`Sitagliptin Phosphate: A DPP-4 Inhibitor for the Treatment of
`Type 2 Diabetes Mellitus
`
`Tina Zerilli, PharmD; and Eunice Y. Pyon, PharmD
`Arnold & Marie Schwartz College of Pharmacy and Health Sciences, Long Island University, Brooklyn,
`New York
`
`ABSTRACT
`Background: Sitagliptin phosphate, the first dipep-
`tidyl peptidase 4 (DPP-4) inhibitor, provides a new
`treatment option for patients with type 2 diabetes.
`Objective: The purpose of this article is to review
`the pharmacology, pharmacokinetics, pharmacody-
`namics, clinical efficacy, adverse effects, and cost of sita-
`gliptin in adults with type 2 diabetes.
`Methods: A literature search of MEDLINE (1966–
`May 10, 2007), Iowa Drug Information Service
`(1966–May 10, 2007), and International Pharmaceutical
`Abstracts (1970–May 10, 2007) was performed using
`the terms sitagliptin and MK-0431. English-language,
`original research and review articles were reviewed, as
`were citations from these articles. The 2005 and 2006
`American Diabetes Association Scientific Abstracts
`were searched, and the US Food and Drug Adminis-
`tration review of the new drug application for
`sitagliptin and select information from the manufac-
`turer were consulted.
`Results: By inhibiting DPP-4, sitagliptin enhances
`postprandial levels of active glucagon-like peptide-1
`(GLP-1), leading to a rise in insulin release and de-
`crease in glucagon secretion from pancreatic α-cells.
`Sitagliptin is 87% orally bioavailable, undergoes mini-
`mal hepatic metabolism, and is primarily excreted un-
`changed (~79%) in the urine. At doses ≥100 mg QD,
`DPP-4 activity is inhibited by >80%, with a conse-
`quent 2-fold rise in active GLP-1 levels. The reduction
`in glycosylated hemoglobin (HbA1c) observed with
`100 mg QD of sitagliptin in Phase III monotherapy
`trials ranged from ~0.5% to 0.6% (P ≤ 0.001 vs
`placebo). In Phase III combination trials, HbA1c was
`reduced by ~0.7% when added to metformin and
`~0.9% with pioglitazone (P < 0.001 vs placebo).
`Markers of β-cell function, including proinsulin/insulin
`ratio and homeostasis model assessment of β-cell func-
`
`tion, were improved with sitagliptin treatment. In
`studies, sitagliptin has been well tolerated; significant
`hypoglycemia and weight gain have not been noted.
`Conclusions: When used alone or in combination
`with metformin or pioglitazone, sitagliptin has been
`associated with significant reductions in HbA1c and
`has been well tolerated. Before its place in therapy can
`be firmly established, long-term studies evaluating the
`safety of prolonged DPP-4 inhibition are necessary.
`(Clin Ther. 2007;29:2614–2634) Copyright © 2007
`Excerpta Medica, Inc.
`Key words: sitagliptin, MK-0431, type 2 diabetes,
`dipeptidyl peptidase-4.
`
`INTRODUCTION
`It is estimated that >180 million people worldwide
`have diabetes mellitus; this figure is expected to more
`than double by the year 2030.1 In the United States
`alone, 7% of the population (20.8 million) has dia-
`betes mellitus, including 1.5 million newly diagnosed
`cases in people ≥20 years of age in 2005.2 Although
`typically associated with older age, type 2 diabetes—
`the most prevalent type, accounting for 90% to 95%
`of all cases—is also now being diagnosed in children
`and adolescents.2
`The increasing prevalence of diabetes is of concern
`because of the morbidity and mortality associated
`with the disease. Complications of uncontrolled type 2
`diabetes include cardiovascular disease and microvas-
`cular complications, such as peripheral neuropathy,
`
`Accepted for publication September 19, 2007.
`doi:10.1016/j.clinthera.2007.12.034
`0149-2918/$32.00
`Printed in the USA. Reproduction in whole or part is not permitted.
`Copyright © 2007 Excerpta Medica, Inc.
