DI A BETES RESE ARCH A ND CLl N IC A L P RA CTICE 81 (2008) 161 1 6 8
`
`available at www.sciencedirect.com
`
`.. -.-,
`·•;" ScienceDjrect
`
`journal homepage: www .elsevier.com/locate/diabres
`
`ll"llnn&iio,nal Oi&betos ~
`
`n
`
`Dose-dependent improvement in glycemia with once-daily
`liraglutide without hypoglycemia or weight gain:
`A double-blind, randomized, controlled trial in Japanese
`patients with type 2 diabetes
`
`Y. Seino a,*, M.F. Rasmussen b, M. Zdravkovic b, K. Kaku c
`
`• Kansai Electric Power Hospital, 2 1 7 Fukushima Fukushima ku Osaka shi, 553 0003 Osaka, Japan
`b Nouo Nordisk NS, Bagsua.erd, Denmark
`c Department of Internal Medicine, Kawasaki Medical School, Okayama, Japan
`
`ARTICLE INFO
`
`ABSTRACT
`
`Article history:
`Received 6 November 2007
`Accepted 10 March 2008
`Published on line 20 May 2008
`
`Keywords:
`Liraglu ti.cl e
`GLP 1
`Type 2 diabetes mellitus
`lncretin
`
`Aims: To evaluate dose response efficacy and safety of once daily human GLP 1 analog
`liraglutide in Japanese subjects with type 2 diabetes.
`Methods : Patients (226, treated with diet with/without OADs, mean HbA1c 8.30%, mean BMl
`23.9 kwm2) were randomized after OAD discontinuation and washout to receive liraglutide
`0.1, 0.3, 0.6 or 0. 9 mg once daily, or placebo in double blind, parallel group design for 14
`weeks.
`Results: Liraglutide dose levels reduced HbA1c versus placebo (by 0.79%, 1.22%, 1.64% and
`1.85%, respectively; p < 0.0001 for linear contrast). Liraglutide 0.9 mw"day resulted in 75% of
`patients achieving HbA1c <7.0% and 57% achieving HbA1 c <6.5%. There were no major or
`minor hypoglycemic events. Llraglutide also reduced, with significant dose response (each
`p < 0.0001 for linear contrast) versus placebo: fasting plasma glucose (up to 2. 5 mmol/L),
`postprandial (0 3 h) glucose excursion (up to 12.8 mmol/(L h)); and increased postprandial
`insulin secretion (up to 23.0 µ.U/(mL h)) and beta cell function as evaluated by HOMA ~ (up
`to around 20.0 (µ,U/mL)/ (mwdL)). Body weight was unchanged; no development of liraglu
`tide antibodies was detected.
`Conclusions: Llraglutide was highly effective and well tolerated at doses up to 0.9 mwday in
`Japanese patients with type 2 diabetes, allowing glycemic control without weight gain or
`hypoglycemia.
`
`© 2008 Elsevier Ireland Ltd. All rights reserved.
`
`1.
`
`Introduction
`
`Type 2 diabetes is a multi.factorial disease, in which individuals,
`to a vaiying degree, exhibit failing beta cell function, weight
`gain and increasing insulin resistance over time. The relative
`
`importance of these factors differs between individuals and
`between populations, and it is well established that the
`pathophysiology of type 2 diabetes differs between Japanese
`and Caucasian patients. Insulin secretory capacity in Japanese
`patients with type 2 diabetes has been shown to be half of that
`
`• Corresponding author.
`E mail address: seino.yutaka@e2.kepco.co.jp (Y. Seino).
`see front matter © 2008 Elsevier Ireland Ltd. All rights reserved.
`0168 8227/$
`doil 0.1016/j. di ab res. 2008 .03.018
`
`MPI EXHIBIT 1038 PAGE 1
`
`MPI EXHIBIT 1038 PAGE 1
`
`Apotex v. Novo - IPR2024-00631
`Petitioner Apotex Exhibit 1038-0001
`
`

`

`162
`
`d i a b e t e s r e s e a r c h a n d c l i n i c a l p r a c t i c e 8 1 ( 2 0 0 8 ) 1 6 1 1 6 8
`
`seen in Caucasian patients, a difference that is particularly
`pronounced for meal related secretion [1 3]. Japanese patients,
`in addition, typically have less insulin resistance compared
`to Caucasians and are generally less obese, with a typical
`body mass index (BMI) of 23 24 kg/m2 [4]. However, deteriorat
`ing glycemic control is often observed resulting in increased
`risk of microvascular and macrovascular complications. Avail
`able treatments for type 2 diabetes are often associated
`with undesirable effects (such as weight gain, hypoglycemia
`or edema)
`that
`limit
`their acceptability and potential
`for reaching treatment targets. There remains an urgent need
`for new therapies that address the multiple dysfunctions in
`type 2 diabetes without limitations of poor tolerability or
`acceptability.
`Glucagon like peptide 1 (GLP 1) is an incretin hormone
`with a broad spectrum of physiological actions. Analogs of
`GLP 1 may potentially be able to modulate the otherwise
`inevitable progression of type 2 diabetes [5 7]. Liraglutide is a
`long acting human GLP 1 analog that has a high degree of
`homology to native GLP 1 but, via acylation to myristic acid,
`achieves a longer plasma half life of 13 h and can thus be
`administered once daily [8 10].
