`Current Status and Future Prospects
`
`Scott R. Drab, Pharm.D.
`
`Incretin-based therapies encompass two new classes of antidiabetic drugs:
`glucagon-like peptide-1 (GLP-1) receptor agonists (e.g., liraglutide, exenatide,
`and exenatide long-acting release), which are structurally related to GLP-1,
`and the dipeptidyl peptidase-4 (DPP-4) inhibitors (e.g., sitagliptin and
`saxagliptin), which limit the breakdown of endogenous GLP-1. To evaluate
`the safety and effectiveness of incretin-based therapies for the treatment of
`type 2 diabetes mellitus and the role of these therapies in clinical practice, a
`MEDLINE search (January 1985–November 2009) was conducted. Relevant
`references from the publications identified were also reviewed. Of 28 studies
`identified, 22 were randomized controlled trials. Data show that these
`therapies affect insulin secretion in a glucose-dependent manner, achieving
`clinically meaningful reductions in hemoglobin A1c levels, with very low rates
`of hypoglycemia. In addition, reductions in body weight have been observed
`with GLP-1 receptor agonists, which also exert a pronounced effect on
`systolic blood pressure. Various human and animal studies show that GLP-1
`improves -cell function and increases -cell proliferation in vitro, which may
`slow disease progression. Thus, incretin-based therapies represent a
`promising addition to the available treatments for type 2 diabetes.
`type 2 diabetes, incretin-based therapies, glucagon-like peptide-
`Key Words:
`1, GLP-1, dipeptidyl peptidase-4, DPP-4.
`(Pharmacotherapy 2010;30(6):609–624)
`
`OUTLINE
`Treatment Potential for Type 2 Diabetes Mellitus
`Goals of Therapy and Limitations of Current
`Treatments
`Incretin-Based Therapies
`Physiologic Effects of Glucagon-Like Peptide-1
`Two Alternative Approaches
`Pharmacology
`Efficacy and Safety: Clinical Studies
`Benefits
`Dosing and Administration
`Place in Therapy
`Conclusion
`
`From the University of Pittsburgh School of Pharmacy,
`Pittsburgh, Pennsylvania, and the University Diabetes Care
`Associates, Jeannette, Pennsylvania.
`For reprints, visit http://www.atypon-link.com/PPI/loi/phco.
`For questions or comments, contact Scott R. Drab,
`Pharm.D., University of Pittsburgh School of Pharmacy, 922
`Salk Hall, Pittsburgh, PA 15261; e-mail: drab@pitt.edu.
`
`The increasing prevalence of diabetes mellitus
`is an acknowledged world health crisis that is
`both a major contributor to patient morbidity
`and mortality and a huge economic burden.1 The
`complex pathophysiology of type 2 diabetes
`makes effective treatment problematic. Hyper-
`glycemia is associated with an increased risk of
`microvascular complications, sensory neuropathy,
`myocardial infarction, stroke, macrovascular
`mortality, and all-cause mortality.2 Type 2 diabetes
`is also linked causally with obesity—in itself a
`global health crisis—which independently increases
`the risk of serious cardiovascular comorbidities.1
`Hypertension, also often associated with type 2
`diabetes, can further increase cardiovascular risk.
`The global economic impact of diabetes, due to
`premature death and complications from the
`disease, is considerable. At least $232 billion were
`spent on treatment and prevention of diabetes
`worldwide in 2007, with three quarters of that
`
`MPI EXHIBIT 1022 PAGE 1
`
`MPI EXHIBIT 1022 PAGE 1
`
`
`
`610
`
`PHARMACOTHERAPY Volume 30, Number 6, 2010
`
`amount spent in industrialized countries on the
`treatment of long-term complications and on
`general care, such as efforts to prevent micro-
`and macrovascular complications.3 The indirect
`costs of the disease must also be included in any
`discussion of its burden. In 2007, it was estimated
`that disability, lost productivity, and premature
`death due to diabetes cost the United States
`economy alone $58 billion.1
`
`Treatment Potential for Type 2 Diabetes
`Mellitus
`Insulin resistance and impaired insulin secretion
`both contribute to the development of type 2
`diabetes, with the shift from normal to impaired
`glucose tolerance and then to diabetes, accom-
`panied by a 40% decrease in insulin sensitivity
`and a 3–4-fold deterioration in insulin secretion
`due to loss of -cell capacity (50–80%).4, 5 However,
`if impaired glucose tolerance is diagnosed early
`enough, overt type 2 diabetes can be prevented or
`at least delayed, with lifestyle interventions
`aimed at reducing calorie intake and body fat6, 7
`or through drug treatment7 to normalize glycemic
`control. Once type 2 diabetes has been diagnosed,
`weight loss alone can greatly improve both
`glycemic control and patients’ cardiovascular risk
`profiles.8 The effects are interrelated: glycemic
`control, coupled with effective management of
`additional risk factors, benefits cardiovascular
`prognosis.9
`Despite treatment guidelines that recommend
`early, aggressive intervention,10, 11 many patients
`fail to reach targets for glycemic control (~43% of
`patients in the United States have hemoglobin
`A1c [A1C] levels ≥ 7%),12 with minority groups
`having greater disease prevalence.13, 14
`Many factors contribute to the failure to manage
`type 2 diabetes successfully. The most fundamental
`of these are the necessary lifestyle modifications,
`which, if not implemented at the time of diagnosis,
`may result in further decline in glycemic control
`and hasten the need for pharmacotherapy. Other
`factors include psychosocial and economic
`influences and shortcomings in the efficacy and
`tolerability profiles of many available antidiabetic
`drugs.
