[Frontiers in Bioscience 13, 3648-3660, May 1, 2008]
`
`Dipeptidyl peptidase IV (DPP IV) and related molecules in type 2 diabetes
`
`Peter R Flatt1, Clifford J Bailey2, Brian D Green3
`
`1School of Biomedical Sciences, University of Ulster, Coleraine, UK, 2School of Life and Health Sciences, Aston University,
`Birmingham, UK, 3School of Biological Sciences, Queens University Belfast, UK
`
`TABLE OF CONTENTS
`
`1. Abstract
`2. Introduction
`3. DPP IV and related enzymes
`3.1. DPP IV (EC 3.4.14.5)
`3.2. Dipeptidyl peptidase II (DPP II)
`3.3. Dipeptidyl peptidase 8 (DPP 8) and dipeptidyl peptidase 9 (DPP 9)
`3.4. Fibroblast Activation protein (FAP)
`4. The incretin hormones, DPP IV and diabetes
`4.1. The incretins
`4.2. Inactivation of the incretins by DPP IV
`4.3. Overcoming DPP IV mediated incretin inactivation
`4.4. Rodent models lacking DPP IV activity
`5. Inhibitors of DPP IV activity
`5.1. Preclinical studies
`5.2. Clinical studies
`5.2.1. Sitagliptin
`5.2.2. Vildagliptin and other gliptins
`5.3. Developmental issues
`5.4. Future outlook for DPP IV inhibitors
`6. Conclusions
`7. Acknowledgements
`7. References
`
`1. ABSTRACT
`
`2. INTRODUCTION
`
`Dipeptidyl peptidase IV (DPP IV) is a widely
`distributed physiological enzyme
`that can be found
`solubilized in blood, or membrane-anchored in tissues.
`DPP IV and related dipeptidase enzymes cleave a wide
`range of physiological peptides and have been associated
`with several disease processes including Crohn’s disease,
`chronic liver disease, osteoporosis, multiple sclerosis,
`eating disorders, rheumatoid arthritis, cancer, and of direct
`relevance to this review, type 2 diabetes. Here, we place
`particular emphasis on two peptide substrates of DPP IV
`with insulin-releasing and antidiabetic actions namely,
`glucagon-like peptide-1 (GLP-1) and glucose-dependent
`insulinotropic polypeptide
`(GIP). The
`rationale
`for
`inhibiting DPP IV activity in type 2 diabetes is that it
`decreases peptide cleavage and
`thereby enhances
`endogenous incretin hormone activity. A multitude of novel
`DPP IV inhibitor compounds have now been developed and
`tested. Here we examine the information available on DPP
`IV and related enzymes, review recent preclinical and
`clinical data for DPP IV inhibitors, and assess their clinical
`significance.
`
` Two primary defects in the pathogenesis of type
`2 diabetes are a relative loss of insulin secretion from
`pancreatic beta cells and a decreased sensitivity of liver and
`peripheral tissues to insulin (1). Drug treatments for type 2
`diabetes have centred therefore on enhancing insulin
`secretion and action. In the case of sulphonylureas (e.g.
`glibenclamide) and meglitinides (e.g. nateglinide) insulin
`secretion
`is
`increased from
`the pancreas (2). The
`biguanides (e.g. metformin) and thiazolidinediones (e.g.
`pioglitazone) improve the body’s sensitivity to insulin (3-
`5). Additionally, synthetic insulin and insulin analogues
`can be administered when oral drugs are no longer
`sufficient to provide adequate blood glucose control.
`
`the
`in particular
`Insulin secretagogues and
`sulphonylureas suffer from a lack of glucose-dependency
`which can lead to episodes of hypoglycaemia (6). Also, as
`the disease progresses and beta-cell function declines,
`several years of use often
`lead
`to declining drug
`effectiveness. With the prevalence of type 2 diabetes
`reaching epidemic proportions (predicted to be about 350
`
`3648
`
`Mylan EX 1011, Page 1
`
`

`
`Dipeptidyl peptidase IV and diabetes
`
`Figure 1. N-terminal cleavage activity and substrate
`specificity of dipeptidyl peptidase IV (DPP IV). DPP IV
`cleaves dipeptides from the N-terminus of peptides and
`polypeptides which have either a proline or an alanine
`residue in the penultimate position (examples of which can
`be found in Table 1). Xaa represents one of the 20
`proteinogenic amino acids.
`
`million by 2025, (7)) new antidiabetic drug treatments are
`urgently required. Incretin hormones are peptides secreted
`from endocrine cells in the small intestine which stimulate
`significant insulin secretion at physiological concentrations
`in a glucose-dependent manner (8-12). The two principal
`incretin hormones are glucagon-like peptide-1 (GLP-1) and
`glucose dependent
`insulinotropic polypeptide
`(GIP).
