`
`Diabetes, Obesity and Metabolism 14 (Suppl. 2): 20–32, 2012.
`© 2012 Blackwell Publishing Ltd
`
`Comparison of liraglutide versus other incretin-related
`anti-hyperglycaemic agents
`L. Blonde1 & E. Montanya2
`1OchsnerDiabetesClinicalResearchUnit,DepartmentofEndocrinology,DiabetesandMetabolism,OchsnerMedicalCenter,NewOrleans,LA,USA
`2HospitalUniversitariBellvitge-IDIBELL,UniversityofBarcelona,CIBERdeDiabetesyEnfermedadesMetab´olicasAsociadas(CIBERDEM),Barcelona,Spain
`
`The two classes of incretin-related therapies, dipeptidyl peptidase-4 (DPP-4) inhibitors and glucagon-like peptide-1 receptor agonists (GLP-1
`RAs), have become important treatment options for patients with type 2 diabetes. Sitagliptin, saxagliptin, vildagliptin and linagliptin, the
`available DPP-4 inhibitors, are oral medications, whereas the GLP-1 RAs—twice-daily exenatide, once-weekly exenatide and once-daily
`liraglutide—are administered subcutaneously. By influencing levels of GLP-1 receptor stimulation, these medications lower plasma glucose
`levels in a glucose-dependent manner with low risk of hypoglycaemia, affecting postprandial plasma glucose more than most other anti-
`hyperglycaemic medications. Use of GLP-1 RAs has been shown to result in greater glycaemic improvements than DPP-4 inhibitors, probably
`because of higher levels of GLP-1 receptor activation. GLP-1 RAs can also produce significant weight loss and may reduce blood pressure and
`have beneficial effects on other cardiovascular risk factors. Although both classes are well tolerated, DPP-4 inhibitors may be associated with
`infections and headaches, whereas GLP-1 RAs are often associated with gastrointestinal disorders, primarily nausea. Pancreatitis has been
`reported with both DPP-4 inhibitors and GLP-1 RAs, but a causal relationship between use of incretin-based therapies and pancreatitis has not
`been established. In clinical trials, liraglutide has shown efficacy and tolerability and resulted in certain significant benefits when compared with
`exenatide and sitagliptin.
`Keywords: DPP-4 inhibitor, GLP-1, linagliptin, liraglutide, saxagliptin, sitagliptin, vildagliptin
`
`Date submitted 1 January 2012; date of final acceptance 11 January 2012
`
`article
`review
`
`Introduction
`Incretin-related therapies have become established as impor-
`tant treatment options for patients with type 2 diabetes
`(T2D) [1] and the number of available options will increase in
`the near future. Knowledge about how the current incretin-
`related anti-hyperglycaemic therapies compare to each other
`can help health care professionals make informed treatment
`choices for individual patients.
`Patients with T2D have an impaired incretin effect, which
`appears to be the result of reductions in the insulinotropic
`and glucagon-suppressive actions of
`the incretin hor-
`mones glucagon-like peptide (GLP-1) and glucose-dependent
`insulinotropic peptide (GIP) [2,3], although declines in the
`release of these hormones have also been reported among
`patients with T2D [2,4]. When these patients receive infusions
`of GLP-1 to supraphysiological levels, the insulin secretory
`response improves, glucagon secretion is suppressed and
`plasma glucose levels can be significantly improved. These
`effects occur in a glucose-dependent manner, that is, only
`when glucose levels are elevated, resulting in a low risk of
`hypoglycaemia [2,5,6].
`
`Correspondenceto: Dr Lawrence Blonde, Ochsner Diabetes Clinical Research Unit, 1514
`Jefferson Highway, New Orleans, LA 70121, USA.
`E-mail: lawrence.blonde@gmail.com
`
`Re-use of this article is permitted in accordance with the Terms and Conditions set out at
`http://wileyonlinelibrary.com/onlineopen#OnlineOpen_Terms
`
`In addition to improving glucose control, raising GLP-1
`levels can provide additional benefits such as slowed gastric
`emptying, decreased acid secretion, increased feeling of satiety
`and reduced energy intake [5,7]. In humans, endothelial
`dysfunction, cardiovascular function and β-cell function also
`appear to improve with a GLP-1 infusion [8–11], while animal
`studies suggest that GLP-1 can stimulate expansion of β-cell
`mass [12], and reduce high blood pressure [13].
`As native GLP-1 is degraded rapidly by the enzyme dipeptidyl
`peptidase-4 (DPP-4), resulting in a half-life of approximately
`2 min following intravenous administration [14], its therapeu-
`tic use is impractical. Two strategies have been employed to
`produce incretin-related therapies. One approach is to inhibit
`the DPP-4 enzyme, resulting in an extended half-life and
`an increase in circulating endogenous GLP-1 and GIP [3].
