`
`The incretin system: glucagon-like peptide-1 receptor
`agonists and dipeptidyl peptidase-4 inhibitors in type 2
`diabetes
`
`Daniel J Drucker, Michael A Nauck
`
`Glucagon-like peptide 1 (GLP-1) is a gut-derived incretin hormone that stimulates insulin and suppresses glucagon
`secretion, inhibits gastric emptying, and reduces appetite and food intake. Therapeutic approaches for enhancing
`incretin action include degradation-resistant GLP-1 receptor agonists (incretin mimetics), and inhibitors of dipeptidyl
`peptidase-4 (DPP-4) activity (incretin enhancers). Clinical trials with the incretin mimetic exenatide (two injections
`per day or long-acting release form once weekly) and liraglutide (one injection per day) show reductions in fasting
`and postprandial glucose concentrations, and haemoglobin A1c (HbA1c) (1–2%), associated with weight loss (2–5 kg).
`The most common adverse event associated with GLP-1 receptor agonists is mild nausea, which lessens over time.
`Orally administered DPP-4 inhibitors, such as sitagliptin and vildagliptin, reduce HbA1c by 0·5–1·0%, with few
`adverse events and no weight gain. These new classes of antidiabetic agents, and incretin mimetics and enhancers,
`also expand β-cell mass in preclinical studies. However, long-term clinical studies are needed to determine the benefi ts
`of targeting the incretin axis for the treatment of type 2 diabetes.
`
`Lancet 2006; 368: 1696–705
`Samuel Lunenfeld Research
`Institute, Mount Sinai
`Hospital, University of Toronto,
`Toronto, Ontario, Canada
`(Prof D J Drucker MD); and
`Diabeteszentrum Bad
`Lauterberg, Bad Lauterberg,
`Germany (Prof M A Nauck MD)
`Correspondence to:
`Prof Daniel J Drucker,
`Samuel Lunenfeld Research
`Institute, Mount Sinai Hospital,
`600 University Avenue, Toronto,
`Ontario, Canada M5G 1X5
`d.drucker@utoronto.ca
`
`Adipose tissue
`
`Glucose uptake ↑
`Glycogen synthesis ↑
`(? indirect actions)
`
`Liver
`
`Muscle
`
`Introduction
`Eating provokes the secretion of multiple gastrointestinal
`hormones involved in the regulation of gut motility,
`secretion of gastric acid and pancreatic enzymes, gall
`bladder contraction, and nutrient absorption. Gut
`hormones also facilitate the disposal of absorbed glucose
`through the stimulation of insulin secretion from the
`
`Brain/nervous system:
`Hypothalamus
` Appetite ↓, satiety ↑, food intake ↓, water intake ↓
`Nucleus tractus solitarii
` GLP–1 production
`Access
`
`CNS: circumventricular organs (circulating GLP-1)
`Autonomic nervous system
` Afferent vagus (GLP-1 from GI tract)
`
`“Hepatoportal” region
`
`Stomach
` Gastric emptying decelerated
` Acid secretion ↓
`
`that enteral
`endocrine pancreas. The observation
`nutrition provided a more potent insulinotropic stimulus
`compared with isoglycaemic intravenous challenge led to
`the development of the incretin concept.1 The fi rst
`incretin to be identifi ed, glucose-dependent insulinotropic
`polypeptide (GIP), was purifi ed from porcine intestinal
`extracts and had weak eff ects on gastric acid secretion
`but more potent insulinotropic actions in human beings.2
`GIP is a 42-aminoacid hormone synthesised in duodenal
`and jejunal enteroendocrine K cells in the proximal small
`bowel.
`A second incretin hormone, glucagon-like peptide-1
`(GLP-1) was identifi ed after the cloning of the cDNAs
`and genes encoding proglucagon (fi gure 1). GLP-1 exists
`in
`two circulating equipotent molecular
`forms,
`GLP-1(7-37) and GLP-1(7-36)amide, although GLP-1(7-
`36)amide is more abundant in the circulation after eating.
`Most GLP-1 is made in enteroendocrine L cells in the
`distal ileum and colon, but plasma levels of GLP-1, like
`GIP, also increase within minutes of eating. Hence a
`combination of endocrine and neural signals probably
`promote the rapid stimulation of GLP-1 secretion well
`before digested food transits through the gut to directly
`engage the L cell in the small bowel and colon. More
`proximally located L cells in the duodenum and jejunum
`have also been described; however,
`the precise
`
`Search strategy and selection criteria
`
`We searched the MEDLINE and PubMed databases
`(1987–2006) with the search terms “glp-1”, “glucagon”,
`“glucagon-like”, “gip”, “incretin”, “dipeptidyl peptidase-4”,
`and “diabetes”. We preferentially selected publications from
`the past 5 years, but did not exclude older publications that
`are commonly referenced or highly regarded. We also
`searched the reference lists of articles identifi ed by this search
`strategy and selected those we judged relevant.