`Expanding knowledge of the incretin hormones and their
`physiological inactivation by dipeptidyl peptidase IV (DPP
`IV) has lead to two new classes of antidiabetic drugs,
`incretin analogues/mimetics and DPP IV inhibitors/gliptins.
`In this article we focus on the enzyme DPP IV and the
`progress made towards the development of DPP IV
`inhibitors.
`
`3. DPP IV AND RELATED ENZYMES
`
`3.1. DPP IV (EC 3.4.14.5)
`DPP IV is classified as a serine protease by virtue
`of the classical consensus motif Gly-Xxx-Ser-Xxx-Gly,
`which is Gly628-Trp629-Ser630-Tyr631-Gly632 in the case of
`human DPP IV (13). DPP IV was one of the earliest
`identified prolyl peptidases and over the years DPP IV has
`become one of the most intensively studied of its class (14).
`The proteolytic activity of DPP IV is relatively selective,
`cleaving only peptide bonds following proline or alanine
`amino acid residues located penultimate to the N-terminus
`(14; See Figure 1) (this also commonly occurs at sites
`following serine (See Table 1)). DPP IV is also the
`lymphocyte cell surface protein CD26 discussed in detail in
`other reviews (15). Structural studies with soluble human
`DPP IV reveal that the inhibitor diprotin A covalently
`bonds to Ser630 in the catalytic triad, irreversibly blocking
`the active site (13). Physiologically the role of DPP IV is
`wide ranging since it is capable of interacting with a
`number of diverse proteins such as collagen (16),
`fibronectin (17), adenosine deaminase (18),
`tyrosine
`
`3649
`
`phosphatase CD45 (19), and a plethora of regulatory
`peptides across a range of physiological systems (20-22).
`Table 1 lists many physiological peptide substrates (or
`potential substrates) of DPP IV showing N-terminal regions
`targeted by the enzyme.
`
` DPP IV is expressed by endothelia and epithelia
`in most tissues, including bone marrow, kidney, intestine,
`pancreas, liver, lymphocytes, placenta, uterus, prostate and
`skin (14). Levels and expression and activity of DPP IV in
`certain tissues and blood plasma are known to vary
`significantly following the onset of disease, injury or
`inflammation
`(reviewed elsewhere 14, 22). Peptide
`substrates of DPP IV have such extensive physiological
`implications, that most body systems are likely to be
`affected including nervous, endocrine, neuroendocrine,
`immune, vascular, digestive, skeletal and reproductive
`systems (see substrates in Table 1). Among the most
`commonly recognized substrates of DPP IV are several
`chemokines that affect the immune system: (RANTES,
`eotaxin, IP-10, MCP-1, MCP-2, MCP-3, SDF-1α, SDF-1β,
`GCP-2 and MDC, see Table 1 (20); several neuropeptides:
`substance P, bradykinin, peptide YY (PYY), neuropeptide
`Y (NPY), and pituitary adenylate cyclase activating peptide
`(PACAP) (20); and glucagon,
`the counter-regulatory
`hormone of insulin (23). As we shall see later in this
`review, much attention has been focused towards the
`incretin hormones GLP-1 and GIP as substrates of DPP IV.
`Considered briefly below are other physiological proteases
`related to DPP IV which possess similar post-proline
`aminopeptidase activity.
`
`3.2. Dipeptidyl peptidase II (DPP II, also known as DPP
`7, or quiescent cell proline dipeptidase (QPP) (E.C.
`3.4.14.2)
`
`Evidence in recent years has suggested that three
`earlier identified proteases DPP II, DPP 7 and QPP are in
`fact the same enzyme (24-26). Discovered in 1968 by the
`extraction of Lys-Ala- hydrolytic activity from the anterior
`pituitary, DPP II appears to be widely distributed across a
`range of mammalian tissues (24, 27). The identification of
`physiological substrates of DPP II has been hindered by
`low purification yields from
`tissue and a
`lack of
`information
`regarding
`the molecular and catalytic
`properties. Although there are no totally selective DPP II
`inhibitors available, compounds such as Ala-ψ[CS-N]-Pyrr
`and Ala-ψ[CS-N]-Thia have more selectivity towards DPP
`II than DPP IV (28).
