Articles —.
`
`Lancet 2009; 373: 473-81
`Published Online
`September 25, 2008
`D01:10.1016/50140-
`6736(08)61246-5
`
`See Comment page 438
`
`*Members listed at end of paper
`
`Baylor College of Medicine,
`Houston,TX, USA
`(A Garber PhD); U niversity of
`Caliornia at San Diego School of
`Medicine, San Diego, CA, USA
`(R Henry MD); Georgetown
`University School of Medicine,
`Hyattsville, MD, USA
`(R Ratner, MD); Hospital
`U niversitario, Monterrey,
`Mexico
`(PAGarcia-Hernandez, MD);
`Instituto Mexicano de
`Investigacion Clinica, Col Roma
`Mexico, Mexico
`(H Rodriguez-Pattzi MD);Centro
`de Atencion a Pacientes
`Osteosol-COMOP, Hipodromo,
`Mexico (I Olvera-Alvarez MD);
`Novo Nordisk Incorporated,
`Princeton, NJ, USA
`(P M Hale, PhD); Novo Nordisk
`A/S, Denmark
`(M Zdravkovic MD); Atlanta
`Diabetes Associates, Atlanta,
`GA, USA (A Bode MD)
`
`Correspondence to:
`Dr Alan Garber, Baylor College of
`Medicine, 1709 Dryden Road,
`Suite 1000, Houston,TX77030,
`USA
`agarber@bcm.tmc.edu
`
`Liraglutide versus glimepiride monotherapy for type 2
`diabetes (LEAD-3 Mono): a randomised, 52-week, phase III,
`double-blind, parallel-treatment trial
`Alan Garber, Robert Henry, Robert Ratner, Pedro A Garcia-Hernandez, Hiromi Rodriguez-Pattzi, Israel Olvera-Alvarez, Paula M Hale,
`Milan Zdravkovic, Bruce Bode, for the LEAD-3 (Mono) Study Group*
`
`Summary
`Background New treatments for type 2 diabetes mellitus are needed to retain insulin—glucose coupling and lower the
`risk of weight gain and hypoglycaemia. We aimed to investigate the safety and efficacy of liraglutide as monotherapy
`for this disorder.
`
`Methods In a double-blind, double-dummy, active-control, parallel-group study, 746 patients with early type 2 diabetes
`were randomly assigned to once daily liraglutide (1.2 mg [n=251] or 1.8 mg [n=247]) or glimepiride 8 mg (n=248) for
`52 weeks. The primary outcome was change in proportion of glycosylated haemoglobin (HbA). Analysis was done by
`intention-to-treat. This trial is registered with ClinicalTrials.gov, number NTC00294723.
`
`Findings At 52 weeks, HbA,c decreased by 0.51% (SD 1.20%) with glimepiride, compared with 0.84% (1.23%) with
`liraglutide 1.2 mg (difference —0.33%; 95% CI —0.53 to —0.13, p=0.0014) and 1.14% (1.24%) with liraglutide 1.8 mg
`(-0.62; —0.83 to —0.42, p<0.0001). Five patients in the liraglutide 1.2 mg, and one in 1.8 mg groups discontinued
`treatment because of vomiting, whereas none in the glimepiride group did so.
`
`Interpretation Liraglutide is safe and effective as initial pharmacological therapy for type 2 diabetes mellitus and leads
`to greater reductions in HbA, weight, hypoglycaemia, and blood pressure than does glimepiride.
`
`Funding Novo Nordisk A/S.
`
`Introduction
`Type 2 diabetes mellitus is a progressive disease; many
`treatments work early in the course of disease but do not
`remain effective:1'2 Glucagon-like peptide 1 (GLP-1) stim-
`ulates glucose-dependent insulin secretion, suppresses
`glucagon secretion, and moderates appetite by delaying
`gastric emptying and reducing hunger.' Endogenous
`GLP-1 has a very short half-life (1.5 min) because of rapid
`degradation by dipeptidyl peptidase 4,' which restricts its
`therapeutic usefulness. Liraglutide is an analogue of
`human GLP-1 with 97% homology to the endogenous
`protein4 and a half life of 13 h, which gives it a
`pharmacokinetic profile suitable for once daily treatment'
`Liraglutide restores glucose-dependent insulin secretion
`after one injection in patients with type 2 diabetes
`mellitus.6 In a 14 week monotherapy trial,7 treatment with
`liraglutide produced substantial and clinically significant
`reductions
`in
`fasting and postprandial glucose
`concentrations and glycosylated haemoglobin (HbA,,),
`resulted in moderate weight loss, and had a very low risk
`of hypoglycaernia. Com mon side-effects of liraglutide
`treatment include gastrointestinal side-effects, such as
`nausea, diarrhoea, and vomiting.
`We investigated the safety and efficacy of two doses of
`liraglutide versus glimepiride over 52 weeks for treatment
`of type 2 diabetes mellitus. We studied patients thought
`to be in the early stages of disease because they were
`either drug-naive, treated with lifestyle modifications, or
`
`had failed to achieve control with a single oral drug at
`less than 50% of maximum approved dose.