`
`Endocrine pancreas:
`Secretion
`
`β cells: insulin secretion ↑
`
`α cells: glucagon secretion ↓
`
`δ cells: somatostatin secretion ↑
`Biosynthesis
`
`(Pro-) insulin↑
`β-cell mass
` Growth, regeneration, neogenesis ↑
` Apoptosis ↓
`
`Ileum
`Site of GLP-1 synthesis from proglucagon,
`GLP-1 secretion ↑ after meals (carbohydrate, fat)
`
`Figure 1: Physiology of GLP-1 secretion and action on GLP-1 receptors in diff erent organs and tissues
`
`1696
`
`www.thelancet.com Vol 368 November 11, 2006
`
`MPI EXHIBIT 1024 PAGE 1
`
`
`
`New Drug Class
`
`GLP-1 (amidated form)
`
`HisAlaGluGlyThrPheThrSerAspValSerSerTyrLeuGluGlyGlnAlaAlaLysGluPheIleAlaTrpLeuValLysGlyArgamide
`7
`10
`15
`20
`25
`30
`35 36
`Proteolytic inactivation (DPP-4)
`
`Exenatide
`
`HisGlyGluGlyThrPheThrSerAspLeuSerLysGlnMetGluGluGluAlaValArgLeuPheIle GluTrpLeuLysAsnGlyGlyProSerSerGlyAlaProProProSeramide
`
`Liraglutide
`
`HisAlaGluGlyThrPheThrSerAspValSerSerTyrLeuGluGlyGlnAlaAlaLysGluPheIle AlaTrpLeuValArgGlyArgGly
`Glu
`
`Albumin
`
`C-16 free fatty acid
`
`N C
`
`F₃
`
`N
`
`N
`
`N
`
`Sitagliptin
`
`F
`
`NH₂
`
`O
`
`F
`
`F
`
`O
`
`N
`
`OH
`
`NC
`
`HN
`
`Vildagliptin
`
`Figure 2: Structure of GLP-1, GLP-1R agonists exenatide and liraglutide, and DPP-4 inhibitors vildagliptin and sitagliptin
`
`contributions of the proximal and distal L cells to the
`early rapid increase in plasma GLP-1 remains unclear.
`Plasma levels of GLP-1 are low in the fasted state, in the
`range of 5–10 pmol/L, and increase rapidly after eating,
`reaching 15–50 pmol/L. The circulating levels of intact
`GLP-1 and GIP decrease rapidly because of enzymatic
`inactivation, mainly dipeptidyl peptidase-4 (DPP-4), and
`renal clearance.3 Whether additional proteases, such as
`human neutral endopeptidase 24·11, are also essential
`determinants of GLP-1 inactivation is being investigated.
`Both GIP and GLP-1 contain alanine at position 2, and
`hence are excellent substrates for DPP-4. Indeed, DPP-4
`is essential for incretin inactivation, and mice with
`targeted inactivation of the DPP-4 gene have raised levels
`of plasma GIP and GLP-1, increased insulin secretion,
`and reduced glucose excursion after glycaemic challenge.4
`As a result of DPP-4 activity, intact, biologically active
`GLP-1 represents only 10–20% of total plasma GLP-1.5
`Both GIP and GLP-1 exert their actions by the
`engagement of structurally distinct G-protein-coupled
`receptors (GPCRs). The GIP receptor is predominantly
`expressed on islet β cells, and to a lesser extent, in adipose
`tissue and in the central nervous system. By contrast, the
`GLP-1 receptor (GLP-1R) is expressed in islet α and β
`cells and in peripheral tissues, including the central and
`peripheral nervous systems, heart, kidney, lung, and
`gastrointestinal tract (fi gure 1). Activation of both incretin
`
`receptors on β cells leads to rapid increases in levels of
`cAMP and intracellular calcium, followed by insulin
`exocytosis,
`in a glucose-dependent manner.6 More
`sustained incretin receptor signalling is associated with
`activation of protein kinase A, induction of gene
`transcription, enhanced levels of insulin biosynthesis,
`and stimulation of β-cell proliferation.7 Both GLP-1R and
`GIP receptor activation also promote resistance to
`apoptosis and enhanced β-cell survival, in both rodent8
`and human islets.9 Consistent with the distribution of
`GLP-1R expression, GLP-1 also
`inhibits glucagon
`secretion, gastric emptying, and food ingestion, and
`promotes enhanced glucose disposal through neural
`mechanisms,10 actions that also contribute to the control
`of glucoregulation. Notably, eff ects on glucagon secretion,
`like those on insulin secretory responses, are glucose-
`dependent, whereas counter-regulatory
`release of
`glucagon in response to hypoglycaemia is fully preserved
`even in the presence of pharmacological concentrations
`of GLP-1.11
`The physiological importance of endogenous GIP and
`GLP-1 for glucose homoeostasis has been investigated in
`studies with receptor antagonists, or gene-knockout
`mice. Acute antagonism of GIP or GLP-1 lowers insulin
`secretion and increases plasma glucose after glycaemic
`challenge in rodents. Similarly, mice with inactivating
`mutations in the GIP or GLP-1 receptors also have
`
`www.thelancet.com Vol 368 November 11, 2006
`
`1697
`
`MPI EXHIBIT 1024 PAGE 2
`
`