`
`Goals of Therapy and Limitations of Current
`Treatments
`To successfully improve patient outcomes and
`minimize the risk of complications and their
`cost, treatment should ideally address deterio-
`rating -cell function, A1C, and fasting plasma
`
`glucose (FPG) and postprandial plasma glucose
`(PPG) levels simultaneously, without increasing
`risk of hypoglycemia, weight gain, or cardio-
`vascular disease contributors. Adequate glycemic
`control is essential, as reduction of A1C levels
`can decrease the risks of microvascular and
`macrovascular complications and mortality
`associated with type 2 diabetes.2 For every 1%
`reduction in A1C, a 37% decrease in risk for
`microvascular complications (p<0.0001) and a
`21% decrease in risk of death related to diabetes
`(p<0.0001) have been observed. 2 An ideal
`treatment would reduce cardiovascular risk
`factors as well as control blood glucose levels.
`Current guidelines for type 2 diabetes from the
`American Diabetes Association (ADA) and the
`European Association for the Study of Diabetes
`(EASD) advocate a stepwise escalation of
`intervention, starting with lifestyle modification
`and metformin 15; however, sulfonylureas, a
`common second-line intervention in type 2
`diabetes treatment regimens, are limited by a
`failure of -cell function that underlies the
`progression of the disease. Although -cell
`function cannot be measured directly in humans,
`several indicators can be used to monitor its
`decline. These include the homeostatic model
`assessment of -cell function (HOMA-B), which
`yields an estimate from fasting plasma insulin and
`glucose concentrations,16 and the proinsulin:insulin
`ratio. The HOMA-B results have shown that
`established oral antidiabetic drug therapies such
`as metformin, sulfonylureas, and thiazolidine-
`diones can improve -cell function during the
`first year of treatment when they are given as
`initial treatment in patients with newly diagnosed
`type 2 diabetes; however, -cell function declines
`progressively thereafter.4, 17, 18 The A1C level
`parallels these changes in -cell function,
`decreasing in the first year of therapy and then
`increasing progressively, highlighting the need for
`therapies that can sustain improvements in -cell
`function.19
`Insulin therapy is necessary once -cell
`secretory capacity becomes insufficient; however,
`alternative antihyperglycemic drugs—the
`incretin-based therapies—are now available that
`may offer advantages over conventional oral
`antidiabetic drug therapies. These incretin-based
`agents were included in the 2009 recommen-
`dations of an American Association of Clinical
`Endocrinologists–American
`College
`of
`Endocrinology consensus panel,20 which sug-
`gested that patients with A1C levels of 7.6–9.0%
`be treated with dual therapy with metformin
`
`MPI EXHIBIT 1022 PAGE 2
`
`MPI EXHIBIT 1022 PAGE 2
`
`
`
`INCRETIN-BASED THERAPIES FOR TYPE 2 DIABETES MELLITUS Drab
`
`611
`
`(unless contraindicated) plus, in order of
`preference, glucagon-like peptide-1 (GLP-1)
`receptor agonists, dipeptidyl peptidase-4 (DPP-4)
`inhibitors, glinides, or sulfonylureas. The panel
`emphasized the lower risk of hypoglycemia with
`GLP-1 receptor agonists and DPP-4 inhibitors
`compared with glinides and sulfonylureas, but
`preferred GLP-1 receptor agonists over DPP-4
`inhibitors because of their greater potential for
`lower PPG level and substantial weight loss. For
`patients requiring triple therapy, metformin, a
`GLP-1 receptor agonist, and a third oral antidia-
`betic drug—a thiazolidinedione, a sulfonylurea,
`or a glinide—are recommended.