`
`Participants and study design
`Participants were aged 18-80 years, had body-mass index
`of 45 kg/m2 or less, and were diagnosed with type 2
`diabetes mellitus. Eligible patients had been treated with
`diet and exercise (36.5% of patients randomised) or up
`to half the highest dose of oral antidiabetic drug
`monotherapy (63.5%) including sulphonylureas, megliti-
`nides, aminoacid derivatives, biguanides, a-glucosidase
`inhibitors, and thiazolidinediones (1500 mg metformin
`or 30 mg pioglitazone were allowed) for at least 2 months.
`Patients had a screening HbA, value of 7-11% if treated
`with diet and exercise or 7-10% with oral antidiabetic
`monotherapy.
`Exclusion criteria were insulin treatment during the
`previous 3 months (except short-term treatment for
`intercurrent illness), treatment with systemic cortico-
`steriods, hypoglycaemia unawareness or recurrent severe
`hypoglycaemia, and impaired liver function (aspartate
`aminotransferase or alanine aminotransferase con-
`centrations >>-2.5 times upper normal range). Local
`institutional review boards approved the protocol, and all
`patients provided written informed consent before
`initiation of any trial-related activities. The study was
`done in accordance with the Declaration of Hclsinki8 and
`
`www.thelancet.com Vol 373 February 7, 2009
`
`473
`
`MPI EXHIBIT 1097 PAGE 1
`
`

`

`Articles
`
`Good Clinical Practice guidelines. This trial is registered
`with ClinicalTrials.gov, number NTC00294723.
`This trial was a 52-week, phase III, multicentre (126 sites
`in the USA and 12 sites in Mexico), double-blind,
`double-dummy, active-control, parallel-group
`study.
`Patients were randomly assigned (1:1:1) to receive once
`daily subcutaneous liraglutide 1.2 mg or 1.8 mg or once
`daily oral glimepiride 8 mg, and stratified by baseline
`diabetes treatment (diet and exercise vs oral antidiabetic
`
`746 patients randomised
`
`1 patient withdrawn for
`non-compliance
`
`745
`
`1 246 patients exposed to
`
`liraalutide 1.8 ma
`
`I
`
`248 patients exposed to
`alimeoiride 8 ma
`
`74 patients withdrew
`18 adverse events
`9ineffectivetherapy
`11 non-compliance
`36 other
`
`96 patients withdrew
`15 adverse events
`25 ineffective therapy
`5 non-compliance
`51 other
`
`251 patients exposed to
`liraolutide 1.2 ma
`
`89 patients withdrew
`25 adverse events
`15 ineffective therapy
`11 non-compliance
`38 other
`
`162 patients completed
`
`173 patients completed
`
`152 patients completed
`
`142 patients per protocol
`
`I 154 patients per protocol
`
`130 patients per protocol
`
`Figure 1: Trial profile
`Analyses were done on the intention-to-treat population exposed to at least one dose of treatment. *Patient
`withdrawn from liraglutide 18 mg group before exposure.
`
`Liraglutide 1.2 mg Liraglutide 1.8 mg Glimepiride 8 mg
`
`Randomised (ITT population)
`
`251
`
`Men
`Age (years)
`
`Race
`White
`
`Black
`
`Asian
`
`Other
`
`Hispanic or Latin American ethnicity
`Body mass index (kg /m')
`
`Weight (kg)
`Duration of diabetes (years)
`
`Prestudy treatment
`Diet and exercise
`
`Oral antidiabetic monotherapy
`
`HbA, (%)
`Fasting plasma glucose (mmol, L)
`
`Postprandial plasma glucose (mmoliL)
`Systolic blood pressure (mm Hg)
`
`Diastolic blood pressure (mm Hg)
`
`117(47%)
`53 7 (11.0)
`
`200(80%)
`34 (14%)
`5 (2%)
`12 (S%)
`81 (32%)
`33 2 (5.6)
`92 5 (19.2)
`52 (5 5)
`
`91 (36%)
`
`160(64%)
`8.3%(1.0%)
`9.3 (2.6)
`11.3 (2.4)
`1276 (14 3)
`785 (8.3)
`
`247
`
`121(49%)
`52.0 (108)
`
`186(75%)
`
`30 (12%)
`
`12 (6%)
`19 (7%)
`
`87(35%)
`32 8 (6 3)
`9 .8 (20 7)
`5.3 (5.1)
`
`87(35%)
`160 (65%)
`
`8.3%(1.1%)
`9.5 (2.6)
`11.4 (2.5)
`128 1(13 9)
`78.8 (34)
`
`248
`
`133 (54%)
`53 4 10.9)
`
`197(77%)
`
`30(12%)
`
`9 (4%)
`
`7 (7%)
`93(38%)
`33 2 (5.6)
`93.419.2)
`56 (5.1)
`
`94(38%)
`
`154(62%)
`
`8 4%(1.2%)
`9.5 (2.6)
`11.4 (2.7)
`1300 (161)
`79.5 (8.6)
`
`Data are mean (SD) or n (%) unless otherwise noted ITT=intention to treat.