`
`New Drug Class
`
`insulin secretion and
`defective glucose-stimulated
`impaired glucose tolerance.12,13 GLP-1, but not GIP, is also
`essential for control of fasting glycaemia, since acute
`antagonism or genetic disruption of GLP-1 action leads
`to increased levels of fasting glucose in rodents.13
`Furthermore, GLP-1 is essential for glucose control in
`human beings: studies with the antagonist exendin(9-39)
`show defective glucose-stimulated insulin secretion,
`reduced glucose clearance, increased levels of glucagon,
`and quicker gastric emptying after disruption of GLP-1
`action.14
`The pleiotropic actions of GLP-1 and GIP on the control
`of blood glucose have fostered considerable interest in
`the use of these agents for the treatment of type 2
`
`diabetes. Whereas in healthy human beings oral glucose
`elicits a considerably higher insulin secretory response
`than does intravenous glucose (even if leading to the
`same glycaemic increments), this incretin eff ect is
`substantially reduced or even lost in patients with type 2
`diabetes.15 As an explanation for the acquired incretin
`defect, GIP but not GLP-1 shows noticeably attenuated
`insulinotropic action in patients with type 2 diabetes.16
`Furthermore, those with type 2 diabetes show a small but
`signifi cant reduction in meal-stimulated levels of GLP-1.17
`Since GLP-1 action remains relatively preserved in
`diabetic patients, most pharmaceutical eff orts directed at
`potentiation of incretin action for the treatment of type 2
`diabetes have focused on GLP-1R agonists.
`
`Biological actions of
`incretin hormone in
`type 2 diabetes
`
`Native GLP-1*
`
`Incretin mimetics
`
`DPP-4 inhibitors
`(eg, vildagliptin,
`sitagliptin)
`
`Lack of biphasic response†
`
`Slow insulin secretory response to meals50
`
`Reduction in or absence of incretin eff ect15
`
`Hyperglucagonaemia52
`
`Hypoglycaemia counter-regulation
`
`Reduced pancreatic β-cell insulin content
`
`Reduced endocrine pancreatic β-cell mass56,57
`
`Characteristic features of type 2 diabetes
`Defective glucose-stimulated insulin secretion Glucose-dependent
`stimulation of insulin
`secretion
`Restoration of biphasic
`responses
`More adequate insulin
`secretory response after
`meals‡
`Replacement of incretin
`activity, greater incretin
`eff ect§
`Suppression of glucagon
`secretion
`Glucagon secretion, when
`plasma glucose is low
`Increased synthesis of
`proinsulin
`Increase in pancreatic islet
`β-cell mass
`Diff erentiation of islet
`precursor cells into β cells
`Inhibition of
`toxin-induced|| β-cell
`apoptosis
`Deceleration in gastric
`emptying
`Suppression of
`appetite/induction of
`satiety
`Weight loss (fi gure 3)
`
`Abnormally high rate of β-cell apoptosis57
`
`Normal, decelerated or accelerated67 gastric
`emptying
`Hypercaloric energy intake/obesity72
`
`Exenatide
`
`Yes45
`
`Yes49
`
`Yes45
`
`Yes‡
`
`Yes45
`
`Yes53
`
`Yes
`
`Yes59
`
`Yes
`
`Yes64
`
`Yes69
`
`Yes74
`
`Yes25-27
`
`Liraglutide
`
`Yes46
`
`Yes47
`
`Not tested
`
`Not tested
`
`Yes46
`
`Yes‡
`
`Yes46
`
`Yes54
`
`Yes
`
`Yes60
`
`Yes
`
`Yes65
`
`Yes46
`
`Yes75
`
`Yes32
`
`Yes47
`
`Not tested, but probable
`
`Yes47
`
`Not tested
`
`Yes
`
`Yes61
`
`Unknown
`
`Probable66
`
`Marginal70,71
`
`No obvious eff ect
`
`No weight change34
`
`Yes16
`
`Yes48
`
`Yes21,51
`
`Yes‡
`
`Yes16
`
`Yes11
`
`Yes55
`
`Yes58
`
`Yes62
`
`Yes8,63
`
`Yes68
`
`Yes73
`
`Yes22
`
`Pharmacological characteristics
`Mode of administration
`
`Frequency of administration
`Predominant adverse event
`
`Intravenous,
`subcutaneous
`Continuous
`Nausea
`
`Subcutaneous
`
`Subcutaneous
`
`Orally
`
`Twice daily
`Nausea
`
`Once daily
`Nausea
`
`Once (or twice) daily
`None noted
`
`*GLP-1 exists as glycine-extended (7–37) and amidated form (7–36)amide, with both forms having similar properties; †Biphasic response is only seen under artifi cial
`conditions leading to rapid rise in glucose concentrations (glucose bolus injection or “squarewave stimulus” when starting hyperglycaemic clamp); ‡As judged by
`improvement (normalisation) of postprandial glucose excursions; §By defi nition, GLP-1 and incretin mimetics replace incretin activity; ¶These actions have only been
`reported from animal or in vitro (eg, islet) studies; methods to assess human β-cell mass in vivo are not available; ||Hydrogen peroxide, free fatty acids, or streptozotocin.
