Protocol. This trial protocol has been provided by the authors to give readers additional information about their work.

Transcription

1 Protocol This trial protocol has been provided by the authors to give readers additional information about their work. Protocol for: Pi-Sunyer X, Astrup A, Fujioka K, et al. A randomized, controlled trial of 3.0 mg of liraglutide in weight management. N Engl J Med 2015;373: DOI: /NEJMoa

2 Table of contents amended protocol for trial conduct and execution Original protocol Statistical analysis plan Changes in the conduct of the trial or planned analyses

3 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 1 of 112 Global Substantial Protocol Amendment no. 6 to Protocol version no.1, dated 08-Jul-2008 Including amendments no. 1 dated 28-Aug-2008 and no. 3 dated 28-Apr-2009 Effect of liraglutide on body weight in non-diabetic obese subjects or overweight subjects with co-morbidities A randomised, double-blind, placebo controlled, parallel group, multicentre, multinational trial with stratification of subjects to either 56 or 160 weeks of treatment based on pre-diabetes status at randomisation Trial Phase: 3a Author: Author s name: XXXX Author s department: XXXX This confidential document is the property of. No unpublished information contained herein may be disclosed without prior written approval from. Access to this document must be restricted to relevant parties.

4 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 2 of 112 Table of Contents Page Table of Contents...2 Table of figures...4 Table of tables Introduction including Rationale for the Substantial Protocol Amendment Changes Frontpage: Table of contents: List of abbreviations: Section 1 Summary Section 2 Flow Chart Section 3 Introduction Section 4 Objectives and endpoints Section 5 Trial design Section 6 Trial population Section 7 Trial schedule Section 8 Methods and assessments Section 9 Trial supplies Section 10 Randomisation, breaking of blinded codes and interactive voice/web response system (IV/WRS) Section 11 Concomitant illnesses/medical history and concomitant medication Section 12 Adverse events and clinical technical complaints pregnancies Section 13 Case report forms Section 14 Monitoring procedures Section 15 Data management Section 16 Computerised systems Section 16 Evaluability of subjects for analysis Section 17 Statistical considerations Primary efficacy endpoints...84 The four primary efficacy endpoints consists of three co-primary endpoints describing weight change and one endpoint describing onset of diabetes:...84 Fasting body weight loss (defined as the change from week 0 to Week 56)...84 The proportion of subjects losing more thanat least 5% of baseline fasting weight (measured at Week 0)...85 The proportion of subjects losing more than 10% of baseline fasting weight (measured at Week 0)...85 Onset of type 2 diabetes in subjects with pre-diabetes at baseline evaluated at week Section 18 Ethics Section 19 Premature termination of the trial/ trial site Section 20 Protocol compliance Section 21 Critical documents...105

5 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 3 of Section 22 Responsibilities Section 23 Reports and publications Section 24 Retention of clinical trial documentation Section 25 Indemnity statements References Appendix A Approval of Substantial Protocol Amendment no. 6 Appendix B Agreement on Substantial Protocol Amendment no. 6

6 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 4 of 112 Table of figures Page Figure 2 1 Trial design diagram...11 Figure 8 1 Classification of hypoglycaemia...56 Table of tables Page Table 5 1 Treatment groups: Main trial (56 weeks) incl week FU period...22 Table 5 2 Treatment groups: Main trial (56 weeks) incl. 12 week off-drug observational period...23 Table 5 3 Treatment groups: Delayed-onset-of-type-2-diabetes part of the trial...23 Table 2 4 ATP-III criterion...93

7 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 5 of Introduction including Rationale for the Substantial Protocol Amendment In this substantial protocol amendment: Any new text is written in italic. Any text deleted from the protocol is written with a strike through. Following the recent approval in the US, Europe and other countries worldwide of liraglutide for the treatment of type 2 diabetes, it has been decided to initiate the two remaining confirmatory phase 3 trials with liraglutide 3.0 mg (SCALE -Obesity and Pre-diabetes, SCALE -Diabetes) to support a weight management indication. The first of the three confirmatory trials (SCALE - Maintenance), conducted in the US and Canada, was recently completed and is currently under reporting. To that end, the NN protocol (SCALE -Obesity and Pre-diabetes) has been carefully reviewed and updated in accordance with recommendations provided by the FDA as well as with current protocol template and procedures. The following significant changes have been implemented: Trial design: WHO diagnostic criteria have been replaced by ADA diagnostic criteria for diabetes and increased risk for diabetes ( pre-diabetes ), in accordance with recently issued recommendations (International Expert Committee Report: International Expert Committee Report on the Role of the A1C Assay in the Diagnosis of Diabetes, Diabetes Care 2009 & ADA Position Statement: Diagnosis and Classification of Diabetes Mellitus, Diabetes Care 2010). A minimum requirement of 1000 randomised subjects with pre-diabetes at baseline, to ensure that the frequency of pre-diabetes in the trial population is representative of that in the general overweight and obese population Follow-up regimen for subjects without pre-diabetes has been simplified: o All subjects will undergo a 12 week re-randomised treatment (Week 56 to Week 68) during which liraglutide placebo subjects will continue liraglutide placebo treatment and active-treated subjects will be re-randomised 1:1 to either continue active treatment ( on-drug ) or change to liraglutide placebo ( off-drug ) Treatment allocation will remain blinded to, Investigator and subject. o All subjects will attend a 2-week off-treatment follow-up visit (Week 70) Exclusion criteria:

