Bioorganic & Medicinal Chemistry Letters 23 (2013) 4011–4018
`
`Contents lists available at SciVerse ScienceDirect
`
`Bioorganic & Medicinal Chemistry Letters
`
`j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / b m c l
`
`BMCL Digest
`Recent progress and future options in the development of GLP-1
`receptor agonists for the treatment of diabesity
`
`Martin Lorenz a,⇑, Andreas Evers b, Michael Wagner a,⇑
`
`a Diabetes Division/Res. & Transl. Med, Sanofi-Aventis Deutschland GmbH, Industriepark Höchst, 65926 Frankfurt am Main, Germany
`b LGCR/Struct., Design & Informatics, Sanofi Deutschland GmbH, 65926 Frankfurt am Main, Germany
`
`a r t i c l e
`
`i n f o
`
`a b s t r a c t
`
`Article history:
`Received 16 March 2013
`Revised 6 May 2013
`Accepted 7 May 2013
`Available online 16 May 2013
`
`Keywords:
`GLP-1 receptor agonists
`Gastro-intestinal hormones
`Diabesity
`Delivery technologies
`
`The dramatic rise of the twin epidemics, type 2 diabetes and obesity is associated with increased mortal-
`ity and morbidity worldwide. Based on this global development there is clinical need for anti-diabetic
`therapies with accompanied weight reduction. From the approved therapies, the injectable glucagon-like
`peptide-1 receptor agonists (GLP-1 RAs) are the only class of agents which are associated with a modest
`weight reduction. Physiological effects of the gastro-intestinal hormone GLP-1 are improvement of gly-
`cemic control as well as a reduction in appetite and food intake. Different approaches are currently under
`clinical evaluation to optimize the therapeutic potential of GLP-1 RAs directed to once-weekly up to once-
`monthly administration. The next generation of peptidic co-agonists comprises the activity of GLP-1 plus
`additional gastro-intestinal hormones with the potential for increased therapeutic benefits compared to
`GLP-1 RAs.
`
`Ó 2013 Elsevier Ltd. All rights reserved.
`
`Introduction. Type 2 diabetes mellitus (T2DM) is a progressive
`chronic disease characterized by hyperglycemia due to defective
`insulin secretion and resistance to insulin action. The disease is
`associated with morbidity and mortality in patients, and is a lead-
`ing cause for cardiovascular (CV) disease, renal failure, blindness,
`amputations and hospitalizations.1,2 T2DM has recently reached
`epidemic proportions in both developed and developing countries
`in conjunction with a substantial increase in obesity.3 Up to 80% of
`people with T2DM are overweight or obese, whereas obesity is
`considered as a major risk factor for T2DM.4 ‘Diabesity’ is a new
`term to characterize this phenomenon of obesity-dependent diabe-
`tes associated with multiple comorbidities (mainly CV disease) and
`in addition describes a rising epidemic.5 An effective approach to
`the management of diabesity is a modest reduction in body weight
`(4–5 kg) resulting in highly beneficial effects on glycemic control
`as well as reduced morbidity and mortality.4 According to the new
`guidelines of the American Diabetes Association (ADA) and the
`European Association for the Study of Diabetes (EASD) out of avail-
`able classes of glucose-lowering agents, the only class associated
`with modest body weight reduction are peptidic glucagon-like
`peptide-1 receptor agonists (GLP-1 RAs). All other agents are either
`weight neutral (Metformin, DPP4-inhibitors) or associated with
`weight gain (basal insulin, thiazolidinediones, sulfonylureas).6
`
`⇑ Corresponding authors. Tel.: +49 69 305 46875 (M.W.); tel.: +49 69 305 12041
`
`(M.L.).
`E-mail addresses: Martin.Lorenz@sanofi.com (M. Lorenz), Andreas.Evers@sanofi.
`com (A. Evers), Michael.Wagner@sanofi.com (M. Wagner).
`
`0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
`http://dx.doi.org/10.1016/j.bmcl.2013.05.022
`
`Therapy with peptidic GLP-1 RAs is based on self-administra-
`tion by subcutaneous injection using a pen device (or a syringe).