`The other approach involves the use of agents resistant to the
`breakdown of DPP-4 that bind to and activate the GLP-1 recep-
`tor, thus producing glucoregulatory effects similar to those of
`GLP-1. This article discusses the clinical profiles and compares
`the available agents in these two classes.
`
`Pharmacological Differences Between
`the Incretin-Related Therapies
`The GLP-1 receptor agonists (GLP-1 RAs) exenatide, exenatide
`once weekly (currently only approved in Europe) and
`liraglutide are peptides, and so they must be given by
`
`
`
`MPI EXHIBIT 1058 PAGE 1
`
`MPI EXHIBIT 1058 PAGE 1
`
`
`
`DIABETES, OBESITY AND METABOLISM
`
`subcutaneous injection: twice daily (BID) for exenatide, within
`60 min before the two main meals and at least 6 h apart; once
`a week on the same day for exenatide once weekly, with or
`without meals; and once a day for liraglutide, independent
`of meals. Although both exenatide and liraglutide are GLP-1
`RAs, exenatide is a mimetic, discovered in the saliva of the
`Gila monster (Heloderma suspectum), with 53% amino acid
`sequence identity to the chemical structure of native GLP-
`1, while liraglutide is an analogue of human GLP-1 with
`97% sequence identity [15,16]. Exenatide once weekly is a
`long-acting formulation of exenatide in which exenatide is
`encapsulated in microspheres of poly(d,l lactic-co-glycolic
`acid) for gradual drug delivery [17]. Liraglutide differs from
`human GLP-1 by the attachment of a palmitic acid via a
`glutamic acid spacer to lysine at position 26 and by the
`replacement of lysine at position 34 with arginine [18]. It
`is thought that the addition of a fatty acid chain to liraglutide’s
`structure allows it to form heptamers when injected, which
`delay its absorption and binding to albumin, increasing its
`resistance to DPP-4 degradation and allowing the maximum
`concentration to be reached at 8–12 h after dosing [19,20].
`In contrast, exenatide reaches its median peak concentration
`in 2 h [21]. Exenatide once weekly takes much longer than
`either liraglutide or exenatide BID to reach maximum
`concentrations. After 2 weeks of administering exenatide
`once weekly, serum concentrations exceed minimal efficacy
`levels and continue to increase over the next 4–5 weeks if
`treatment is maintained [22]. Clinical studies have confirmed
`that administration of exenatide and liraglutide results in dose-
`dependent decreases in hyperglycaemia through insulinotropic
`activity and suppression of glucagon secretion, both occurring
`in a glucose-dependent manner [23–25]. It is probable that
`there is a greater degree of GLP-1 receptor stimulation with
`GLP-1 RAs than that resulting from the two- to threefold
`increase in GLP-1 levels with DPP-4 inhibitors [3,26].
`DPP-4 inhibitors are small molecule oral medications. There
`are presently three available in the USA: sitagliptin (also
`produced in a single pill combination tablet with metformin),
`saxagliptin (also produced in a single pill combination tablet
`with metformin extended release) and linagliptin. All are
`administered once a day at any time, except for the sitagliptin
`combination tablet with metformin, which should be taken
`BID with meals. In Europe, another twice-daily DPP-4
`inhibitor can be prescribed: vildagliptin (also produced in
`a single pill combination tablet with metformin). Sitagliptin
`is a phenethylamine type of DPP-4 inhibitor, saxagliptin
`and vildagliptin are cyanopyrrolidines [27] and linagliptin is
`xanthine-derived [28]. As saxagliptin has been observed in
`the laboratory to have strong interactions with the DPP-4
`residues Ser630, Glu205 and Glu206, which are essential to the
`enzyme’s catalytic activity, its potency for inhibiting DPP-4
`activity is considered more robust than that of both sitagliptin
`and vildagliptin [29]. However, linagliptin has demonstrated
`more potent inhibition of DPP-4 when compared to the three
`other DPP-4 inhibitors under identical in vitro conditions [28].
`Despite the variable selectivity of the currently approved DPP-4
`inhibitors, it is not clear that there is much clinical difference
`between them. When the recommended dosages for each agent
`
`review article
`
`are administered, the median time to maximum concentration
`(Tmax) ranges from 1.7 h for vildagliptin, 2 h for saxagliptin,
`1–4 h for sitagliptin and 1.5 h for linagliptin [30–33]. The
`duration of DPP-4 inhibition is claimed to be 24 h with each
`agent; however, the terminal half-life (t1/2) for saxagliptin is
`2.5 h, vildagliptin 3 h and sitagliptin 12.4 h [30–32]. Although
`the t1/2 for linagliptin is more than 100 h, the effective half-life
`is approximately 12 h [33]. Only saxagliptin appears to have a
`pharmacologically active metabolite, 5-hydroxy saxagliptin,
`which has a t1/2 of 3.1 h [31]. Following an oral glucose
`load or meal in patients receiving DPP-4 inhibitors, there
`is an increase in the circulating levels of GLP-1, a reduction
`in glucagon concentration and an enhancement of glucose-
`dependent insulin secretion [30–33].