8 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 6 of 112 Consistent with ADA diagnostic criteria, in addition to fasting plasma glucose (FPG) and 2- hr plasma glucose (PG) criteria, subjects with HbA 1c 6.5% at screening will be excluded from the trial Consistent with FDA requirement for GLP-1 analogues in development, subjects with a screening calcitonin value 50 ng/l, family or personal history of multiple endocrine neoplasia type 2 (MEN2) or familial medullary thyroid carcinoma (FMTC), or personal history of non-familial medullary thyroid carcinoma will be excluded from trial Medical History/Concomitant Illness: To ensure systematic collection and recording of relevant medical history/concomitant illness, in addition to standard entries, specific information related to signs or symptoms of diabetic complications, concomitant cardiovascular disease, history of gallbladder disease, and history of psychiatric disorders will be collected in the ecrf at Screening visit 1. Assessments/Endpoints (efficacy): Four (4) non-fasting visits will be replaced by fasting visits to increase the number of body weight and FPG assessments available for analysis of the 3 co-primary endpoints relating to body weight (fasting body weight loss, % of subjects losing at least 5% or more than 10% of baseline body weight) and the secondary endpoints related to glycaemic control (FPG, proportion of subjects with change in antidiabetic medication Urinary Albumin-to-Creatinine Ratio and PAI-1 have been added to list of endpoints Assessments/Endpoints (safety): Key adverse events will be adjudicated: o Death o Cardiovascular events according to Endpoints and Standardized Data Collection for Cardiovascular Outcomes Trials: Draft Recommendations, CDER July o Acute pancreatitis, or acute severe and persistent pain leading to suspicion of pancreatitis o Neoplasms o Thyroid disease requiring thyroidectomy Establishment of an external Calcitonin Monitoring Committee to provide recommendations to Investigator in case of elevated calcitonin values during trial conduct Subjects undergoing thyroidectomy will be consented to provide biological material (thyroid tissue, blood) for genetic testing as requested by FDA (please refer to description under Additional Subject Information/Informed Consent below) List of Medical Events of Special Interest has been updated Routine assessment of ECG implemented as requested by FDA ECGs will undergo central reading All hypoglycaemic episodes must be reported as AE as required by FDA

9 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 7 of 112 The Subject Information/Informed Consent forms have updated to reflect changes in this amendment. An additional Subject Information/Informed Consent for subjects who are subject to thyroidectomy during the trial has been added. Subjects will be consented to provide a thyroid tissue sample, and in case of confirmed thyroid C-cell pathology (i.e., hyperplasia or neoplasia), a blood sample to allow for genetic testing of activating mutations in RET. In subjects with confirmed thyroid C-cell pathology, it will be determined if RET is activated (phosphorylated at Y1062) in hyperplastic/neoplastic thyroid C-cells. These tests have been included at the request of FDA. The two flow charts Appendix E and Appendix F have been updated to reflect the changes in this amendment. The Statistical Considerations section has been updated to reflect changes in this amendment.

10 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 8 of Changes Changes throughout the document: Definition of fasting will be changed from i.e. having consumed only water since midnight to i.e., at least eight hours overnight fast without food and/or drink intake, except for water CRF will be changed to ecrf Sponsor will be changed to 2.1 Frontpage: ELIOT General Effect of Liraglutide In Obesity Treatment SCALE Obesity and Pre-diabetes Satiety and Clinical Adiposity- Liraglutide Evidence in Nondiabetic and Diabetic Subjects 2.2 Table of contents: Appendix E Flow chart: main trial including a 13 week follow-up/12 week off-drug observational period 12 week re-randomised treatment period Appendix F Appendix I Appendix J Flow chart: delayed-onset-of-type-2-diabetes trial (subjects with prediabetes) Calcitonin Monitoring Committee Screening of Calcitonin (CT Levels Medical Events of Special Interest (MESI) 2.3 List of abbreviations: CABG coronary artery bypass graft surgery ecrf electronic case report from FMTC familial medullary thyroid carcinoma MEN2 multiple endocrine neoplasia type 2 PAI-1 plasminogen activator inhibitor-1 PCI percutaneous coronary intervention

11 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 9 of 112 SIF TRIMm-Weight safety information form Treatment Related Impact measure - Weight 2.4 Section 1 Summary Secondary objective: To investigate the long term efficacy of 3.0 mg liraglutide versus liraglutide placebo on cardiovascular risk markers such as blood pressure, lipids, glucose parameters and the metabolic syndrome as well as effects on quality of life Trial design: This is a randomised, double-blind, placebo controlled, parallel group, multi-centre, multinational trial in non-diabetic obese subjects and overweight subjects with co-morbidities. Subjects will be randomised in a 2:1 manner to receive either 3.0 mg liraglutide or liraglutide placebo, and based on body mass index (BMI [kg/m 2 ]) and pre-diabetes status at randomisation subjects will be stratified to either 56 or 160 weeks of treatment. Treatment allocation for subjects with pre-diabetes will be double-blind for the first 56 weeks and single blind for the remaining 104 weeks (i.e. sponsor is unblinded, whereas Investigator and subjects remain blinded). Pre-diabetes is defined as: IFG (FPG 110 mg/dl / 6.1 mmol/l, and < 126 mg/dl / 7.0 mmol/l) or IGT (2 hr post OGTT plasma glucose 140 mg/dl / 7.8 mmol/l, and < 200 mg/dl / 11.1 mmol/l). This is a randomised, double-blind, placebo controlled, parallel group, multi-centre, multinational trial in non-diabetic obese subjects and overweight subjects with co-morbidities. Subjects will be randomised in a 2:1 manner to receive either 3.0 mg liraglutide or liraglutide placebo, and based on body mass index (BMI [kg/m 2 ]) and pre-diabetes status at randomisation subjects will be stratified to either 56 or 160 weeks of treatment (160 week treatment will only applicable to subjects with pre-diabetes at screening). Treatment will be stratified by body mass index (BMI [kg/m 2 ]) (BMI 30 kg/m 2, or BMI < 30 kg/m 2 ) and pre-diabetes status at baseline (Yes/No, see below for definition of pre-diabetes). Treatment allocation for subjects with pre-diabetes will be double-blind for the first 56 weeks (, Investigator, and subject blinded), after which time point sponsor is un-blinded (Investigator and subject remain blinded to treatment allocation). A subject will be classified as having pre-diabetes if falling within the following categories of increased risk for diabetes: Fasting plasma glucose 100 mg/dl ( 5.6 mmol/l) to 125 mg/dl (6.9 mmol/l) (IFG), 2-hour post-challenge (OGTT) glucose 140 mg/dl (7.8 mmol/l) to 199 mg/dl (11.0 mmol/l) (IGT), or HbA 1c % both inclusive.