`
`2614
`
`Volume 29 Number 12
`
`Mylan EX 1012, Page 1
`
`
`
`nephropathy, and retinopathy. These life-threatening
`complications have made diabetes the fifth-leading
`cause of death in the United States2 and accounted for
`US $24.6 billion of the $92 billion of direct medical
`expenditures attributed to the disease in 2002.3
`The societal and economic burdens of type 2 diabetes
`highlight the importance of tight glycemic control and
`prevention and management of diabetic complica-
`tions. In addition to lifestyle modifications, several
`classes of pharmacologic agents are available that
`lower blood glucose levels by various mechanisms of ac-
`tion. These include α-glucosidase inhibitors, biguanides
`(eg, metformin), meglitinides, sulfonylureas, thiazo-
`lidinediones, insulin, amylin agonists (eg, pramlin-
`tide), and glucagon-like peptide-1 (GLP-1) analogues
`(eg, exenatide).4 The glucose-lowering effectiveness of
`each of these antidiabetic interventions, when used as
`monotherapy, varies, as described in a consensus state-
`ment on the management of hyperglycemia in patients
`with type 2 diabetes developed by the American Diabetes
`Association (ADA) and the European Association for
`the Study of Diabetes (Table I).4
`In October 2006, the US Food and Drug Adminis-
`tration (FDA) approved sitagliptin phosphate* for use
`as monotherapy or in combination with metformin or
`thiazolidinediones to improve glycemic control in pa-
`tients with type 2 diabetes in conjunction with diet
`and exercise.5 Sitagliptin was the first agent world-
`wide in a new class of medications called dipeptidyl
`peptidase-4 (DPP-4) inhibitors, providing a new oral
`therapeutic option. The purpose of this article is to re-
`view the pharmacology, pharmacokinetics, pharmaco-
`dynamics, clinical efficacy, adverse effects, and cost of
`therapy of sitagliptin phosphate in adult patients with
`type 2 diabetes mellitus.
`
`MATERIALS AND METHODS
`A literature search of MEDLINE (1966–May 10,
`2007), Iowa Drug Information Service (1966–May 10,
`2007), and International Pharmaceutical Abstracts
`(1970–May 10, 2007) was performed using the search
`terms sitagliptin and MK-0431. English-language,
`original research articles and review articles were
`identified and evaluated. Citations from these articles
`were also reviewed. Using the same search terms, the
`2005 and 2006 ADA Scientific Abstracts were also
`
`*Trademark: Januvia® (Merck & Co. Inc., Whitehouse Station,
`New Jersey).
`
`Table I. Glycemia-lowering effectiveness of antidia-
`betic interventions used as monotherapy in
`patients with type 2 diabetes mellitus.
`
`Expected
`Decrease in
`HbA1c, %
`
`1–2
`0.5–0.8
`1.5
`1–1.5
`1.5
`0.5–1.4
`1.5–2.5
`0.5–1.0
`0.5–1.0
`
`Intervention
`
`Lifestyle modifications to decrease
`weight and increase activity
`α-Glucosidase inhibitors
`Metformin
`Meglitinides
`Sulfonylureas
`Thiazolidinediones
`Insulin
`Pramlintide
`Exenatide
`
`HbA1c = glycosylated hemoglobin.
`Adapted with permission.4
`
`searched for pertinent abstracts. The FDA review of
`the new drug application (NDA) for sitagliptin was
`also consulted, as was select information provided by
`the manufacturer.
`
`CLINICAL PHARMACOLOGY
`Sitagliptin phosphate is chemically described as 7-[(3R)-
`3-amino-1-oxo-4-(2,4,5-trifluorophenyl)butyl]-5,6,7,
`8-tetrahydro-3-(trifluoromethyl)-1,2,4-triazolo[4,3-a]
`pyrazine phosphate (1:1) monohydrate. The com-
`pound has a molecular weight of 523.32 Da and a
`molecular formula of C16H15F6N5O·H3PO4·H2O.
`The chemical structure is depicted in the figure.