`In the USA, all FDA-approved incretin-related therapies
`can be used as monotherapy (although liraglutide is not rec-
`ommended as first-line therapy) [33–37], whereas in Europe,
`only sitagliptin and linagliptin are approved as monother-
`apy [32,33]. In the USA, all incretin therapies can also be used in
`dual- or triple-combination therapy with metformin, sulpho-
`nylureas (SUs) and/or thiazolidinediones (TZDs) [33–37].
`However, the approved combined uses for these medications
`are slightly more restrictive in Europe, where liraglutide is
`approved for use in dual combination with metformin or SUs,
`and exenatide, exenatide once weekly, saxagliptin, sitagliptin
`and vildagliptin can be used in dual combination with met-
`formin, SUs or TZDs [20–22,30–32]. Exenatide, exenatide
`once weekly, liraglutide and sitagliptin are the only medica-
`tions approved for triple combination with metformin and
`a SU or metformin and a TZD in Europe [20–22,32]. In
`both the USA and Europe, sitagliptin is approved for use
`with insulin [32,34]. The efficacy and safety of exenatide and
`liraglutide in conjunction with insulin have been studied, but
`only limited approval has been obtained for any combined use
`[38,39]. In one 52-week trial, the addition of insulin detemir to
`liraglutide 1.8 mg and metformin in patients not achieving gly-
`caemic targets led to decreases in HbA1c, sustained weight loss
`and a small increase in minor hypoglycaemic events [40,41]. In
`Europe, insulin detemir may be used as add-on therapy with
`liraglutide [41]. The addition of liraglutide in patients already
`treated with insulin has not been evaluated. The addition of
`exenatide following insulin optimisation led to reduction in
`HbA1c, modest weight decrease, and no change in hypogly-
`caemic rates [38], and in the USA, exenatide may be added
`on to therapy with insulin glargine [37]. No clinical data exist
`regarding the combination of GLP-1 RAs with DPP-4 inhibitors
`in treatment, and it is not currently recommended, although
`data in minipigs have suggested pharmacokinetics are unlikely
`to be altered by the combination [42].
`As expected from the glucose-dependent pharmacoki-
`netic/pharmacodynamic profiles of incretin-related therapies,
`these treatments lead to improvements in controlling postpran-
`dial glucose (PPG) excursions [43,44]. Twice-daily exenatide
`is dosed to peak with PPG concentrations, unlike the other
`incretin-related therapies, which can be administered without
`regard for meals, and provides control of postprandial excur-
`sions. In head-to-head trials, exenatide taken BID lowered PPG
`to a greater degree than did sitagliptin and liraglutide [15,45].
`
`Volume 14 No. (Suppl. 2) April 2012
`
`doi:10.1111/j.1463-1326.2012.01575.x 21
`
`
`
`MPI EXHIBIT 1058 PAGE 2
`
`MPI EXHIBIT 1058 PAGE 2
`
`
`
`review article
`
`Improvements in HbA1c demonstrated by incretin-related
`therapies when added to metformin are shown in Table 1.
`
`Effect on Gastric Emptying and Weight Loss
`Patients with T2D often have accelerated gastric emptying,
`which may contribute to postprandial hyperglycaemia [51,52].
`Native GLP-1 can slow accelerated emptying of the stomach,
`and slow acid secretion, contributing to its effectiveness at
`lowering postprandial hyperglycaemia [5,7]. Clinical studies
`have shown that GLP-1 RAs produce the same effect as
`native GLP-1 [23,25], while DPP-4 inhibitors do not [3,53].
`The slowed gastric emptying observed with GLP-1 RAs
`may contribute
`to the most
`common gastrointestinal
`adverse event
`reported for
`these therapies
`in clinical
`trials—nausea [15,26,43,54–58]. Gastrointestinal problems
`are infrequent with DPP-4 inhibitors [43].