12 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 10 of 112 The maximum overall duration of the main part of the trial from screening to follow-up will be weeks including the 13/12 12 week follow-up/off-drug observational re-randomised treatment period. For subjects enrolled in the delayed-onset-of-type-2-diabetes part of the trial the maximum total duration of the trial will be weeks. Trial population: 4800 subjects are expected to be screened in order to meet the randomisation target of minimum 3600 subjects in total, including a minimum of randomised subjects with pre-diabetes. If the number of subjects with pre-diabetes equals at least after randomisation of 3600 subjects, no further randomisation will be done. The screening and randomisation of subjects with pre-diabetes will be monitored closely throughout the recruitment period. If the number of subjects with pre-diabetes is lower than expected (approximately 30% of screened population), screening will continue with the purpose of identifying and enrolling the minimum number of subjects with pre-diabetes. If 3600 subjects have been randomised without reaching the minimum number of subjects with pre-diabetes (N = 1000), only subjects with pre-diabetes will be randomised. Approximately 900 subjects without pre-diabetes participating in the 56 week main trial will be included in a 13 week follow-up period starting at week 56. The subjects selected for participation in the 13 week follow-up period are the subjects first to complete the main trial (56 weeks). Subjects receiving placebo throughout the main part of the trial will continue on liraglutide placebo treatment for the follow-up period (approximately 300 subjects), whereas subjects receiving active liraglutide treatment (approximately 600 subjects) will be randomised 1:1 to either continue active liraglutide treatment (approximately 300 subjects) or switch from active liraglutide treatment to placebo (approximately 300 subjects). At week 56 all subjects who did not have pre-diabetes at baseline will continue in a blinded 12 week re-randomised treatment period followed by a 2 week off-treatment period for assessment of liraglutide antibody development. Subjects who have received treatment with liraglutide during the previous 56 weeks will be re-randomised 1:1 to either continue on liraglutide treatment or be switched to liraglutide placebo (for a blinded off active treatment evaluation). Subjects who had been randomised to liraglutide placebo will continue on liraglutide placebo during the 12 week re-randomised treatment period. The remaining (approximately 2100) non-pre-diabetic subjects completing the 56 week main trial will continue in a 12 week off-drug observational period starting at Week 56. This period will be unblinded. Key exclusion criteria: HbA 1c % (Screening visit 1) or FPG 126 mg/dl (7.0 mmol/l) (Screening visit 2) or 2 hr post-challenge (OGTT) plasma glucose 200 mg/dl (11.1 mmol/l) (Screening visit 2) Screening calcitonin 50 ng/l

13 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 11 of 112 Family or personal history of multiple endocrine neoplasia type 2 (MEN2) or familial medullary thyroid carcinoma (FMTC) Personal history of non-familial medullary thyroid carcinoma Assessments: Endpoints to support the The primary efficacy objectives include change from baseline in endpoints are body weight loss at 56 weeks as well as the proportion of subjects losing more than at least 5% and or more than 10% of baseline body weight at 56 weeks respectively. Furthermore, a time to event analysis of time from randomisation to diagnosis of type 2 diabetes in subjects who entered the trial having pre-diabetes will be performed. Moreover, the proportion of subjects who have converted to overt diabetes at 160 weeks is included as a co-primary endpoint for subjects who had pre-diabetes at baseline. The secondary efficacy endpoints include change from baseline in body are weight loss at 160 weeks of treatment in subjects with pre-diabetes, the proportion of subjects losing more than at least 5% and or more than 10% of baseline body weight at Wweek 160 and change from baseline in respectively, waist circumference, glucose related control parameters (HbA 1c, FPG, fasting insulin, C-peptide, 2 hr OGTT-derived parameters values, pre-diabetes status), vital signs blood pressure, cardiovascular biomarkers, lipids, urinary albumin-to-creatinine ratio (UACR) and Patient Reported Outcome (PRO)., and metabolic syndrome status (ATP-III.) at 56 weeks. Key assessments of safety are include physical examination, ECG, pulse, adverse events, haematology, biochemistry (including amylase, lipase and calcitonin), liraglutide antibodies and ratings based on mental health assessments questionnaires. 2.5 Section 2 Flow Chart Flow charts (Appendix E and F) has been updated to reflect the changes described in this amendment. Please refer to Appendices E and F. Figure 2 1 Trial design diagram

14 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 12 of Section 3 Introduction Glucagon-like peptide-1 (GLP-1) GLP-1 is an incretin hormone secreted from the L-cells in the lower gut in response to meal ingestion, which stimulates endogenous insulin secretion in a glucose-dependent manner. GLP-1 also lowers decreases blood glucagon levels and reduces delays gastric emptying by decreasing gastric motility and increasing satiety and subsequently reducing food intake, and has been shown to promote β-cell growth and proliferation in animal models (6-8).GLP-1 reduces appetite in lean and normal weight individuals, as well as in obese individuals (9-11), and has been shown to reduce body weight in people with type 2 diabetes (12). The underlying mechanism mediating the weight-reducing effects of GLP-1 is most likely a combination of effects on the gastrointestinal tract and the central nervous system, i.e. decreased gastric motility and reduced appetite/increased satiety with a subsequent reduction in food intake. The combination of these mechanisms makes GLP-1 receptor stimulation an attractive mechanism to investigate for weight management and diabetes prevention. 6, Liraglutide Liraglutide is a long-acting GLP-1 analogue under development by A/S. Compared to human GLP-1, liraglutide has a C16 fatty (palmitic) acid chain attached at position 26 (lysine) of the peptide, and has lysine at position 34 replaced by arginine. When administered subcutaneously, these structural modifications result in a compound with protracted kinetic properties suitable for once daily injection. In vitro receptor studies have shown that liraglutide is a selective, potent and full agonist of the cloned human GLP-1 receptor. The effects of liraglutide include delayed gastric