`
`3.3. Dipeptidyl peptidase 8 (DPP 8) and dipeptidyl
`peptidase 9 (DPP 9)
`DPP 8 and DPP 9 are relative newcomers to the
`DPP IV enzyme family and their inadvertent inhibition
`appears to be responsible for at least some of the toxic side-
`effects of DPP
`IV
`inhibitors
`including alopecia,
`thrombocytopenia, anaemia, enlarged spleen, and multiple
`histological pathologies,
`including skin
`lesions and
`premature mortality in animals (29, 30). DPP 8 and 9 are
`monomeric
`soluble
`cytoplasmic
`enzymes
`sharing
`approximately 50% sequence similarity with human DPP
`IV (14). Like DPP IV they are widely distributed across
`human tissues and although they have not yet been
`
`Mylan EX 1011, Page 2
`
`

`
`Dipeptidyl peptidase IV and diabetes
`
`Table 1. Physiological regulatory peptides identified as substrates of DPP IV
`Physiological System
`Peptide
`Nutrient metabolism and glucose
`GLP-1 (7-36)amide
`homeostasis
`GIP (1-42)
`GLP-1 (7-37)
`Glucagon
`GLP-2 (1-33)
`Trypsinogen
`
`Digestive system
`
`Trypsinogen pro-peptide
`Gastrin releasing peptide (GRP)
`Pro-colipase
`Enterostatin
`β-Casomorphin
`Aprotinin
`Insulin-like growth factor-1 (IGF-1)
`Growth hormone releasing factor (GHRF)
`Growth hormone-releasing hormone
`(GRH (1-29))
`GRH (1-44)
`PACAP (1-27)
`PACAP (1-38)
`Substance P
`Neuropeptide Y
`Peptide YY (1-36)
`Enkephalins
`Corticotropin-like intermediate lobe peptide
`Endomorphin-2
`Bradykinin
`Human chorionic gonadotrophin α (hCGα)
`Leutinising hormone α chain (LHα)
`Prolactin
`Interleukin-2
`Interleukin-1β
`α1-Microglobulin
`RANTES
`Granulocyte chemtactic protein-2 (GCP-2)
`Stromal cell-derived factor-1α (SDF-1α)
`SDF-1β
`Macrophage-derived chemokine (MDC)
`Monocyte chemotactic protein-1 (MCP-1)
`MCP-2
`MCP-3
`Eotaxin
`Interferon-γ-inducible protein-10 (IP-10)
`Thyrotropin α
`Vasostatin-1
`Peptide histidine methionine
`Tyr-Melanostatin
`
`Growth and development
`
`Neuroendocrine system
`
`Nervous system
`
`Vascular system
`Reproductive system
`
`Immune system
`
`Endocrine system
`
`Other
`
`N-terminus
`His-Ala-Glu-
`Tyr-Ala-Asp-
`His-Ala-Glu-
`His-Ser-Gln-
`His-Ala-Asp-
`Phe-Pro-Thr-
`
`Phe-Pro-Thr-
`Val-Pro-Leu-
`Val-Pro-Asp-
`Val-Pro-Asp-
`Tyr-Pro-Phe-
`Arg-Pro-Asp-
`Gly-Pro-Glu-
`Tyr-Ala-Glu-
`Tyr-Ala-Asp-
`
`Tyr-Ala-Asp-
`His-Ser-Asp-
`His-Ser-Asp-
`Arg-Pro-Lys-
`Tyr-Pro-Ser-
`Tyr-Pro-Ile
`Tyr-Pro-Val-
`Arg-Pro-Val-
`Tyr-Pro-Phe-
`Arg-Pro-Pro-
`Ala-Pro-Asp-
`Phe-Pro-Asn-
`Thr-Pro-Val-
`Ala-Pro-Thr-
`Ala-Pro-Val-
`Gly-Pro-Val-
`Ser-Pro-Tyr-
`Gly-Pro-Val-
`Lys-Pro-Val-
`Lys-Pro-Val-
`Gly-Pro-Tyr-
`Glu-Pro-Asp-
`Glu-Pro-Asp-
`Glu-Pro-Val-
`Gly-Pro-Ala-
`Val-Pro-Leu-
`Phe-Pro-Asp-
`Leu-Pro-Val-
`His-Ala-Asp-
`Tyr-Pro-Leu-
`
`assigned any particular biological function, the undesirable
`consequences of unselective DPP IV inhibition may
`provide important clues. Their post-proline aminopeptidase
`activity has been evidenced by the hydrolysis of H-Ala-
`Pro- and H-Gly-Pro derived substrates (31). Selective
`inhibitors of DPP 8 and DPP 9 are already available and
`have been used to characterise DPP8/9 activity in human
`leukocytes (32).
`
`3.4. Fibroblast Activation Protein (FAP)
`The DPP IV-like activity of FAP has been
`confirmed by the rapid cleavage of an Ala-Pro-NH F3 Mec
`substrate (33). FAP has 52% sequence similarity to DPP IV
`and has been linked with liver injury and chronic liver
`disease (31, 33). FAP is not as widely expressed as DPP
`IV and has been identified in serum and pancreatic alpha
`cells. The active site which carries out N-terminal dipeptide
`cleavage also possesses collagenolytic activity degrading
`
`gelatine and type 1 collagen (34). Selective fluorescent
`probes have been developed to detect and differentiate
`between the proteolytic activities of FAP and DPP IV (35).