`Previous studies with liraglutide in Japanese subjects have
`included single dose and 21 day stepwise dose escalation (up
`to 15 mg/kg) studies in healthy subjects and a 14 day dose
`escalation study (to 10 mg/kg) in subjects with type 2 diabetes
`[11]. Studies in non Japanese subjects with type 2 diabetes
`have shown that liraglutide is able to achieve sustained
`improvements in glycemic control with significant reduction
`in body weight and with a very low risk of hypoglycemia [7,12
`14]. Liraglutide has shown favorable effects on several
`parameters of beta cell function [10,15 17], and animals
`treated with liraglutide have shown increases in beta cell
`mass [18]. Recent data in non Japanese subjects with type 2
`diabetes also suggest potentially beneficial effects of liraglu
`tide on systolic blood pressure (SBP),
`triglycerides and
`cardiovascular biomarkers [19].
`The present study was conducted to assess the efficacy,
`safety and optimal dose for liraglutide during sustained
`treatment in Japanese subjects with type 2 diabetes.
`
`2.
`
`Subjects and methods
`
`This was a multicenter, double blind, randomized, parallel
`group phase 2 trial to evaluate the dose response relationship
`on glycemic control of four doses of liraglutide and placebo in
`Japanese subjects with type 2 diabetes. The primary efficacy
`endpoint was HbA1c after 14 weeks of treatment, with other
`measures of glycemic control (fasting plasma glucose (FPG),
`postprandial plasma glucose (PPG), self monitored plasma
`glucose profile) as secondary endpoints. The study was
`performed at 63 centers in Japan between March 2005 and
`May 2006. The study was performed in accordance with the
`Declaration of Helsinki, with informed consent of subjects and
`with approval of relevant ethics committees. Trial registration
`numbers were NCT00154414 and JapicCTI 050131.
`Patients included were to have type 2 diabetes treated
`with diet therapy with or without oral antidiabetic drug
`(OAD) monotherapy, HbA1c 7.0% and <10.0%, to be aged
`
`between 20 and 75 years and to have BMI <30 kg/m2. Patients
`treated with insulin or insulin sensitizer within 16 weeks, or
`receiving or expected to receive systemic corticosteroids,
`were excluded, as were those with impaired hepatic or renal
`function (serum glutamic oxaloacetic transaminase or
`serum glutamic pyruvic transaminase >80 IU/L, or serum
`creatinine 1.7 mg/dL), congestive heart failure (New York
`Heart Association class III or IV), unstable angina pectoris or
`myocardial
`infarction within 12 months, uncontrolled
`hypertension (SBP > 160 mmHg or diastolic blood pressure
`>100 mmHg), non stabilised proliferative retinopathy or
`maculopathy.
`The trial consisted of an 8 week run in period following
`screening (at which point OAD therapy, if any, was discontinued
`and during which an FPG <7.5 or 11.1 mmol/L at either of two
`visits was cause for exclusion), a 2 week dose titration period
`and a 12 week maintenance phase. Patients were randomized
`at the end of the run in period to liraglutide or placebo in four
`cohorts, each planned to contain 50 patients; in each cohort a
`planned 40 patients received one of four liraglutide doses
`(0.1 mg/day; 0.3 mg/day; 0.6 mg/day or 0.9 mg/day) and 10
`received placebo. Liraglutide doses in the 0.6 and 0.9 mg/day
`cohorts were increased from a starting dose of 0.3 0.6 mg/day
`after 1 week and, in the 0.9 mg/day cohort, by a further 0.3 mg/
`day after 2 weeks. Randomization was stratified by pre
`treatment (with or without OAD monotherapy) and was
`performed by sealed code by a central telephone registration
`centre; allocation to liraglutide or placebo in each cohort was
`blinded to subject and investigator. Dynamic allocation was
`employed in order to guarantee a balanced allocation within
`strata of pre trial treatment. Liraglutide was supplied by Novo
`Nordisk A/S (Copenhagen, Denmark) as a 6.25 mg/mL solution,
`and visually indistinguishable liraglutide vehicle as placebo.
`Trial medication was administered by abdominal subcutaneous
`injection using pre filled pen and needle set (FlexPen1,
`PenNeedle1, Novo Nordisk), once daily in the evening (at the
`same time every day for each subject, as far as possible).