`
`Table: Type 2 diabetes and biological actions of GLP-1, incretin mimetics, and DPP-4 inhibitors
`
`1698
`
`www.thelancet.com Vol 368 November 11, 2006
`
`MPI EXHIBIT 1024 PAGE 3
`
`
`
`New Drug Class
`
`Antidiabetic actions of GLP-1
`of GLP-1
`Short-term
`intravenous
`infusions
`(1–1·2 pmol kg–¹ min–¹, leading to pharmacological
`plasma concentrations of total GLP-1 of 70–150 pmol/L,
`and of intact biologically active GLP-1 of 10–20 pmol/L)
`lowers blood glucose in patients with type 2 diabetes
`through a transient glucose-dependent stimulation of
`insulin and suppression of glucagon secretion and
`gastric emptying.18–21 A 6-week subcutaneous infusion of
`GLP-1 in patients with type 2 diabetes, achieving plasma
`levels of GLP-1 in the 60–70 pmol/L range,22 produced
`substantial improvements in insulin secretory capacity,
`insulin sensitivity, a reduction in HbA1c of 1·2% and
`modest weight loss (1·9 kg).22 Although intravenous or
`subcutaneous GLP-1 infusions could be useful for the
`short-term control of hyperglycaemia,23,24 the long-term
`treatment of type 2 diabetes needs a more feasible
`approach to achieve sustained activation of GLP-1
`receptors. The effi cacy of injectable GLP-1 receptor
`agonists (degradation-resistant peptides or larger pro-
`teins with more suitable pharmacokinetic properties,
`fi gure 2) and DPP-4 inhibitors (small molecules with
`good oral bioavailability, webtable),25–42 has been assessed
`in clinical trials.
`
`GLP-1R agonists
`Exenatide
`Exenatide (synthetic exendin-4) was discovered in the
`search for biologically active peptides in lizard venom.43
`Exendin-4 shares roughly 50% of
`its aminoacid
`sequence with mammalian GLP-1 , is encoded by a
`unique gene in the lizard,44 and is a potent degradation-
`resistant agonist at the mammalian GLP-1R (fi gure 2).
`Exenatide has been developed for the treatment of
`type 2 diabetes (table).47–75 Exenatide has a circulating
`half-life of 60–90 min,69 with increases in plasma
`exenatide concentrations lasting 4–6 h after a single
`subcutaneous injection.76,77
`Phase III trials investigated the effi cacy of adding
`exenatide (5 or 10 μg by subcutaneous injection twice
`daily) to ongoing therapy in patients suboptimally
`controlled on oral antidiabetic agents (metformin,25
`sulphonylureas,26 a combination of both,27 or thiazo-
`lidinediones28). The starting dose of exenatide is 5 μg
`twice daily for 4 weeks, followed by an increase to 10 μg
`twice daily.78,79 Exenatide reduced HbA1c concentrations
`by 0·8–1·0 % (fi gure 3)25–42 over 30 weeks, with prevention
`of weight gain or modest weight loss of 1·5–3 kg.
`Patients continuing in an open-label extension lost more
`weight, with the total weight loss reaching 4–5 kg after
`80 weeks.80 The commonest adverse events with
`exenatide were gastrointestinal (nausea, or more rarely
`vomiting or diarrhoea25–27) (fi gure 3). However, exenatide
`was rarely discontinued because of side-eff ects, and the
`occurrence of nausea lessened the longer the duration
`of therapy.25–27 An increased number of mild to moderate
`hypoglycaemic events was noted in patients given
`
`exenatide and sulphonylureas,26,27 but not in those given
`exenatide and metformin,25 despite a similar reduction
`in glycaemia.