15 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 13 of 112 emptying, reduced sensation of hunger and increased satiety leading to decreased food intake and subsequent weight loss. As of 31 January 2008, 38 clinical trials with liraglutide have been completed (i.e. a final clinical trial report exists).this includes 26 clinical trials as part of an extensive clinical pharmacology programme. The clinical trials were conducted world-wide, with the majority being conducted in Europe (EU). A total of 6498 subjects were randomised and exposed to trial products in the 38 trials subjects were exposed to liraglutide, 1122 subjects were exposed to liraglutide placebo, and 1165 subjects exposed to an active comparator. The majority of the subjects exposed to liraglutide had type 2 diabetes (3328 subjects (79%)) and most of these were from the therapeutic confirmatory trials. Data from finalised trials have shown liraglutide to have a pharmacokinetic profile suitable for once daily administration, as evidenced by a relatively slow absorption ([tmax] =8-12 hours) with a terminal elimination half-life (t½) of approximately 13 hours. The pharmacokinetic profile is comparable between healthy subjects and subjects with type 2 diabetes. Results from a phase 2 trial in obese subjects without type 2 diabetes showed a dose dependent weight loss ranging from 3.8 to 7.8 kg with liraglutide doses of mg administered for 52 weeks (20 weeks as double blind and 32 weeks of open-label treatment). In addition to weight lowering, there was a decrease in systolic blood pressure and an impact on the number of subjects with pre-diabetes. In addition, of the approximately 30% of subjects that had pre-diabetes at baseline, 85 % had normal glucose tolerance after 20 weeks compared to 45% of placebo subjects. Of the 70% of subjects with normal glucose tolerance at baseline, 20% of placebo subjects developed pre-diabetes, whereas only 2-4% of liraglutide treated subjects had pre-diabetes after 20 weeks. The safety evaluation was favourable with the main tolerability finding being gastrointestinal side effects (liraglutide obesity Investigators Brochure (IB), 1st Edition, 2008). To explore the mechanism behind the observed weight loss with liraglutide, the effect of liraglutide on various body weight related parameters known to be affected by native GLP-1, have been investigated. Results from various trials have shown a minor delay in gastric emptying following administration of liraglutide ( 0.9 mg). Furthermore, liraglutide 1.8 mg seems to exert a mild suppression of hunger ratings and increase postprandial fullness, as indicated by some appetite rating endpoints. Liraglutide is generally well tolerated, and the Adverse Events (AEs) most frequently reported are from the gastro-intestinal system (e.g. nausea, diarrhoea and vomiting) and central and peripheral nervous system (e.g. headache). The gastrointestinal side effects are primarily reported within the first few weeks of treatment and can be mitigated using a stepwise dose-escalation.

16 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 14 of 112 Thyroid C-cell tumours were seen in the 104-week carcinogenicity studies in mice and rats. Based on mechanistic studies and data from the literature, the C-cell proliferative changes seen in rodents are associated with liraglutide. It is possible that chronic activation of the GLP-1 receptor present on thyroid C-cells by liraglutide can lead to hyperplastic and neoplastic changes in the thyroid. This may be associated with an increase in plasma calcitonin levels in rodents. A mode of action analysis is underway to assess the human relevance of the C-cell tumours observed in the rodent carcinogenicity studies. A comprehensive assessment of the human relevance of the findings is described in the liraglutide obesity Investigators Brochure (IB), 1st Edition, The C-cell tumors induced in mice and rats by dosing of liraglutide are thought to be caused by a nongenotoxic, specific receptor mediated mechanism. The thyroid tumours and C-cell proliferative changes induced by liraglutide in rodents is considered of uncertain relevance to human subjects. Few cases of acute pancreatitis (inflammation of the pancreas) presenting with persistent severe abdominal pain (usually accompanied by vomiting) have been reported with liraglutide and exenatide (a drug belonging to the same drug-class as liraglutide, and currently approved for the treatment of type 2 diabetes in the US and in Europe). If the investigator suspects acute pancreatitis, all suspected drugs should be discontinued until confirmatory tests have been conducted and appropriate treatment should be initiated. Subjects that are diagnosed with acute pancreatitis (as a minimum 2 of 3: characteristic abdominal pain, amylase and/or lipase >3x UNR or characteristic findings on CT/MRI), should be withdrawn from the study. Further information can be obtained in the liraglutide obesity Investigator Brochure (IB), 1st edition of 2008 or any updates hereof. Liraglutide is a long-acting GLP-1 analogue developed by A/S and recently approved for the treatment of type 2 diabetes in the US, EU, Japan, and other countries worldwide under the brand name Victoza (1.2 mg or 1.8 mg once-daily). Development of liraglutide 3.0 mg once-daily for weight management is currently ongoing. Compared to human GLP-1, liraglutide has a C16 fatty (palmitic) acid chain attached at position 26 (lysine) of the peptide, and has lysine at position 34 replaced by arginine. When administered subcutaneously, these structural modifications result in a compound with protracted kinetic properties suitable for once daily injection. In vitro receptor studies have shown that liraglutide is a selective, potent and full agonist of the cloned human GLP-1 receptor. The effects of liraglutide include delayed gastric emptying, reduced sensation of hunger and increased satiety leading to decreased food intake and subsequent weight loss. A total of 50 clinical trials with liraglutide have been completed (includes doses up to 3.0 mg). The trials were conducted world-wide, with most being conducted in Europe. Out of more than subjects, more than 7000 subjects were exposed to liraglutide (including 850 subjects treated for