`
`Table 1: Demographic and baseline characteristics
`
`monotherapy). Previous treatment with oral antidiabetic
`drugs was discontinued at randomisation. After random-
`isation, patients underwent forced titration: doses of
`liraglutide were increased every week from 0.6 mg to
`1.2 mg to 1.8 mg and glimepiride (or placebo) was
`increased over 2 weeks (2 mg to 4 mg to 8 mg) . Glimepiride
`(active and placebo) was to be taken orally once daily in
`the morning before or with the first meal of the day.
`Liraglutide (active or placebo) was injected once daily at
`any time of day in the upper arm, abdomen, or thigh with
`a prefilled pen injection device with 30 gauge or 31 gauge
`needle. Participants were encouraged to inject liraglutide
`at the same time each day. Doses of study drugs were
`maintained for 52 weeks, including the titration period.
`Randomisation was done with telephone-based or
`web-based systems. Participants were randomly assigned
`to the lowest available number. Recruitment began on
`Feb 7, 2006, with the last patient visit for this portion of
`the study on Nov 7, 2007. Participants completing the
`study could enrol, subject to eligibility, into a continuing,
`open-label extension period.
`The primary outcome was change in value of HbA1c from
`baseline to 52 weeks. Secondary outcomes included
`changes in body weight, fasting plasma glucose, self-
`measured eight-point plasma-glucose profiles (measured
`before each meal, 90 rain after the start of each meal, at
`bedtime, and at 0300 h), blood pressure, B-cell function
`(proinsulin to insulin ratio and two models of B-cell
`function: homoeostasis model assessment [HOMA]-B and
`HOMA-IR [insulin resistance]), fasting glucagon, and
`patients' reported assessment of quality-of-life. Laboratory
`analyses were done by central laboratories (MDS Pharma
`Services in Canada, Germany, and Switzerland). Parti-
`cipants used MediSense Precision Xtra/MediSense
`Optium (Abbott Diagnostics Inc, Abbott Park, IL, USA)
`glucose metres calibrated to plasma glucose to determine
`self-measured plasma glucose and recorded these values
`in diaries. Patients' reported outcome measures were
`developed by the validated Phase V Health Outcomes
`Information System (Phase V Technologies Inc, Wellesley
`Hills, MA, USA). A self-administered questionnaire was
`completed at randomisation and at weeks 28 and 52.
`Key safety assessments were tolerability (including
`nausea and other gastrointestinal adverse events),
`serum calcitonin, and hypoglycaemic episodes (defined
`as measured plasma glucose <3.1 mmol/L). We defined
`self-treated episodes of hypoglycaemia as minor and
`those that needed third-party assistance as major.
`Calcitonin concentrations were measured on the basis
`of C-cell tumour findings in the rodent carcinogenicity
`studies (Novo Nordisk, unpublished) with liraglutide.
`
`Statistical analysis
`163 participants were needed in each group to achieve
`85% of power to detect a difference of 0.4% in HbA1c,
`(SD of 1.2% and a two-sample one-sided a of 0.025).
`With the assumption of a 30% drop-out rate, we enrolled
`
`474
`
`www.thelancet.com Vol 373 February 7, 2009
`
`MPI EXHIBIT 1097 PAGE 2
`
`

`

`Articles -.
`
`•
`f
`
`Liraglutide 12 mg
`Liraglutide 1.8 mg
`Glimepiride 8.0 mg
`
`-----
`
`•
`
`-----
`
`--------
`
`----
`
`-------
`
`r
`
`9.00 --
`
`8 50
`
`8.0
`
`7.5
`
`7.0
`
`6.5
`
`6•o
`
`9.00
`
`8.50
`
`8.0
`
`= 7.0
`
`6.5
`
`6.0
`
`9.00
`
`8.50
`
`8.0
`
`= 7.0
`
`6.5
`
`6.0
`
`0
`
`4
`
`8
`
`12
`
`16
`
`20
`
`24
`Week
`
`28
`
`32
`
`36
`
`40
`
`44
`
`48
`
`52
`
`Figure 2: Efficacy of 9lycaemic control shown by HbA,, profiles
`(A) all particpants. (B) drug-naive participants. (C) participants previously treated with one oral antidiabetic drug.
`Data are mean (SE).
`
`between glimepiride and liraglutide 1.8 mg of -0.62%
`(95% CI -0.83 to -0.42, p<O.0001) and liraglutide 1.2 mg
`of -0.33% (-0.53 to -0.13, p=0.0014). Additionally, the
`reduction with liraglutide 1.8 mg was significantly
`
`702 subjects (234 per arm). Furthermore, this sample
`size would be large enough to detect differences in
`bodyweight between treatment groups (3% of difference
`in percent change from baseline).