`40–50% of patients receiving exenatide develop
`antibodies with weak binding affi nity and low titres.25–27
`Antibody formation has not been associated with
`impaired antidiabetic eff ectiveness of exenatide in most
`of those treated. However, the drug might not be as
`eff ective in the few patients with high-titre antibodies.
`Exenatide has been compared with insulin glargine in
`an open-label study as additional treatment for diabetic
`patients not achieving eff ective glucose control on
`metformin and a sulphonylurea.29 Fasting glucose
`concentrations were reduced more in patients receiving
`insulin glargine, but postprandial glucose reduction was
`greater with exenatide, especially after breakfast and
`dinner. Both exenatide and insulin glargine reduced
`levels of HbA1c by 1·1% over 26 weeks.29 No signifi cant
`diff erences in overall rates of hypoglycaemia were seen
`in the diff erent treatment groups, although nocturnal
`hypoglycaemia was less frequent with exenatide and
`daytime hypoglycemia was less common in patients
`given insulin glargine. Gastrointestinal side-eff ects, such
`as nausea and vomiting, were more often reported with
`exenatide than with insulin glargine, and the dropout
`rate was also higher in the exenatide-treated cohort.
`However, patients receiving insulin glargine gained an
`average of 1·8 kg compared with a 2·3 kg weight loss in
`exenatide-treated patients.29 Exenatide was approved by
`the US Food and Drug Administration for the treatment
`of type 2 diabetes in April, 2005. In Europe, exenatide is
`expected to be approved by the end of 2006 or early 2007
`for use in patients with type 2 diabetes that is not well
`controlled on oral agents.
`
`Liraglutide
`Liraglutide, a partly DPP-4-resistant GLP-1 analogue,
`contains a Arg34Lys substitution, and a glutamic acid
`and 16-C free-fatty-acid addition to Lys26 (fi gure 2).81 The
`acyl moiety promotes non-covalent binding to albumin
`with 1–2% of liraglutide circulating as the non-albumin-
`bound free peptide.82 Liraglutide has a half-life of about
`10–14 h after subcutaneous administration in human
`beings,83,84 and can be given as a once daily injection.
`Early phase II studies were done with up to 0·75 mg per
`day of liraglutide,31,85 but more recent studies with weekly
`escalating dose-titration have investigated the effi cacy of
`doses up to 2·0 mg.32 Liraglutide reduces fasting and
`postprandial glucose, and levels of HbA1c by up to 1·75 %
`(fi gure 3),33 while preventing weight gain or inducing
`modest but signifi cant weight loss.32,33 Nausea, vomiting,
`and diarrhoea were the most prominent adverse events
`but were generally mild, transient, and rarely caused
`discontinuation of liraglutide treatment.32,33 So far, no
`studies of exposure to liraglutide have reported antibody
`formation, and phase III testing was started earlier this
`year.
`
`See Online for webtable
`
`www.thelancet.com Vol 368 November 11, 2006
`
`1699
`
`MPI EXHIBIT 1024 PAGE 4
`
`
`
`New Drug Class
`
`Exenatide (twice daily)
`Exenatide LAR (once weekly)
`Liraglutide (once daily)
`Placebo
`
`Insulin glargine
`Vildagliptin
`Sitagliptin
`
`Metformin
`Glimepiride
`Rosiglitazone
`
`Incretin mimetics
`
`DPP-4 inhibitors
`
`50 mg
`
`Up to 200 mg
`100 mg
`25 mg Up to 100 mg
`100 mg
`
`Duration (weeks):
`12 52 12 52
`
`24
`
`18
`
`24
`
`Drug-naive patientsMetformin
`
`
`
`(washed out)
`
`MetforminPioglitazone
`Diet or OAD
`
`Metformin
`
`1·0
`
`0·5
`
`0·0
`
`–0·5
`
`–1·0
`
`–1·5
`–2·0
`
`20
`
`0
`
`–20
`
`–40
`
`–60
`