`HbA1c and FPG were measured at baseline, after 2, 6 and 10
`weeks and at the end of the trial. A meal test, using a standard
`Japanese style breakfast, was performed at baseline and at
`the end of the trial; plasma glucose, insulin and glucagon
`being measured and the pre breakfast plasma glucose being
`taken as measure of FPG. Fasting pro insulin and C peptide
`levels were also recorded. All analyses were carried out by a
`central
`laboratory (Mitsubishi Kagaku BCL,
`Inc., Tokyo,
`Japan), except for a seven point plasma glucose profile which
`was measured before and approximately 2 h after each meal,
`and at bedtime by self monitoring at home using glucose
`meters (Glutest Ace1, Glutest PRO1, Sanwa Kagaku, Nagoya,
`Japan; Glucocard Diameter or Glucocard Diameter a, Arkray
`KDK Corp., Kyoto, Japan) before start of treatment and end of
`study, furthermore plasma liraglutide concentrations were
`measured using ELISA by Capio Diagnostics (Copenhagen,
`Denmark), and liraglutide antibodies were measured by
`radioimmunoassay by Novo Nordisk A/S Immunochemistry
`(Ma˚ løv, Denmark). Safety assessments included thyroid
`ultrasonography at screening and end of study, electrocar
`diography and clinical
`laboratory assessments including
`calcitonin. Beta cell function and insulin resistance were
`assessed from fasting insulin and plasma glucose values
`
`
`
`MPI EXHIBIT 1038 PAGE 2
`
`MPI EXHIBIT 1038 PAGE 2
`
`Apotex v. Novo - IPR2024-00631
`Petitioner Apotex Exhibit 1038-0002
`
`

`

`D l ABE T E S R E S E ARC H AN D C LJ N I CA L P R AC T I C E 8r ( 2 008) 16 r r68
`
`163
`
`using the homeostasis model assessment (HOMA) 13 and
`HOMA R models, respectively (20):
`
`HOMA 13
`
`360 x fastinginsulin/(FPG 63)
`
`HOMA R
`
`(f?istinginsulin X FPG)/'WS
`
`where units of insulin and glucose µU/mL and mg/d.L, respec
`tively.
`Efficacy endpoints were analyzed using data from all
`patients who received trial product and for whom any efficacy
`data were recorded. Safety endpoints were recorded for any
`patient receiving any dose of trial product. HbA1c and other
`efficacy endpoints were analyzed using an ANOVA model with
`dose group and pre trial treatment (with and without OAD) as
`fixed effects and value at baseline as covariate. The existence
`of a monotonic dose response relationship was assessed for
`each endpoint using an F test for linear contrast with contrast
`coefficient of ( 2, 1, 0, 1, 2). A sequentially rejecting Dunnett
`test was used to perform pairwise comparison of HbA1c levels
`in liraglutide groups versus placebo. Required sample size was
`calculated as 32 per dose group, based on a two sided 5%
`significance level for the F test for linear contrast in the
`primary endpoint with 80% power. Randomization of a
`minimum of 200 subjects was planned, allowing for an
`expected dropout rate of 15%.
`
`3.
`
`Results
`
`A total of 372 patients were screened, of whom 226 were
`randomized to study cohorts and exposed to treatment
`
`(Table 1). Mean HbA1c was 8.30%; baseline characteristics
`were comparable across study groups (Table 1). Sixteen
`patients withdrew from the study, the largest number from
`the groups allocated to placebo (eight withdrawals).
`Mean HbA1c was reduced by all liraglutide dose levels
`relative to placebo (Fig. 1), the magnitude of the treatment
`effect ranging from 0.79 percentage points in the 0.1 mg/day
`group to 1.85 percentage points with 0.9 mg/day liraglutide
`(Table 2). HbA1c levels below ?.CJ% were achieved by 22%, 43%,
`62% and 75% of patients receiving liraglutide, 0.1, 0.3, 0.6 and
`0.9 mg/day, respectively, and 9% of those receiving placebo.
`The proportion of patients achievingHbA1c <6.5% ranged from
`7% to 57% in the liraglutide dose groups, and 2% in the placebo
`group (Table 2). A monotonic dose response relationship was
`confirmed in HbA1c level, with p < 0.0001 for linear contrast.
`Pairwise comparison showed statistically significant differ
`ences in HbA1c levels versus placebo in all liraglutide groups at
`14weeks.
`Self monitored seven point plasma glucose profiles
`showed reductions in mean glucose levels across the day;
`mean AUC values for glucose across the seven point profiles
`were lower than placebo with all liraglutide doses above
`0.1 mg/day (Table 2). Dose response relationship was con
`firmed with p < 0.0001 for linear contrast. FPG levels also
`showed reductions with all liraglutide dose levels versus
`placebo ranging from 0.76 to 2.48 mmol/L, with p < 0.0001 for
`linear contrast. A reduction in FPG was already evident at the
`2 week visit, which was the first timepoint following start of
`li.raglutide administration.
`Glucose levels following a standard breakfast after 14
`weeks showed a significant dose response in glucose AUCo-
`
`Table 1 - Patient flow and baseline characteristics of treatment groups
`Screened
`372
`
`Screening failure
`Met ~ 1 exclusion criterion
`Failed ;::1 inclusion criterion
`Other
`
`Randomized
`
`Withdrew
`Adverse event
`Non compliance with protocol
`Ineffective therapy
`Other
`Completed
`
`100
`39
`15
`
`Placebo
`46
`
`1
`0
`1
`6
`38
`
`0.1mg
`45
`
`0.3mg
`46
`
`0.6mg
`45
`
`0.9mg
`44
`
`0
`1
`1
`0
`43
`
`0
`1
`1
`1
`43
`
`0
`0
`1
`0
`44
`
`1
`0
`0
`1
`42
`
`Age (years)
`Mean (S.D.)
`Gender: male/female (n/n)
`Body weight (kg), mean (S.D.)
`BM! (kg/m2), mean (S.D.)