`Analysis of efficacy outcomes was based on the
`intention-to-treat population, defined as participants
`exposed to at least one dose. Each endpoint was analysed
`with an ANCOVA model with treatment, country, and
`previous antidiabetic treatment as fixed effects, and
`baseline as the covariate. Missing data were imputed
`with last observation carried forward.
`To reduce type 1 error, we did hierarchical tests for
`non-inferiority and superiority of liraglutide. We also
`compared the two dose groups of liraglutide, although
`this analysis was in addition to the primary analysis of
`comparison to glimepiride. Other efficacy endpoints
`were analysed with the ANCOVA model described above.
`The proportions of patients achieving HbA,, targets
`(American Diabetes Association: <7%; International
`Diabetes Federation/American Association of Clinical
`Endocrinologists: <6.5%) were compared between
`treatments with a logistic regression model with treat-
`ment and baseline HbA,, as covariates. Hypoglycaemic
`episodes were analysed with a generalised linear model
`that included treatment and country as fixed effects.
`Other safety data were compared with descriptive
`statistics. Results are means (SD) unless otherwise
`noted.
`
`Role of funding source
`The study was funded by Novo Nordisk, the manu-
`facturer of liraglutide. In collaboration with the
`investigators, Novo Nordisk was responsible for the
`study design, protocol, statistical analysis plans and
`analysis, oversight, and reporting of results. Data were
`recorded at participating clinical centres and maintained
`by the sponsor. The LEAD-3 monotherapy study group
`had full access to the data. The authors had final respon-
`sibility for the decision to submit for publication.
`
`Results
`The three treatment groups were well balanced at base-
`line (figure 1, table 1). In the liraglutide treatment
`groups, most participants who withdrew did so because
`of other reasons or adverse events, whereas in the
`glimepiride group, other or ineffective therapy were the
`most common reasons for withdrawal. Mean baseline
`HbAk and fasting plasma glucose values were 8.2%
`and 9.5 mmol/L, respectively. Mean baseline weight
`was 92.6 kg and mean blood pressure was
`129/79 mm Hg.
`HbA,, values decreased from baseline by 0.84%
`(SD 1.23) with liraglutide 1.2 mg, 1.14% (1.24) with
`liraglutide 1.8 mg, and 0.51% (1.20) with glimepiride.
`Decreases in proportion of HbA,, in the liraglutide
`treatment groups were significantly greater than those in
`the glimepiride group, as shown by the differences
`
`www.thelancet.com Vol 373 February 7, 2009
`
`475
`
`MPI EXHIBIT 1097 PAGE 3
`
`

`

`Articles
`
`greater than that with liraglutide 1.2 mg (-0.29%;
`-0.50 to -0.09, p=0.(J046).
`Figure 2 shows mean HbA,, values over time for all
`participants (all points after baseline with last observation
`carried forward) and for those previously treated with diet
`
`Diet and exercise Oral antidiabetic monotherapy
`
`Liraglutide12 mg
`Liraglutide1.8 mg
`Glimepiride
`
`-119%(0.15)*
`-1.60%(0.15)t
`-0.88% (0.13)
`
`-0•47%(0.10)t
`-0.71%(0.09)#
`-0.17% (0.08)
`
`Data are mean (SE).Comparedwith change wit, glimepirlde*p=0.0234.
`tp-0.0215, and #peo•0001.
`
`Table 2: Decreases in HbA5 at 52 weeks for each trial intervention by
`previous treatment
`
`and exercise or monotherapy. HbA,, values generally
`decline over the first 8-12 weeks of treatment. From
`week 12 to week 52, HbA,a values increased slightly but
`significantly for participants treated with liraglutide
`1.2 mg (p=0.0071) and glimepiride (p-0.0006); however,
`HbA,, values had not changed significantly at week 52
`with liraglutide 1.8 mg (p-0.33). Participants previously
`treated with diet and exercise had greater decreases in
`HbA„ than did those switched from an oral antidiabetic
`drug to liraglutide (table 2). Participants who had never
`had any antidiabetic drugs and those previously treated
`showed significant decreases in HbA,, after starting
`liraglutide.
`At the end of the study, 28% of participants treated
`with liraglutide 1.2 mg and 38% treated with liraglutide
`1.8 mg reached the International Diabetes Federation/
`
`St
`S
`C a
`.ra.
`
`B
`
`70 1
`
`60-I
`
`58.3
`
`62.0
`
`50
`
`40
`
`C a
`
`30
`
`20
`
`10
`
`Liragludde
`1.2 mg
`746•%%
`
`L~ragluade
`18 mg
`717"%
`
`Glinrepiridr
`780%
`
`Lira9lvlide
`1.2 mg
`723%
`
`Lira9.ulide
`1-8 mg
`6.95%
`
`Glimreprr.de
`772%
`
`Figure 3: Participants achieving HbA targets of less than 7.0% (ADA) and less than or equal to6.5%(IDF/RACE)
`(A) all participants. (B) drug-naive participants. Percentages under each treatment group are mean final HbA„ values.