`The following GLP-1 RAs are approved for the treatment of diabe-
`tes: twice-daily exenatide BID (ByettaÒ, BMS), once-daily lixisena-
`tide (LyxumiaÒ, Sanofi/Zealand), once-daily liraglutide (VictozaÒ,
`Novo Nordisk) and once-weekly exenatide LAR (BydureonÒ, BMS).
`Current optimization of peptidic GLP-1 RAs aims to improve
`efficacy, tolerability and compliance such as frequency and ease
`of administration. Furthermore, significant efforts are currently
`undertaken to identify therapeutic peptide approaches with the
`potential for greater body weight reduction compared to existing
`GLP-1 RAs. The present article reviews key data on marketed and
`novel long-acting GLP-1 RAs currently in late clinical development,
`describes novel techniques suitable for once-weekly up to once-
`monthly administration and discusses potential novel peptide ap-
`proaches based on further gut hormones. However, recent research
`activities have resulted also in peptidic and small molecule GLP-1
`RAs which might be suited for oral application7–among those
`TTP054 from Transtech Pharma as most advanced in phase 2.8
`Those still early developments in the field of orally available
`GLP-1 RAs are not covered by this review.
`GLP-1 RAs: overview. Natural human glucagon-like peptide-1
`(GLP-1) is an incretin hormone derived from the transcription
`product of the proglucagon gene. It is produced by intestinal L-cells
`and released in response to meal intake to induce insulin secretion
`from pancreatic b-cells. The biologically active forms of GLP-1 are
`GLP-1(7–37) and GLP-1(7–36)NH2, which exert their action by
`activation of the GLP-1 receptor, a class B G-protein-coupled
`receptor. GLP-1 shows a helical character in solution and in the
`
`MPI EXHIBIT 1139 PAGE 1
`
`

`

`4012
`
`M. Lorenz et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4011–4018
`
`receptor-bound conformation (see Fig. 1).9,10 Exendin-4 is another
`potent peptidic GLP-1 RA, which was isolated from the venom of
`the Gila monster. Although it shares only 53% sequence identity
`with native GLP-1, both peptides show a helical character with
`many of the conserved amino acids facing the interaction site of
`the receptor (see Fig. 1).11 All GLP-1 RAs on the market or in clinical
`development are either derived from the natural GLP-1 or exendin-
`4 peptide. Based on their PK-profile, marketed GLP-1 RAs exenatide
`BID and lixisenatide could be classified as shorter-acting agents
`compared with longer-acting agents liraglutide and exenatide LAR.
`GLP-1 and GLP-1 RAs have 3 major pharmacological activi-
`ties12,13 to improve glycemic control in patients with T2DM by
`reducing fasting and postprandial glucose (FPG and PPG): (i) in-
`creased glucose-dependent insulin secretion (improved first- and
`second-phase), (ii) glucagon suppressing activity under hypergly-
`cemic conditions, (iii) delay of gastric emptying rate resulting in re-
`tarded absorption of meal-derived glucose. Treatment with GLP-1
`RAs for 12–52 weeks leads to reduction in glycosylated hemoglo-
`bin (HbA1c), a diagnostic marker reflecting the average glucose
`levels of the last 2–3 months (and serving as primary efficacy end-
`point in phase 2b–3 studies) of 1.1–1.6% due to reduced postpran-
`dial and fasting blood glucose.14 Unlike insulin or
`insulin
`secretagogues (sulfonylureas) the insulinotropic effect of GLP-1
`RAs is strictly glucose-dependent resulting in a generally low risk
`for hypoglycemia. Beyond anti-hyperglycemic actions of GLP-1
`RAs, central and possibly peripheral effects (e.g. vagal afferents)
`which modulate feelings of appetite and satiety are considered to
`reduce food intake and finally account for a moderate body weight
`reduction in the range of 1–3 kg.15
`In addition, beneficial reduction in systolic and diastolic blood
`pressure as well as reduced cholesterol plasma levels were ob-
`served in clinical trials.16 Preclinical data illustrate different cardio-
`protective effects of GLP-1 RAs (e.g. on cardiomyocytes, blood
`vessels, adipocytes and lipids).17 These effects might impair the on-
`set or progression of atherosclerotic disease. Furthermore, animal
`and initial clinical studies have shown that GLP-1 RAs could reduce
`ischemia–reperfusion injury. However, treatment with longer-
`
`acting GLP-1 RAs (liraglutide and exenatide LAR) is accompanied