`
`Mechanism of Action
`Sitagliptin enhances the effects of the incretin hor-
`mones glucose-dependent insulinotropic peptide (also
`known as gastric inhibitory polypeptide [GIP]) and
`GLP-1. Secreted in the intestine in response to food,
`GIP and GLP-1 have a role in the regulation of glu-
`cose homeostasis. Activation of GIP and GLP-1 recep-
`tors on pancreatic β-cells leads to increased levels of
`cyclic adenosine monophosphate and intracellular cal-
`cium, with subsequent glucose-dependent insulin se-
`cretion.6 In addition, sustained receptor activation is
`associated with insulin biosynthesis and stimulation
`of β-cell proliferation.6 Animal and in vitro data fur-
`
`December 2007
`
`2615
`
`Mylan EX 1012, Page 2
`
`
`
`F
`
`F
`
`F
`
`NH2
`
`O
`
`N
`
`N
`
`N
`
`N
`
`CF3
`
`Figure. Chemical structure of sitagliptin. Reprinted
`with permission.10
`
`ther suggest that activation of GIP and GLP-1 recep-
`tors promotes β-cell resistance to apoptosis, prolifera-
`tion, and neogenesis, resulting in enhanced β-cell
`function.6,7 Additional functions of GLP-1 include in-
`hibition of glucagon secretion from pancreatic α-cells,
`resulting in decreased hepatic glucose production;
`slowing of gastric emptying; suppression of food in-
`take; and enhancement of glucose disposal via neural
`mechanisms.6,7
`In patients with type 2 diabetes, the response to the
`incretin hormones is defective, largely because the in-
`sulinotropic activity of GIP, but not GLP-1, is attenu-
`ated.6 Data suggest that there is also a statistically
`significant decline in meal-stimulated GLP-1 levels in
`patients with type 2 diabetes compared with those with
`normal glucose tolerance (mean [SD], 2482 [145] vs
`3101 [198] pmol/L/240 min; P = 0.024).8 Exogenous
`administration of GLP-1, therefore, would in theory
`appear to be an attractive therapeutic modality for
`type 2 diabetes. This approach, however, is limited by
`the rapid inactivation of the incretin hormones by the
`DPP-4 enzyme, which preferentially cleaves substrates
`that have a proline or alanine in the penultimate posi-
`tion.9 In addition to and independent of its enzymatic
`activity in plasma, DPP-4 is a membrane-spanning
`peptidase that is widely distributed in numerous tis-
`sues and T-cells, B-cells, and natural killer cells.7 Also
`known as CD26, DPP-4 serves as a T-cell costimula-
`tor, playing a functional role in T-cell activation and
`proliferation.7,9
`Two viable methods to enhance GLP-1 effects in
`vivo include administration of agents that mimic the
`effects of the incretins but are resistant to degradation
`by DPP-4 (eg, exenatide) and agents that prevent in-
`cretin degradation. Sitagliptin exerts its therapeutic
`effect via the latter mechanism. Thus, following ad-
`ministration of sitagliptin, postprandial levels of ac-
`
`tive GLP-1 are increased and activity is prolonged,
`with a resultant rise in insulin release and decrease in
`glucagon secretion from the pancreatic α-cells.5
`Sitagliptin is a potent, reversible competitive in-
`hibitor of DPP-4. Results from in vitro studies that
`evaluated the DPP-4 inhibitory properties of sitagliptin,
`among other compounds, have suggested that sita-
`gliptin exhibits high selectivity for DPP-4 (IC50, 18 nM).
`Affinity for other proline-specific peptidases, DPP-8
`(IC50, 48,000 nM) and DPP-9 (IC50, >100,000 nM), is
`low.10 Low affinity for these peptidases is of particu-
`lar importance since in preclinical studies, inhibition
`of DPP-8 and DPP-9 has been associated with severe
`toxicities, including alopecia, blood dyscrasias, multi-
`organ histopathologic changes, and mortality in rats;
`gastrointestinal toxicity in dogs; and attenuation of
`T-cell function in human in vitro models.11 Notably,
`the effects on the immune system were not seen with
`a DPP–4-selective compound.11 Likewise, whereas other
`nonselective DPP-4 inhibitors have been associated
`with the development of necrotic skin lesions in pre-
`clinical studies involving monkeys, no treatment-related
`skin toxicity was observed in a 3-month study in
`monkeys treated with sitagliptin (personal communi-
`cation, Carol Teutsch, MD, Merck & Co. Inc., Sep-
`tember 7, 2007).