`
`4. THE INCRETIN HORMONES, DPP IV AND
`DIABETES
`
`4.1. The incretins
` Although the concept of targeting DPP IV in
`type 2 diabetes is relatively recent, the origin can be traced
`back to early work establishing the importance of the gut in
`regulating post-prandial glucose homeostasis (36, 37). The
`gut contributes neural and endocrine signals that account
`for the enhanced physiological insulin response after a
`meal, a signalling pathway known as the enteroinsular axis
`(36, 37). GLP-1 and GIP secreted from intestinal L- and K-
`cells, respectively, account for most of the enteroinsular (or
`“incretin”) effect (38-42). A list of well characterised
`
`3650
`
`Mylan EX 1011, Page 3
`
`

`
`Dipeptidyl peptidase IV and diabetes
`
`actions of GLP-1 and GIP relevant to type 2 diabetes can
`be found in Table 2. The glucose-dependent nature of their
`insulinotropic activity provides a clear advantage to
`enhance postprandial insulin secretion and reduce the risk
`of interprandial hypoglycaemia (8-10). Insulin biosynthesis
`is also increased. Furthermore, GLP-1 and GIP possess
`extrapancreatic mechanisms which contribute to limit
`hyperglycaemia, e.g. reducing hepatic insulin extraction
`(43, 44), reported ‘insulin-like’ effects on skeletal muscle,
`liver and adipose tissue (45-49), and a reduction of gastric
`acid secretion or gastric emptying (50-52). Incretin
`hormones have additional potential benefits over other
`insulin-releasing drugs through improved islet morphology
`and protective and proliferative effects on the pancreatic
`beta-cell (53-59). Such properties might help to counter the
`characteristic age-related decline of beta-cell mass in
`diabetes. Although these properties are evident in animal
`models, substantiating effects of incretin hormones on beta-
`cell morphology in humans remains problematic. Finally,
`an important yet occasionally overlooked effect is the
`suppression of glucagon secretion by GLP-1 (60).
`
`4.2. Inactivation of the incretins by DPP IV
`Mentlein and co-workers were amongst the first
`to demonstrate degradation of GLP-1 and GIP by DPP IV
`in vitro (61). Using DPP IV purified from human placenta
`they observed the enzymatic removal of N-terminal
`dipeptides His7-Ala8 and Tyr1-Ala2 from GLP-1 and GIP,
`respectively. More significantly, they observed that this
`degradation also took place when GLP-1 and GIP were
`incubated in human serum (61). It was subsequently
`confirmed that DPP IV-mediated metabolism of GLP-1 and
`GIP did indeed occur in vivo (40, 62). The action of DPP
`IV in vivo reduces the half-life of GLP-1 and GIP to <2
`min (40, 62, 63). Since the predominant and active forms of
`incretin hormones are GLP-1(7-36)amide and GIP(1-42),
`degradation by DPP IV leads to major degradation
`fragments GLP-1(9-36)amide and GIP(3-42), respectively.
`
`The activities and binding characteristics of these
`fragments have been elucidated. Since GLP-1(9-36)amide
`and GIP(3-42) are both non-insulinotropic peptides, it was
`initially suggested that these were relatively inert and
`inactive metabolites (64, 65). Although the affinity of GLP-
`1(9-36)amide for the GLP-1 receptor is 100-fold lower than
`the parent molecule it appears to act as a weak receptor
`antagonist (64, 66, 67). While GLP-1(9-36)amide does not
`antagonise the insulinotropic activity of GLP-1(7-36)amide
`in vivo, there is evidence that this metabolite possesses
`weak antihyperglycaemic activity through a mechanism not
`involving insulin secretion (68). This concept currently
`remains controversial due to conflicting findings in mice,
`pigs and humans (68-70).
`
`The receptor affinity of GIP fairs comparatively
`better following truncation by DPP IV. The binding affinity
`of GIP(3-42) is approximately 4-fold lower than that of
`GIP(1-42) (71). In vitro studies have demonstrated that
`GIP(3-42) antagonises the GIP receptor (72, 73). However,
`in vivo studies have been conflicting (70-73). A recent
`study confirmed that GIP(3-42) antagonises GIP-stimulated
`cAMP production and insulin secretion, but found that in
`
`vivo it does not behave as an antagonist at physiological
`concentrations (71). However, pharmacological doses of
`GIP(3-42) (25 nmol/kg) administered to obese diabetic
`(ob/ob) mice once daily for 14 days enhanced insulin
`sensitivity and improved glycaemic control (70). This
`appears to involve extrapancreatic mechanisms which lead
`to improved insulin sensitivity. During this study neither
`GIP(3-42) or GLP-1(9-36)amide affected body weight,
`food intake, pancreatic insulin content or islet morphology
`(70).