`
`00
`
`-0.4
`
`-0.6
`
`-06
`3'
`E E -10
`
`-1.2
`
`-1.4
`
`-1.6
`
`-1.8
`
`-2.0
`
`fl
`
`r
`
`fl
`
`u
`
`E
`
`==]
`029
`
`u
`
`-0.84
`
`-142
`
`p=00223
`
`P-0.0001
`
`Liragludde
`1.2 mg
`
`p=0027
`Liraglutide
`1.8mg
`
`Glimepiride
`8.0mg
`
`11.0
`
`10.5
`
`10.0
`
`9.5
`
`9.0
`
`E
`
`85
`80
`
`75
`
`7.0
`
`o 1'
`
`-4 - Liragutide 1.2 mg
`-*- Liragiudde 1.8 mg
`-♦-
`Glimepiride 8.0mg
`
`---+---+.---------+-------------------------------------------
`k -
`'---
`
`a
`
`0 4 8 12 16 20 24 28 32 36 40 44 48 52
`Week
`
`Figure4: Change in fasting plasma glucose (FPG)
`Data fro, laboratory values: difference from baseline to end-of-study with last observation cam led forward (A) and change over time (B). Data are nrean (SE).
`
`476
`
`www.thelancet.cor, Vol 373 February 7, 2009
`
`MPI EXHIBIT 1097 PAGE 4
`
`

`

`Articles -.
`
`American Association of Clinical Fndocrinologists
`target HbA,~ of 6.5% or less, compared with 16% in
`those on glimepiride (p=0.0025 and p<0.0001 for
`liraglutide 1.2 mg and 1.8 mg, respectively; figure 3).
`Although participants previously treated with diet and
`exercise had higher baseline HbA, values, a greater
`proportion achieved the HbA, target (figure 3) than did
`those previously treated with oral antidiabetic mono-
`therapy. Overall, compared with 28% in the glimepiride
`group, 43% of participants treated with liraglutide
`1.2 mg (p=0.0007) and 51% on liraglutide 1.8 mg
`(p<0.0001) reached the American Diabetes Association
`target HbA1c of less than 7.0% (figure 3). The proportion
`of participants achieving these targets with liraglutide
`1.8 mg was significantly higher than with liraglutide
`1.2 mg.
`Fasting plasma glucose concentrations (from labora-
`tory values) fell during the first 2 weeks after
`randomisation in the liraglutide groups and 4 weeks in
`the glimepiride group and thereafter remained stable.
`At the end of the study, fasting plasma glucose con-
`centrations were 8.65 mmol/L (SD 3.17), 8.25 mmol/L
`(2.75), and 9.27 mmol/L (2.99) in the liraglutide
`1.2 mg, liraglutide 1.8 mg, and glimepiride groups,
`respectively. Decreases in fasting plasma glucose from
`baseline for the liraglutide groups were significantly
`greater than those in the glimepiride group (figure 4).
`A greater proportion of participants in the liraglutide
`groups achieved the American Diabetes Association
`fasting plasma glucose target (5.0-7.2 mmol/L) than in
`the glimepiride group (37.6% and 41.4% vs 222% for
`the liraglutide 1.2 mg and 1.8 mg vs glimepiride group,
`respectively, p<0.0001 for each comparison).
`Postprandial glucose concentrations, from self-moni-
`tored eight-point plasma-glucose profiles, decreased in all
`three treatment groups (figure 5).
`HOMA-IR and fasting glucagon showed significant
`decreases with liraglutide but mean increases with
`glimepiride. Insulin resistance was reduced by 0.65
`absolute percentage points in the liraglutide 1.2 mg group
`and 1.35% in the 1.8 mg group, but increased 0.85% in
`the glimepiride group (p=0.0249 and p=0.0011 for
`liraglutide 1.2 mg and 1.8 mg, respectively; vs glimepiride).
`The proinsulin to insulin ratio and HOMA-B showed no
`significant differences between treatments. Table 3 shows
`the ratios of proinsulin to insulin at baseline and end-of-
`study. These results suggest an improvement in insulin
`resistance, which could indicate either the effects of a
`GT P-1 agonist on the hyperglucagonaemia of type 2
`diabetes or the limitations of HOMA methodology to
`assess B-cell function, or both.