`
`Pharmacokinetics
`Several studies characterizing the pharmacokinetic
`properties of sitagliptin in animals, healthy subjects,
`and patients with type 2 diabetes have been pub-
`lished.10,12–17 Key pharmacokinetic parameters in
`healthy subjects, as provided by the manufacturer,5
`are summarized in Table II. In general, the pharma-
`cokinetics of sitagliptin in healthy subjects are compa-
`rable with those observed in patients with type 2
`diabetes.5,14,17
`Following administration of an oral 100-mg dose in
`healthy volunteers, sitagliptin was rapidly absorbed,
`with a median Tmax of 1 to 4 hours.5 Plasma AUC was
`8.52 µM · h and Cmax was 950 nM.5 Sitagliptin plas-
`ma AUC has been found to be increased in an approx-
`imate dose-dependent manner in both single-dose
`(1.5–600 mg)14 and multiple-dose (25–600 mg QD
`and 300 mg BID)15 studies in healthy volunteers,
`whereas Cmax increased in a slightly greater than dose-
`proportional manner. The administration of a high-fat
`breakfast prior to a single oral 25-mg dose of
`sitagliptin has not been found to influence the plasma
`
`2616
`
`Volume 29 Number 12
`
`Mylan EX 1012, Page 3
`
`
`
`Table II. Pharmacokinetic parameters of sitagliptin
`in healthy subjects.5
`
`Parameter
`
`Bioavailability
`Volume of distribution
`Protein binding
`Tmax
`Metabolism
`
`Elimination
`
`Apparent terminal t1/2
`Renal clearance
`
`Value
`
`87%
`~198 L
`38%
`1–4 h
`Minimal hepatic
`metabolism
`87% Urine
`(~79% unchanged);
`13% feces
`~12.4 h
`~350 mL/min
`
`AUC0–∞; the ratio of the least-squares (LS) mean
`(95% CI) ratio (fed/fasted) was 1.01 (0.94–1.10).14
`An increase in Cmax of ~20% was observed in the fed
`state; however, this difference was not statistically sig-
`nificant versus that observed in the fasting state (LS
`mean ratio, 1.21 [95% CI, 1.00–1.45]).14 Since no
`pharmacokinetic parameters are appreciably influenced
`by food, sitagliptin may be dosed without regard to
`meals.5 Steady-state concentrations of sitagliptin are
`achieved within 2 to 3 days of administration.15,16
`The mean volume of distribution of sitagliptin, as de-
`termined after administration of a single 100-mg IV
`dose in healthy subjects, is ~198 L, and 38% of the
`drug is reversibly bound to plasma proteins.5
`Sitagliptin does not appear to undergo extensive
`metabolism. Data from a study by Vincent et al13 evalu-
`ating the metabolism and excretion of [14C]sitagliptin
`in 6 healthy male subjects suggest that after a single
`oral dose, the parent drug comprised the majority of
`plasma (78%–90%) and urinary (~84%–88%) radio-
`activity. Six metabolites (M1–M6) were detected in
`small amounts, each comprising <1% to 8% of the
`circulating plasma radioactivity and <1% to 5% of
`total urinary radioactivity. In vitro experiments found
`that cytochrome P450 (CYP) isozyme 3A4, and to a
`lesser extent CYP2C8, were the major isozymes asso-
`ciated with the limited sitagliptin metabolism.13 Due
`to the low levels in plasma and low affinity for the
`DPP-4 enzyme (M1, M2, and M5 were tested for
`DPP-4 inhibition and found to be ~300-, 1000-, 1000-
`fold less active, respectively, than the parent drug),13
`the metabolites are not believed to contribute to the
`
`pharmacologic activity of sitagliptin.5,13,18 In the same
`study, the majority (87%) of the radioactive dose was
`recovered in the urine within 1 week of dosing; 13%
`of the administered dose was excreted via the feces.13
`The apparent terminal t1/2 of sitagliptin is ~12.4 hours.
`Both the fraction of the oral dose excreted unchanged
`in the urine (~79%) and the renal clearance (ClR
`(~350 mL/min) are independent of dose.14,15 Sita-
`gliptin undergoes active tubular secretion, as evidenced
`by the fact that ClR exceeds creatinine clearance
`(CrCl).5 The compound is a substrate for human
`organic anion transporter 3, the organic anion trans-
`porting polypeptide OATP4C1, and the efflux trans-
`porter P-glycoprotein.19
`
`Special Populations
`In a single-dose open-label study, Bergman et al20
`evaluated the effects of varying degrees of renal
`impairment on the pharmacokinetics of sitagliptin.