`
`4.3. Overcoming DPP IV mediated incretin inactivation
`As knowledge of incretin hormone inactivation
`expanded greater emphasis was placed on ways
`to
`overcome this problem. The pharmacological strategies
`adopted have led to the development of two fundamentally
`new ways to treat type 2 diabetes. The first approach has
`involved generating GLP-1 and GIP agonists resistant to
`the action of DPP IV (reviewed elsewhere (74)).
`Production and testing of numerous modified forms of the
`incretin hormones have generated effective DPP IV-
`resistant analogues of GLP-1 and GIP (74). In human
`subjects GLP-1 analogues have demonstrated sustained
`improvements in glycaemic control in type 2 diabetes. To
`date one GLP-1 agonist, exendin (exenatide/Byetta), has
`been clinically approved, and another, liraglutide (NN2211)
`is in phase III clinical trials (75). A second and more
`recently adopted approach, which is the focus of this
`review, has been the development of DPP IV inhibitors (22,
`76, 77). As illustrated in Figure 2 the concept of DPP IV
`inhibitors is to enhance endogenous incretin activity by
`preventing the rapid inactivation of incretin hormones. The
`preclinical and clinical data for DPP IV inhibitors (or
`‘gliptins’ as they are termed) is reviewed later in this
`article.
`
`4.4. Rodent models lacking DPP IV activity
`lacking
`The generation of
`rodent models
`functional DPP IV has brought major advances in our
`understanding of this enzyme’s role in metabolism, and has
`strengthened the rationale for developing specific inhibitors
`of DPP IV (78-81). Of particular note is the fact that mice
`lacking DPP IV activity have significantly reduced
`glycaemic excursions, greater levels of glucose-stimulated
`insulin, while the degradation of both GLP-1 and GIP is
`reduced (78). Similarly, DPP IV-deficient rats have
`improved glucose tolerance, enhanced insulin release and
`higher levels of active GLP-1 (79). Evidence gathered from
`these rodent models underpins the role of DPP IV in
`regulating
`incretin activity and consequently glucose
`homeostasis.
`
`IV
`that DPP
`interesting
`is especially
` It
`‘knockout’ mice are relatively resistant to the development
`of glucose intolerance and diabetes following 20 weeks on
`a high fat diet (80). These mice exhibited reduced food
`intake and enhanced metabolic energy expenditure and did
`not develop obesity (80). This has been substantiated by
`similar observations in DPP IV-deficient Fischer rats (81).
`Furthermore, DPP IV ‘knockout’ appears
`to confer
`protection from the diabetogenic effects of modest amounts
`of the beta-cell toxin streptozotocin (80).
`
`3651
`
`Mylan EX 1011, Page 4
`
`

`
`[Frontiers in Bioscience 13, 3648-3647, May 1, 2008]
`
`Table 2. Characterised actions of GLP-1 and GIP relevant to type 2 diabetes
`
`Released in response to a mixed meal
`Lower blood glucose
`Glucose-dependent stimulation of insulin secretion
`Suppress glucagon secretion
`Extrapancreatic glucose-lowering actions
`Extend beta cell mass and survival
`Suppress gastric acid secretion
`Inhibition of gastric emptying
`Inhibition of hepatic insulin extraction
`Enhance satiety
`Reduce body weight
`
`Effect
`√
`√
`√
`√
`√
`√
`-
`√
`√
`√
`√
`
`GLP-1
`
`Reference
`111, 112
`113
`8
`60
`45-47
`53-56
`
`-
`
`50, 51
`43
`116
`117
`
`GIP
`
`Effect
`√
`√
`√
`-
`√
`√
`√
`-
`√
`-
`-
`
`Ref
`111, 112
`114, 115
`9, 10
`-
`48, 49
`57-59
`52
`-
`44
`-
`-
`
`Figure 2. Incretin hormone inactivation and DPP IV inhibitor mode of action. The incretin hormones (GLP-1 and GIP) are
`released from the intestine following meal ingestion. Incretins circulate to the pancreas where they stimulate insulin-release
`leading to a lowering of plasma glucose concentrations. However, enzymatic cleavage by ubiquitous DPP IV renders them non-
`insulinotropic. DPP IV inhibitors prevent processing by DPP IV and therefore enhance endogenous incretin hormone activity.
`
`lacking DPP IV
` Although, animal models
`activity are viable and appear relatively normal, recent
`evidence is emerging of some neurological, immunological
`and inflammatory alterations (82-87). Mice lacking DPP IV
`have shortened latencies to nociceptive stimuli, perhaps due
`to observed higher plasma levels of substance P (82). In the
`context of experimental asthma, rats lacking DPP IV
`demonstrate decreased T-cell recruitment associated with
`significantly reduced ovalbumin-specific IgE-titres (83).