`–80
`
`3 2 1
`
`0
`–1
`–2
`–3
`–4
`–5
`
`100
`90
`80
`70
`60
`50
`40
`30
`20
`10
`0
`
`100
`90
`80
`70
`60
`50
`40
`30
`20
`10
`0
`
`Karasik et al, 2006
`Rosenstock et al, 2006
`
`Raz et al, 2006Aschner et al, 2006
`(12 weeks)
`Garber et al, 2006
`Åhren et al, 2005
`Rosenstock et al, 2006
`Dejager et al, 2006
`(52 weeks)
`Pratley et al, 2004
`
`www.thelancet.com Vol 368 November 11, 2006
`
`Up to 2 mg
`weekly
`Up to
` 0·75, 2·0, 1·9 mg
`once daily
`
`Up to 10 μg
`twice daily
`
`Duration (weeks):
`30
`15 26 15 12
`
`5
`
`14
`
`Metformin and/or sulfonylurea
`Metformin+sulfonylureaSulfonylurea
`TZD (+metformin)
`
`
`
`Diet ± OADMetformin
`
`Diet or OAD(washed out)
`
`
`Metformin
`
`1·0
`
`0·5
`
`0·0
`
`–0·5
`
`–1·0
`
`–1·5
`–2·0
`
`20
`
`0
`
`–20
`
`–40
`
`–60
`
`–80
`
`3 2 1
`
`0
`–1
`–2
`–3
`–4
`–5
`
`100
`90
`80
`70
`60
`50
`40
`30
`20
`10
`0
`
`HbA 1c (%)
`
`Change in fasting glucose (mg/dL)
`
`Change in bodyweight (kg)
`
`Hypoglycaemia (%)
`
`Nausea (%)
`
`A
`
`B
`
`C
`
`D
`
`E
`
`100
`90
`80
`70
`60
`50
`40
`30
`20
`10
`0
`
`DeFronzo et al, 2005
`Kendall et al, 2005
`Heine et al, 2005
`Buse et al, 2004
`Nauck et al, 2006
`Kim et al, 2006
`VilsbØll et al, 2006
`Zinman et al, 2006
`Madsbad et al, 2004
`
`Figure 3: Clinical eff ects of
`GLP-1R agonists (incretin
`mimetics) and DPP-4
`inhibitors (incretin
`enhancers) on HbA1c, fasting
`glucose concentrations,
`bodyweight, hypoglycaemic
`episodes, and nausea
`Doses are indicated in top
`panels (A); concomitant
`medication is shown in lower
`panel (D). Results are from
`phase II or III studies on:
`exenatide;25–29 exenatide LAR;30
`liraglutide;31-33 vildagliptin;34-38
`and sitagliptin.39-42 *Signifi cant
`diff erences to placebo or
`respective comparator; if no
`comparator is shown, results
`are depicted as placebo-
`subtracted diff erences. Bars are
`mean and SE. OAD=oral
`antidiabetic agents.
`
`1700
`
`MPI EXHIBIT 1024 PAGE 5
`
`
`
`New Drug Class
`
`Long-acting GLP-1R agonists
`Because one subcutaneous injection of exenatide does
`not produce eff ective glucose control for more than
`6–8 h, there is considerable interest in the development
`of long-acting GLP-1R agonists that need less frequent
`parenteral administration. Exenatide long-acting release
`(LAR) is a polylactide-glycolide microsphere suspension
`containing 3% exendin-4 peptide that shows sustained
`dose-dependent glycaemic control in diabetic fatty
`Zucker rats for up to 28 days after one subcutaneous
`injection.86 Preliminary experience with exenatide LAR
`in 45 patients with type 2 diabetes indicates a much
`greater reduction in fasting glucose concentrations and
`HbA1c after once weekly administrations of exenatide
`LAR for 15 weeks compared with exenatide twice daily.30
`However, long-term experience with the drug in larger
`numbers of patients has not yet been reported. Exenatide
`LAR is currently being assessed in a phase III head-to-
`head trial against twice-daily exenatide.
`Additional strategies for development of long-acting
`GLP-1R agonists include the use of chemical linkers to
`form covalent bonds between GLP-1 (CJC-1131) or
`exendin-4 (CJC-1134).87 Similarly, recombinant albumin-
`GLP-1 proteins have been developed that mimic the full
`range of GLP-1 actions in preclinical studies.88 Although
`these drugs are expected
`to have an extended
`pharmacokinetic profi le suitable for once weekly dosing
`in diabetic patients, little clinical information is available
`about the effi cacy and safety of these albumin-based drugs
`in human beings.