`Thirty otherwise healthy participants (18–75 years of
`age) with either mild (CrCl, 50–80 mL/min), moder-
`ate (CrCl, 30–50 mL/min), or severe (CrCl, <30 mL/
`min) renal insufficiency, end-stage renal disease (ESRD)
`receiving hemodialysis, or normal renal function
`(CrCl, >80 mL/min) were included in the study (6 in
`each group). Subjects with normal renal function and
`patients with mild to severe renal insufficiency re-
`ceived a single 50-mg oral dose. Patients with ESRD
`received a single 50-mg dose of sitagliptin 48 hours
`prior to their normally scheduled hemodialysis ses-
`sion. To quantify the amount of sitagliptin removed
`by hemodialysis, patients with ESRD received a sec-
`ond 50-mg dose after a 1-week washout period. Hemo-
`dialysis was performed 4 hours postdose. Healthy sub-
`jects (n = 145) from 11 other studies were included
`in the historical control group to supplement the sub-
`jects in the study with normal renal function. A <2-fold
`increase in plasma AUC0–∞ was considered by the in-
`vestigators to be not clinically meaningful. This asser-
`tion was based on the fact that in prior studies,14,15
`sitagliptin was well tolerated in healthy subjects re-
`ceiving doses of up to 600 mg.20
`Compared with subjects with normal renal func-
`tion (n = 151), sitagliptin AUC0–∞ values were 1.6,
`2.3, 3.8, and 4.5-fold higher in patients with mild,
`moderate, and severe renal insufficiency and ESRD,
`respectively.20 The geometric LS mean (90% CI) ratios
`for Cmax were 1.35 (1.15–1.58) in patients with mild
`renal insufficiency, 1.43 (1.23–1.67) in patients with
`
`December 2007
`
`2617
`
`Mylan EX 1012, Page 4
`
`
`
`moderate renal insufficiency, 1.75 (1.51–2.03) in pa-
`tients with severe renal insufficiency, and 1.42 (1.22–
`1.65) in patients with ESRD. Compared with values
`in subjects with normal renal function (13.1 hours),
`the terminal t1/2 values of sitagliptin in those with
`mild, moderate, and severe renal insufficiency and
`ESRD were 16.1, 19.1, 22.5, and 28.4 hours, respec-
`tively (P = 0.011 for group with mild renal insufficien-
`cy; P < 0.001 for all other sitagliptin groups). ClR of
`sitagliptin was approximately proportional to CrCl.
`The fractions of the sitagliptin dose removed by he-
`modialysis initiated at 4 and 48 hours postdose were
`13.5% and 3.5%, respectively. As reported by the in-
`vestigators, sitagliptin was well tolerated in all groups,
`although no specific data on adverse events were pro-
`vided.20 In light of these findings, dosage adjustments
`are recommended in patients with moderate or severe
`renal insufficiency or ESRD (see Dosage and Adminis-
`tration section).5 Caution may be needed, however, when
`prescribing sitagliptin in patients with mild renal im-
`pairment, although no dosage adjustment is recom-
`mended by the manufacturer.5
`The effects of moderate hepatic impairment on
`sitagliptin pharmacokinetics have also been evaluated
`in a study published only as an abstract.21 Ten pa-
`tients with Child-Pugh scores ranging from 7 to 9 and
`10 healthy matched controls each received a single
`100-mg oral sitagliptin dose in an open-label fashion.
`The prespecified range of bounds for clinical non-
`significance for the AUC was 0.5 to 2.00. Compared
`with healthy subjects, ~21% and ~13% increases in
`mean plasma AUC0–∞ and Cmax, respectively, were ob-
`served in patients with hepatic insufficiency; however,
`both parameters fell within the prespecified bounds
`for clinical nonsignificance (actual values not provid-
`ed). There were no statistically significant differences
`in Tmax, apparent terminal t1/2, fraction of the oral
`dose excreted into urine, or ClR between the 2 groups.
`Sitagliptin was well tolerated in both groups.21 As
`such, no dosage adjustment is recommended by the
`manufacturer in patients with moderate hepatic im-
`pairment.5 However, due to the small sample size of
`the study, more data may be necessary to evaluate the
`true effect of hepatic impairment on sitagliptin phar-
`macokinetics. Studies assessing the effects of severe
`hepatic impairment on sitagliptin pharmacokinetics
`were not identified in the literature search.
`In another study published only as an abstract, the
`effects of age, sex, and obesity on sitagliptin pharma-
`
`cokinetics were assessed.22 Eight healthy, young (age,
`18–45 years), nonobese women; 10 healthy, elderly
`(age, 65–80 years), nonobese men; 10 healthy, elderly,
`nonobese women; and 10 healthy, young adult, obese
`(body mass index [BMI], 30–40 kg/m2) subjects were
`enrolled in the study. Within each group, 2 partici-
`pants received placebo, while the others received a sin-
`gle oral 50-mg sitagliptin dose. Pharmacokinetic data
`from a previous study in which 6 healthy, young,
`nonobese male subjects received a single oral 50-mg
`dose of sitagliptin were also included in the analysis.22
`Pharmacokinetic parameters that were significantly
`different between groups were as follows. The AUC0–∞
`geometric mean ratio (GMRs) (90% CI) comparing
`elderly and young (pooled across sex) and young male
`obese and nonobese subjects were 1.31 (1.19–1.43)
`and 0.77 (0.69–0.86), respectively. The Cmax GMR
`(90% CI) comparing elderly and young was 1.23
`(1.04–1.46); for the comparison between female and
`male subjects (pooled across age), GMR (90% CI)
`Cmax was 1.46 (1.23–1.73). These differences in plas-
`ma pharmacokinetics, however, were not considered
`by the investigators to be clinically meaningful.22
`Notably, as per the prescribing information, the dose
`of sitagliptin does not require adjustment for age, sex,
`or obesity.5
`In a multicenter, randomized, double-blind, placebo-
`controlled study, 32 middle-aged (45–65 years), obese
`(mean BMI, 33.7 kg/m2 [range, 30.2–39.8 kg/m2]) sub-
`jects received sitagliptin 200 mg BID (n = 24) or place-
`bo (n = 8) for 28 days.23 Sitagliptin pharmacokinetic
`parameters were similar to those obtained from single-
`dose14 and multiple-dose15 studies in healthy male sub-
`jects. A search of the literature did not yield any
`studies assessing the use of sitagliptin in the pediatric
`population.