`Furthermore, marked changes in the cytokine responses of
`
`interleukins and tumour necrosis factor have been observed
`(84; 85). DPP IV deficient rats are more susceptible to
`angiooedema caused by ACE inhibitor administration (86)
`and it has been reported that DPP IV inhibition in some
`species (e.g. dogs and monkeys) cause gastrointestinal
`disturbances and skin lesions, although these remain to be
`confirmed (31). Finally, the severity of antigen-induced
`arthritis is increased in DPP IV-deficient mice which may
`be due to increased levels of circulating active stromal cell-
`derived factor-1 (87).
`
`3652
`
`Mylan EX 1011, Page 5
`
`

`
`[Frontiers in Bioscience 13, 3648-3647, May 1, 2008]
`
`Company
`
`Clinical Phase
`
`Specificity data
`
`Table 3. DPP IV inhibitors in clinical development
`(Proprietary
`Inhibitor
`(reference
`Generic
`name
`number)
`name)
`Sitagliptina (Januvia)
`IC50 = 18 nM Ki = 9 nM
`4
`Merck
`MK-0431
`Vildagliptinb (Galvus)
`IC50 = 3.5 nM Ki = 17 nM
`3
`Novartis
`LAF237
`IC50 = 26nM Ki = <1 nM
`3
`Bristol-Myers Squib and AstraZeneca
`BMS-477118
`Saxagliptin
`-
`3
`Takeda
`SYR-322
`Alogliptin
`Ki = 0.1 nM
`2
`Roche
`R1438
`Aminomethylpyridine deivative
`Ki = 80 nM
`2
`Probiodrug
`P32/98
`Isoleucine thiazolidide
`-
`2
`Prosidion
`PSN9301
`-
`IC50 = 1.6 nM
`2
`Glenmark
`GRC-8200
`Carbazole compound
`-
`2
`Phenomix
`PHX-1149
`-
`-
`1
`Sanofi-Aventis
`SSR-162369
`Pyrrolidinoxanthine derivative
`-
`1
`Alantos/Amgen
`ALS 2-0426
`-
`-
`?
`NovoNordisk
`NN-7201
`-
`a Sitagliptin was launched in Mexico in 2006, and in USA, UK and several other European countries in 2007, b A New Drug
`Application (NDA) for vildagliptin was submitted to FDA (American Food and Drug Administration) and EMEA (European
`Medicines Evaluation Agency) in 2006, ‘-’ = information not available.
`
`5. INHIBITORS OF DPP IV ACTIVITY
`
`5.1. Preclinical studies
`In the late 1990s the first DPP IV inhibitors were
`developed and tested in animal models of type 2 diabetes.
`One of
`the
`initial studies demonstrated
`that oral
`administration of P32/98
`(isoleucine
`thiazolidide),
`improved glucose tolerance in Zucker fatty rats (88). The
`effectiveness of another inhibitor vildagliptin (LAF-237) to
`control glycaemia was similarly demonstrated (89). These
`studies indicated that effects were dose-dependent and
`caused no tachyphylaxis over the 3 weeks treatment period
`(89). NVP-DPP728 similarly improved glucose tolerance
`and the mechanism was verified by the presence of higher
`plasma levels of GLP-1(7-36)amide (90). In high fat fed
`mice, valine-pyrrolidide potentiated the plasma insulin
`response to intra-gastric glucose and improved glucose
`tolerance (91). Interestingly, valine-pyrrolidide did not
`affect glucose-stimulated insulin secretion from isolated
`islets, suggesting the mechanism was indirect, although
`enhancement of GLP-1 (91) and GIP (92) activity was
`confirmed. Investigation of DPP IV inhibitor in isolated
`perfused porcine ileum revealed that active levels of
`secreted GLP-1 could be increased from ~50% to over 85%
`(93).
`
`Chronic treatment (12 weeks) of Zucker rats with
`P32/98 improved glucose tolerance and increased plasma
`insulin levels, but did not lead to any measurable changes
`in beta cell mass or islet morphology (94). P32/98 reduced
`body weight gain, an effect which was associated with
`GLP-1 activity, but there was no change in food intake. An
`almost identical P32/98 study indicated that hepatic and
`peripheral insulin sensitivity was significantly improved
`(95).
`
`Preclinical studies have also demonstrated that
`DPP IV inhibitors are most effective in mild to moderate
`models of diabetes but fail to control glycaemia in severely
`hyperglycaemic mice (79, 96). The effects of DPP IV
`inhibitors in combination with other antidiabetic drugs have
`been examined in diabetic mouse models. Combining
`vildagliptin with rosiglitazone did not provide any
`additional efficacy to rosiglitazone alone, but it did appear
`to reduce common side-effects such as weight gain and
`haemodilution (96). Studies combining a DPP IV inhibitor
`
`with an alpha-glucosidase inhibitor have demonstrated
`additive improvements of glucose tolerance and increased
`active GLP-1 levels (97).