`in humans by a slight increase in heart rate of 2–4 beats per min-
`ute. Ongoing cardiovascular outcome studies for the marketed
`agents will reveal if the potential harm of an increased heart rate
`will be outweighed by the reported cardiovascular benefits.17
`The most common side effects of GLP-RAs are nausea, vomiting
`and diarrhea, which are mostly transient and can be at least partly
`reduced by gradual uptitration of the dose. Beyond gastro-intesti-
`nal side effects an association of long-term GLP-1RA therapy with
`an increased risk for pancreatitis is under discussion. Some rare
`cases of acute pancreatitis have been reported as postmarketing
`surveillance activities for GLP-1 RAs. It is not clear if such cases
`are therapy-related or due to a general 2–3-fold increased risk
`for acute pancreatitis associated with T2DM.18 In rodents malig-
`nant thyroid C-cell carcinomas were observed following treatment
`with liraglutide. The human relevance of these findings in rodents
`could not be determined by clinical or nonclinical studies.18
`As peptide based drugs, GLP-1 RAs have the potential to induce
`an immune response and antibody formation was reported for all
`marketed GLP-1 RAs but does not appear to impact efficacy or
`safety of these agents. The level of glycemic control (HbA1c) is
`overall reported similar regardless of the antibody status, although
`for a rare number of patients high-titer antibody formation might
`be associated with attenuated glycemic response (reported for 1–
`4% of total patients treated with exenatide BID).19 There seems to
`be slightly increased antibody development and injection site reac-
`tions with exendin-4 compared to the GLP-1 analog liraglutide,20
`presumably due to the greater differences in sequence identity
`compared to native GLP-1.
`A differentiating characteristic within the GLP-1 RA class impor-
`tant for glycemic control is the propensity to slow gastric empty-
`ing, which is dependent on the pharmacokinetics of these agents.
`For longer-acting agents such as liraglutide and exenatide LAR, de-
`layed gastric emptying fades following the first dose probably
`reflecting tachyphylaxis.21,22 These agents strongly reduce fasting
`blood glucose probably via continuous increased insulin levels in
`the fasting state and cause an apparently better efficacy than
`
`Figure 1. Binding hypotheses of GLP-1 RAs at the GLP-1R. Amino acid changes versus native GLP-1 are shown in orange. Binding mode of GLP-1 from RCSB PDB (PDB ID 3iol),
`binding modes of the remaining peptides are computational models.
`
`MPI EXHIBIT 1139 PAGE 2
`
`

`

`M. Lorenz et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4011–4018
`
`4013
`
`shorter-acting GLP-1 RAs.14 In contrast, the shorter-acting agents,
`exenatide BID and lixisenatide are both proven to strongly slow
`gastric emptying in patients with T2DM without any signs for
`desensitization with continuous treatment.23,24 As consequence,
`shorter-acting GLP-1 RAs are associated with a pronounced reduc-
`tion in PPG. As HbA1c levels decrease under therapy, the contribu-
`tion of postprandial glucose to HbA1c increases over that of fasting
`plasma glucose.25 For patients with improved HbA1c under ther-
`apy (including those on insulin treatment), who have a greater
`dependency on postprandial glucose, shorter-acting agents may
`be the better choice. This greater PPG-lowering effect of shorter-
`acting GLP-1 RAs would be complementary to the glucose lowering
`effects of basal insulin, which primarily targets fasting glucose.
`Hence, a combination therapy of shorter-acting GLP-1 RAs with ba-
`sal insulin may offer the advantage to lower basal insulin require-
`ments and result in beneficial weight loss effects.26
`GLP-1 agonists on the market and in clinical development. Thera-
`peutic utility of natural GLP-1 is limited by its rapid degradation by
`serum proteases, predominantly dipeptidyl peptidase IV (DPPIV),
`but also other enzymes such as neutral endopeptidase (NEP), plas-
`ma kallikrein or plasmin.27 Therefore, GLP-1 has a very short half-
`life of 2 min following intravenous administration. A variety of
`strategies have been applied to provide novel GLP-1 RAs with
`longer-term in vivo activity following subcutaneous administra-
`tion, which are described below.