17 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 15 of weeks in completed trials). A total of 986 obese subjects without type 2 diabetes (< 9% of all subjects) have been included to date in the obesity clinical development programme for liraglutide in the completed phase 2 trial NN and the completed phase 3a trial NN (of which 305 subjects were randomised to liraglutide 3.0 mg). A further 48 obese subjects were randomised in the ongoing phase 1 trial NN Data from finalised trials have shown liraglutide to have a pharmacokinetic profile suitable for one time daily administration, as evidenced by a relatively slow absorption ([tmax] =8-12 hours) with a terminal elimination half-life (t½) of approximately 13 hours. The pharmacokinetic profile is comparable between healthy subjects and subjects with type 2 diabetes. Results from a phase 2 trial in obese subjects without type 2 diabetes (NN ) showed a dose dependent weight loss ranging from 3.8 to 7.8 kg with liraglutide doses of mg administered for 52 weeks (20 weeks as double-blind and 32 weeks of open-label treatment (sponsor unblinded at 20 weeks). In addition to weight lowering, there was a decrease in systolic blood pressure and an impact on the number of subjects with pre-diabetes. In addition, of the approximately 30% of subjects who had pre-diabetes at baseline, 85% did not have pre-diabetes after 20 weeks compared to 45% of placebo subjects. Of the 70% of subjects without pre-diabetes at baseline, 20% of placebo subjects developed pre-diabetes, whereas only 2-4% of liraglutide-treated subjects had pre-diabetes after 20 weeks. The safety evaluation was favourable with the main tolerability finding being gastrointestinal side effects (please refer to liraglutide obesity Investigators Brochure (IB), 3rd Edition, 2010). The first of three confirmatory phase 3 trials within the liraglutide obesity development programme (Trial NN , or SCALE -Maintenance) was recently completed. Reporting is ongoing. The trial was a 56-week randomised, double-blind, placebo-controlled trial investigating treatment of liraglutide 3.0 mg versus placebo as an adjunct to diet and exercise in obese subjects or overweight subjects with co-morbidities who had already lost at least 5% of their body weight during a 4 to 12-week run-in period on a low calorie diet. The mean weight loss for subjects in the run-in period was approximately 6 kg. From a body weight of approximately 100 kg at randomisation, treatment with liraglutide for 56 weeks provided an additional estimated mean weight loss of 6.11% (-5.7 kg), compared to weight-neutrality or maintenance in the placebo group (+0.16 kg vs. baseline). Eighty one (81) percent of liraglutide-treated subjects maintained their run-in weight loss compared to 48% in the placebo group. Moreover, 51% of liraglutide-treated subjects lost additional 5% or more of their baseline body weight, compared to 22% in the placebo group. Treatment with liraglutide maintained and in some instances further improved beneficial effects on markers of glycaemic control and cardiovascular risk. Treatment with liraglutide was generally well-tolerated, with high completion rates in groups (75% in liraglutide group, 70% in placebo group). The number of withdrawals due to adverse events was evenly distributed between groups (8.5% in the liraglutide group vs. 8.6% in placebo group).

18 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 16 of 112 Serious adverse events were relatively uncommon, but were more frequent in liraglutide-treated subjects (4.2%) compared to placebo (2.4%). There were no events of pancreatitis or medullary thyroid cancer, and no treatment-related increases in blood calcitonin levels. Consistent with previous trials with liraglutide, the most commonly reported adverse events were from the gastrointestinal system, with nausea reported by 47% of subjects in the liraglutide group compared to 17% in the placebo groups, and vomiting by 17% vs. 2%, respectively. As in previous trials, the majority of events were reported in the first 6-8 weeks, were mild or moderate in severity and transient in nature. Please refer to liraglutide obesity Investigators Brochure, 3rd Edition, 2010, or any updates hereof. Patients treated with liraglutide should be advised of the potential risk of dehydration in relation to gastrointestinal side effects and take precautions to avoid fluid depletion. Few cases of acute pancreatitis (inflammation of the pancreas) presenting with persistent severe abdominal pain (usually accompanied by vomiting) have been reported with liraglutide and exenatide. Post-marketing surveillance identified at least 30 cases of pancreatitis with exenatide(15). However, a health services registry-based study found no increased frequency of pancreatitis among exenatide users (16). If the investigator suspects acute pancreatitis, all suspected drugs should be discontinued until confirmatory tests have been conducted, and appropriate treatment should be initiated. Subjects diagnosed with acute pancreatitis (as a minimum 2 of 3: characteristic abdominal pain, amylase and/or lipase >3x UNR or characteristic findings on CT/MRI) should be withdrawn from the study. Liraglutide causes dose-dependent and treatment-duration-dependent thyroid C-cell tumors at clinically relevant exposures in both genders of rats and mice. It is unknown whether liraglutide causes thyroid C-cell tumors, including medullary thyroid carcinoma (MTC), in humans, as human relevance could not be ruled out by clinical or nonclinical studies. Liraglutide is contraindicated in patients with a personal or family history of MTC and in patients with Multiple Endocrine Neoplasia syndrome type 2 (MEN 2). Based on the findings in rodents, monitoring with serum calcitonin or thyroid ultrasound was performed during clinical trials, but this may have increased the number of unnecessary thyroid surgeries. It is unknown whether monitoring with serum calcitonin or thyroid ultrasound will mitigate human risk of thyroid C-cell tumors. In a 2-year repeat subcutaneous dose carcinogenicity study of liraglutide injected once a day in CD-1 mice, a treatment-related increase in fibrosarcomas was seen on the dorsal skin and subcutis, the body surface used for drug injection, in males in the 3 mg/kg/day group. These fibrosarcomas were attributed to the high local concentration of drug near the injection site. The liraglutide concentration in the clinical formulation (6 mg/ml) is 10 times higher than the concentration in the formulation used to administer 3 mg/kg/day liraglutide to mice in the carcinogenicity study (0.6 mg/ml).

19 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 17 of 112 Further information can be obtained in the Liraglutide Investigator's Brochure Obesity, Edition 3, October 2010 or any updates hereof. 3.2 Rationale for the trial A t Treatment duration of 56 weeks is considered to be sufficient to demonstrate weight loss and subsequent weight maintenance. In order to assess the effects of drug cessation on appetite and weight control, possible withdrawal effects and rebound will be ascertained in the week follow-up/12 week off drug observational re-randomised treatment period.. 13 week follow-up period: To allow for a blinded assessment of the withdrawal effects, a fraction of subjects (of the sub-group without pre-diabetes) will maintain dietary measures and physical activity while keeping the blind. The subjects selected for participation in the 13 week follow-up period are the subjects first (approximately 900 subjects) to complete the 56 week main trial. Subjects receiving placebo throughout the main part of the trial will continue on liraglutide placebo treatment for the follow-up period (approximately 300 subjects), whereas subjects receiving active liraglutide treatment (approximately 600 subjects) will be randomised 1:1 to either continue active liraglutide treatment (approximate 300 subjects) or switch from active liraglutide treatment to placebo (approximately 300 subjects). 12 week off-drug observational period: The remaining non-pre-diabetic subjects who complete the 56 week main trial will continue in a 12 week off-drug observational period and will contribute with additional information on the effects of drug cessation on weight control and potential weight rebound. 12 week re-randomised treatment period: To allow for a blinded assessment of potential rebound and withdrawal effects, all subjects who did not have pre-diabetes at baseline will continue into a 12-week blinded re-randomised treatment period. Subjects who have received active treatment (liraglutide) throughout the 56 week treatment period will be re-randomised 1:1 to either continue active treatment ( on-drug ) or switch to liraglutide placebo ( off-drug ). Subjects who have received liraglutide placebo treatment throughout the 56 week treatment period will continue on liraglutide placebo in the re-randomised treatment period. Subjects will continue to receive guidance on dietary measures and physical activity throughout this period. After the 12 week blinded re-randomised tretament period, randomised treatment will be discontinued and all subjects will be followed for an additional 2 weeks off-treatment for assessment of liraglutide antibody formation. A major health risk associated with being obese is development of type 2 diabetes. This risk is particularly pronounced in subjects with pre-diabetes. In order to investigate whether liraglutide not only safely and efficaciously lowers body weight but also prevents delays the development onset of