`Duration of diabetes (years) , mean (S.D,)
`
`57.5
`(8.7)
`29/17
`62.78 (10.88)
`23.77 (2.63)
`7.48 (5.65)
`
`56.5
`(8.4)
`3V14
`64.82 (10.29)
`24.26 (277)
`7.15 (5.14)
`
`56.8
`(8.8)
`32/14
`62.42 (11.18)
`23.93 (3.09)
`6.78 (4.69)
`
`60.0
`(7.0)
`28/17
`61.97 (9.40)
`23.74 (2.78)
`8.87 (6.77)
`
`55.5
`(7.6)
`31/13
`62.36 (10.65)
`23.59 (3.04)
`7.62 (4.92)
`
`Diabetic complications
`Retinopathy (n)
`Nephropathy (n)
`Neuropathy (n)
`Other (n)
`
`Treatment includes OAD (n)
`
`10
`6
`12
`2
`
`22
`
`17
`9
`14
`0
`
`21
`
`13
`9
`12
`0
`
`22
`
`11
`8
`7
`1
`
`19
`
`5
`3
`11
`0
`
`20
`
`MPI EXHIBIT 1038 PAGE 3
`
`MPI EXHIBIT 1038 PAGE 3
`
`Apotex v. Novo - IPR2024-00631
`Petitioner Apotex Exhibit 1038-0003
`
`

`

`164
`
`DI AB E TE S R E SE AR C H AN D CLI N I C A L PRA CTI CE 81 (2008) 161 r 6 8
`
`9.0
`
`8.5 I
`
`8.0
`
`l 1.s
`
`(.) 7.0
`<(
`.0 6.5
`I
`6.0
`
`5.5
`
`0
`
`I
`
`Placebo
`
`0.1
`0.6
`0.3
`Liraglutide dose ( mg/day)
`
`0.9
`
`Fig. 1 - Dose-response for HbA1c with increasing doses of
`liraglutide. Data are means ± S.E. for HbA1c after 14 w eeks.
`
`3 h , with reductions versus placebo at all liraglutide dose levels
`and p < 0.0001 for linear contrast (Table 2). One hour PPG
`levels showed reductions with all liraglutide doses except for
`the lowest, while all liraglutide doses showed reductions in 2
`and 3 h PPG versus placebo (Fig. 2).
`Insulin levels during the 3 h following the standard
`breakfast were also increased by the t hree highest liraglutide
`dose levels (Table 2), also showing dose response (p < 0.0001
`
`for linear contrast). This was accompanied by significant dose
`relationship in beta cell function (HOMA B), with increases
`versus placebo at 0.3, 0.6 and 0.9 mg/day liraglutide (also
`p < 0.0001 for linear contrast). Pro insulin:insulin ratio and
`pro insulin:C peptide ratios were decreased in all liraglutide
`dose groups {p 0.0008 and p < 0.0001 for linear contrast,
`respectively). There was no evidence of change in insulin
`resistance (HOMA R).
`No relevant changes in body weight occurred during the
`study, a reduction of 0.95 kg (from baseline 62 kg) occurring in
`the placebo group and changes in the liraglutide groups
`ranging from +0.13 kg to 0.48 kg versus baseline (Table 2).
`No major hypoglycemic events were reported in any study
`group during the trial. Likewise, no episodes of minor
`hypoglycemia occurred (symptomatic events confirmed by
`plasma glucose <3.1 mmoVL).
`Plasma levels of liraglutide remained steady in all dose
`groups from 2 to 14 weeks; mean (S.D.) plasma concentrations
`at 14 weeks were 1.98 (0.93), 4.31 (1.68), 9.42 {4.13) and 10.08
`{4.21) nmoVL in subjects receiving 0.1, 0.3, 0.6 and 0.9 m!q'day
`liraglutide, respectively. No treatment related increase in
`liraglutide antibodies occurred during the study. An assay
`specific normal range was defined from a phase 2 study in
`Caucasians with type 2 diabetes (21). In this trial of Japanese
`subjects, antibody levels were above t he pre defined assay
`cut off level in 11/226 patients at baseline and in 13/207
`
`Table 2 - Effect of liraglutide on measures of glycemia, beta-cell function and body weight
`Placebo
`0.1mg
`0.3mg
`0.6mg
`
`0.9mg
`
`8.43 (1.02)
`8.52 (1.23)
`
`8.50 (0.84)
`7.78 (0.91)
`0.79 ( 1.08,
`0.50)
`
`HbA,c (%), mean (S.D.)
`Baseline
`Week 14
`Liraglutide placebo,
`mean (95% CQ
`Patients achieving HbA1c levels at week 14, n (%)
`< 5.8%
`0 (0%)
`0(0%)
`~ 5.8%, < 6.5%
`1 (2.2%)
`3 (6.7%)
`;,:6.5%, < 7.0%
`3 (6.5%)
`7 (15.6%)
`34 (73.9%)
`33 (73.3%)
`~ 7.0%
`7 point SMPG profile: AUC(7 point) (mmoV(L h)), mean (S.D.)
`Baseline
`170.8 (320)
`173.7 (32.0)
`Week 14
`165.6 (35.1)
`153.1 (30.3)
`Liraglutide. placebo,
`13.7 ( 25.3,
`mean (95% CQ
`2.1)
`Fasting plasma glucose (mmoVL), mean (S.D.)
`Baseline
`9.99 (1.71)
`10.03 (1. 73)
`Week 14
`9.77 (2.47)
`9.03 (1.76)
`Llraglutide placebo,
`0.76 ( 1.40,
`mean (95% CQ
`0.11)
`1 h postprandial plasma glucose (mmoVL), mean (S.D.)