`Peptides stabilized against DPPIV by amino acid exchange. One
`strategy to prolong in vivo half-life is stabilization towards degra-
`dation by DPPIV, which preferably cleaves N-terminal Xaa–Pro or
`Xaa–Ala dipeptide sequences. Alteration of that N-terminal se-
`quence, especially the second amino acid, has proven to reduce
`degradation by DPPIV.28,29
`Exenatide BID. In exendin-4, the second amino acid is a Gly ren-
`dering it resistant to DPPIV mediated degradation. Furthermore,
`the Leu21–Ser39 span of exendin-4 forms a compact tertiary fold
`(the Trp-cage) which shields the side chain of Trp25 from solvent
`exposure, leading to enhanced helicity and stability of the peptide
`(see Fig. 1).9 Exenatide BID (Amylin, now BMS), a synthetic version
`of exendin-4, represents the first GLP-1 RA approved in 2005 as
`antidiabetic therapy for the treatment of T2DM. It has a terminal
`half-life of 2.4 h after subcutaneous administration and is applied
`twice daily (10 lg). With this dosing regimen as monotherapy or in
`combination with oral antidiabetics, HbA1c reductions of 0.8–0.9%
`are typically observed accompanied by a body weight reduction in
`the range of 1.8–2.6 kg following a treatment duration of 30 weeks.
`In an open label extension the combination of exenatide BID and
`metformin showed a HbA1c reduction of 1.1% and a body weight
`reduction of 4.4 kg after 82 weeks.30 Exenatide BID and insulin
`glargine did not demonstrate a significant difference in HbA1c
`reduction (1.25% vs 1.26%) after 26 weeks treatment. However,
`insulin glargine resulted in a weight gain of 3 kg compared to a
`reduction of 2.7 kg seen with exenatide BID.31
`Lixisenatide. Lixisenatide, developed by Sanofi/Zealand, is a syn-
`thetic analog of exendin-4 (Fig. 1). Compared to exendin-4, six Lys
`residues have been added to the C-terminus (also amidated), while
`one Pro in the C-terminal region has been deleted. Lixisenatide has
`a mean terminal half-life of approximately 3 h in humans. In a
`dose-ranging study in patients with T2DM, inadequately con-
`trolled with metformin, lixisenatide at a dose of 20 lg once-daily
`demonstrated the best efficacy-to-tolerability ratio.32 Lixisenatide
`was applied as once-daily administration (20 lg) in phase 3 trials
`and has demonstrated efficacy as monotherapy, in combination
`with oral antidiabetic drugs, and as add-on to basal insulin, with
`particular efficacy in reducing postprandial glucose excursion
`(HbA1c reduction up to 0.9%). In combination with oral antidia-
`betic drugs, lixisenatide resulted in a sustained body weight reduc-
`tion from baseline in controlled studies (24 weeks) in a range from
`
`1.8–3 kg. Comparison of lixisenatide with exenatide BID as add-on
`to metformin demonstrated noninferior improvements in HbA1c,
`but with less hypoglycemia, slightly less weight loss and a more
`favorable gastro-intestinal tolerability profile after 24 weeks.33 Lix-
`isenatide significantly delays gastric emptying, a process which is
`accompanied by strong post-prandial glucose lowering.34 These
`attributes are very attractive with regard to a combination therapy
`with basal insulins, for example, insulin glargine, where Lixisena-
`tide demonstrated significantly reduced HbA1C (0.7–0.8%) and
`body weight (1.8 kg).34 In February 2013, the EMA granted market-
`ing authorization in Europe for lixisenatide for the treatment of
`adults with T2DM.35
`Sustained release of DPPIV stabilized peptides. DPPIV-resistant
`peptidic GLP-1 RAs are still subjected to renal clearance as illus-
`trated by the rather moderate half-lives of exenatide and lixisena-
`tide. Therefore, several approaches have been explored to further
`prolong the duration of action by either creating a depot (e.g. in
`the subcutaneous space), from which the peptide is slowly re-
`leased, or by conjugating peptidic GLP-1 agonists to carrier mole-
`cules in order to reduce renal clearance.
`Taspoglutide. Taspoglutide developed by Roche (in collaboration
`with Ipsen) is a close analog of natural GLP-1(7–36) in which the
`unnatural amino acid aminoisobutyric acid (Aib) has been intro-
`duced in position 8 and 35 in order to avoid degradation by DPPIV,
`but also by other serine proteases such as plasma kallikrein and
`plasmin.36 NMR-studies with taspoglutide showed a similar sec-
`ondary structure compared to native GLP-1, but clearly an in-
`creased a-helicity in the C-terminal part of
`the peptide.