20 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 18 of 112 type 2 diabetes an additional 104 week treatment period for subjects at high risk of developing diabetes (i.e. subjects with pre-diabetes at baseline) has been included. A treatment period of 160 weeks is predicted to be sufficient to demonstrate a 70 60% relative risk reduction in the development of type 2 diabetes 8, 9, Section 4 Objectives and endpoints Section 4.1 Objective(s) Secondary objective: To investigate the long term efficacy of 3.0 mg liraglutide versus liraglutide placebo on cardiovascular risk markers such as blood pressure, lipids, glucose parameters, urinaryalbumin-to-creatinine-ratio (UACR), and the metabolic syndrome as well as effects on quality of life, Patient Reported Outcomes (PRO) and Binge Eating Scale (BES) ratings 4.2 Endpoint(s): Primary efficacy endpoints: Change from baseline in Body weight (fasting) loss at 56 weeks Proportion of subjects losing more than at least 5% of baseline body weight at 56 weeks Proportion of subjects losing more than 10% of baseline body weight at 56 weeks Proportion of subjects with Oonset of type 2 diabetes at 160 weeks (among subjects having with pre-diabetes evaluated at 160 weeks at baseline) Secondary efficacy endpoints: Change from baseline to Week 56 in: Glucose related parameters (HbA 1c, FPG, fasting insulin, C-peptide, endpoints from 2 hr Oral Glucose Tolerance test (OGTT) values parameters, Development proportion of subjects with of type 2 diabetes (yes/no), pre-diabetes status, and homeostasis model assessment parameters [HOMA]) 11 Urinary-albumin-to-creatinine ratio (UACR) Vital signs Systolic and diastolic blood pressure(vital signs) and pulse Cardiovascular biomarkers (high sensitivity C reactive protein [hscrp], adiponectin, fibrinogen, PAI-1) Metabolic syndrome status (ATP-III) at 56 weeks PRO assessed by IWQoL-Lite, SF-36 and Treatment Related Impact Mmeasure - Weight (TRIMm-Weight) Proportion of subjects with change Change in concomitant medication from baseline to Week 56 in: anti-hypertensives drugs

21 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 19 of 112 lipid lowering agents drugs oral antidiabetic drugs Mean change in body weight loss from baseline to Week 160 (subjects with pre-diabetes at baseline) Proportion of subjects losing more than at least 5% and more than 10% of baseline body weight at 160 weeks (subjects with pre-diabetes at baseline) Change from baseline to Week 160 in other secondary endpoints (subjects with pre-diabetes at baseline) week re-randomised treatment/12 week off-drug observational period efficacy endpoints: Change from Week 56 to week 68 in: Urinary Albumin-to-creatinine ratio (UACR) Vital signs Systolic and diastolic blood pressure (vital signs) Change from baseline to Week 68 in: Urinary Albumin-to-creatinine ratio (UACR) Vital signs Systolic and diastolic blood pressure (vital signs) Change from baseline to Week 56 and in binge eating (assessed by BES) Safety endpoints: Pulse (vital sign) week re-randomised treatment /12 week observational period safety endpoints: Adverse events Pulse (vital sign) ECG Haematology and biochemistry including amylase, lipase, calcitonin Change from baseline to Week 56 and 58 in binge eating (assessed by BES) Mental health assessed by the Columbia Suicidality Severity Rating Scale (C-SSRS) and Patient Health Questionnaire (PHQ-9) 2.8 Section 5 Trial design 5.1 Type of trial This is a randomised, double-blind, placebo controlled, parallel group, multi-centre, multinational trial in non-diabetic obese subjects or overweight subjects with co-morbidities. Subjects will be stratified to either 56 weeks (plus 12 weeks re-randomised treatment period and 2 weeks follow-up period) or 160 weeks plus 2 weeks follow-up period) of treatment based on pre-diabetes status at

22 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 20 of 112 randomisation. Treatment allocation for subjects with pre-diabetes will be double-blind for the first year (, Investigator and subject are blinded), and single blind for the remaining 2 years (i.e. is unblinded, whereas Investigator and subjects remain blinded).within these groups subjects are furthermore stratified according to BMI (BMI 30 kg/m 2 or BMI < 30 kg/m 2 ). Pre-diabetes is defined as: IFG (FPG mg/dl / mmol/l, and mg/dl / mmol/l) and/or IGT (2 hr post-challenge OGTT) plasma glucose 140 mg/dl / 7.8 mmol/l, and < mg/dl / mmol/l) or HbA 1c % both inclusive (21) subjects are expected to be screened in order to meet the randomisation target of a minimum of 3600 subjects in total, including a minimum of subjects with pre-diabetes. These numbers are based on an anticipated screen failure rate of 25% and an expected incidence of prediabetes in the trial population of 16-18% 30% (please refer to section 17). The duration of the main part of the trial from screening to end follow-up will be a maximum of weeks including a week re-randomised treatment /12 week off-drug observational period and a 2 weeks follow-up period, whereas while for subjects enrolled in the delayed-onset-of-type-2- diabetes part of the trial the total duration of the trial will be weeks. Approximately 900 subjects continuing in the 13 week follow-up period will be re-randomised at Visit 17a, attend Visit 18a one week after re-randomisation and then afterwards attend the clinic at 4 week intervals. At week 56 (Visit 17) all subjects who did not have pre-diabetes at baseline and had been randomised to liraglutide during the first 56 weeks will be re-randomised. They will attend Visit 18a two weeks after re-randomisation, attend Visit 19a two weeks after Visit 18a and then afterwards attend the clinic at 4 week intervals. In addition they will attend a final follow-up visit (Visit 22a) at week 70. The primary purpose of this visit is to collect blood samples for analysis of liraglutide antibodies. The remaining (approximately 2100) non-pre-diabetic subjects will continue in a 12 week off-drug observational period starting at Week 56 (Visit 17). One week after Visit 17 subjects will attend Visit 18a where they will have blood samples taken for assessment of antibodies. During the observational period subjects will attend the clinic at 4 week intervals. After a successful completion of dose escalation, subjects with pre-diabetes at baseline will attend the clinic at four week intervals (Visits 6 to 17 + Visits 18b to 43b). A safety follow-up visit (Visit 44b) is performed one included two weeks after Visit 43b (End of 160 week randomised Treatment visit). The primary purpose of this visit is to collect blood samples for liragluitde antibody analysis. Subjects who have discontinued the trial prematurely will be asked to attend a visit (17x) one year 56 weeks ± 5 days after their randomisation date. The purpose of this visit will be recording of the body weight and assessment of MESIs. In addition the subjects will be asked to complete mental