`
`DPP-4 inhibitors
`The observation that GLP-1 is rapidly degraded by
`DPP-45,89,90 has fostered the development of specifi c
`protease inhibitors that prevent the rapid fall of GLP-1 in
`circulating plasma after eating. DPP-4 is a ubiquitous
`membrane-spanning cell-surface aminopeptidase widely
`expressed in many tissues, such as liver, lung, kidney,
`intestinal brush-border membranes, lymphocytes, and
`endothelial cells.91–93 The extracellular domain of DPP-4
`can also be cleaved from its membrane-anchored form
`and circulate in plasma, where it retains its full enzymatic
`activity. DPP-4 preferentially cleaves peptides with a
`proline or alanine residue in the second aminoterminal
`position. Many gastrointestinal hormones, neuropeptides,
`cytokines, and chemokines are substrates for DPP-4,91,93
`among them both GIP89,90,94 and GLP-1.5,89,90,95 In preclinical
`studies, DPP-4 inhibitors mimic many of the actions
`ascribed to GLP-1R agonists, including stimulation of
`insulin and
`inhibition of glucagon secretion, and
`preservation of β-cell mass through stimulation of cell
`proliferation and inhibition of apoptosis.7,96 By contrast,
`DPP-4 inhibitors are generally not associated with a
`deceleration of gastric emptying or weight loss, perhaps
`due to the modest stabilisation of postprandial levels of
`intact biologically active plasma GLP-1 (doubled to
`15–25 pmol/L) seen after DPP-4 inhibition (table).
`
`Many small-molecule DPP-4 inhibitors have been
`developed that specifi cally and potently inhibit DPP-4
`activity after oral administration. Typically, these agents
`reduce serum DPP-4 activity by more than 80%, with
`some inhibition maintained for 24 h after one dose or with
`once daily treatment.47,97 DPP-4 inhibition is accompanied
`by a rise in postprandial levels of intact GLP-1.47,97,98 Most
`published studies used vildagliptin.99
`
`Vildagliptin and sitagliptin
`At a dose of 100 mg once daily, fasting and postprandial
`glucose concentrations were reduced after 4 weeks of
`vildagliptin treatment.47 Plasma glucagon concentrations
`were suppressed after vildagliptin treatment, together
`with an increase in the ratio of insulin to glucose.47 In
`clinical studies of longer duration, the addition of
`vildagliptin to patients already given metformin reduced
`HbA1c by 0·8% after 12 weeks, compared with placebo,34
`and this diff erence was maintained during an open-label
`extension for 52 weeks34 (fi gure 3). Indirect evidence
`from modelling experiments suggests that β-cell function
`is improved with vildagliptin treatment over 1 year in
`patients with type 2 diabetes.100 Preliminary reports of
`longer phase III clinical studies with vildagliptin
`monotherapy, either 50 mg twice daily or 100 mg once
`daily, showed sustained effi cacy but slight non-inferiority
`compared with metformin36 after 1 year of therapy,
`although vildagliptin was better tolerated than metformin.
`Similarly, vildagliptin was as eff ective as rosiglitazone in
`direct comparison monotherapy study37 and also produced
`signifi cant reductions in HbA1c when used in combination
`with metformin38 (fi gure 3).
`Clinical studies have also been reported for sitagliptin97
`(fi gure 3). Phase III clinical trial data presented at the
`American Diabetes Association meeting in June, 2006,
`indicated that sitagliptin is well-tolerated at doses of
`100 mg once daily, either as monotherapy, or in
`combination with metformin or pioglitazone, without
`signifi cant hypoglycaemia or weight gain.40–42 Fewer data
`are available for other DPP-4 inhibitors in development
`such as saxagliptin101 or denagliptin.102 Thus whether
`various chemically distinct DPP-4 inhibitors will show
`signifi cant diff erences in pharmacokinetic profi les, side-
`eff ects, or clinical activity cannot be predicted. Sitagliptin
`was approved for the treatment of type 2 diabetes in the
`USA in October, 2006.
`No characteristic pattern of adverse events has been
`associated with the use of vildagliptin34,47 or other DPP-4
`inhibitors,103-105 despite the large number of potential
`substrates for DPP-4.91,93 In view of the widespread
`expression of DPP-4 on many cell types, including
`lymphocytes, there is considerable interest in the long-
`term safety profi le of DPP-4 inhibitors. Although highly
`selective DPP-4 inhibition seems to be well tolerated in
`preclinical studies and DPP-4
`inhibitors do not
`substantially inhibit cell proliferation in experiments
`with human
`lymphocytes
`in vitro,106 considerable
`
`www.thelancet.com Vol 368 November 11, 2006
`
`1701
`
`MPI EXHIBIT 1024 PAGE 6
`
`
`
`New Drug Class
`
`additional clinical experience with these agents will be
`needed before any theoretical safety concerns emerge.