`Taspoglutide was developed as a sustained release formulation
`containing zinc chloride suitable for once-weekly administration.
`When injected into the subcutaneous tissue, taspoglutide precipi-
`tates and forms a depot. Taspoglutide was studied in several phase
`3 studies with a 10 and 20 mg once-weekly dosing regimen, and
`showed consistent HbA1c and weight reductions.37–41 Significant
`weight reductions were only observed with the 20 mg dose.42
`Once-weekly taspoglutide was superior to twice-daily exenatide
`with respect to glycemic control, while resulting in a similar
`weight loss after treatment for 24 and 52 weeks. The overall safety
`profile comprising gastro-intestinal tolerability, systemic allergic
`reactions and injection-site reactions was clearly worse, especially
`the gastro-intestinal side effects, which resulted in a roughly dou-
`bled discontinuation rate in patients treated with taspoglutide
`compared to exenatide (34 vs 16%). Based on this phase 3 data
`Roche decided to stop the development of taspoglutide.43
`Exenatide LAR–Bydureon. Further strategies to create a subcuta-
`neous depot from which the peptidic GLP-1 RA of choice can be
`slowly released utilize polymeric matrices. Exenatide LAR (devel-
`oped by Amylin/Lilly/Alkermes–now BMS) is a once-weekly for-
`mulation of exenatide,
`in which exenatide is noncovalently
`entrapped into a biodegradable polymeric matrix consisting of
`poly(D,L-lactide-co-glycolide) (PLG) forming microspheres.44 Slow
`release from the polymeric matrix takes place through diffusion
`and microsphere breakdown. Exenatide LAR has a half-life in hu-
`mans of 5–6 days. After a 2 mg s.c. injection, steady state plasma
`levels of exendin-4 are typically obtained after 6–10 weeks.13 At
`that dose HbA1c reductions of 1.3–1.9% were observed. A direct
`comparison with exenatide BID revealed a better reduction in
`HbA1c (1.9 vs 1.5%) with a similar reduction in body weight (3.6
`vs 3.7 kg).45 Different from this efficacy data, a head-to-head com-
`parison with liraglutide showed an improved glycemic control fol-
`lowing 26 weeks of treatment associated with increased weight
`reduction in favor of liraglutide (HbA1c reduction 1.5% vs 1.3%,
`body weight reduction 3.6 kg vs 2.7 kg).46 One drawback of exena-
`tide LAR is the relatively large needle size (23 gauge), which is used
`for administration due to the viscosity of the polymeric suspen-
`sion, as well as the quite laborious preparation prior to injection.
`
`MPI EXHIBIT 1139 PAGE 3
`
`

`

`4014
`
`M. Lorenz et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4011–4018
`
`Fatty acid conjugation. The concept of fatty acid conjugation is a
`well established strategy to prolong the action of peptides by facil-
`itating binding to serum albumin thereby reducing their renal
`clearance. Landmark achievements in this area have been the
`development of long acting insulin analogs.47,48
`Liraglutide. Liraglutide (VictozaÒ) developed by Novo Nordisk is
`a close structural homolog to GLP-1(7–37) with 97% sequence
`identity to the native hormone. Lys in position 34 is substituted
`by Arg and a palmitic acid is conjugated to Lys in position 26 via
`a glutamate spacer (see Fig. 2a).20 This combination of spacer
`and fatty acid length turned out to be optimal
`in terms of
`in vitro activity and in vivo prolonged duration of action in
`pigs.49,50 The mechanisms for prolongation of action are manifold.