23 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 21 of 112 health questionnaires. The assessment will be done by asking the subject if he/she has experienced any MESI (Pancreatitis, Psychiatric disorders, Thyroid disorders, Bile duct disorders, Neoplasms, Cardiovascular Disorders, Immunogenicity - see section and Appendix J) since the last contact. The subjects should attend Visit 17x fasting. If the subject is not willing to attend Visit 17x, it should be documented in the patient medical record that effort has been made to contact the subject and that the subject has refused to attend the visit. 5.2 Rationale for the trial design Results from a phase 2 trial in obese subjects without type 2 diabetes showed a dose-dependent weight loss ranging from 3.8 to 7.8 kg with liraglutide doses of mg administered for 52 weeks (20 weeks as double blind and 32 weeks of open-label treatment (i.e. unblinded at 20 weeks). In addition to weight lowering, there was a decrease in systolic blood pressure and an impact on the a reduction in the number of subjects with pre-diabetes. The safety evaluation was favourable with the main tolerability finding being gastro-intestinal side effects. The present trial is a confirmatory trial and a double-blind, randomised, parallel group, placebo controlled trial design will be employed. Treatment allocation for subjects with pre-diabetes will be double blind for the first 56 weeks (i.e., Investigator and subjects blinded), and single blind (i.e. unblinded, whereas Investigator and subjects remain blinded) for the remaining 104 weeks (i.e. sponsor is unblinded, whereas Investigator and subjects remain blinded). The trial has been designed to further investigate the potential of liraglutide to safely induce longterm weight loss in non-diabetic obese subjects, as well as in overweight subjects with comorbidities. A treatment duration of 56 weeks should be sufficient to demonstrate weight loss and subsequent weight maintenance. In order to assess the effects of drug cessation on appetite and weight control, possible withdrawal effects and rebound will be ascertained in the 13 a blinded 12 week follow-up/12 week off-drug observational re-randomised treatment period (12 weeks on liraglutide or liraglutide placebo treatment plus two weeks off treatment). The re-randomised treatment period is only applicable to subjects who did not have pre-diabetes at baseline as subjects with pre-diabetes at baseline will continue in the 104 week extension. Subjects who have been on active treatment during the first 56 weeks of the trial will be re-randomised in a 1:1 manner to either continue on active treatment or switch to liraglutide placebo during the re-randomised treatment period. Subjects who have received placebo treatment during the first 56 weeks of the trial will continue on liraglutide placebo treatment during the follow-up period. During this period, treatment allocation will remain blinded to, Investigator and subjects. After the blinded 12 week re-randomised treatment period all subjects will discontinue randomised treatment and will be followed for an additional 2 weeks off-treatment to assess potential antibodies to liraglutide. A major health risk associated with being obese is development of type 2 diabetes. This risk is particularly pronounced in subjects with pre-diabetes (IFG and/or IGT or HbA 1c 5.7-

24 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 22 of % both inclusive (21). In order to demonstrate that liraglutide not only safely and efficaciously lowers body weight, but also prevents delays the development onset of type 2 diabetes, an additional blinded (subjects and Investigator are blinded to treatment allocation)104 week extension period for subjects at high risk of developing diabetes has been included. For this population a total treatment period of 160 weeks is predicted to be sufficient to demonstrate a 70 60% relative risk reduction in the development proportion of subjects with new onset of type 2 diabetes. 5.3 Treatment of subjects: Subjects will attend Screening visit 1 to assess their eligibility. If found eligible the subjects will return at Screening visit 2, where they will have laboratory samples taken for assessment of prediabetes/diabetes status (FPG and OGTT) and fasting lipids. At Visit 3 subjects will be randomised to one of the two treatment groups as shown in Table 5-1/Table 5-2 and Table 5-3. If subjects based on either FPG, 2-hr post-challenge plasma glucose or HbA 1c are classified as having pre-diabetes In the event of IFG or IGT (see section 5.1 for definition) at Screening visit 2, subjects will be asked to participate in the delayed-onset-of-type-2-diabetes part of the trial lasting for weeks. Furthermore, subjects will be asked to sign an additional informed consent form no later than at Visit 3 confirming this. All other subjects (not classified as having pre-diabetes) (with normal glucose tolerance and fasting glucose) will be randomised to the 56 week main trial plus a 12 week re-randomised treatment period and a two weeks follow-up period only. Approximately 900 subjects participating in the 56 week main trial will be re-randomised at Week 56 to continue in a 13 week follow-up period (see Table 5-1). The remaining (approximately 2100) non-pre-diabetic subjects will continue in a 12 week off-drug observational period (see Table 5-2.) In addition, a subset population of approximately 60 subjects from US and Europe (UK and the Netherlands) will be selected to participate in the PK sub-study (Visit 8) as well. A separate informed consent form will be signed for the participation in the sub-study (see section 8.4.2). Table 5 1 Treatment groups: Main trial (56 weeks) incl week re-randomised treatment period and a two weeks FU period* * (please note that this table does not include subjects with pre-diabetes at baseline. For treatment of subjects with pre-diabetes at baseline, please refer to Table 5-2)