`Baseline
`16.91 (2.42)
`16.57 (261)
`Week 14
`16.03 (2.72)
`15.00 (270)
`Llraglutide placebo,
`0.83 ( 1.81,
`mean (95% CQ
`0.15)
`2 h postprandial plasma glucose (mmoVL), mean (S.D.)
`Baseline
`15.97 (292)
`16.05 (3.04)
`Week 14
`14 .95 (3.49)
`13.57 (3.15)
`
`8.24 (0.92)
`7.17 (1.01)
`1.22 ( 1.50,
`0.93)
`
`8.21 (0.83)
`6.71 (0.92)
`1.64 ( 1.93,
`1.35)
`
`8.12 (0.98)
`6.45 (0.n)
`1.85 ( 2 14,
`1.56)
`
`1 (2.2%)
`12 (26,1%)
`7 (15.2%)
`23 (50.0%)
`
`164.7 (32.2)
`137.8 (29.0)
`25.0 ( 36.6,
`13.3)
`
`9.84 (1.73)
`8.42 (1.96)
`1.26 ( 1.89,
`0.61)
`
`16.08 (2.61)
`13.78 (2.87)
`1 .75 ( 2.73,
`0.76)
`
`6 (13.3%)
`18 (40.0%)
`4 (8.9%)
`15 (33.3%)
`
`9 (20.5%)
`16 (36.4%)
`8 (18.2%)
`9 (20.5%)
`
`167 .0 (35.1)
`123.5 (23.6)
`40.2 ( 51.9,
`28.6)
`
`1.55.5 (27. 7)
`116.8 (27.4)
`41.9 ( 53.8,
`29.9)
`
`9.84 (1.23)
`7.41 (1.23)
`2.27 ( 2.92,
`1.63)
`
`16.47 (2.99)
`12.99 (2.78)
`2.78 ( 3.74,
`1.81)
`
`9.34 (1.36)
`6.91 (122)
`248 ( 3.13,
`1.82)
`
`15.61 (289)
`11.39 (2.68)
`3.85 ( 4.84,
`2.86)
`
`14.n (2.74)
`11.48 (2.89)
`
`16.01 (3.78)
`10.74 (294)
`
`14.42 (3.44)
`8.89 (3.16)
`
`F test for
`linear
`contrast
`
`p < 0.0001
`
`Not applicable
`
`p < 0.0001
`
`p < 0.0001
`
`Not tested
`
`MPI EXHIBIT 1038 PAGE 4
`
`MPI EXHIBIT 1038 PAGE 4
`
`Apotex v. Novo - IPR2024-00631
`Petitioner Apotex Exhibit 1038-0004
`
`

`

`DI ABE T ES RESEAR CH AN D CLIN IC AL PRAC TI CE 8 r (200 8) 1 6 r r6 8
`
`165
`
`Table 2 (Continued)
`
`Placebo
`
`0.1mg
`
`0.3mg
`
`0.6mg
`
`0.9mg
`
`liraglutide placebo,
`mean (95% CI)
`
`1.43 ( 2.58,
`0.29)
`
`2.83 ( 3.98,
`1.68)
`
`4.22 ( 5.35,
`3.09)
`
`5.23 ( 6.39,
`1.07)
`
`3 h postprandial plasma glucose mmol/L), mean (S.D.)
`Baseline
`13.86 (3.09)
`13.34 (2.91)
`Week 14
`12.57 (3.10)
`11.24 (2.68)
`liraglutide placebo,
`1 .56 ( 2.53,
`mean (95% CI)
`0.58)
`
`AUC(0 3 h) plasma glucose (mmol/(L h)), mean (S.D.)
`Baseline
`44.35 (6.88)
`44.50 (7.19)
`Week 14
`41.86 (8.19)
`38.64 (7.40)
`liraglutide placebo,
`3.30 ( 5.88,
`mean (95% CI)
`0.73)
`
`AUC (0 3 h) insulin (µ.U/(mL h)), mean (S.D.)
`Baseline
`58.84 (32.47)
`54.12 (30.98)
`Week 14
`59.38 (33.72)
`64.23 (33.17)
`liraglutide placebo,
`10.69 ( 6.00,
`mean (95% CI)
`27.39)
`
`Beta cell function (HOMA 13) ((µU/IllQ/(mg/dL)) , mean (S.D.)
`Baseline
`21.45 (14.75)
`19.36 (13.34)
`Week 14
`22.34 (21.77)
`27.56 (19.39)
`liraglutide placebo,
`7.79 ( 0.11,
`mean (95% CI)
`15.70)
`
`Insulin resistance (HOMA R), mean (S.D.)
`Baseline
`2 78 (1. 90)
`Week 14
`2.35 (1.33)
`liraglutide placebo,
`mean (95% CI)
`
`2.68 (1.69)
`2 .82 (1.98)
`0.54 ( 0.15,
`1.23)
`
`Pro insulin:insulin ratio ((pmol/L)/(µU/mL)), mean (S.D.)
`Baseline
`1 .56 (1.32)
`1.52 {1.17)
`Week 14
`1.58 (1.25)
`1.04 (0.66)
`liraglutide placebo,
`0.53 ( 0.87,
`mean (95% CI)
`0.20)
`
`Pro insulin:C peptide ratio ((pmol/L)/(ng/mL)), mean (S.D.)