`
`Incretin-Based Therapies
`
`Physiologic Effects of Glucagon-Like Peptide-1
`The “incretin effect” describes the phenomenon
`whereby a glucose load delivered orally produces
`a much greater insulin secretion than the same
`glucose load administered intravenously.21 This
`effect is mediated by two incretin hormones
`secreted by intestinal cells.22 Glucose-dependent
`insulinotropic polypeptide (GIP) was identified
`first,22 followed by GLP-1.23
`It is now thought
`that healthy individuals may derive up to 70% of
`their prandial insulin secretory response from the
`incretin effect.24 Under normal conditions, the
`incretin peptides are secreted as needed, in
`response to ingested nutrients, and have a short
`plasma half-life due to degradation by the
`ubiquitous DPP-4 enzyme.25
`In people with type 2 diabetes, pancreatic
`responsiveness to GLP-1 is impaired,26 but the
`insulin-secretory response can be restored with
`pharmacologic doses of human GLP-1.27 Further
`studies in vitro and in animals show that GLP-1
`
`promotes -cell neogenesis and preservation28–30
`and inhibits -cell apoptosis.30, 31 Also, GLP-1
`has potentially beneficial effects on cardiac
`function:
`it has been shown to be cardio-
`protective in ischemia-reperfusion models in rats
`in vivo,32 and to reduce blood pressure in Dahl
`salt-sensitive rats.33 In humans, GLP-1 improves
`left ventricular function as well as endothelial
`function in patients with high cardiac risk.34, 35
`Along with these benefits, GLP-1 has been shown
`to slow the rate of gastric emptying in humans36
`and to reduce appetite. 37 These effects are
`particularly important, since type 2 diabetes is
`often associated with obesity and excessive
`caloric intake.
`It is clear that drugs acting
`through the physiologic GLP-1 pathways might
`have important ancillary benefits for the
`treatment of type 2 diabetes.
`A key feature of GLP-1 action is the glucose-
`dependent stimulation of insulin secretion and
`concomitant suppression of glucagon. When
`patients with type 2 diabetes are infused with
`human GLP-1, insulin levels return toward basal
`levels when normal FPG concentrations are
`reached, despite ongoing infusion of GLP-1.38
`Continuous subcutaneous GLP-1 infusion for 6
`weeks also reduced A1C by 1.3% and FPG by 77
`mg/dl, slowed gastric emptying, and reduced
`appetite and body weight by 1.9 kg.39 Therefore,
`GLP-1 and therapies that activate GLP-1
`receptors only stimulate increases in insulin
`secretion when glucose levels are elevated and, as
`a result, carry a very low risk of hypoglycemia
`(Figure 1) unless combined with a sulfonylurea
`or insulin.
`Unfortunately, therapeutic use of human GLP-
`1 is not practical due to its rapid degradation by
`DPP-4 enzyme,25 with an elimination half-life of
`
`Figure 1. Effects of glucagon-like peptide-1 (GLP-1) or placebo infusion on insulin, glucagon, and glucose levels. *p<0.05.
`(From reference 38 with permission.)
`
`MPI EXHIBIT 1022 PAGE 3
`
`MPI EXHIBIT 1022 PAGE 3
`
`
`
`612
`
`PHARMACOTHERAPY Volume 30, Number 6, 2010
`
`about 2 minutes.27 As a result, two alternative
`approaches to harnessing the therapeutic poten-
`tial of the incretin system have been pursued—
`GLP-1 receptor agonists and DPP-4 inhibitors.
`
`Two Alternative Approaches
`
`The GLP-1 receptor agonists are designed to
`fulfill the role of native GLP-1 while exhibiting
`resistance to degradation by DPP-4, and are
`present for therapeutic use at supraphysiologic
`levels (equivalent to 6–10-fold normal GLP-1
`levels). These drugs include exenatide (available
`since 2005), exenatide long-acting release (LAR),
`which is undergoing regulatory review by the
`U.S. Food and Drug Administration (FDA); and
`liraglutide, which is available in the United
`States, Europe, and Japan. Other drugs (AVE-
`0010 bid/ZP10, albiglutide, and taspoglutide/
`R1583) have recently entered phase III trials;
`their data will not be reviewed.
`The DPP-4 inhibitors act to impair the activity
`of the DPP-4 enzyme, thereby increasing the half-
`life of endogenous GLP-1. Compared with the
`GLP-1 receptor agonists, DPP-4 inhibitors
`produce modestly elevated levels of GLP-1 and
`exert their effects principally through interaction
`with receptors on afferent nerves rather than
`through receptors on target organs.40 The first of
`these drugs to enter the market was sitagliptin,
`which was approved in the United States in 2006.
`A second DPP-4 inhibitor, vildagliptin, has been
`approved for use in Europe; however, since its
`manufacturer is not pursuing approval in the
`United States, vildagliptin will not be discussed
`in detail. Saxagliptin has received recent European
`Medicines Agency and FDA approval, but U.S.
`approval of alogliptin has been delayed because
`of the requirement for an additional cardiovascular
`study.
`In December 2008, the FDA issued the
`recommendation that all therapies for the
`treatment of type 2 diabetes should provide
`evidence that they do not increase the risk of
`cardiovascular events. Hence, alogliptin will not
`be reviewed.