`Furthermore, given the large number of chemically
`distinct DPP-4 inhibitors under clinical development, it
`seems likely that one or more of these agents could be
`associated with adverse events arising as a result of unique
`properties attributable to the individual chemical structure,
`as opposed to a class eff ect arising as a consequence of
`inhibition of DPP-4 activity. Non-selectivity for actions on
`the related enzymes DPP-8, DPP-9, or both, could be of
`particular importance.106
`
`Contrasting properties of GLP-1R agonists and
`DPP-4 inhibitors
`Twice daily exenatide through subcutaneous injection is
`indicated for the treatment of patients with type 2 diabetes
`mellitus in whom one or more oral agents do not work,
`often as an alternative to insulin treatment. By contrast,
`once daily DPP-4 inhibitors could be used as fi rst-line
`therapy, or as add-on therapy to patients failing one or
`more oral agents. While there does not seem to be a great
`diff erence in the HbA1c-lowering capacity of GLP-1R
`agonists compared with DPP-4 inhibitors, the obvious
`diff erence between these classes of drugs is their eff ect on
`bodyweight. Weight loss is a common outcome of therapy
`with native GLP-1,22 exenatide,25–27 and liraglutide,32,75
`whereas treatment with DPP-4 inhibitors is associated
`with prevention of weight gain34,47,103,104 (fi gure 3). By
`contrast, gastrointestinal side-eff ects, predominantly
`nausea, are often reported after treatment with injectable
`GLP-1R agonists but have not been described with DPP-4
`inhibition.34,47,102–105 These diff erences might be explained in
`part by the relatively modest stabilisation of postprandial
`GLP-1 seen after DPP-4 inhibition, compared with the
`pharmacological increases in circulating levels of GLP-1R
`agonists exemplifi ed by exenatide. Hence therapy with
`DPP-4 inhibitors might not be associated with weight loss
`perhaps partly because of the relative levels of GLP-1
`achieved after treatment with these agents. Although
`nausea is a common side-eff ect of exenatide therapy, many
`patients lose weight independently of nausea. Consistent
`with the above diff erences in circulating levels of GLP-1,
`GLP-1R agonists, but not DPP-4 inhibitors, greatly
`decelerate gastric emptying.69,107,108
`
`Future developments
`Liraglutide and exenatide are fi rst-generation GLP-1
`receptor agonists, requiring once or twice daily parenteral
`administration, respectively. Much eff ort continues to be
`directed towards improvement of the pharmacokinetic
`profi le of GLP-1R agonists, to minimise peak levels of the
`drug and thus reduce the extent of nausea. Longer-acting
`GLP-1R agonists should ideally provide more uniform and
`sustained GLP-1R activation over a 24-h period, but require
`less frequent administration.
`Furthermore, there is great interest in determining
`whether chronic therapy with GLP-1R agonists will be
`
`associated with sustained long-term control of HbA1c and
`improvement in β-cell function beyond that achievable
`with existing agents. Similar questions pertain to the
`DPP-4 inhibitors, which also indirectly target β cells;
`however, long-term clinical data assessing the durability
`and effi cacy of these agents in the treatment of type 2
`diabetes are not yet available. Because patients with
`type 2 diabetes have increased risks of cardiovascular
`morbidity and mortality, the observation that GLP-1R
`agonists improve myocardial function in human patients
`after myocardial infarction109 highlights the need for
`studies that assess cardiovascular endpoints in patients
`treated with DPP-4 inhibitors or GLP-1R agonists.
`Overall, agents that enhance incretin action show great
`promise for the treatment of type 2 diabetes by
`recruitment of new, often physiologically based
`mechanisms of action for glucoregulation, in the context
`of a currently favourable safety profi le. Never theless,
`long-term clinical studies are needed to compare these
`agents with existing oral therapies or insulin, or both, to
`permit a greater understanding of the true benefi ts and
`role of these drugs for the treatment of diabetes
`mellitus.
`Confl ict of interest statement
`D J Drucker is an inventor or co-inventor on patents related to the fi eld
`of type 2 diabetes that are licensed to Amylin Pharmaceuticals Inc or
`Arisaph Pharmaceuticals Inc. He has served as a consultant or adviser
`within the past 12 months to Abbott Laboratories, Amgen Inc, Amylin
`Pharmaceuticals, Bayer Inc, Chugai Inc, Conjuchem Inc, Eli Lilly Inc,
`GlaxoSmithKline, Glenmark Inc, Johnson & Johnson, Merck Research
`Laboratories, Novartis Pharmaceuticals, NPS Pharmaceuticals Inc, PPD
`Inc, Takeda Inc, Transition Pharmaceuticals Inc, and Arisaph
`Pharmaceuticals Inc. M A Nauck has received grants for the support of
`product-related studies from Amylin Pharmaceuticals, Eli Lilly & Co,
`Novartis Pharma, and NovoNordisk and consulting honoraria from
`Amgen, Amylin Pharmaceuticals, Bayer, ConjuChem, Eli Lilly & Co,
`GlaxoSmithKline, Merck, Sharp & Dohme, Novartis Pharma,
`NovoNordisk, and Takeda. He has received speaker’s honorar