`Baseline
`428 (2. 90)
`3.98 (2.45)
`Week 14
`3.87 (2.32)
`3.22 (2.35)
`liraglutide placebo,
`0.54 ( 1.25,
`mean (95% CI)
`0.18)
`
`12.49 (2.96)
`9.39 (2.86)
`2.79 ( 3.77,
`1.82)
`
`42.02 (6.72)
`34.16 (7.12)
`6.42 ( 9.01,
`3.84)
`
`71.69 (48.06)
`95.29 (69.12)
`19.69 (3.01,
`36.36)
`
`23.97 (18.51)
`36.82 (28.16)
`11.40 (3.58,
`19.21)
`
`3.02 (2.00)
`3.34 (3.12)
`0.83 (0.15,
`1.51)
`
`1.44 (1.03)
`1.04 (0.77)
`0.50 ( 0.83,
`0.17)
`
`4.05 (2.28)
`294 (2.08)
`0.84 ( 1.55,
`0.13)
`
`13.52 (3.48)
`8.13 (228)
`4.53 ( 5.49,
`3.57)
`
`12.27 (2.92)
`6.86 (1.88)
`5.23 ( 6.21,
`4.26)
`
`44.18 (8.54)
`31.46 (6.39)
`10.30 ( 12.85,
`7.76)
`
`40.84 (7.85)
`27.12 (6.41)
`12.80 ( 15.40,
`10.20)
`
`60.53 (32.04)
`102.56 (78.51)
`41.34 (25.12,
`57,55)
`
`22.06 (13.69)
`44.06 (30.27)
`20.99 (13.17,
`28.80)
`
`293 (1.63)
`261 (1.81)
`0.14 ( 0.54,
`0.82)
`
`1.50 (127)
`0.86 (0.65)
`0.69 ( 1.02,
`0.36)
`
`4.16 (3.56)
`241 (1.47)
`1.40 ( 2 11,
`0.69)
`
`63.83 (45.73)
`88.97 (67.74
`23.02 (6.52,
`39.52)
`
`22.99 (18.62)
`44.04 (32.44)
`19.82 (11.96,
`27.67)
`
`2.58 (1.86)
`2.05 (1.47)
`o.v ( 0.85,
`0.52)
`
`1.51 (1.01)
`1.00 (0.94)
`0.56 ( 0.89,
`0.23)
`
`3.60 (1.97)
`2.28 (1.53)
`1.32 ( 2.04,
`0.61)
`
`F test for
`linear
`contrast
`
`Not tested
`
`Not tested
`
`p < 0.0001
`
`p < 0.0001
`
`p < 0.0001
`
`p = 0.3521
`
`p = 0.0008
`
`p < 0.0001
`
`Body weight (kg), mean (S.D.)
`Baseline
`6200 (10.97)
`Week 14
`61.05 (10.89)
`liraglutide placebo,
`mean (95% CI)
`Values for baseline and week 14 are mean (S.D.). Liraglutide placebo differences (at week 14) were calculated using an ANOVA model with dose
`group and pre treatment as fixed effects and baseline value as covariate. SMPG, self measured plasma glucose.
`
`61.48 (10.55)
`61.00 (11.07)
`0.46 ( 022, 1.14)
`
`p = 0.2481
`
`64.26 (10.46)
`64.21 (10.67)
`0.87 (0.19, 1.55)
`
`61.54 (10.98)
`61.67 (11.39)
`1.08 (0.41, 1.75)
`
`61.52 (9.46)
`61.42 (9.68)
`0.84 (0.16, 1.51)
`
`patients at week 15 (post treatment). No subject's antibody
`levels at week 15 exceeded the maximum level recorded at
`baseline.
`A total of 154 patients (68%) experienced adverse events
`during the study, most commonly infections and gastro
`intestinal disorders. Distribution of adverse events was similar
`across treatment groups for infection/infestation disorders;
`incidences of gastrointestinal disorders were placebo, 24%;
`liraglutide 0.1 mg/day, 18%; 0.3 mg/day, 15%; 0.6 mg/day, 31 %;
`0.9 mg/day, 30%. Two patients, one receiving placebo, and one
`receiving 0.9 mg/day liraglutide, withdrew due to adverse
`
`events (abdominal discomfort and viral gastroenteritis; upper
`abdominal pain, respectively).
`Seven patients entering the study experienced serious
`adverse events; in five cases these events occurred before
`receiving any study medication. The two remaining events
`concerned a subject with suspected papillary thyroid carci
`noma (in the 0.6 mg/day liraglutide group) and a subject who
`suffered an alcohol related fall (in the 0.9 mg/day liraglutide
`group). Neither event was considered by investigators to be
`related to study treatment and both patients completed the
`study.
`
`MPI EXHIBIT 1038 PAGE 5
`
`MPI EXHIBIT 1038 PAGE 5
`
`Apotex v. Novo - IPR2024-00631
`Petitioner Apotex Exhibit 1038-0005
`
`

`

`166
`
`d i a b e t e s r e s e a r c h a n d c l i n i c a l p r a c t i c e 8 1 ( 2 0 0 8 ) 1 6 1 1 6 8
`
`o Placebo
`.t. 0.1 mg/day
`• 0.3 mg/clay
`♦ 0.6 mg/day
`■ 0.9 mg/day
`
`___
`--
`
`2---__
`
`P ~;=:~~=:;
`----------- ~-----------
`---
`~---- -----~
`----- ----..