`Following subcutaneous injection, the peptide is slowly released
`from the injection site due to self-association.51 Once it enters
`the bloodstream, liraglutide is extensively bound to serum albu-
`min (99%), which leads to an increased enzymatic stability to-
`wards DPPIV and NEP while reducing renal clearance. The plasma
`half-life in humans is 11–13 h.20 Liraglutide was approved in 2009
`for the treatment of T2DM. The therapeutic standard dose is 1.2 mg
`once-daily with possible increase to 1.8 mg to further improve gly-
`cemic control. The HbA1c reductions observed with these dosing
`regimens in phase 3 trials were 1.1–1.8% accompanied by a weight
`loss of around 2–3 kg (26 week treatment).52 In a direct compari-
`son liraglutide once-daily showed significant improvements in gly-
`cemic control compared with exenatide twice-daily (HbA1c
`reduction of 1.12% vs 0.79%).53 Using higher doses (3 mg) liraglu-
`tide is also under development (phase 3) for the treatment of obes-
`ity54 as well as of obese patients with T2DM.55 In a phase 3a study
`liraglutide 3 mg treatment resulted in 6% weight reduction in
`overweight to obese people compared to placebo.56 Among the
`class of GLP-1 RAs, liraglutide currently seems to be the most effec-
`tive agent for the treatment of obesity.
`Semaglutide. Semaglutide is a next generation GLP-1 analog
`from Novo Nordisk, currently in phase 3 clinical development for
`
`T2DM as a once-weekly injection. Semaglutide’s structure is based
`on liraglutide, with two further modifications: Gly in position 8 is
`replaced by Aib, a beneficial modification already applied in taspo-
`glutide. Furthermore, the fatty acid side chain has been modified
`towards a N6-[N-(17-carboxy-1-oxoheptadecyl-L-c-glutamyl[2-
`(2-aminoethoxy)ethoxy]acetyl[2-(2-aminoethoxy)ethoxy]acetyl]
`residue (see Fig. 2a). The human half-life is 160 h. In a 12 week
`phase 2 trial in T2DM patients, semaglutide was tested at 5 doses
`(0.1–1.6 mg) once-weekly. Semaglutide P0.2 mg dose depen-
`dently reduced HbA1c from baseline up to 1.7% (vs 0.5% reduction
`for placebo) and for doses P0.8 mg also body weight by up to
`4.8 kg (vs 1.2 kg reduction for placebo).57
`Conjugation to albumin. Another half-life prolonging principle is
`the (genetic) fusion to recombinant albumin. Human serum albu-
`min (HSA) has a molecular weight of 67 kDa.58 The half-life of
`albumin in humans is 19 days. PH-dependent recycling mediated
`by the neonatal FC receptor (FcRN) has been shown to contribute
`to that long half-life.59 When albumin is fused to therapeutic pep-
`tides such as GLP-1 RAs, both FcRN mediated recycling as well as
`reduced clearance due to the increased molecular weight are
`responsible for half-life prolongation.
`Albiglutide. In albiglutide, developed by GlaxoSmithKline (GSK),
`two copies of GLP-1 are fused as tandem repeat to the N-terminus
`of albumin. DPPIV-resistance is achieved by a single substitution,
`Ala for Gly, at the DPPIV cleavage site (see Fig. 2b). The tandem re-
`peat unit was developed to overcome reduced potency as was seen
`when only one GLP-1 moiety was directly fused to albumin, mean-
`ing that one GLP-1 moiety of the tandem repeat serves as a spacer
`to allow potent binding to the GLP-1 receptor.60 Albiglutide has a
`half-life of 6–8 days in humans and is currently in phase 3 for
`the treatment of T2DM as once-weekly injection. In a 32-week
`head-to-head study comparing albiglutide (50 mg) to once-daily
`liraglutide (1.8 mg) (HARMONY 7), albiglutide demonstrated a sta-
`tistically significant reduction in HbA1c (0.78%) from baseline.61
`The rather moderate body weight loss of 0.6 kg seen for
`
`Figure 2. Approaches for half-life extension of GLP-1 agonists. (a) Liraglutide and semaglutide carry fatty acids, which facilitate binding to serum albumin thereby reducing
`their renal clearance. (b) In albiglutide, two copies of modified GLP-1 are fused as tandem repeat to the N-terminus of albumin. (c) Similar to albumin fusion, peptides can be
`linked to the Fc region of immunoglobulin G (IgG) as applied in dulaglutide.