`In clinical studies, GLP-1 RAs have been associated with
`dose-dependent weight loss [49,59], which has generally not
`been seen with DPP-4 inhibitors as the latter appear to be
`weight neutral [60,61]. The higher levels of GLP-1 receptor
`stimulation achieved with GLP-1 RAs compared to DPP-4
`inhibitors are probably the most important factor responsible
`for the difference in weight effect observed between the
`two kinds of incretin-related therapies. The weight loss is
`not primarily related to gastrointestinal symptoms such as
`nausea, as many patients using GLP-1 RAs lose weight
`without experiencing any nausea, and the nausea is typically
`transient [3]. Preclinical and clinical studies have shown that
`these agents can dose-dependently increase the feeling of satiety,
`reduce meal size and lower energy intake in a manner similar to
`that of native GLP-1 [7,62–65]. The slowed gastric emptying
`by GLP-1 RAs may also contribute to increased satiety, reduced
`food intake and the resultant clinically significant weight loss
`shown in many studies with these agents [3,15,49,55–59].
`Another possible explanation for the lack of DPP-4 inhibitor
`effect on weight may be that they inhibit cleavage of the gut
`hormone peptide YY (PYY). Thus, levels of intact PYY1-36,
`which stimulates food intake, may be increased while levels
`of the active form PYY3-36, which reduces food intake, may
`be reduced [66]. In a recent 12-week, randomised, placebo-
`controlled clinical study, sitagliptin decreased PYY3-36 while
`increasing intact PYY1-36. The depression of PYY3-36 levels
`with DPP-4 inhibitor treatment may thus contribute to the
`difference in weight response between the two classes of incretin
`therapies [67]. Lastly, in animal studies, GIP has been linked
`with obesity through over-nutrition [68]. Although T2D is a
`GIP-resistant state [2], DPP-4 inhibitors raise GIP, as well as
`GLP-1 levels, by blocking the activity of the DPP-4 enzyme.
`This effect might also have a role in the weight neutrality (rather
`than weight loss) that is seen with DPP-4 inhibitor therapy.
`
`Blood Pressure, Lipids and Other
`Cardiovascular Risk Factors
`GLP-1 RAs have been shown to potentially improve multiple
`cardiovascular risk factors, but the mechanisms for these
`additional benefits are not yet clear. Reductions in systolic
`
`DIABETES, OBESITY AND METABOLISM
`
`blood pressure (SBP) that range from 2 to 7 mmHg over 26
`weeks [15,54–58] have been shown to precede any significant
`weight loss [69,70]. Nevertheless, weight loss due to GLP-1 RAs
`may be responsible for some of the observed improvements
`in blood pressure and lipids. After 3.5 years of exenatide
`twice-daily treatment in an open-label study, the quarter of
`patients who experienced the largest mean weight loss (12.8 kg)
`also had the greatest mean changes in SBP (−8.1 mmHg),
`high-density lipoprotein cholesterol (HDL-C) (+10.6 mg/dl)
`and triglycerides (−104.2 mg/dl) [71], despite a minimal
`correlation in the overall results between weight loss and lipid
`changes [71]. Further research is necessary to determine the
`actual mechanism for the improvements in blood pressure and
`the modest but significant reductions in triglycerides, free fatty
`acids and low-density lipoprotein cholesterol (LDL-C) levels
`that result with GLP-1 RA treatments [54,72]. Furthermore, a
`significant increase in HDL-C has also been observed in some
`studies [71,73]. In addition, liraglutide treatment has been
`associated with significant decreases in levels of plasminogen
`activator inhibitor-1 (PAI-1) and B-type natriuretic peptide
`(BNP), both of which are considered as biomarkers for
`cardiovascular risk [74]. The mechanisms for these effects also
`remain to be shown.
`Although blood pressure reductions and improved lipid
`profiles similar to those experienced with GLP-1 RAs have
`not been seen in clinical trials of DPP-4 inhibitors, a modest
`reduction in blood pressure has been reported for sitagliptin
`and vildagliptin in some studies [75–77]. A retrospective study
`of a large cohort database that found an association between
`sitagliptin treatment, slight weight loss and a small decrease
`in blood pressure suggests that the improvement in blood
`pressure is connected to weight loss [75], but more study
`of the underlying mechanism is warranted because DPP-4
`inhibitors are generally considered weight neutral [60,61]. A
`few studies have also recorded modest beneficial effects on lipid
`profiles with sitagliptin and vildagliptin [75,78,79]. One clinical
`trial concluded that vildagliptin may counteract postprandial
`hyperlipidaemia by either decreasing chylomicron production,
`increasing chylomicron clearance or both [79]. However, more
`study is needed to confirm whether DPP-4 inhibitors have any
`clinically significant effect on lipid levels and, if so, what the
`possible mechanisms would be.