`To identify relevant literature on the incretin-
`based therapies, a MEDLINE search (January
`1985–November 2009) was performed by using
`the following key words:
`type 2 diabetes,
`incretin, GLP-1, DPP-4, liraglutide, exenatide,
`sitagliptin, and saxagliptin. Relevant references
`from the publications identified during this
`search were also reviewed. All studies that were
`published in the English language and that
`evaluated the safety and efficacy of incretin-based
`
`therapies were analyzed. Priority was given to
`randomized clinical trials and retrospective
`studies in clinical settings.
`
`Pharmacology
`
`Glucagon-Like Peptide-1 Receptor Agonists
`Exenatide. This drug is a recombinant peptide
`based on exendin-4, which is found in the saliva
`of Heloderma suspectum (i.e., Gila monster
`lizard). Exenatide shares 53% of its amino acid
`sequence identity with human GLP-1 and
`functions as a full agonist of the GLP-1
`receptor.41
`It is, however, more resistant to
`degradation by DPP-4 enzyme, due to the
`presence of a glycine residue on the penultimate
`NH2 group.42
`Studies investigating the pharmacokinetic
`profile of exenatide show that it reaches peak
`plasma levels at about 2 hours after subcutaneous
`injection, has a plasma half-life of 3–4 hours, and
`provides glucose reduction over 5–7 hours.43, 44
`This means that exenatide requires twice-daily
`dosing (0–60 min before breakfast and dinner)
`and exerts its predominant effect on PPG levels.44
`A new drug application for a long-acting (slow-
`absorption) formulation requiring once-weekly
`dosing (exenatide LAR) was submitted in May
`2009.
`In March 2010, a response letter was
`issued by the FDA requesting the finalization of
`the product labeling, risk evaluation and
`mitigation strategy, and clarification of existing
`manufacturing processes.
`Preliminary data
`indicate that exenatide LAR may have better
`efficacy than twice-daily exenatide.45 Exenatide
`is renally excreted,46 and although no dosage
`adjustment is needed in cases of mild or
`moderate renal impairment, exenatide is not
`recommended for use in patients with severe
`renal impairment (creatinine clearance < 30
`ml/min) or end-stage renal disease ([ESRD] i.e.,
`chronic, irreversible renal failure). This advice
`follows the finding that single doses of exenatide
`5 µg caused gastrointestinal adverse effects in
`patients with ESRD who were receiving dialysis.47
`Gastrointestinal adverse events in these patients
`may worsen renal function.
`In November 2009,
`the FDA issued guidance advocating caution
`when starting or uptitrating exenatide in patients
`with moderate renal impairment (creatinine
`clearance 30–50 ml/min).48
`Elimination of exenatide is not dependent on
`hepatic function, however, and it is therefore not
`expected to interact with hepatically metabolized
`drugs as a result of competition for liver
`
`MPI EXHIBIT 1022 PAGE 4
`
`MPI EXHIBIT 1022 PAGE 4
`
`
`
`INCRETIN-BASED THERAPIES FOR TYPE 2 DIABETES MELLITUS Drab
`
`613
`
`enzymes. One way in which exenatide and other
`GLP-1 receptor agonists might interact with
`other drugs is through their effects on gastric
`motility and, hence, absorption kinetics. Studies
`examining the possibility of such an effect on
`digoxin, warfarin, and a statin do not suggest that
`their kinetics will be altered to a clinically
`meaningful extent by concomitant exenatide.49–51
`
`Liraglutide. This GLP-1 receptor agonist is
`rather different from exenatide in its structure
`(Figure 2). It is an analog of human GLP-1 with
`97% primary amino acid sequence identity, but
`with a fatty acid side chain attached through a
`glutamic acid linker.52 This fatty acid structure
`enables reversible plasma albumin binding,
`which may contribute to the drug’s partial resis-
`tance to DPP-4 degradation and its subsequent
`prolonged action.53 The primary mechanism for
`prolonging the action of liraglutide, however, is
`thought to be through slowed absorption due to
`the self-association (as heptameric aggregates) of
`liraglutide molecules within the injection depot.53
`These heptamers are thought to be too large to
`pass through capillary membranes readily, so
`
`Figure 2. The structures of native glucagon-like peptide-1
`(GLP-1), liraglutide, and exenatide.
`(From reference 52
`with permission.)
`
`absorption is delayed pending the dissociation of
`liraglutide through dilution at the absorption
`site. Although liraglutide is partially resistant to
`DPP-4 degradation, elimination studies show that
`it is nevertheless degraded almost completely
`into small molecules by this enzyme and by
`endopeptidases, as with endogenous GLP-1.54, 55
`This means that its elimination is not dependent
`on hepatic or renal function, which may prove to
`be clinically advantageous.
