbA1c was 8.2%, weight 102 kg, body mass index
`(BMI) 35 kg/m2 and duration of diabetes 6–7 years. The
`reduction in HbA1c was greater in the exenatide QW group
`(−1.9%) compared with—1.5% in the exenatide BID-treated
`
`DIABETES, OBESITY AND METABOLISM
`
`DURATION-1
`HbA1c=8.2%
`
`DURATION-2
`HbA1c=8.5%
`
`DURATION-3
`HbA1c=8.3%
`
`DURATION-5
`HbA1c=8.4%
`
`-1
`
`1.5
`
`1.9
`
`-2
`HbA1c
`(%)
`
`0.9
`
`1.2
`
`1.5
`
`1.3
`
`1.5
`
`0.9
`
`1.6
`
`Exenatide QW
`
`Sitagliptin
`
`Glargine
`
`Exenatide BID
`
`Piogitazone
`
`Figure 2. The effect of exenatide once weekly compared with oral
`antidiabetic agents and insulin glargine on changes in haemoglobin A1c
`(HbA1c) from baseline. Baseline HbA1c in the individual studies is also
`given.
`
`patients (figure 2). Mean difference was 0.33% in HbA1c.
`Most of the patients reached HbA1c ≤7.0 (77% for QW vs.
`61% for BID), and 49% in QW reached an HbA1c ≤6.5 and
`25% reached HbA1c ≤6.0%. The reduction in fasting plasma
`glucose was −2.3 and −1.4 mmol/l in QW and BID groups,
`respectively. Also fasting plasma glucagon was reduced more
`with QW, while reduction in postprandial glucose excursions
`and slowing in gastric emptying were less pronounced in QW
`compared with BID. The weight loss did not differ between
`the two groups by 30 weeks (−3.7 kg for QW vs. −3.6 kg for
`BID), and about 75% of the patients lost weight (figure 3). Both
`treatments were associated with reduction in triglycerides and
`blood pressure [35].
`treated with exenatide QW
`After 30 weeks, patients
`continued treatment, while patients who were treated with
`exenatide BID shifted to QW for 22 weeks [39]. Two hundred
`and twenty-eight of the initial 295 patients entered the open-
`label extension. Patients continuing exenatide QW maintained
`HbA1c improvement through 52 weeks (−2.0%). Patients
`switching from exenatide BID to QW achieved further
`improvements in HbA1c, and both groups displayed the
`same reduction and mean HbA1c (6.6%) at week 52, and
`71 versus 54% achieved HbA1c ≤7.0 and ≤6.5%, respectively.
`In patients with a basal HbA1c >9.0%, the reduction in
`HbA1c was 2.6–2.8%. The mean reduction in weight was
`about −3.6 to −3.7 kg (figure 3) [39]. After 1 year about 78%
`achieved reduction in both HbA1c and weight. The reduction
`in fasting plasma glucose was −2.5 mmol/l, but during the shift
`at week 30 the BID patients experienced a transient increase in
`fasting plasma glucose for few weeks, which was followed by a
`further improvement the following weeks. After 52 weeks, the
`reductions in systolic and diastolic blood pressures were −6.2
`and −2.8 mm Hg, respectively. Significant reductions in lipids,
`especially triglycerides, were shown [39].
`
`Safety and Tolerability. In DURATION-1, the incidence of
`nausea (26 vs. 35%) and vomiting (11 vs. 19%) was lower
`in QW compared with BID [35]. Injection site pruritus, or
`erythema, or induration or pain was observed in 18% of
`exenatide QW [35]. Most patients developed antibodies to
`exenatide QW (110 of 148) compared with 71 of 147 patients
`in the BID group. Antibodies to exenatide peaked in week
`
`396 Madsbad et al.