`
`Pharmacodynamics
`In clinical studies involving healthy volunteers,
`treatment with sitagliptin was associated with dose-
`dependent inhibition of DPP-4 activity.14,15 The per-
`centage inhibition of DPP-4 activity that has cor-
`related with near-maximal glucose-lowering effects
`has been found to be 80% or greater in rodent mod-
`els.10 This degree of inhibition has been observed in
`subjects treated with sitagliptin in pharmacodynamic
`studies. In a single-dose study (1.5–600 mg), the
`weighted average inhibition (WAI) of DPP-4 activity
`was at least 80% with doses ≥50 mg over a 12-hour
`
`2618
`
`Volume 29 Number 12
`
`Mylan EX 1012, Page 5
`
`
`
`period and with doses ≥100 mg over a 24-hour period.14
`Following multiple-dose administration (25–600 mg
`QD or 300 mg BID), the WAI of DPP-4 activity over
`24 hours was 79.9% (95% CI, 71.2%–89.2%) with
`the 50-mg QD dosage relative to placebo and >80%
`for daily doses of 100 mg or more.15 Inhibition of
`DPP-4 activity by sitagliptin at steady-state trough was
`≥80% at dosages of 100 QD or greater.15
`As expected by the physiology of the incretin sys-
`tem, inhibition of DPP-4 activity due to sitagliptin
`treatment has been associated with alterations in
`GLP-1 levels. Normally, because of DPP-4 activity, in-
`tact, biologically active GLP-1 represents 10% to 20%
`of total plasma GLP-1.6 However, compared with
`placebo, treatment with sitagliptin daily doses ranging
`from 25 to 600 mg has been associated with an ~2-fold
`increase in weighted average concentrations of active
`GLP-1 through 2 hours after administration of stan-
`dardized meals, which were given at various times
`postdose.15,23 Although active levels of GLP-1 were
`increased, no consistent change in total GLP-1 levels
`has been noted, supporting that the pharmacologic ef-
`fect of sitagliptin is due to lack of degradation of ac-
`tive GLP-1 rather than an increase in secretion of
`GLP-1.14,15,23 In healthy subjects, sitagliptin has not
`been associated with postprandial effects on glucose,
`insulin, C-peptide, or glucagon levels.14,15
`The pharmacodynamic effects of sitagliptin in pa-
`tients with type 2 diabetes have been described in a
`randomized, double-blind, placebo-controlled, 3-period,
`single-dose, crossover study involving 56 patients.17
`After an overnight fast, patients received, in random-
`ized order, sitagliptin 25 or 200 mg and placebo, with
`a 7-day washout period in between treatments. As in
`healthy subjects, in patients with type 2 diabetes,
`treatment with sitagliptin was associated with dose-
`dependent inhibition of plasma DPP-4 activity. The
`mean (95% CI) percentages of inhibition of plasma
`DPP-4 activity over a 24-hour period with the 25- and
`200-mg doses and placebo were 68.1% (66.6% to
`69.6%), 91.4% (90.9% to 91.8%), and 2.1% (–2.8%
`to 6.7%), respectively; the difference between either
`sitagliptin dose and placebo was statistically signifi-
`cant (P < 0.001). The difference between sitagliptin
`doses was also significant (P < 0.05). After an oral
`glucose tolerance test (OGTT) administered 2 hours
`postdose, the weighted average augmentation (WAA)
`active GLP-1 and GIP levels in either sitagliptin dose
`were both ~2-fold greater than those observed with
`
`placebo (P < 0.001). Higher levels of WAA active
`GLP-1 (1.3- and 1.9-fold for the 25- and 200-mg doses,
`respectively) and WAA active GIP (1.4- and 2-fold for
`the 25- and 200-mg doses, respectively) compared with
`placebo were also observed following an OGTT given
`24 hours postdose (P < 0.001).15 The differences be-
`tween the 25- and 200-mg sitagliptin doses were also
`significant for both the WAA active GLP-1 and WAA
`active GIP levels (P < 0.001).