`Studies that investigated the pharmacokinetic
`profile of liraglutide have shown that it reaches
`peak plasma levels 9–12 hours after dosing by
`subcutaneous injection, with a subsequent mean
`elimination half-life of 11–15 hours.56 Steady
`state is reached after 3 days, and therapeutic
`plasma liraglutide concentrations are observed
`for up to 24 hours after a single injection. These
`characteristics make liraglutide compatible with
`once-daily administration, and it may be possible
`to choose the dosing time without regard to
`meals.56, 57 As with exenatide, drug-drug inter-
`action studies have been performed to investigate
`liraglutide’s potential to affect the kinetics of
`drugs that are sensitive to gastric transit time due
`to their absorption properties. Again, it was
`concluded that the effect of liraglutide on
`atorvastatin, lisinopril, digoxin, and griseofulvin
`was unlikely to be clinically relevant.58
`
`Dipeptidyl Peptidase-4 Inhibitors
`Sitagliptin. This drug is an orally absorbed small
`molecule that produces at least 80% inhibition of
`DPP-4 activity for 12 hours at doses of 50 mg or
`more, and for 24 hours at doses of 100 mg or
`more.59 Therefore, the recommended dosage is
`100 mg once/day. Compared with placebo,
`sitagliptin results in at least a 2-fold increase in
`postprandial GLP-1 levels, both after single doses
`and at steady state, reducing postprandial glycemia
`and not causing hypoglycemia.59, 60 Steady state
`is reached after about 2 days.
`Inhibition of
`glucagon secretion is also seen with DPP-4
`inhibition, but this class of drugs does not seem
`to affect gastric motility to the extent of the GLP-
`1 receptor agonists, possibly because of the
`relatively modest increase in GLP-1 signal that
`DPP-4 inhibition alone induces.61 This might
`also account for the observation that weight
`tends to remain unchanged rather than reduced
`in response to DPP-4 inhibition. It is also worth
`noting that the DPP-4 enzyme degrades other
`peptides besides the incretins,
`including
`cytokines and chemokines,61 and inhibition of
`
`MPI EXHIBIT 1022 PAGE 5
`
`MPI EXHIBIT 1022 PAGE 5
`
`
`
`614
`
`PHARMACOTHERAPY Volume 30, Number 6, 2010
`
`this process might account for some of the
`adverse effects seen with DPP-4 inhibitors, which
`are elaborated on below.
`Sitagliptin is primarily (79%) eliminated
`unchanged by the kidney. This process involves
`active tubular excretion, so dosage adjustment is
`recommended in patients with renal impairment
`to ensure that plasma sitagliptin levels do not
`accumulate.62 Dosing should be reduced to 50
`mg once/day in patients with moderate renal
`insufficiency and to 25 mg once/day in cases of
`severe renal impairment or ESRD. Only about
`16% of sitagliptin undergoes hepatic metabolism;
`hence, its pharmacokinetics have been shown to
`be unaffected by mild-to-moderate hepatic
`failure.63 Drug interactions are not anticipated to
`be a problem with sitagliptin, and studies have
`shown that it does not alter the kinetics (to a
`clinically relevant extent) of other antidiabetic
`drugs, simvastatin, warfarin, digoxin, and oral
`contraceptives,64, 65 although a possible cytochrome
`P450 (CYP) interaction with lovastatin has been
`reported.66 The concomitant use of cyclosporine
`produced only modest changes in the pharmaco-
`kinetics of sitagliptin.65
`
`Saxagliptin. This is the second small-molecule
`once-daily DPP-4 inhibitor to become available
`in the United States. The pharmacokinetic and
`pharmacodynamic properties of saxagliptin have
`been investigated in two 2-week studies of a wide
`range of doses given once/day in both healthy
`subjects (40–400 mg)67 and patients with type 2
`diabetes (2.5–50 mg).67 The level of DPP-4
`inhibition with saxagliptin 24 hours after dosing
`was 50% of predose levels in the 2.5-mg group,
`and 79% of predose levels in the 400-mg group.
`Doses of 150 mg or greater provided the same
`level of inhibition, and no dose-related adverse
`events, including hypoglycemia, were observed
`during the study. All doses of saxagliptin
`increased postprandial levels of intact GLP-1 by
`1.5–3 times compared with placebo, but there
`was no dose-related trend in this effect.68 The
`recommended dosage is therefore 2.5 or 5 mg/day.
`Saxagliptin is extensively metabolized in the
`liver by the CYP system, and its primary
`metabolite is active and has about 50% of the
`DPP-4 inhibitory potency of
`the parent
`compound. Both the parent and the metabolite
`are renally excreted,69 and accumulation can
`occur in patients with renal
`impairment,
`necessitating a daily dose limit of 2.5 mg.