`
`Volume 13 No. 5 May 2011
`
`
`
`MPI EXHIBIT 1052 PAGE 3
`
`MPI EXHIBIT 1052 PAGE 3
`
`
`
`review article
`
`DURATION-2
`Most patients with type 2 diabetes often begin pharmacotherapy
`with metformin, but eventually need additional treatment. In
`DURATION-2, exenatide QW was compared with pioglitazone
`and sitagliptin to assess the potential differences between these
`antidiabetic drugs as add-on therapy to metformin [42]. The
`average baseline HbA1c was 8.5%, fasting plasma glucose
`9.1 mmol/l and BMI 32 kg/m2. Patients were randomly
`assigned to receive 2 mg exenatide QW (n = 170), 100 mg
`sitagliptin (n = 172) or 45 mg pioglitazone (n = 172) for
`26 weeks. Treatment with exenatide QW reduced HbA1c
`(−1.5%) significantly more than sitagliptin (−0.9%) or
`pioglitazone (−1.2%) (figure 2). The final HbA1c levels
`were 7.2, 7.7 and 7.4%, respectively. Significantly more
`patients reached HbA1c <7.0% with exenatide compared
`with sitagliptin or pioglitazone. The reduction in fasting
`plasma glucose was
`significantly greater with exenatide
`(−1.8 mmol/l) than with sitagliptin (−0.9 mmol/l) but not
`with pioglitazone (−1.5 mmol/l). Weight loss with exenatide
`(−2.3 kg) was
`significantly greater than with sitagliptin
`(−1.5 kg) or pioglitazone (+2.8 kg) (figure 3). The reduction in
`systolic blood pressure was significantly greater with exenatide
`(−4 mm Hg) compared with sitagliptin, but not pioglitazone.
`Diastolic blood pressure did not differ between the groups. The
`improvement in high-density lipoprotein (HDL) and reduction
`in triglycerides were greatest with pioglitazone. As in other
`studies with GLP-1 receptor agonists, a reduction in B-type
`natriuretic peptide as well as microalbuminuria was observed
`in the exenatide-treated group [13]. No major hypoglycaemia
`occurred in any group. About 24 and 10% registered nausea
`with exenatide and sitagliptin, while diarrhoea was observed
`in 18 and 10%, respectively. Fewer patients withdrew from
`treatment with sitagliptin (13%) than with exenatide (21%) or
`pioglitazone (21%). The improvement in treatment satisfaction
`was greatest with exenatide QW. Thus, the addition of exenatide
`QW to metformin achieved better glycaemic control and weight
`loss than sitagliptin and pioglitazone (figures 2 and 3) [42].
`It is relevant to compare these data with the results obtained
`during a 26-week head-to-head comparison between liraglutide
`and sitagliptin added to metformin in type 2 patients with a
`baseline HbA1c of 8.5%, fasting plasma glucose 10.0 mmol/l,
`BMI 33 kg/m2 and mean duration of diabetes 6–7 years [43].
`The lowering of HbA1c with liraglutide 1.2 and 1.8 mg was
`−1.24 and −1.50%, respectively, and −0.90% with sitagliptin
`100 mg. Nausea was more common with liraglutide 1.2 mg
`(21%) and 1.8 mg (27%) than with sitagliptin (5%). These
`findings may suggest that the efficacies of exenatide QW and
`liraglutide 1.8 mg once daily are similar. Currently, a study
`comparing exenatide QW and liraglutide once daily is ongoing
`(further information about the design of the study can be
`obtained at NCT01029886).
`
`DURATION-3
`Both exenatide BID and liraglutide once daily have been
`compared with insulin glargine [13].
`In the open-label
`DURATION-3 trial, once-weekly exenatide QW (2 mg) was
`compared with once-daily insulin glargine [44]. Seventy
`
`DIABETES, OBESITY AND METABOLISM
`
`DURATION-1
`BMI=35
`
`DURATION-2
`BMI=32
`
`DURATION-3
`BMI=32
`
`DURATION-5
`BMI=33
`
`2.8
`
`1.5
`
`2.3
`
`1.4
`
`2.6
`
`1.4
`
`2.3
`
`3.6
`
`3.7
`
`Kg
`
`Exenatide QW
`
`Sitagliptin
`
`Glargine
`
`Exenatide BID
`
`Piogitazone
`
`3
`
`12
`
`-1
`
`-2
`
`-3
`
`Figure 3. The effect of exenatide once weekly compared with oral
`antidiabetic agents and insulin glargine on changes in weight (kg) from
`baseline. Baseline body mass index (BMI) in the individual studies is also
`given.
`
`6 for both treatments, but the titres were about three times
`higher during exenatide QW compared with BID [35]. Overall,
`the titre of antibodies was not predictive of individual HbA1c
`change or adverse events [35].