`Participants in the liraglutide groups lost weight
`whereas
`those
`taking glimepiride gained weight
`(figure 6). Weight loss in the first 16 weeks was sustained
`throughout the 52-week study. To determine if persistent
`nausea was a factor in weight loss, participants were
`analysed by the number of days they had nausea (>7 days
`
`-1.0
`
`0 E
`E
`ft 1.5
`
`-2.0
`
`-2.5
`
`12.5
`
`12,0
`
`115
`
`E 11.0
`10.5
`10.0
`5
`9
`
`a
`a
`d B'S
`c 8.0
`
`N 7.5
`
`7.0
`
`o'(
`
`I
`
`Liraglutide 1.2 mg
`
`p=0.1616
`Liraglutide 1.8 mg
`
`Glimepiride 8.0 mg
`
`Baseline
`-0-
`--D-
`O
`
`Liraglutide1.2mg
`-- -Liraglutidel•8mg
`-4- Glimepiride 8.0 mg
`
`~~
`
`f
`
`v
`
`e
`
`Time
`
`Figure 5: Change in postprandial glucose
`Data from eight-point self-monitored plasma glucose (SMPG) values averaged for all three meals) from baselineto
`end-of-study with last observation carried forward (A); end-of-study eight-point SMPG (B). Data are mean (SE).
`
`or <7 days). Participants who had nausea for more than
`7 days (29 on liraglutide 1.2 mg, 38 on liraglutide
`1.8 mg, and nine on glimepiride) had a mean weight
`change of -3.24 kg, -3.39 kg, and -1.43 kg, compared
`with -1.85 kg, -2.26 kg, and +1.22 kg, respectively, for
`those with no nausea or up to 7 days of nausea (the
`differences were not significant for any treatment
`
`Liraglutide 1.2 mg
`
`Liraglutide 1.8 mg
`
`Glimepiride 8 mg
`
`Baseline
`
`End-of-study
`
`0.368 (0231)
`
`0.379 (0398)
`
`0375 (0223)
`0377 (0292)
`
`0373 (0376)
`0.418 (0242)
`
`Data are mean (SD).
`
`Table3: Proinsulinto insulinratios
`
`www.thelancet.com Vo1373 February 7, 2009
`
`477
`
`MPI EXHIBIT 1097 PAGE 5
`
`

`

`Articles
`
`A
`2
`
`rn 1
`
`0 0
`
`-1
`
`-: Li
`
`p=0.0001
`
`Li
`
`-3
`
`B
`
`98-,
`
`0 1
`
`Liraglutide 1.2 mg
`
`Liraglutide 1.8 mg
`
`Glimepiride 8.0 mg
`
`p=0.2584
`
`p=0.0001
`
`f
`f
`♦
`
`12 mg
`'Liraglutide 1.8 mg
`Glimepiride 8.0 mg
`
`0
`
`4
`
`8 12
`
`16
`
`20
`
`24 28 32
`Week
`
`36 40 44 48
`
`52
`
`Figure6: Change in bodyweight
`Changefrom baseline to end-of-study with last observation carried forward (A) and change over time (B). Data are
`mean (SE).
`
`group). Patients randomly assigned to liraglutide 1.8 mg
`reported improved quality of life scoring for physical
`and emotional domains compared with those assigned
`to glimepiride (p=0.02). These improvements seemed
`largely to result from improvements in weight image
`and weight concern (p<0.01). These results have been
`noted in a separate report of this study.'°
`Systolic blood pressure fell by 07 mm Hg (SD 13.7)
`in the glimepiride group compared with 2.1 mm Hg
`(SD 14.2) in the liraglutide 1.2 mg group (p=0.2912)
`and 3.6 mm Hg (14.1) in the liraglutide 1.8 mg group
`(p<0.0118). Mean diastolic blood pressure fell slightly
`but not significantly for all treatment groups.
`No events of major hypoglycaemia (requiring tlrird-
`party assistance) occurred; 12% and 8% of participants
`in the liraglutide 1 .2 mg and 1 •8 mg groups, respectively,
`had minor hypoglycaemia (plasma glucose <3.1 mmol/L),
`compared with 24% in the glimepiride group. The rate
`of minor hypoglycaemia was significantly
`lower
`(p<O.0001)
`for both
`liraglutide
`treatment groups
`(0-30 and 0-25 events per year for liraglutide 1-2 mg
`
`and 1.8 mg, respectively, compared with 1.96 events per
`year for glimepiride).
`275% and 293% of participants in the liraglutide
`1.2 mg and 1.8 mg groups, respectively, reported nausea,
`compared with 8.5% in the glimepiride group (p<O. 0001
`for both comparisons). Nausea generally occurred early
`during treatment and less than 10% of participants in
`the liraglutide 1.8 mg group had this side-effect by
`week 4. 9.3%, 12.4%, and 3 - 6% of participants in the
`liraglutide 1.2 mg and 1.8 mg, and glimepiride groups,
`respectively, reported vomiting (p<0.0001 for both
`comparisons with glimepiride). 8.9% in the glimepiride
`group reported diarrhoea compared with 15.5% and in
`the liraglutide 1.2 mg group (p=0.0283) and 18.7% in
`the liraglutide 1.8 mg group (p=0.0017). 11(4%) of 251
`and 6 (2%) of 246 participants taking liraglutide 1.2 mg
`and 18 mg, respectively; withdrew from the study
`because of vomiting, nausea, or diarrhoea, compared
`with none of 248 in the glimepiride group. Five
`participants in the 1.2 mg and one in the 1.8 mg groups
`withdrew specifically because of vomiting. 'lable 4
`summarises all adverse events reported by more than
`5% of participants.