`Although some accumulation can occur with
`hepatic impairment, dosage adjustment does not
`
`appear necessary in patients with hepatic
`impairment but healthy renal function.70
`Because of its hepatic metabolism, exposure to
`saxagliptin is potentially affected by inducers or
`inhibitors of the CYP system. However, dosage
`adjustments are not considered necessary to
`compensate for increased elimination caused by
`inducers such as rifampin, nor to avoid accumu-
`lation due to moderate inhibitors such as
`amprenavir, aprepitant, erythromycin, fluconazole,
`fosamprenavir, grapefruit juice, and verapamil.
`The daily dose of saxagliptin should be limited to
`2.5 mg when used concomitantly with strong
`CYP inhibitors such as ketoconazole.70
`
`Efficacy and Safety: Clinical Studies
`The reviewed clinical studies investigating the
`efficacy and safety of incretin-based therapies are
`summarized in Table 145, 71–85 and Table 2.86–93
`
`Glucagon-Like Peptide-1 Receptor Agonists
`Exenatide. Pivotal phase III efficacy and safety
`data for exenatide were derived from three key
`Diabetes Management for Improving Glucose
`Outcomes (AMIGO) trials in patients with type 2
`diabetes who were inadequately controlled with
`either a sulfonylurea, 72 metformin, 73 or a
`sulfonylurea plus metformin75 (Table 1). In each
`of these trials, twice-daily exenatide was shown
`to reduce A1C significantly, by approximately
`0.9% from baseline over the 30-week trial period
`compared with placebo.72, 73, 75 These large, well-
`designed studies (double- or triple-blind,
`placebo-controlled, 30-wk trials with well-
`matched treatment and placebo groups) provided
`proof of the ability of exenatide to achieve
`significant A1C reductions over the time period
`studied. A pooled analysis of trial data plus two
`52-week extension studies, which were
`completed by 314 of 1446 patients in an intent-
`to-treat group from the 30-week randomized
`trials, found that the reductions in A1C were
`sustained (–1.1% over 2 yrs). 94 The same
`analysis showed that 48% of patients achieved
`A1C less than 7%, and that FPG level was
`reduced by 25.2 mg/dl.
`It should be noted that
`the 48% of patients reaching an A1C less than
`7.0% were those with baseline A1C values above
`7.0%. The subgroup with a baseline A1C of 9.0%
`or greater showed a mean ± SD reduction in A1C
`from baseline of 2.0 ± 0.2%, compared with 0.8 ±
`0.1% for the subgroup with baseline A1C less
`than 9.0%.94 Although this was an open-label
`noncontrolled trial, these results suggest that the
`
`MPI EXHIBIT 1022 PAGE 6
`
`MPI EXHIBIT 1022 PAGE 6
`
`
`
`INCRETIN-BASED THERAPIES FOR TYPE 2 DIABETES MELLITUS Drab
`
`615
`
`Table 1. Summary of Randomized Controlled Trials of the Glucagon-Like Peptide-1 Receptor Agonists
`Change in Change in Change in
`A1C
`Weight
`SBP
`(%)
`(kg)
`(mm Hg)
`
`Population
`
`Treatment Groups
`
`Study Duration
`Exenatide
`24 wks71
`
`232 drug-naïve patients,
`A1C 7.8%, weight 86 kg
`
`30 wks
`(AMIGO 2)72
`
`377 patients taking a sulfonylurea,
`A1C 8.6%, weight 96 kg
`
`30 wks
`(AMIGO 1)73
`
`336 patients taking metformin,
`A1C 8.2%, weight 100 kg
`
`26 wks76
`
`52 wks77
`
`52 wks78
`
`15 wks45
`
`2 x 16-wk periods 138 patients taking metformin
`(crossover
`or a sulfonylurea, A1C 8.95%,
`design)74
`weight 84.8 kg
`30 wks
`733 patients taking metformin +
`(AMIGO 3)75
`a sulfonylurea, A1C 8.5%,
`weight 98 kg
`551 patients taking metformin +
`a sulfonylurea, A1C ~8.25%,
`weight 88 kg
`501 patients taking metformin +
`a sulfonylurea, A1C 8.6%,
`weight ~84 kg
`69 patients taking metformin,
`A1C 7.5%, weight ~91 kg
`45 patients taking metformin
`(60%) or implementing dietary
`changes, A1C 8.5%, weight
`106 kg
`295 patients implementing dietary
`changes and exercise treated with
`metformin, a sulfonylurea,
`a thiazolidinedione, or any
`combination of 2 of these agents;
`A1C 8.3%, weight 102 kg
`
`30 wks79
`
`Liraglutide
`52 wks
`(LEAD 3)80
`
`26 wks
`(LEAD 1)81
`
`746 patients naïve to oral
`antidiabetic drugs, A1C 8.2%,
`weight 92.6 kg
`1041 patients taking a sulfonylurea,
`A1C 8.4%, weight 81.6 kg
`
`26 wks
`(LEAD 2)82
`
`1091 patients taking metformin,
`A1C 8.4%, weight 88.6 kg
`
`Exenatide 5 µg b.i.d.