`The DURATION-1 study illustrates that exenatide QW
`is more effective in reducing HbA1c and fasting plasma
`glucose than BID, while the reduction in weight did not
`differ. Treatment with exenatide QW was generally well
`tolerated, and the only incidences of hypoglycaemia occurred
`in patients concomitantly receiving SU. It is also noteworthy
`that patients switching from the short-acting exenatide BID
`to QW experience a transient deterioration in glycaemic
`control, which generally improved 2 weeks after initiating
`exenatide QW [39]. Furthermore, the reduction in HbA1c
`was maintained during the 52 weeks and resulted in a
`mean HbA1c of 6.6%. The greater reduction in HbA1c with
`exenatide QW compared with BID illustrates the effect of a
`continuous exposure for exenatide during all 24 h compared
`with the intermediate exposure obtained with exenatide BID
`with deterioration of control during night and lunch time.
`During night time, lower glucagon levels during QW treatment
`are likely to contribute to the improvement in fasting glucose
`level. Conversely, although both therapies reduced postprandial
`glucose excursions, the absolute reduction in postprandial
`glucose excursion and inhibition of gastric emptying were
`greater with exenatide BID than QW [35]. Thus, acute exposure
`to exenatide produces greater inhibition of gastric emptying
`than that seen with continuous GLP-1 receptor activation,
`probably illustrating the development of tachyphylaxis during
`continuous exposure [40]. Both groups experienced significant
`improvements in treatment satisfaction and quality of life,
`but patients who switched from BID to QW administration
`reported further significant improvement after 30 weeks [41].
`The mean difference between exenatide QW and BID in
`reduction in HbA1c (0.33%) did not differ from the 0.33%
`difference in reduction in HbA1c between exenatide BID and
`liraglutide 1.8 mg once daily [31].
`
`Volume 13 No. 5 May 2011
`
`doi:10.1111/j.1463-1326.2011.01357.x 397
`
`
`
`MPI EXHIBIT 1052 PAGE 4
`
`MPI EXHIBIT 1052 PAGE 4
`
`
`
`review article
`
`percent of the patients were treated with metformin and 30%
`metformin plus SU. Starting dose for insulin glargine increased
`from baseline 10 to 31 IU/day, targeting a fasting glucose range
`of 4.0–5.5 mmol/l following a prespecified titration algorithm.
`Average baseline HbA1c was 8.3%, fasting plasma glucose
`9.8 mmol/l, BMI 32 kg/m2 and duration of diabetes about
`8.0 years. The reduction in HbA1c was greater in the exenatide
`group (−1.5%) than in those taken insulin glargine (−1.3%)
`(figure 2). Endpoint HbA1c was 6.8 versus 7.0%, and 60 versus
`48% reached an HbA1c <7.0%. Mean weight changes were
`−2.6 kg in the exenatide group and +1.4 kg in the insulin
`glargine-treated patients (figure 3). Seventy-nine percent of
`the patients allocated to exenatide had both a reduction in
`HbA1c and weight, whereas 63% of the patients receiving
`insulin glargine had a reduction in HbA1c paired with a
`weight gain [44]. Fasting plasma glucose was reduced in both
`groups (exenatide −2.1 mmol/l, insulin glargine −2.8 mmol/l,
`p < 0.001). Mean heart rate at week 26 was raised compared
`with baseline in the exenatide but not in the insulin glargine
`group. No other cardiovascular risk factors including lipid
`concentrations differed between the groups. One hundred and
`twenty-seven of 233 patients assigned to exenatide developed
`antiexenatide antibodies, and a lower mean reduction in HbA1c
`was observed in the antibody-positive group compared with
`patients not developing antibodies (−1.3 vs. −1.6%) [44].
`Minor hypoglycaemia was reported in 19 of 233 exenatide
`patients (46 events) compared with 58 of 233 insulin glargine
`patients (135 events), which was significantly different. One
`patient taking exenatide developed pancreatitis. Calcitonin
`concentrations were measured in few patients and were within
`normal limits in all patients. The number of patients, who
`discontinued treatment because of adverse effects, was 5
`versus 1%, respectively. More patients discontinued exenatide
`QW than insulin glargine due to nausea and injection site
`reactions [44].