`
`Metabolism and Tolerability
`Sitagliptin and saxagliptin are eliminated from the body primar-
`ily through renal excretion; in Europe, sitagliptin is therefore
`not recommended for patients with moderate and severe renal
`insufficiency, while saxagliptin can be used in these populations
`with dose reductions. In the USA, lower dosages should be used
`with both treatments in these populations [31,32,34,35]. Exe-
`natide is also eliminated by the kidneys, and both exenatide
`twice daily and once weekly are contraindicated in patients
`with severe renal impairment or end-stage renal disease (CrCl
`<30 ml/min) [21,22,37]. Caution should also be applied when
`initiating or escalating doses of exenatide BID in patients
`with moderate renal impairment (CrCl 30–50 ml/min) and
`in patients with renal transplantation [21,37]. Exenatide once
`
`22 Blonde and Montanya
`
`Volume 14 No. (Suppl. 2) April 2012
`
`
`
`MPI EXHIBIT 1058 PAGE 3
`
`MPI EXHIBIT 1058 PAGE 3
`
`
`
`DIABETES, OBESITY AND METABOLISM
`
`review article
`
`NS
`
`diarrhoea
`
`Nausea,URTI,
`
`vomiting
`
`Nausea,diarrhoea,
`
`diarrhoea
`
`Nausea,URTI,
`
`hypoglycaemia
`hyperhidrosis,
`
`Tremor,
`
`headache,dizziness
`
`Pio
`
`Nasopharyngitis,
`
`nausea/diarrhoea
`headache,
`
`5.3¶
`
`5.3¶
`
`4.5¶
`
`assistance)
`requiring
`incidents
`hypoglycaemic
`are10severe
`16.2†(included
`
`1.7§
`
`NS
`
`Nasopharyngitis,
`
`5%minor§
`
`open-labelmet
`medicationplus
`blindedstudy
`inadditionto
`upwardto45mg)
`couldbetitrated
`pio15mg(which
`receivedopen-label
`extensionand
`enrolledintrial
`
`Patientsimmediately
`
`headache/influenza
`nasopharyngitis,
`
`Diarrhoea,
`
`headache,URTI
`
`Nasopharyngitis,
`
`headache
`diarrhoea,
`
`Nasopharyngitis,
`
`headache
`nasopharyngitis,
`
`Diarrhoea,
`increased
`UTI/headache/BG
`nasopharyngitis,
`
`Hyperglycaemia,
`
`influenza
`nasopharyngitis,
`
`SU
`
`Hyperglycaemia,
`
`allowed
`medicationif
`Rescue
`
`therapyreported
`commonAEsper
`Threemost
`
`5.0†
`
`3.9†
`
`5.2†
`
`7.8†
`
`∗
`
`2.3
`
`(%)
`hypoglycaemia
`moderate
`mildor
`Frequencyof
`
`∗
`
`0.4
`
`−0.3
`−2.8
`−1.6
`
`+1.6
`−0.2
`
`−1.0
`
`effect
`placebo
`similarto
`Allchanges
`
`effect
`placebo
`similarto
`Allchanges
`
`(kg)
`baseline
`from
` weight
`
`+0.1
`−0.8
`−0.4
`
`−0.5
`−0.4
`
`−0.9‡
`
`+0.1
`−0.6
`
`−0.7
`
`−0.6
`
`+0.2
`
`34
`
`34
`
`31.7
`
`31.8
`
`32.6
`
`31.6
`
`31.1
`
`31.2
`
`31.7
`
`30.1
`
`−0.5
`(%)
`frombaseline
` HbA1c
`
`29.9
`
`(kg/m2)
`BMI
`Baseline
`
`34
`
`8.2
`
`8.2
`
`8.3
`
`7.3
`
`7.3
`
`8.5
`
`8.1
`
`8.0
`
`8.1
`
`8.1
`
`8.0
`
`8.1
`
`(%)
`HbA1c
`Baseline
`
`6.6
`
`4.9
`
`6.2
`
`5.8
`
`5.7
`
`6.3
`
`6.7
`
`6.3
`
`6.4
`
`6.7
`
`(53%)
`n=93
`>5years
`
`(56%)
`n=285
`>5years
`
`(years)
`ofdiabetes
`Duration
`
`PlaceboBID113
`
`113
`
`110
`
`10μgBID
`
`5μgBID
`
`1393
`
`≤6mg/day
`Glimepiride
`
`1396
`
`50mgBID
`
`219
`
`100mgOD
`
`179
`
`PlaceboOD
`
`181
`
`10mgOD
`
`191
`
`5mgOD
`
`≥1500
`
`≥1500
`
`≥1500
`
`[49]
`
`30
`
`Exenatide
`
`[48]
`
`52
`
`Vildagliptin
`
`26
`
`192
`
`2.5mgOD
`
`≥1500
`
`[47]
`
`24
`
`Saxagliptin
`
`177
`
`PlaceboOD
`
`523
`
`5mgOD
`
`(n)
`Participants
`
`addition
`Therapy
`
`≥1500
`(mg/day)
`dosage
`Met
`
`[46]
`
`24
`
`Linagliptin
`
`(weeks)
`duration
`Trial
`
`comefromseparatetrialswithdifferentpatientpopulationcharacteristicsanduniquestudydesigns,anddonotsupportdirectcomparison.