`Exenatide 10 µg b.i.d.
`Placebo
`Exenatide 5 µg b.i.d.
`Exenatide 10 µg b.i.d.
`Placebo
`Exenatide 5 µg b.i.d.
`Exenatide 10 µg b.i.d.
`Placebo
`Exenatide 10 µg b.i.d.
`Titrated insulin glargine
`
`Exenatide 5 µg b.i.d.
`Exenatide 10 µg b.i.d.
`Placebo
`Exenatide 10 µg b.i.d.
`Titrated insulin glargine
`
`Exenatide 10 µg b.i.d.
`Titrated premix insulin
`analog
`Exenatide to 20 µg t.i.d.
`Titrated insulin glargine
`Exenatide LAR 0.8 mg q wk
`Exenatide LAR 2.0 mg q wk
`Placebo
`
`Exenatide LAR 2.0 mg q wk
`Exenatide 10 µg b.i.d.
`
`Liraglutide 1.2 mg q.d.
`Liraglutide 1.8 mg q.d.
`Glimepiride 8 mg q.d.
`Liraglutide 0.6 mg q.d.
`Liraglutide 1.2 mg q.d.
`Liraglutide 1.8 mg q.d.
`Placebo
`Rosiglitazone 4 mg/day
`Liraglutide 0.6 mg q.d.
`Liraglutide 1.2 mg q.d.
`Liraglutide 1.8 mg q.d.
`Glimepiride
`Liraglutide 1.2 mg q.d.
`Liraglutide 1.8 mg q.d.
`Placebo
`Liraglutide 1.8 mg q.d.
`Placebo
`Insulin glargine
`Liraglutide 1.8 mg q.d.
`Exenatide 10 µg b.i.d.
`
`–0.7
`–0.9
`–0.2
`–0.46
`–0.86
`+0.12
`–0.4
`–0.8
`+0.1
`–1.36
`–1.36
`
`–0.6
`–0.8
`+0.2
`–1.11
`–1.11
`
`–1.04
`–0.89
`
`–0.8
`–0.7
`–1.4
`–1.7
`+0.4
`
`–1.9
`–1.5
`
`–0.84
`–1.14
`–0.51
`–0.60
`–1.08
`–1.13
`+0.23
`–0.44
`–0.69
`–0.97
`–1.00
`–0.98
`–1.48
`–1.48
`–0.54
`–1.33
`–0.24
`–1.09
`–1.12
`–0.79
`
`NA
`
`–2.8
`–3.1
`–1.4
`–0.9
`–1.6
`–0.6
`–1.6
`–2.8
`–0.3
`–2.2
`(exenatide
`vs glargine)
`–1.6
`–1.6
`–0.9
`–2.3
`+1.8
`
`–2.5
`+2.9
`
`–3.6
`+1.0
`–0.0
`–3.8
`–0.0
`
`–3.7
`–3.6
`
`–2.05
`–2.45
`+1.12
`+0.72
`+0.32
`–0.23
`–0.10
`+2.11
`–1.78
`–2.58
`–2.79
`+0.95
`–1.02
`–2.02
`+0.60
`–1.81
`–0.42
`+1.62
`–3.24
`–2.87
`
`–2.1
`–3.6
`–0.7
`–0.9
`–2.6
`–2.8
`–2.3
`–0.9
`–0.6
`–2.8
`–2.3
`+0.4
`–6.7
`–5.6
`–1.1
`–4.0
`–1.4
`+0.5
`–2.51
`–2.00
`
`26 wks
`(LEAD 4)83
`
`26 wks
`(LEAD 5)84
`
`26 wks
`(LEAD 6)85
`
`533 patients taking metformin +
`a thiazolidinedione, A1C 8.5%,
`weight 97.0 kg
`581 patients taking metformin +
`a sulfonylurea, A1C 8.2%,
`weight 85.4 kg
`464 patients taking metformin,
`a sulfonylurea, or both;
`A1C 8.2%, weight 93 kg
`A1C = hemoglobin A1c; SBP = systolic blood pressure; NA = not applicable; LAR = long-acting release.
`
`MPI EXHIBIT 1022 PAGE 7
`
`MPI EXHIBIT 1022 PAGE 7
`
`
`
`616
`
`PHARMACOTHERAPY Volume 30, Number 6, 2010
`
`Table 2. Summary of Randomized Controlled Trials of the Dipeptidyl Peptidase-4 Inhibitors
`
`Study Duration
`Sitagliptin
`18 wks86
`
`521 patients naïve to oral antidiabetic
`drugs, A1C 8.1%
`
`Population
`
`Treatment Groups
`
`24 wks87
`
`24 wks88
`
`741 patients either treated with or
`naïve to oral antidiabetic drugs,
`A1C 8.0%
`701 patients treated with metformin,
`A1C 8.0%
`
`24 wks89
`
`175 patients treated with a
`thiazolidinedione, A1C 8.1%
`
`24 wks90
`
`1091 patients naïve to oral antidiabetic
`drugs, A1C 8.8%
`
`30 wks91
`
`Saxagliptin
`12 wks92
`
`190 patients treated with metformin,
`AIC 9.2%
`
`Low-dose group: 338 patients naïve
`to oral antidiabetic drugs, A1C 7.9%
`
`6 wks92
`
`24 wks93
`
`High-dose group: 85 patients naïve
`to oral antidiabetic drugs, A1C 7.8%
`743 patients treated with metformin.