`--------
`
`.
`1>
`
`18
`
`16
`
`14
`
`12
`
`10
`
`8
`
`6
`
`4
`
`-:co
`0
`E
`E
`Q)
`(/)
`0
`()
`:J
`0)
`
`(v
`E
`(/)
`(v
`
`0::
`
`2
`o~~ - - - - - -~ - - - - -~ - - - - -~
`1 hour
`Before
`2 hours
`3 hours
`breakfast
`after breakfast
`
`Fig. 2 – Postprandial plasma glucose profiles following a
`standard breakfast after 14 weeks’ treatment with
`liraglutide. Data are means W S.E.
`
`4.
`
`Discussion
`
`In this study, we investigated efficacy and safety from a dose
`response perspective with four doses of liraglutide and
`placebo in Japanese patients with type 2 diabetes. We
`demonstrated significantly dose dependent improvements
`in key parameters of glycemia HbA1c, FPG, PPG, seven point
`self monitored plasma glucose
`and beta cell
`function
`following 14 weeks’ treatment with liraglutide. Treatment
`was weight neutral, no major or minor hypoglycemic episodes
`occurred, and there was no treatment related development of
`liraglutide antibodies during the trial.
`The magnitude of HbA1c reduction with liraglutide in this
`study is comparable to those reported in non Japanese
`subjects by Vilsbøll and colleagues, where reductions of up
`to 1.74% points were achieved from similar baseline HbA1c
`levels of 8.1 8.5% [14]. Madsbad and colleagues reported a
`0.74% point reduction in HbA1c with liraglutide, 0.75 mg/day,
`from a somewhat better controlled baseline of 7.4% [13].
`A clear dose response relationship for HbA1c reduction was
`shown in this study. All doses of liraglutide tested showed a
`degree of efficacy, with a plateau of dose response at 0.9 mg/
`day. A dose of 0.9 mg/day appears to be highly effective and, of
`the doses tested, offers the optimal efficacy/tolerability
`balance in Japanese patients with type 2 diabetes. Indeed,
`the reductions in HbA1c in this study allowed a large majority
`of patients receiving the higher doses of liraglutide to achieve
`satisfactory glycaemic control: in the 0.9 mg/day group 75% of
`patients achieved HbA1c <7.0% and more than 50% achieved
`HbA1c <6.5% (‘good’ or ‘excellent’ categories in current Japan
`Diabetes Society guidance) without safety concerns or limited
`tolerability [22]. As in previous studies with liraglutide in non
`Japanese subjects [7,12,13,21], the risk of hypoglycemia is low
`with this agent, with no confirmed minor or major hypogly
`cemic episodes reported in this study. This combination of
`efficacy and hypoglycemic tolerability is highly encouraging.
`Previous longer term studies with liraglutide, largely in
`Caucasians, have reported significant weight
`reduction
`[7,12,14]. In the present study, body weight remained stable.
`
`This is likely to reflect the very different initial characteristics
`typical for Japanese patients: initial BMI values in the studies
`by Feinglos et al. and Nauck et al. were 34.5 kg/m2 and 31.3
`32.4 kg/m2, respectively, compared with 23.9 kg/m2 in our
`Japanese cohort [7,12].
`Japanese and Caucasian patients also differ, typically, in
`the nature of their type 2 diabetes. Japanese subjects have
`greatly impaired first phase insulin secretion from an early
`stage in type 2 diabetes, which accelerates the onset of
`postprandial hyperglycemia, whereas in some other groups
`type 2 diabetes is a disease of profound insulin resistance [23
`26]. The improvements in overall metabolic control as well as
`in PPG levels with liraglutide in Japanese individuals are
`therefore welcome and consistent with previous data with
`liraglutide in non Japanese subjects [15]. Taken together,
`these data show that
`liraglutide is highly effective in
`populations where either increased insulin resistance or
`impaired insulin secretion is the main contributor to meta
`bolic derangement. The 31% reduction in postprandial glucose
`AUC seen in this study with 0.9 mg/day liraglutide compares
`favorably with the 23% reduction reported in Caucasian
`subjects by Juhl et al.
`[15]. Evidence from this study of
`improvements in beta cell function was also encouraging.
`Pro insulin:insulin ratio, a postulated indicator of insulin
`secretory dysfunction [27,28] was reduced by all doses of
`liraglutide versus placebo, as seen in previous studies with
`liraglutide [13]. Beta cell function (HOMA B) was improved by
`liraglutide, as seen by Madsbad et al.
`[13]; the lack of
`improvement in HOMA R is likely to reflect the lack of effect
`on body weight and the low initial levels of insulin resistance
`in these Japanese subjects.
`Liraglutide was well tolerated in this study, with adverse
`event profiles comparable to placebo and few withdrawals
`from treatment. No safety concerns were raised.