`Table1.Studyresultsofincretin-relatedtherapiesamongpatientsinadequatelycontrolledwithmetformininrandomisedtrials.ExceptforthePratleystudycomparingsitagliptinandliraglutide,resultsshown
`
`Volume 14 No. (Suppl. 2) April 2012
`
`doi:10.1111/j.1463-1326.2012.01575.x 23
`
`
`
`MPI EXHIBIT 1058 PAGE 4
`
`MPI EXHIBIT 1058 PAGE 4
`
`
`
`review article
`
`DIABETES, OBESITY AND METABOLISM
`
`headache/diarrhoea
`nasopharyngitis,
`
`Nausea,
`
`headache
`nasopharyngitis,
`
`NS
`
`Nausea,
`
`oedema,diarrhoea
`
`URTI,peripheral
`
`headache
`
`Nausea,diarrhoea,
`
`vomiting
`
`NS
`
`Nausea,diarrhoea,
`
`allowed
`medicationif
`Rescue
`
`therapyreported
`commonAEsper
`Threemost
`
`5%minor§
`occurred)
`assistance
`eventrequiring
`hypoglycaemic
`major
`additionone
`5%minor§(in
`
`1
`
`3
`
`1
`
`(%)
`hypoglycaemia
`moderate
`mildor
`Frequencyof
`
`−2.9
`2.8
`
`−0.8
`
`−1.2
`−1.2
`−0.9
`
`−2.3
`(kg)
`baseline
`from
` weight
`
`−1.5
`(%)
`baseline
`from
` HbA1c
`
`¶Symptomsreportedconsistentwithpossiblyconfirmedplasmaglucose<3.3mmol/l.
`§Plasmaglucose<3.1mmol/l.
`‡TherelativehighefficacyofsitagliptinincomparisontotheefficacyoftheotherDDP-4inhibitorsoccurredinpatientswhowerenottreatmentna¨ıve.
`†Mildormoderatereportedhypoglycaemiadidnotrequiretreatmentormedicalintervention.
`Plasmaglucose≤3.9mmol/l.
`respiratorytractinfection.
`AE,adverseevent;BG,bloodglucose;BID,twicedaily;BMI,bodymassindex;met,metformin;NS,notspecified;OD,oncedaily,pio,pioglitazone;SU,sulphonylurea;UTI,urinarytractinfection;URTI,upper
`
`∗
`
`−3.4
`
`−1.5
`
`33.1
`
`8.4
`
`6.4
`
`221
`
`1.8mgOD
`
`32.6
`
`32
`
`32
`
`32
`
`8.4
`
`8.5
`
`8.5
`
`8.6
`
`6
`
`5
`
`6
`
`165
`
`166
`
`160
`
`(kg/m2)
`BMI
`Baseline
`
`(%)
`HbA1c
`Baseline
`
`(years)
`diabetes
`Durationof
`
`(n)
`Participants
`
`45mgpio
`
`sitagliptin
`
`100mg
`
`weekly
`2mgonce
`
`addition
`Therapy
`
`study
`throughout
`
`weekly[50]
`once
`
`Stabledoses
`
`26
`
`Exenatide
`
`(mg/day)
`dosage
`Met
`
`(weeks)
`duration
`Trial
`
`Table1.Continued.
`
`6.0
`
`225
`
`1.2mgOD
`
`≥1500
`
`[26]
`
`26
`
`Liraglutide
`
`24 Blonde and Montanya
`
`Volume 14 No. (Suppl. 2) April 2012
`
`
`
`MPI EXHIBIT 1058 PAGE 5
`
`MPI EXHIBIT 1058 PAGE 5
`
`
`
`DIABETES, OBESITY AND METABOLISM
`
`weekly is not recommended in patients with moderate renal
`impairment [22]. Vildagliptin, which is eliminated primarily
`by the kidneys, is not recommended in patients with moderate,
`severe or end-stage renal disease [30], nor should it be used in
`patients with hepatic impairment, including those who have
`alanine aminotransferase (ALT) or aspartate aminotransferase
`(AST) levels three times the normal upper limit. Liver function
`tests for ALT and AST should be performed before treatment,
`and periodically thereafter [30].
`In contrast to the other incretin-related therapies, linagliptin
`is mainly eliminated unchanged via faeces, so no dose
`reduction is recommended in the case of renal or hepatic
`impairment [33,80]. Liraglutide degrades in the body by
`enzymatic activity like GLP-1, only much more slowly [81],
`and should therefore pose a lower risk of accumulation in the
`case of renal disease.