`
`5.2. Clinical studies
`Following the encouraging results of studies in
`animal models, there is now a substantial list of DPP IV
`inhibitors in clinical development with at least 8 others at
`earlier stages (Table 3). These have received the class
`name of ‘gliptins’ or ‘incretin enhancers’ (21, 30, 77, 98).
`The first gliptin to be approved for clinical use and
`marketed as a treatment for type 2 diabetes is sitagliptin
`(Januvia), launched in Mexico in 2006, and in the USA and
`several European countries in 2007. Vildagliptin (Galvus)
`was submitted for evaluation by the regulatory agencies in
`2006. Two further gliptins, saxagliptin and alogliptin, are
`advanced in phase III clinical development.
`
`As discussed above, the antidiabetic effect of
`DPP IV inhibitors relates mainly to increased nutrient-
`stimulated (prandial) insulin secretion. This is particularly
`effective in lowering postprandial hyperglycaemia, but
`there is a substantial carry-over effect to benefit the control
`of interprandial glycaemia (21, 77, 98). However, DPP IV
`inhibitors do not increase basal insulin secretion, and only
`suppress glucagon secretion in the hyperglycaemic state, so
`there
`is
`low risk of
`interprandial ‘over-shoot’
`into
`hypoglycaemia. To assess overall long-term glycaemic
`control in diabetic patients it is customary to measure the
`percentage of glycated haemoglobin (HbA1c): the normal
`range is <6.0%, and the target for treatment of diabetic
`patients is usually to achieve a sustained HbA1c < 6.5-
`7.5% depending on country and regional guidelines.
`
`5.2.1. Sitagliptin
`Sitagliptin is a piperazine derivative with a high
`(~87%) bioavailability, rapid absorption in man (Tmax 1-
`4h) and low (~38%) plasma protein binding. Near
`complete inhibition of DPP IV activity for about 12h is
`achieved in man with a single daily dose of 100 mg.
`Clinical trials to assess the efficacy, tolerability and safety
`of sitagliptin in type 2 diabetes patients have recently been
`reported (98, 99). The long-term effects on HbA1c when
`sitagliptin is administered as monotherapy or add-on
`therapy to certain other types of antidiabetic agents are
`listed in Table 4 (100-104). Sitagliptin typically reduced
`
`3653
`
`Mylan EX 1011, Page 6
`
`

`
`Dipeptidyl peptidase IV and diabetes
`
`RDBPC
`
`24
`
`741
`
`Table 4. Long-term (>24 weeks) clinical trials with the DPP IV inhibitor sitagliptin (Januvia) in patients with type 2 diabetes
`Study design
`Duration (weeks)
`n
`Dose (mg/day)
`Other antidiabetic
`Mean
`Baseline
`Placebo subtracted decrease
`Ref
`treatment
`HbA1c
`in HbA1c
`Diet only
`8.0 %
`↓ 0.79 %
`100
`Diet only
`8.0 %
`↓ 0.94 %
`200
`101
`Diet + metformin
`7.96 %
`↓ 0.65 %
`100
`701
`24
`RDBPC
`102
`Diet + pioglitazone
`8.05 %
`↓ 0.70 %
`100
`353
`24
`RDBPC
`340a
`103
`Diet only
`8.8 %
`↓ 0.83 %
`100
`24
`RDBPC
`104
`Diet + metformin
`7.5%
`↓ 0.67 %
`100
`1,172
`52
`RDBAC
`Adapted with permission from 98, aNumber is the diet+placebo and the diet+sitagliptin arms only, bSame decrease in HbA1c
`(0.67%) for diet+metformin+sitagliptin as for diet+metformin+glipizide. RDBPC = randomised double blind placebo control.
`RDBAC = randomised double blind active comparator.
`
`100
`
`52
`24
`
`24
`52
`24
`
`RDBPC
`RDBAC
`RDBPC
`
`107
`463
`
`607b
`780
`544
`
`Table 5. Long-term (>24 weeks) clinical trials with the DPP IV inhibitor vildagliptin (Galvus) in patients with type 2 diabetes
`Study
`Duration (weeks)
`n
`Dose
`Other antidiabetic
`Mean
`Baseline
`Placebo
`subtracted
`Ref
`design
`(mg/day)
`treatment
`HbA1c
`decrease in HbA1c
`RDBPC
`100
`Diet + metformin
`7.7 %
`↓ 1.1 %
`8.7 %a
`RDBPC
`50
`Diet + pioglitazone
`↓ 0.8 %
`8.7 %a
`100
`Diet + pioglitazone
`↓ 1.0 %
`100
`Diet only
`8.7 %
`↓ 1.1 %
`↓ 1.0 %c
`100
`Diet + metformin
`8.7 %
`50
`Diet + metformin
`≥7.4 %
`↓ 0.7 %
`100
`Diet + metformin
`↓ 1.1 %
`Adapted with permission from 98, aApproximate HbA1c value estimated from illustration, bNumber is the diet+placebo and the
`diet+vildagliptin arms only, cDecrease in HbA1c was 1.0% diet+vildagliptin versus 1.4% for diet+metformin. RDBPC =
`randomised double blind placebo control. RDBAC = randomised double blind active comparator.