`Thus, the exenatide once-weekly treatment resulted in
`greater HbA1c reduction after 26 weeks than insulin glargine.
`Insulin glargine produces greater reduction in fasting glucose
`than did exenatide, while significantly greater reductions in
`postprandial glucose excursions were obtained with exenatide.
`Risk of hypoglycaemia was reduced with exenatide, irrespective
`of background treatment. A notable strength of the study is
`that it included a standard next step (insulin treatment) in the
`treatment of patients not responding to two oral antidiabetic
`agents as an active comparator. An extension period planned
`for 2.5 years is in progress.
`Exenatide BID has previously been compared with insulin
`glargine in a 6-month trial, where the reduction in HbA1c
`did not differ between the groups (i.e. reduction was 1.1% in
`both groups) [45]. Liraglutide has also been compared with
`insulin glargine in a 6-month study, with a difference in HbA1c
`treatment effect (0.2%) and body weight in favour of liraglutide
`(LEAD 5) [46].
`
`DURATION-4
`In the fourth of the series of DURATION studies (DURATION-
`4), exenatide once weekly is compared with sitagliptin
`100 mg, pioglitazone 45 mg or metformin up to 2500 mg,
`
`DIABETES, OBESITY AND METABOLISM
`
`in monotherapy (the design of the study is given at
`all
`NCT00876338). No data have been published.
`
`DURATION-5
`DURATION-5, like DURATION-1, compared exenatide QW
`and BID during a 26-week study in 252 type 2 patients with
`an average baseline HbA1c of 8.4%, fasting plasma glucose
`9.1 mmol/l and weight 96 kg [47]. Patients were drug na¨ıve
`(19%) or treated with one (47%) or a combination of (34%)
`oral antidiabetic drugs. After 26 weeks, the reduction in HbA1c
`was greater in QW (−1.6%) than in BID (−0.9%) (figure 2).
`Fifty-nine percent versus 30% reached the goal of <7.0%.
`Weight loss was −2.3 versus −1.4 kg after 24 weeks (p = NS)
`(figure 3). Nausea occurred less frequently with QW (14%)
`than with BID (35%), and was transient and mild or moderate
`in intensity in most patients. Injection site reactions were
`more common with QW. No change in mean calcitonin
`concentrations was observed, but one patient withdrew due
`to pancreatitis. Thus, also in DURATION-5 exenatide QW
`provides superior control compared to exenatide BID [47].
`
`DURATION-6
`Is a head-to-head comparison between exenatide QW and
`once-daily liraglutide 1.8 mg, including approximately 900
`patients, estimated completion in 2011 (NCT01029886).
`Figures 2 and 3 summarize the changes in HbA1c and weight
`in the DURATION-1, -2, -3 and -5 studies.
`
`Regulatory Affairs
`In a response letter in October 2010, US Food and Drug
`Administration (FDA) requested a thorough QT interval study
`with exposures of exenatide higher than typical therapeutic
`levels of exenatide QW. The background for the request could
`be that after a single dose of 10 mcg of exenatide in healthy
`subjects, a slight positive correlation between plasma exenatide
`concentrations and changes from baseline in QT interval has
`been observed [48]. Additionally, the FDA has requested the
`results of DURATION-5 study to evaluate the efficacy and
`the labelling of the safety and effectiveness of the commercial
`formulation of exenatide QW. The Amylin, Lilly and Alkermes’
`goal is to submit their reply to the response letter by the end
`of 2011. Based on the requirements for additional data, the
`resubmission will likely require a 6-month review by FDA.
`The decision from the European Medical Agency (EMA) about
`exenatide QW can be expected in 2011.
`
`Taspoglutide
`The human GLP-1 receptor agonist taspoglutide (Roche,
`Basel, Switzerland; Ipsen, Paris, France) has 93% homology
`with the native hormone [49]. Taspoglutide contains two
`α-aminoisobutyric acid substitutions replacing Ala8 and Gly35
`of hGLP-1(7-36)NH2 [49]. Taspoglutide is fully resistant to
`DPP-4 degradation [49]. The biological actions have been
`shown to be similar to those of native GLP-1, and after a single
`dose of 30 mg, a glucose-lowering effect was found for up to 2
`
`398 Madsbad et al.