`In a study of 30 patients who were given a single dose of
`liraglutide 0.75 mg, no clear change in pharmacokinetics was
`evident among the 24 patients with varying degrees of renal
`dysfunction compared with the 6 patients with normal renal
`function [82]. In a meta-analysis of 2783 patients, of whom
`486 receiving liraglutide treatment had mild renal impairment,
`similar decreases in HbA1c occurred in patients with mild
`renal impairment (60 ml/min ≤ CrCl ≤ 89 ml/min), compared
`with patients with normal renal function (CrCl > 89 ml/min)
`who received either liraglutide 1.2 or 1.8 mg as monotherapy
`or in combination with an oral anti-hyperglycaemic for 26
`weeks [83]. Changes in serum creatinine were not significantly
`different
`from baseline for either treatment groups [83];
`however, there is limited experience with liraglutide in patients
`with moderate renal impairment and no experience in patients
`with severe renal impairment. Consequently, liraglutide is not
`recommended for these two patient groups in Europe [20].
`In the USA, there have been post-marketing reports of acute
`renal failure and worsening of chronic renal failure, usually in
`association with nausea, vomiting, diarrhoea or dehydration,
`which may sometimes require haemodialysis, so caution is
`warranted when initiating or escalating doses in patients with
`renal impairment [36].
`Very few drug interactions have been noted with incretin
`therapies [84]. When used with an SU, a reduction in the dosage
`of the SU should be considered to reduce the risk of hypogly-
`caemia with most incretin-related therapies [20–22,31–33].
`In the case of vildagliptin, dosage should be reduced when
`combined with a SU [30]. When saxagliptin is used concur-
`rently with a strong cytochrome P450 3A4/5 (CYP3A4/5)
`inhibitor, for example, ketoconazole, plasma concentrations of
`saxagliptin will increase, so a dosage reduction of saxagliptin
`is recommended in the USA, although none is advised in
`Europe [31,35]. Linagliptin’s efficacy may be reduced by a
`strong P-gp or CYP3A4 inducer, such as rifampicin, so
`an alternative treatment is advised [33]. Unlike some other
`pharmacologic treatments for diabetes, particularly SUs and
`insulin, hypoglycaemia is less common with incretin-related
`therapies [40,85]. Use of DPP-4 inhibitors may be associated
`with an increased risk of infections, such as urinary tract
`infections, nasopharyngitis, upper respiratory tract infections
`and headaches [33–35,43]. With exenatide and liraglutide a
`
`review article
`
`few cases of angiooedema have been reported [20,21]. More
`common with GLP-1 RAs are gastrointestinal adverse events
`(primarily nausea), which are usually transient and occur
`mostly during the first 4 weeks of treatment [54,56]. A phase
`2 study that used a validated scale system, the Gastrointesti-
`nal System Rating Scale (GSRS), to evaluate the quantity and
`intensity of gastrointestinal adverse events from the patients’
`perspective confirmed that such disorders are usually tran-
`sient and of mild-to-moderate severity [86]. Titrating the
`dosage of GLP-1 RAs according to instructions can help
`to reduce the incidence of gastrointestinal effects. Pancre-
`atitis has been reported with both DPP-4 inhibitors and
`GLP-1 RAs [20–22,32,33]; however, patients with T2D have
`a nearly three-times greater risk of developing pancreatitis
`than individuals without diabetes [87], and a causal rela-
`tionship between use of incretin-based therapies and pan-
`creatitis has not been established. The evidence regarding the
`risk for pancreatitis with incretin-related therapies compared
`with non-incretin-related therapies is, at present, conflict-
`ing [88–91].
`In rodent studies, thyroid C-cell tumours have resulted from
`exposure to high doses of exenatide and liraglutide [20–22,92],
`but the relevance of these findings for humans is not yet
`known [84]. The ability of GLP-1 RAs to stimulate calcitonin
`release appears to be species-dependent. In studies with non-
`human primates exposed for up to 87 weeks with doses 60-fold
`greater than recommended for humans and in clinical trials
`with up to 2 years’ exposure,
`increased calcitonin release
`did not occur with liraglutide treatment [92]. In a meta-
`analysis of clinical trials lasting no more than 2 years with
`over 5000 patients receiving either liraglutide or control
`therapy, 3-month measurements of serum calcitonin showed
`that mean serum calcitonin concentrations were at the low
`end of the normal range in all treatment groups at baseline
`and remained low throughout the trials [93]. The European
`label for liraglutide cautions patients against possible thyroid
`adverse events [20], while the US prescribing instructions state
`that liraglutide is contraindicated in patients who have a
`personal or family history of medullary thyroid carcinoma
`or of Multiple Endocrine Neoplasia syndrome type 2 (MEN
`2) [36]. Monitoring with serum calcitonin tests or thyroid
`ultrasounds is not advised by the American Association of
`Clinical Endocrinologists (AACE) because it may increase
`unnecessary procedures, because of low test specificity for
`serum calcitonin and a high background incidence of thyroid
`disease. AACE also does not recommend obtaining baseline
`serum calcitonin levels or thyroid ultrasounds unless thyroid
`nodules are detected on initial physical examination, or
`if nodules occur during administration of liraglutide [94].