`
`105
`106
`
`107,108
`109
`110
`
`HbA1c (from a baseline of ~ 8%) by about 0.7 – 0.8% after
`24-52 weeks. Efficacy was similar whether the sitagliptin was
`taken as monotherapy or add-on therapy to metformin or a
`thiazolidinedione. In these trials fasting plasma glucose
`concentrations were reduced by about 1.0 – 1.5 mmol/L, and
`postprandial glucose levels measured 2 hours after a standard
`mixed meal were usually reduced by about 3 mmol/L.
`
`in nutrient-
`increase
`the
`Consistent with
`stimulated (but not basal) insulin secretion, sitagliptin
`therapy did not cause a clinically significant increase in the
`incidence of reported hypoglycaemia in any of the trials.
`Indeed, a combination of sitagliptin + metformin for 52
`weeks was associated with only 4.9% of patients reporting
`hypoglycaemia events compared with 32% of patients
`receiving glipizide + metformin (104). This was achieved
`with similar overall levels of glycaemic control in the two
`groups as indicated by HbA1c. It is also noteworthy that
`[delete space] sitagliptin did not increase body weight
`compared to placebo in any of the trials.
`
`Parameters of tolerability and the adverse events
`profile for sitagliptin were generally similar to the placebo or
`comparator groups in these trials, providing no signals for
`concern in these patient groups over these time periods.
`Despite the potential to slow gastric emptying, this does not
`seem to be a clinical issue as there was little reporting of
`abdominal discomfort or nausea. Patients receiving sitagliptin
`are required to have good renal function since the drug is
`mostly eliminated unchanged in the urine. However, the drug
`does not appear to induce or inhibit P450 isoforms, so it should
`have little effect on the metabolism of other drugs and can be
`used in patients with mild liver disease.
`
`5.2.2. Vildagliptin and other gliptins
`Vildagliptin is a cyanopyrrolidine: absorption is
`rapid, bioavailability is >90%, and there is extensive hepatic
`metabolism of the drug to metabolites that are mostly
`eliminated in the urine. Inhibition of DPP IV by >90% persists
`for more than 12h after a single daily dose of 100 mg, which is
`likely to be the preferred therapeutic dose. Vildagliptin has
`received a similar battery of clinical trials to sitagliptin and
`shown similar results when used as monotherapy or add-on to
`metformin or a thiazolidinedione (105-110; Table 5). Slightly
`greater lowering of HbA1c in some studies with vildagliptin
`may relate in part to a higher baseline (starting) HbA1c. As
`with sitagliptin there was low risk of hypoglycaemia and no
`effect on body weight compared to placebo.
`
`5.3. Developmental issues
`Initial preclinical development of DPP IV inhibitors
`as antidiabetic agents focused on their selectivity: studies in
`rodents, dogs and monkeys have suggested that possible
`interference with DPP 8/9 or other dipeptidyl peptidases could
`cause blood dyscrasias, gastrointestinal disturbances and
`various histopathological changes including skin lesions (29,
`30). Inhibitors of DPP 8/9 have also reduced T-cell activation
`in human in vitro models. However the DPP IV inhibitors
`presently advanced in clinical development have shown highly
`selective and potent inhibition of DPP IV. They have
`exhibited sufficient potency and duration of action that
`therapeutic doses achieved nearly complete inhibition of DPP
`IV activity for >12h with a diminishing effect thereafter.
`
`Although DPP IV inhibition could potentially
`affect a wide range of biological peptides, no apparent
`serious untoward effects have emerged during
`the
`
`3654
`
`Mylan EX 1011, Page 7
`
`

`
`Dipeptidyl peptidase IV and diabetes
`
`and vildagliptin.
`sitagliptin
`trials with
`clinical
`Nevertheless, the urinary metabolites of vildagliptin and
`possible interference with DPP 8/9 are receiving further
`investigation. Available information regarding other
`gliptins such as saxagliptin and alogliptin remain
`preliminary as these agents are proceeding in phase III
`clinical trials. As with all new drugs, phase IV
`pharmacovigilance will watch for any evidence of
`possible long