`Mean pulse rate increased by 3.2, 1.6, and 0.4 beats
`per min for liraglutide 1.2 mg and 1.8 mg, and
`glirepiride group, respectively (p=0.0027 and p=0.1422
`for liraglutide 1.2 mg and 1.8 mg, respectively, vs
`glimepiride). After 52 weeks, calcitonin concentrations
`did not differ in participants taking liraglutide and those
`taking glimepiride.
`Eight participants receiving liraglutide 1.8 mg had nine
`serious adverse events, 16 receiving liraglutide 1.2 mg
`had 18, and 13 receiving glimepiride had 17, including one
`death (an automobile accident classified as not related to
`treatment). Two participants had pancreatitis, one after 197
`(liraglutide 1.2 mg) and another after 333 (liraglutide
`1.8 mg) days of treatment. Both patients recovered; one
`continued in the study (1.2 mg). Despite confounding
`medical factors and the small number of events, a weak
`association between development of pancreatitis and
`treatment with liraglutide cannot be excluded.
`
`Discussion
`Treatment with liraglutide as monotherapy provided
`better glycaemic control for 52 weeks than did glimepiride,
`a traditional first-line secretagogue therapy for type 2
`diabetes mellitus, in participants previously treated with
`either diet and exercise or oral antidiabetic monotherapy.
`T.iraglutide improved glycaemic control with a low rate of
`hypoglycaemia.
`Liraglutide led to decreases in both fasting and
`postprandial plasma glucose, and its 13 h half-life
`makes it suitable for once-daily use. Control of both
`fasting and postprandial plasma is needed for patients
`with type 2 diabetes mellitus to achieve HbA,~ goals."
`Additionally, participants treated with liraglutide had
`significant weight loss and decreases in systolic blood
`
`478
`
`www.thelancet.com Vol 373 February 7, 2009
`
`MPI EXHIBIT 1097 PAGE 6
`
`

`

`Articles —.
`
`Liraglutide 1.2 mg (n=251)
`
`Liraglutide 18 mg (n=246)
`
`Glimepiride 8 mg (n=248)
`
`n (%)
`
`Events
`
`n (%)
`
`Gastrointestinal disorders
`
`Constipation
`
`Diarrhoea
`
`Flatulence
`
`Nausea
`Vomiting
`General disorders and administration site
`conditions
`
`Infections and infestations
`Influenza
`
`Nasopharyngitis
`Sinusitis
`Upper respiratory tract infection
`Urinary tract infection
`
`n (%)
`
`122 (49'.0)
`21(8%)
`39 (16%)
`4(2%)
`
`69(27%)
`31(12%)
`33 (13%)
`
`119 (47%)
`
`17(7%)
`17 (7%)
`15(6%)
`23(9%)
`20(8%)
`
`Injury, poisoning, and procedural complications
`
`22 (9%)
`
`Investigations
`Metabolism and nutrition disorders
`
`Musculoskeletal and connective tissue disorders
`Back pain
`
`Nervous system disorders
`
`Dizziness
`
`Headache
`
`Psychiatric disorders
`Respiratory, thoracic, and mediastinal disorders
`
`Skin and subcutaneous tissuedisorders
`
`Vascular disorders
`
`Hypertension
`
`16(6%)
`38 (15%)
`
`48(19%)
`14 (6%)
`56(22%)
`13 (5%)
`27 (11%)
`21(8%)
`21 (8%)
`
`23(9%)
`11 (4%)
`7 (3%)
`
`Events
`
`282
`
`24
`60
`
`4
`91
`35
`41
`
`207
`20
`
`18
`16
`28
`
`24
`26
`
`21
`46
`
`63
`16
`101
`
`18
`
`47
`25
`31
`
`26
`
`12
`
`7
`
`126 (51%)
`258 (11%)
`46 (19%)
`
`13(5%)
`72(29%)
`23 (9%)
`41 (17%)
`
`102 (41%)
`20(8%)
`
`9 (4%)
`13(5%)
`24(10%)
`10(4%)
`
`24(10%)
`
`23 (9%)
`
`35 (14%)
`46 (19%)
`11(5%)
`
`49(20%)
`16 (6%)
`18 (7%)
`21(9%)
`28 (11%)
`24 (10%)
`15 (6%)
`8 (3%)
`
`332
`32
`61
`
`14
`107
`32
`59
`
`184
`
`25
`10
`18
`30
`
`13
`27
`
`28
`42
`
`59
`11
`71
`
`18
`
`25
`21
`
`39
`26
`
`15
`8
`
`64 (26%)
`12 (5%)
`22 (9%)
`
`4(2%)
`21(8%)
`9(4%)
`37 (15%)
`
`Events
`
`139
`
`12
`
`34
`
`4
`28
`10
`44
`
`90 (36%)
`
`153
`
`9(4%)
`13 (5%)
`15(6%)
`14(6%)
`10 (4%)
`29(12%)
`18 (7%)
`28(11%)
`
`38(15%)
`
`11(4%)
`
`55(22%)
`13 (5%)
`23 (9%)
`14(5%)
`28 (11%)
`
`17(7%)
`
`17(7%)
`15 (6%)
`
`15
`14
`17
`21
`11
`
`33
`
`24
`30
`
`55
`11
`78
`
`14
`30
`
`17
`
`35
`19
`
`21
`
`17
`
`A treatment-emergent adverse event is defined as an even t occurring between first and last dose +7 days or starting before first dose with increasing severity during treatment.