`Thyroid nodules in liraglutide-treated patients should be
`managed in the usual manner [94].
`
`Clinical Trials Comparing DPP-4 Inhibitors
`and GLP-1 RAs
`So far, few trials have been conducted directly comparing the
`two classes of incretin-related therapies. The available data are
`included here.
`
`Volume 14 No. (Suppl. 2) April 2012
`
`doi:10.1111/j.1463-1326.2012.01575.x 25
`
`
`
`MPI EXHIBIT 1058 PAGE 6
`
`MPI EXHIBIT 1058 PAGE 6
`
`
`
`review article
`
`Exenatide taken BID was compared with sitagliptin in a
`4-week, randomised, crossover study in 61 patients with T2D
`with an average HbA1c of 8.5 ± 1.2% and an average 2-h
`PPG of 245 ± 65 mg/dl [45]. Patients received exenatide (5 μg
`BID for 1 week, then 10 μg BID for 1 week) or sitagliptin
`[100 mg once every morning (QAM)] for 2 weeks. After
`2 weeks, patients crossed over to the alternate therapy. At
`the end of the study, the change in fasting plasma glucose
`(FPG) was similar with the two therapies (−15 ± 4 mg/dl
`for exenatide vs. −19 ± 4 mg/dl for sitagliptin; p = 0.3234),
`but
`the 2-h PPG levels were significantly lower with
`exenatide (−133 ± 6 mg/dl vs. −208 ± 6 mg/dl, respectively;
`p < 0.0001). When patients switched from exenatide to
`sitagliptin, 2-h PPG rose by 73 ± 11 mg/dl, and when
`patients switched from sitagliptin to exenatide, 2-h PPG
`decreased by 76 ± 10 mg/dl. Compared to sitagliptin, exenatide
`showed greater improvement in the insulinogenic index of
`insulin secretion (ratio 1.50 ± 0.26; p = 0.0239), produced
`a greater reduction in postprandial glucagon [AUC0 – 240 min
`(pg min/ml) geometric LS mean ratio 0.88 ± 0.03; p = 0.0011]
`and total caloric intake (−134 ± 97 kcal vs. +130 ± 97 kcal;
`p = 0.0227), and slowed gastric emptying to a greater extent
`[acetaminophen AUC0 – 240 min (mg min/dl) LS mean ratio
`0.56 ± 0.05; p < 0.0001]. With both treatments, mild-to-
`moderate gastrointestinal symptoms were the most common
`adverse events reported.
`In a double-blind comparison of exenatide once weekly with
`sitagliptin or the TZD treatment pioglitazone, patients were
`randomised to 2 mg exenatide as a once-weekly injection plus
`oral placebo once daily (n = 160), 100 mg sitagliptin once
`daily plus once-weekly placebo injection (n = 166), or 45 mg
`pioglitazone once daily plus once-weekly placebo injection
`(n = 165) for 26 weeks [50]. Baseline HbA1c levels were 8.6, 8.5
`and 8.5%, respectively. At trial end, the reduction in HbA1c was
`−1.5% with exenatide once weekly, −0.9% with sitagliptin and
`−1.2% with pioglitazone with treatment differences of −0.6%
`(95% CI [−0.9; −0.4]; adjusted p < 0.0001) for exenatide
`once weekly vs. sitagliptin, and −0.3% (95% CI [−0.6; −0.1];
`adjusted p = 0.0165) vs. pioglitazone. Weight change with
`exenatide once weekly was −2.3 kg, compared to −0.8 kg with
`sitagliptin (p = 0.0002) or +2.8 kg with pioglitazone (p <
`0.0001) [50]. There was also a significant difference in reduction
`of SBP between exenatide once weekly and sitagliptin (−4
`mmHg; p = 0.0055), but none between exenatide once weekly
`and pioglitazone [50]. The most common adverse events
`patients experienced for exenatide once weekly and sitagliptin
`were nausea (24 and 10%, respectively) and diarrhoea (18 and
`10%, respectively). With pioglitazone the most common events
`were upper-respiratory tract infections (10%) and peripheral
`oedema (8%) [50].
`The other direct comparison between GLP-1 RAs and DPP-4
`inhibitors occurred in a 26-week trial in which 658 patients
`inadequately controlled with 1500 mg or more of metformin
`were randomised to receive, in addition, 1.2 mg liraglutide,
`1.8 mg liraglutide or