`In contrast to the lack of an effect on glucoregula-
`tory hormones and glucose excursion following a
`standardized meal in healthy subjects treated with
`sitagliptin,14,15 these parameters were significantly al-
`tered in patients with type 2 diabetes treated with
`sitagliptin. Compared with patients who received place-
`bo, those given an OGTT 2 hours after receiving
`sitagliptin 25 or 200 mg experienced a reduction in
`mean incremental glucose AUC0–240 min (22% vs 26%;
`P ≤ 0.001), increased insulin concentration (22% vs
`21%; P ≤ 0.001), increased C-peptide concentration
`(13% vs 21%; P ≤ 0.001), and decreased glucagon
`concentration (7% vs 14%; P < 0.05). Compared
`with the placebo group, sitagliptin-treated patients re-
`ceiving the 200-mg dose experienced an 18% reduc-
`tion in glucose excursion following an OGTT given
`24 hours postdose (P ≤ 0.001). The 9% reduction in
`glucose excursion seen in the 25-mg sitagliptin group
`was not significantly different from the placebo group.17
`In addition to GLP-1 and GIP, a number of other
`bioactive peptides are cleaved by DPP-4 in vitro, includ-
`ing growth hormone–releasing hormone (GHRH).24
`Thus, inactivation of DPP-4 may potentially lead to
`increased GHRH concentrations in vivo, with sub-
`sequent stimulation of the growth hormone (GH)/
`insulin-like growth factor (IGF)-1 axis. Preclinical data
`suggest that DPP-4 inhibition with an analogue of
`sitagliptin does not alter circulating levels of IGF-1.25
`Likewise, treatment with sitagliptin has not been asso-
`ciated with an appreciable rise in IGF-1 or IGF bind-
`ing protein 3 levels in studies involving normogly-
`cemic subjects.15,23
`Progressive decline in β-cell function is a hallmark
`characteristic of type 2 diabetes, with ~50% of func-
`tion lost by the time of diagnosis.26 Any agent that can
`restore β-cell mass and/or function or delay its decline
`would therefore be desirable. Positive effects on β-cell
`function have been noted in preclinical studies of
`des-fluoro-sitagliptin, an analogue of sitagliptin with
`similar potency, selectivity, and pharmacokinetic prop-
`
`December 2007
`
`2619
`
`Mylan EX 1012, Page 6
`
`
`
`erties.27,28 Likewise, in a model-based analysis using
`data from 3 Phase III clinical studies, sitagliptin treat-
`ment was found to be associated with improved β-cell
`function in both the fasting and postprandial states.29
`A detailed description of the effect of sitagliptin treat-
`ment on markers of β-cell function in clinical studies
`is provided in the next section.
`
`EFFICACY AND TOLERABILITY
`Sitagliptin has been evaluated in numerous clinical tri-
`als30–39 as monotherapy and as part of multiple-drug
`therapy for the treatment of type 2 diabetes.
`
`Monotherapy
`Phase II Dose-Ranging Trials
`A 12-week, multinational, randomized, double-
`blind, placebo- and active-controlled, parallel-group
`study was conducted to assess the dose response to
`sitagliptin in patients between the ages of 21 and
`75 years with type 2 diabetes experiencing inadequate
`glycemic control with diet and exercise.30 Patients not
`receiving oral antihyperglycemic agents (OHAs) and
`with glycosylated hemoglobin (HbA1c) ≥6.5% and
`≤10% entered a 2- to 6-week diet and exercise period.
`Patients receiving OHA monotherapy (excluding thia-
`≥6% and ≤9% dis-
`zolidinediones) and with HbA1c
`continued therapy and entered a 6-week diet and
`exercise period. After the diet and exercise period, pa-
`≥6.5% and ≤10% and fasting plas-
`tients with HbA1c
`ma glucose (FPG) ≥130 and ≤240 mg/dL entered a
`2-week, single-blind, placebo run-in period. A total of
`743 patients completed the placebo run-in period and
`were randomized to 1 of 6 treatment groups: sitagliptin
`5 mg BID (n = 125), sitagliptin 12.5 mg BID (n = 123),
`sitagliptin 25 mg BID (n = 123), sitagliptin 50 mg BID
`(n = 124), glipizide 5 mg QD (titrated to a maximum
`daily dose of 20 mg; n = 123), or placebo (n = 125).