`
`Volume 13 No. 5 May 2011
`
`
`
`MPI EXHIBIT 1052 PAGE 5
`
`MPI EXHIBIT 1052 PAGE 5
`
`
`
`DIABETES, OBESITY AND METABOLISM
`
`weeks [49]. Taspoglutide has been shown to protect β cells by
`reducing apoptosis in Zucker diabetic fatty (ZDF) rats, a rodent
`model of type 2 diabetes [50]. In type 2 patients, taspoglutide
`restored both first- and second-phase insulin secretion [51].
`Roche licensed the drug in 2006 from Ipsen SA.
`At present, only two trials have been published as full
`articles [52,53]. In a phase 2 study, type 2 diabetic patients
`(n = 306, mean age 55 years, BMI 32.7 kg/m2 and duration of
`disease 5 years) treated with metformin were randomized to
`8-week treatment with placebo, taspoglutide, either 5, 10 or
`20 mg once weekly or 10 or 20 mg once every second week [52].
`Baseline HbA1c was 7.9%. The reduction in HbA1c was −1.0%
`(5 mg QW), −1.2% (10 mg QW), −1.2% (20 mg QW) and
`−0.9% (10 mg Q2W) and −1.0 (20 mg Q2W) versus—0.2%
`with placebo. The greatest reduction in fasting plasma glucose
`was observed with 10 and 20 mg QW (−2.5 mmol/l compared
`with −0.8 mmol/l with placebo). After 8 weeks, weight
`loss was greater in the 10 mg QW (−1.9 kg), 20 mg QW
`(−2.8 kg) and 20 mg Q2W (−1.9 kg) than in the placebo
`group (−0.8 kg) [52]. Taspoglutide has also been investigated
`in a smaller and shorter phase 2 studies [53,54].
`During the 2010 meetings in the American Diabetes
`Association Meeting and the European Association for the
`Study of Diabetes, five phase 3 studies from the T-emerge (effect
`of human weekly GLP-1 for glycaemic control) programme
`were presented. All studies were of 24 weeks’ duration, but
`with an extension to 52 weeks. The results after 24 weeks
`were presented. About 6000 patients have been enrolled in the
`T-emerge programme.
`
`T-emerge 1
`The T-emerge 1 trial is a double-blinded placebo-controlled
`study in drug na¨ıve patients [55]. Patients (mean age 55 years,
`BMI 32 kg/m2, baseline HbA1c 7.6% and duration of diabetes
`about 3 years) were randomized to 10 mg taspoglutide QW
`(n = 112), 10 mg taspoglutide QW for 4 weeks titrated to 20 mg
`QW (n = 127) or placebo (n = 115) for 24 weeks. Reduction
`in HbA1c (−1.0, −1.2 vs. −0.1 %) (figure 4) and fasting
`plasma glucose (−1.6, −1.9 vs. −0.1 mmol/l) after 24 weeks
`was significantly greater in the taspoglutide groups compared
`with placebo, while weight reduction was significantly greater
`only in 20 mg taspoglutide QW compared with placebo (−1.5,
`−2.2 vs. −1.2 kg) (figure 5). An HbA1c target of ≤7.0 was
`reached by 76, 80 and 37%, respectively. Nausea was observed
`in 26, 31 and 4%, vomiting in 17, 18 versus 0% and diarrhoea in
`14, 10 and 4% of the patients in the three groups, respectively.
`Withdrawal due to gastrointestinal side effects occurred in
`5.2, 7.8 and 0.8% of the patients, respectively. Incidence of
`injection site nodules, or indurations, or pruritus was 4–12%
`in the taspoglutide groups versus about 1% in the placebo
`group. Incidence of hypoglycaemia did not differ between the
`groups [55].
`
`T-emerge 2
`In the T-emerge 2 trial, taspoglutide was compared with
`exenatide BID in type 2 patients inadequately controlled
`with metformin +/− TZD [56]. In this open-label trial,
`
`review article
`
`T-emerge 1
`HbA1c=7.6%
`
`T-emerge 2
`HbA1c=8.1%
`
`T-emerge 4
`HbA1c=8.0%
`
`T-emerge 5
`HbA1c= 8.3 %
`
`T-emerge 7
`HbA1c=7.6%
`
`0.1
`
`0.1
`
`