`
`MPI EXHIBIT 1139 PAGE 4
`
`

`

`M. Lorenz et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4011–4018
`
`4015
`
`albiglutide is thought to be linked to limited actions in the CNS due
`to its large molecular size.13 A regulatory application for albiglu-
`tide was submitted in the US in January 2013.62
`CJC-1134-PC. Likewise, in ConjuChem’s CJC-1134-PC, exendin-4
`is coupled via its C-terminus to albumin by a chemical linker bear-
`ing a terminal maleimide which is used for the chemical conjuga-
`tion to a single cystein residue on albumin (see Fig. 2b).63 CJC-
`1134-PC has a half-life of 8 days in humans and is currently in
`phase 2 trials for the treatment of T2DM as once-weekly injection.
`CJC-1134-PC seems to be more potent compared to albiglutide. In
`phase 2 trials the once-weekly application of 2 mg led to a HbA1c
`reduction of 1.4%. As observed for albiglutide, the effects on weight
`were only moderate.64
`FC fusion. Similar to albumin fusion, peptides can be linked to
`the constant region of immunoglobulin G (IgG), the Fc region
`(see Fig. 2c).65 The Fc region of IgG has a half-life of 22 days.66
`Likewise, when fusing peptides to the FC region of IgG the protrac-
`tion principle is based on reduced renal clearance and FcRN med-
`iated receptor recycling.
`Dulaglutide. Dulaglutide (Eli Lilly) is a recombinant fusion pro-
`tein, which consists of two GLP-1 peptides covalently linked by a
`small peptide [tetraglycyl-L-seryltetraglycyl-L-seryltetraglycyl-L-
`seryl-L-alanyl] to a human IgG4-Fc heavy chain variant. Compared
`to natural GLP-1, the GLP-1 moieties contain amino acid substitu-
`tions (Ala8?Gly, Gly26?Glu, Arg36?Gly) to ensure protection
`from DPPIV cleavage as well as maintenance of the potency of
`the construct. Dulaglutide has a half-life of 4 days and is in ad-
`vanced clinical trials as once-weekly injection for the treatment
`of T2DM.67 In phase 2 clinical trials dulaglutide showed significant
`dose-dependent reduction in HbA1c (1.5% for the 1.5 mg dose after
`12 weeks) and dose-dependent reductions in body weight. How-
`ever, the decrease in body weight was not statistically significant
`when compared to the placebo group.68 Currently, a large phase
`3 program (AWARD studies Nos. 1–7) is ongoing, in which dulaglu-
`tide has shown a superior glycemic control compared to exenatide
`BID after 6 month of treatment.69 Detailed reports regarding the
`efficacy in weight loss in longer studies are not yet available.
`Langlenatide (HM11260C). In langlenatide, developed by Hanmi
`Pharmaceuticals, an exendin-4 analog is fused via a short polyethyl-
`eneglycol (PEG) linker to a nonglycosylated Fc region (‘LAPS-carrier’).
`In contrast to dulaglutide only one peptide copy is fused to the Fc-car-
`rier. Such monomeric peptide-Fc-fusion constructs hold promise to
`show even higher in vivo efficacy and longer duration of action. In
`fact, the half-life of langlenatide has been reported to be 150 h.70
`Langlenatide is currently being investigated in phase 2 trials for the
`treatment of T2DM as once-weekly injection (1–4 mg doses), but also
`for its suitability as once-monthly application (8–16 mg).71
`Peptides coupled to biological polymers–Xtenylation. Xtenylation,
`which is the genetic fusion of an unstructured recombinant poly-
`peptide to a peptide or protein, is a further generic approach to ex-
`tend plasma half-life of peptides, again by reducing renal
`clearance. The Xten protein is a nonrepetitive amino acid polymer,
`which comprises 864 amino acids selected from Ser, Ala, Pro, Thr,
`Glu and Gly, with a huge hydrodynamic volume and good serum
`stability.72 In contrast to polyethyleneglycol (PEG), Xten is highly
`biodegradable which might be beneficial with respect to safety.
`VRS-859: In VRS-859, Xten is genetically fused to exenatide. The
`product shows a half-life in cynomolgous monkeys of 3 days.
`VRS-859 is currently being developed by Diartis Pharmaceuticals
`as once-monthly treatment for patients with T2DM. In October
`2012, phase 1 results in T2DM patients were disclosed showing
`statistically significant reductions in HbA1c levels at 30 days after
`only one 200 mg dose of VRS-859, supporting the once-monthly
`dosing regimen.73
`Novel approaches and combination of multiple hormone actions.