`To serve as a benchmark to provide information on
`the efficacy and tolerability profile of a sulfonylurea in
`the same study population, glipizide was added as a
`treatment arm; however, it was not a direct compara-
`tor agent in this trial.30
`After 12 weeks of therapy, all sitagliptin groups were
`found to have significant mean reductions in HbA1c
`compared with placebo (all, P < 0.001). The placebo-
`subtracted LS mean (95% CI) changes in HbA1c from
`baseline to week 12 were as follows: –0.38% (–0.58%
`to –0.19%) in the 5-mg BID group; –0.64% (–0.84%
`to –0.45%) in the 12.5-mg BID group; –0.66%
`
`(–0.85% to –0.47%) in the 25-mg BID group; and
`–0.77% (–0.96% to –0.58%) in the 50-mg BID group
`(P < 0.001 for each sitagliptin group compared with
`placebo). In patients who received glipizide, the LS
`mean change from baseline in HbA1c compared with
`patients receiving placebo was –1.00% (–1.19% to
`–0.80%). The authors stated that, based on placebo-
`subtracted HbA1c changes from baseline, a stepwise
`increase in efficacy was observed across the dose
`range through the 50-mg BID group, although statis-
`tical significance was not achieved. Pairwise compari-
`sons in changes in HbA1c between the sitagliptin
`groups did not achieve statistical significance except
`for comparisons with the 5-mg BID group (all, P <
`0.01). The incidences of adverse events were similar
`between the sitagliptin and placebo groups. Clinical
`adverse events determined by the investigator to be
`possibly, probably, or definitely related to treatment
`occurred in 8.9%, 16.3%, 13.8%, 12.3%, and 9.6% of
`patients who received sitagliptin 5, 12.5, 25, or 50 mg
`BID or placebo, respectively. The incidence of drug-
`related clinical adverse events was 27.6% in patients
`who received glipizide.30
`Although not identified in published literature, re-
`sults of a second Phase II trial of similar design were
`found in the FDA review of the new drug application
`for sitagliptin.31 Sitagliptin study P014 (as notated in
`the FDA review) was a 12-week, multinational, ran-
`domized, placebo-controlled, dose-ranging study evalu-
`ating the efficacy and tolerability of sitagliptin 25 mg
`QD, 50 mg QD, 100 mg QD, and 50 mg BID in 555 pa-
`tients between the ages of 21 and 75 years with type 2 dia-
`betes. The inclusion criteria were similar to the criteria
`≥6.5% and ≤10%
`described in the study above30 (HbA1c
`≥6%
`in patients not receiving OHA therapy and HbA1c
`and ≤9% in patients receiving OHA monotherapy).
`The LS mean (95% CI) changes in HbA1c from base-
`line at week 12 were –0.28% (–0.42% to –0.14%),
`–0.44% (–0.58% to –0.30%), –0.44% (–0.58% to
`–0.30%), –0.43% (–0.56% to –0.29%), and 0.12%
`(–0.02% to 0.26%) in the sitagliptin 25-mg QD, 50-mg
`QD, 100-mg QD, 50-mg BID, and placebo groups, re-
`spectively (P < 0.001 for all sitagliptin groups com-
`pared with baseline). Results suggested there was no
`significant difference in change in HbA1c between the
`100-mg QD and 50-mg BID groups. In addition, HbA1c
`changes at week 12 were not significantly different be-
`tween patients receiving 50 or 100 mg QD. Based on
`the results of the Phase II trials, there was a recom-
`
`2620
`
`Volume 29 Number 12
`
`Mylan EX 1012, Page 7
`
`
`
`mendation from the FDA to the manufacturer to in-
`clude the 50-mg QD dose in the Phase III clinical trials.
`However, the Phase III clinical trials described below
`focus on the 100- and 200-mg QD dosages, presumably
`based on modeling of DPP-4 inhibition (≥80% inhibi-
`tion achieved with daily doses ≥100 mg), serum concen-
`trations, and the tolerability of the increased doses.31,32
`
`Phase III Trials
`Aschner et al33 conducted a 24-week, multicenter,
`double-blind, randomized, placebo-controlled study
`in 741 patients between the ages of 21 and 75 years
`with inadequately controlled type 2 diabetes. Patients
`could be enrolled in the study in 1 of 3 ways: patients
`with HbA1c of 7% to 10% and not receiving an OHA
`were entered into a 2-week, single-blind, placebo run-
`in period; patients with