`A1C 8.1%
`
`Change in AIC
`(%)
`
`–0.60 vs placebo
`–0.48 vs placebo
`
`–0.79 vs placebo
`–0.94 vs placebo
`
`–0.65 vs placebo
`
`–0.70 vs placebo
`
`–0.83 vs placebo
`–1.57 vs placebo
`
`–2.07 vs placebo
`
`–0.99 vs placebo
`–1.30 vs placebo
`
`–1.0 vs placebo
`
`–0.72
`–0.90
`–0.81
`–0.74
`–0.80
`–0.27
`–1.09
`–0.36
`–0.73 vs placebo
`–0.83 vs placebo
`–0.72 vs placebo
`
`Change in
`Weight
`(kg)
`
`–0.2
`–0.1
`–1.1
`
`–0.94
`–0.23
`–1.28
`–0.11
`+0.51
`–1.03
`–0.20
`–0.85
`–1.5
`–0.9
`–0.5
`–1.0
`
`Sitagliptin 100 mg q.d.
`Sitagliptin 200 mg q.d.
`Placebo
`Sitagliptin 100 mg q.d.
`Sitagliptin 200 mg q.d.
`Placebo
`Sitagliptin 100 mg q.d.a
`Placeboa
`Pioglitazone
`Rescue therapy permitted
`(metformin)
`Sitagliptin 100 mg q.d.
`Placebo
`Metformin
`Rescue therapy permitted
`(pioglitazone)
`Sitagliptin 100 mg q.d.
`Sitagliptin 100 mg q.d. +
`metformin 1000 mg q.d.
`Sitagliptin 100 mg q.d. +
`metformin 2000 mg q.d.
`Metformin 1000 mg q.d.
`Metformin 2000 mg q.d.
`Placebo
`Sitagliptin 100 mg q.d.
`Placebo
`
`Saxagliptin 2.5 mg q.d.
`Saxagliptin 5 mg q.d.
`Saxagliptin 10 mg q.d.
`Saxagliptin 20 mg q.d.
`Saxagliptin 40 mg q.d.
`Placebo
`Saxagliptin 100 mg q.d.
`Placebo
`Saxagliptin 2.5 mg q.d.
`Saxagliptin 5 mg q.d.
`Saxagliptin 10 mg q.d.
`Placebo
`
`A1C = hemoglobin A1c.
`aChange in systolic blood pressure was −0.6 mm Hg for sitagliptin 100-mg group and −0.6 mm Hg for placebo group.
`
`initial glycemic control achieved with exenatide
`over a 30-week period is sustainable in the long
`term.
`Data from studies that used the once-weekly
`formulation (exenatide LAR) suggest that its
`efficacy is at least comparable to that of twice-
`daily exenatide.
`In a 15-week, placebo-
`controlled phase II trial in 45 patients, exenatide
`0.8 mg or 2.0 mg/week reduced mean A1C by
`1.4% and 1.7%, respectively (p<0.0001 vs
`placebo for both doses),45 enabling 36% and 86%
`of patients to achieve an A1C of 7% or less.
`In
`
`another open-label, 30-week, phase III study in
`295 patients previously managed with dietary
`changes and exercise and/or oral antidiabetic
`drugs, exenatide LAR 2 mg once/week achieved a
`greater reduction in A1C (–1.9%) than exenatide
`10 µg twice/day (–1.5%, p=0.002), as well as
`greater reductions in FPG level (42 vs 25 mg/dl,
`p<0.0001).79
`Exenatide was also found to significantly
`reduce body weight during the AMIGO trials, by
`up to 2.8 kg from baseline in patients who were
`also receiving metformin,73 and by up to 3.1 kg
`
`MPI EXHIBIT 1022 PAGE 8
`
`MPI EXHIBIT 1022 PAGE 8
`
`
`
`INCRETIN-BASED THERAPIES FOR TYPE 2 DIABETES MELLITUS Drab
`
`617
`
`from baseline (vs 1.4 kg with placebo) in a more
`recent study of drug-naïve patients.71 These
`reductions in weight are apparently unrelated to
`nausea (a common adverse effect with this
`therapy), as patients who did not experience this
`adverse effect also lost weight.73 Body weight
`reductions were further