`In conclusion, we have shown that liraglutide, at doses of
`up to 0.9 mg/day, is effective in improving glycaemic control
`and well tolerated in Japanese patients with type 2 diabetes,
`with encouraging indications of beneficial effects on beta cell
`function. An agent that allows three quarters of a population
`of subjects with type 2 diabetes to achieve good glycaemic
`control, with a low risk of hypoglycemia and no weight gain, is
`well tolerated and requires only a single daily administration
`will be welcome in clinical practice.
`
`Conflict of interest statement
`
`We have a competing interest to declare. Professor Seino and
`Professor Kaku and currently on an advisory board for Novo
`Nordisk Pharma Ltd., Tokyo,
`Japan. Mads Rasmussen is
`employed by, and holds shares in, Novo Nordisk A/S,
`Copenhagen, Denmark.
`
`Acknowledgments
`
`This study was conducted as part of the development program
`for liraglutide, and was supported by Novo Nordisk Pharma
`Ltd., Japan. A list of participating investigators is presented in
`Appendix A.
`
`
`
`MPI EXHIBIT 1038 PAGE 6
`
`MPI EXHIBIT 1038 PAGE 6
`
`Apotex v. Novo - IPR2024-00631
`Petitioner Apotex Exhibit 1038-0006
`
`

`

`d i a b e t e s r e s e a r c h a n d c l i n i c a l p r a c t i c e 8 1 ( 2 0 0 8 ) 1 6 1 1 6 8
`
`167
`
`Appendix A. Investigators
`
`Takashi Sasaki, Director, Keijinkai Sasaki Hospital, Sap
`poro City, Hokkaido; Fuminobu Okuguchi, Director, Okuguchi
`Clinic of Internal Medicine, Sendai City, Miyagi; Kenichi
`Yamada, Director, Kenichi Yamada Clinic, Tagajo City, Miyagi;
`Hiroaki Seino, Director, Diabetic Centre, Ohta Nishinouchi
`Hospital, Koriyama City, Fukushima; Takeshi Osonoi, Direc
`tor, Iryo Houjin Kenseikai, Naka Memorial Clinic, Naka City,
`Ibaraki; Yukiko Onishi, Director of Clinical Research Division,
`The Institute for Adult Diseases Asahi Life Foundation
`Marunouchi Hospital, Tokyo; Yasuhiko Iwamoto, Director,
`Diabetic Centre, Tokyo Women’s Medical University School of
`Medicine, Tokyo; Kouichi Hirao, President, H.E.C Science
`Clinic, Yokohama City, Kanagawa; Rika Usuda, Director,
`Toyama Prefectural Central Hospital, Toyama City, Toyama;
`Toshinari Takamura, Assistant Professor, Kanazawa Univer
`sity Hospital Endocrinology and Metabolism Department,
`Kanazawa City Ishikawa; Tatsuhide Inoue, Director of Live
`hood Support, Shizuoka Kenritsu Sougo Hospital, Shizuoka
`City Shizuoka; Nobuyuki Abe, Director, Internal Medicine, ABE
`Hospital, Ohita City Ohita; Hideaki Jinnouchi, Sub Director,
`Diabetes Care Center,
`Jinnouchi Clinic, Kumamoto City,
`Kumamoto; Shinji Taneda, Director, Manda Memorial Hospi
`tal, Sapporo City, Hokkaido; Masato Odawara, Professor,
`Tokyo Medical University Hospital, Tokyo; Keiko Arai, Direc
`tor, Arai Clinic, Yokohama City, Kanagawa; Hiroshi Tanaka,
`Dept. Chief Director, Social
`Insurance Chukyo Hospital
`Endocrinology and Metabolism Department, Nagoya City,
`Aichi; Syuji Nakamura, Director, Iryo Houjin Shadan Kowakai
`Heiwadai
`Hospital, Miyazaki City, Miyazaki;
`Hiroki
`Yokoyama, Director,
`Jiyugaoka Clinic,
`Internal Medicine,
`Obihiro City, Hokkaido; Takashi Sasaki, Director, Kashiwa
`Hospital,
`Jikei University School of Medicine Diabetes,
`Metabolism and Endocrinology Dept., Kashiwa City, Chiba;
`Masami Nemoto, Chief Director, Jikei University School of
`Medicine Diabetes, Metabolism and Endocrinology Dept.,
`Tokyo; Takashi Kadowaki, Professor, University of Tokyo
`Hospital Department of Metabolic Diseases, Tokyo; Kiyokazu
`Matoba, President, Matoba Diabetes Clinic, Ebina City, Kana
`gawa; Kunikazu Kondo, Director, Aichi Prefectural Welfare
`Federation of Agricultural Co operative Associations Anjo
`Kosei Hospital, Anjo City, Aichi; Hiroshi Sobajima, Internal
`Department Director, Ogaki Municipal Hospital Department of
`Diabetology and Nephrology, Ogaki City, Gifu; Yutaka Noto,
`Director, National Hospital Organisation Kanazawa Medical
`Center, Kanazawa City, Ishikawa; Yukihiro Nagai, Director,
`Nagai Internal Medicine Clinic, Kanazawa City,
`Ishikawa;
`Akira Okada, Director, Okada Clinic, Fukuoka City, Fukuoka;
`Yoshifumi Yokomizo, Director, Iryo Houjin Yokomizo Internal
`Medicine Clinic, Kitakyusyu City Fukuoka; Shigeru Miyazaki,
`Internal Depart