`Approximately 70–80% of severely obese patients with T2DM
`
`who undergo bariatric surgery show significant and fast improve-
`ment of glycemic control even up to a complete remission of dia-
`betes, often resulting in discontinuation of insulin or oral anti-
`diabetic therapy.74 The mechanism seems partly independent of
`weight reduction and related to alterations in the release of gas-
`tro-intestinal hormones, for example increased plasma levels of
`GLP-1, oxyntomodulin and peptide YY (PYY). In addition to the ef-
`fects of GLP-1, these peptides are known to be involved in the reg-
`ulation of glucose homeostasis,
`food intake
`and energy
`expenditure.75 Based on the impressive improvements of T2DM
`after bariatric surgery, trying to mimic some alterations in plasma
`levels of these peptide hormones is a highly attractive research ef-
`fort to identify novel therapeutic options for obese patients with
`T2DM. Combining dual or triple agonistic effects in a single peptide
`(feasible if peptides are sufficiently similar) or combining 2 distinct
`peptides in a formulation approach are possible options. Slow re-
`lease formulations/delivery approaches as mentioned above could
`prolong the duration of action from once-daily up to a once-weekly
`administration regimens.
`Below we discuss promising combinations of peptides, which
`have the potential for improved glycemic control associated with
`more effective weight reduction compared to pure GLP-1 RAs.
`GLP-1/glucagon co-agonists. Oxyntomodulin is a 37 amino acid
`peptide (comprising the entire 29 amino acid sequence from gluca-
`gon plus a C-terminal extension of 8 residues) secreted by intesti-
`nal L-cells (together with GLP-1 and PYY) following meal ingestion.
`The dual agonistic activity of oxyntomodulin is weaker for the GLP-
`1 and glucagon receptors when compared to the cognate native li-
`gands GLP-1 and glucagon.76 Originally glucagon was discovered as
`a hormone with effects counter to those of insulin, raising blood
`glucose levels by stimulating gluconeogenesis and glycogenolysis
`to circumvent a hypoglycemic state. However, more recent data
`in rodents and humans reveal that glucagon could have beneficial
`effects on energy balance, body fat mass and nutrient intake. Those
`effects seem to be mediated at least in part by FGF21-dependent
`pathways.77,78
`In overweight and obese people native oxyntomodulin was
`shown to significantly reduce body weight by 1.7 kg vs. placebo
`following three-times subcutaneous administration (to compen-
`sate for the short half-life of the native peptide) for 4 weeks.79
`Furthermore, oxyntomodulin was mechanistically proven to re-
`duce food intake after an ad libitum test meal and increase energy
`expenditure in humans.80 Thus, combining the actions of GLP-1
`and glucagon in one molecule like in oxyntomodulin might lead
`to a therapeutic principle with anti-diabetic action related to the
`GLP-1 component and a pronounced weight lowering effect related
`to glucagon receptor stimulation. Pioneering work in the field was
`carried out by Bloom and co-workers,81 DiMarchi and co-workers82
`and Merck Research Laboratories.83 Chemical strategies in order to
`identify such GLP-1/glucagon co-agonists started either from the
`sequence of oxyntomodulin or glucagon by carefully incorporating
`amino acids known to be relevant for GLP-1 agonism but also ele-
`ments to prolong half-life. A Tschöp, model peptide is shown in
`Figure 3, which is a hybrid peptide derived from native GLP-1
`and glucagon. Further chemical modifications to increase the sta-
`bility and in vivo half-life include a lactam bond between the
`side-chains of Glu in position 16 and Lys in position 20 (‘stapling’)
`and the addition of a linear 40 kDa PEG chain to Cys in position 24.
`For the design of novel stabilized peptides the selection of the
`activity-ratio for the GLP-1 and the glucagon receptor is important
`for adjusting the right balance between anti-diabetic and anti-
`obesity effects. High potency at the glucagon receptor can be
`anticipated to strongly increase weight loss but potentially at the
`expense of elevated glucose levels. Data in mice suggest that a
`balanced ratio of GLP-1/glucagon-receptor activation (1:1) is
`optimal to ensure glucose control associated with improved
`
`MPI EXHIBIT 1139 PAGE 5
`
`

`

`4016
`
`M. Lorenz et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4011–4018
`
`Figure 3. Amino acid sequence of native GLP-1, native glucagon and