`
`Table4: Treatment-emergent adverse events reported by more than S% of participants, by system organ class and preferred term
`
`study of patients naive to oral antidiabetic drugs,"
`pressure. The additional benefit of weight loss early in
`sitagliptin reduced HbA,, by 0.79% and 0.94% (sitagliptin
`the course of treatment might have long-term effects as
`doses of 100 mg and 200 mg, respectively) and fasting
`shown in the UKPDS2 and Look AHEAD12 studies.
`plasma glucose by 1.0 mmol/L and 1.2 mmol/L. Unlike
`Exenatide, another GLP-1 receptor agonist,
`is a
`synthetic exendin-4, with a half-life of 2.4 h, requiring GLP-1 receptor agonists, such as liraglutide and exenatide,
`twice daily administration before meals. A recent
`sitagliptin did not lead to weight change.
`24-week study" of patients naive to oral antidiabetic
`Traditional first-line therapy might not be appropriate
`treatment who initiated exenatide therapy reported
`for all patients. Metformin is poorly tolerated by about
`5% of patients" and is contraindicated for renal reasons,
`reductions in HbA,~ of 0.7% and 0.9% with 5 pg and
`10 µg, twice daily, respectively. Fasting serum glucose which affect almost 30% of patients within 15 years of
`declined by 1.0 mmol/L for both groups, and bodyweight diagnosis of type 2 diabetes mellitus.'' Exenatide and
`decreased by 2.8 kg and 3.1 kg, respectively. Two
`sitagliptin require dose adjustments in patients with
`previously published 30-week studies of exenatide added
`renal impairment,'9'2° and exenatide is contraindicated in
`to previous monotherapy (4 mg glimepiride'^ or 1500 mg patients with severe renal impairment. A recent
`metformin15) reported decreases in HbA, of 0.4-0.9% pharmacoldnetic study of Liraglutide reported no need
`and weight loss of 0.9-1.6 kg (+sulphonylurea), and
`for dose adjustment when comparing healthy participants
`1.6-2.8 kg (+metformin). Fasting plasma glucose with those with severe renal impairment.2'
`decreased by 0.3-0.6 mmol/I., with
`significant
`Similar concerns are evident for other treatments. In
`reductions in postprandial plasma glucose as shown by a
`the ADOPT (A Diabetes Outcome Progression Trial)
`standardised meal-tolerance test in the study with
`study,' glibenclamide treatment produced a 39% incidence
`metformin.
`of hypoglycaemia and had a higher dropout rate than
`The oral antidiabetic sitagliptin is an inhibitor of other therapies (44% vs 37% for rosiglitazone and 38% for
`dipeptidyl peptidase 4 given once daily. In a 24-week metformin), despite having the greatest initial reductions
`
`www.thelancet.com Vol 373 February 7, 2009
`
`479
`
`MPI EXHIBIT 1097 PAGE 7
`
`

`

`Articles
`
`in HbAL and fasting plasma glucose after 6 months of
`treatment. Durability during the 5-year study was not
`seen with glibenclamide; and HbA,~ control, as indicated
`by the Kaplan-Meier cumulative incidence, failed in
`34% of patients after 5 years (compared with 21% of those
`given metformin). The 4-year extension of the LEAD-3
`study, as well as other continuing long-term studies of
`liraglutide, will help us assess the durability of liraglutide
`treatment and identify subpopulations that best respond
`to liraglutide treatment.
`Liraglutide treatment as initial monotherapy is a safe
`and effective option for treatment for patients with type 2
`diabetes early in the course of disease. Improvement in
`key efficacy endpoints, such as HbA,~, fasting plasma
`glucose, and blood pressure happen quickly after
`initiation of liraglutide treatment. With liraglutide
`18 mg, the decrease in HbA,~ and weight remained
`stable through the 52 weeks for patients who had not had
`antidiabetic drugs. Because of
`the
`low
`rate of
`hypoglycaemia with liraglutide monotherapy, there is no
`greater need for glucose monitoring for safety concerns
`than with other treatments. The increased insulin
`secretion with liraglutide, being glucose-dependent,
`retains more physiological stimulus-secretion coupling
`between glucose and insulin than does a sulphonylurea
`that acts by potassium-channel closure and produces
`more hypoglycaemia than does a GLP-1 agonist. However,
`type