25 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 23 of 112 Trial periods Visits in each period Screening V1-V2 Randomisa tion/ Dose escalation V3-V5 Maintenance V6, V6c-V16 17 Rerandomisat ion FU rerandomised treatment period End Treatment FU V17a V18a-20a 21a V21a V22a Duration of each period 2 weeks 4 weeks weeks 6 weeks 8 12 weeks 4 weeks 1 2 weeks Treatment arm N Liraglutide Liraglutide Liraglutide placebo Screening Screening Screening Liraglutide mg Liraglutide mg Liraglutide placebo Liraglutide 3.0 mg Liraglutide 3.0 mg Liraglutide placebo Liraglutide 3.0 mg Liraglutide Liraglutide placebo Liraglutide 3.0 mg Liraglutide placebo Liraglutide placebo Liraglutide 3.0 mg FU Liraglutide placebo Liraglutide placebo FU FU Table 5 2 Trial periods Visits in each period Duration of each period Treatment N arm Treatment groups: Main trial (56 weeks) incl. 12 week off-drug observational period Screening Liraglutide 1400 Screening Liraglutide placebo Randomisation/ Dose escalation Maintenance End of Treatment V1-V2 V3-V5 V6, V6c-V16 V17 V18a-21a 2 weeks 4 weeks 46 weeks 6 weeks 12 weeks 700 Screening Liraglutide mg Liraglutide placebo Liraglutide 3.0 mg Liraglutide placebo Liraglutide 3.0 mg Liraglutide placebo Observational period Off-drug Off-drug Table 5 3 Treatment groups: Delayed-onset-of-type-2-diabetes part of the trial* * (applies to subjects with pre-diabetes at baseline)

26 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 24 of 112 Trial periods Screening Randomisation/ Dose escalation Maintenance End of trial End of 160 week randomised treatment Follow-up Visits in each period V1-V2 V3-V5 V6, V6c-V43b V44b Duration of each period 2 weeks 4 weeks 156 weeks 1 2 weeks Treatment arm N* Liraglutide 400 Liraglutide Liraglutide Screening mg 3.0 mg Follow-up Liraglutide placebo Screening Liraglutide placebo Liraglutide placebo Follow-up * N refers to the minimum number of randomised subjects. Actual number will depend on prediabetes prevalence in the randomised population Liraglutide/liraglutide placebo Liraglutide will be available in a concentration of 6.0 mg/ml and supplied in 3 ml FlexPen. Treatment allocation will be blinded to subjects, (irrespective of pre-diabetes status at baseline) Investigator and throughout the main part of the trial (56 weeks). However, after database release database lock of the week-data (56-week-data plus 13/12 12 week followup/off-drug observational re-randomised treatment data and two weeks follow-up data) the sponsor will be unblinded. is no longer blinded to the treatment allocation. As a consequence, only subjects and Investigator remain blinded to Treatment treatment allocation in the 104-week extension (plus 2-week follow up) period. for subjects with pre-diabetes will be double blind for the first 56 weeks, and single blind for the remaining 104 weeks. The Investigator will be notified of the treatment allocation when the week trial has ended and the results have been released. Subjects will be instructed to perform an air shot before the first injection with the FlexPen. For further information please see the handling instruction for liraglutide FlexPen. Handling instructions will be provided in local language together with the trial product. The Investigator must instruct subjects how to inject liraglutide or liraglutide placebo and must ensure that subjects are familiar with the instructions. It must be documented in the subject s medical record that the subject has been instructed in the use of the FlexPen. The treatment duration (for a subject without pre-diabetes at baseline completing the trial) will be up to 56 weeks in the main trial plus 12 weeks in the re-randomised treatment period and 2 weeks follow-up (a total of 70 weeks). for subjects continuing in the 12 week off-drug observational period and up to 68 weeks for a subject included in the 13 week follow-up period. The treatment duration for a subject with pre-diabetes at baseline who completes completing the delayed-onset-of-type-2- diabetes part of the trial will be up to weeks.

27 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 25 of 112 The exact date, time, dose and site of injection of the last dose of trial product on three consecutive days prior to Visit 4, 7, 8 (sub-study) and Visit 11, must be noted in the case report form (CRF) due to PK sampling. The exact date, time, dose and site of injection of the 3 doses of trial product immediately prior to Visit 4, 7, 8 (sub-study) and Visit 11, must be noted in the electronic case report form (ecrf) for evaluation of PK. Subjects should be encouraged to preferably use the same injection site three days prior to blood sampling for PK measurement. Subjects should be encouraged to preferably use the same injection site three days prior to blood sampling for PK measurement on Visit 8 (sub-study). Section Counselling on diet and physical activity All subjects are instructed by dieticians to keep a 3-day food diary every second month between Visits 2 and 3, Visits 5 and 6c, Visits 7 and 8, Visits 9 and 10, Visits 11 and 12, Visit 13 and 14, Visits 16 and 17 for the main part of the trial. For subjects continuing in During the 13 week follow-up re-randomised treatment period, the 3-day food diary will be used to register food intake between Visits 16 and 17a and between Visit 19a and 20a. The subject s dietary compliance and the average daily level of physical activity will be recorded every second month. The subject will be questioned whether asked if they performed less than half an hour, between half an hour and one hour or more than 1 hour of physical activity per day. For the main part of the trial this applies to Visit 3 (only average level of physical activity), 6c, 8, 10, 12, 14 and 17. For the 13 week follow-up re-randomised treatment period, the dietary compliance and physical activity will also in addition be recorded at Visit 20a. For the delayed-onset-of-type-2- diabetes part of the trial, the subject s dietary compliance and average daily level of physical activity will be recorded at Visit 3, 6c, 8, 10, 12, 14, 17, 19b, 21b, 23b, 25b, 27b, 29b, 31b, 33b, 35b 37b, 39b, 41b and 43b. An increase in physical activity (recommended minimum150 minutes/week) will be encouraged and re-enforced by use of pedometers. 5.4 Rationale for treatment Based on a balanced evaluation of the efficacy and safety/tolerability results from the doubleblinded phase 2 dose-range finding trial (NN ) and subsequent 32 week open label extension (Investigator and subjects remained blinded), a dose of 3.0 mg has been chosen as the optimal dose for weight management indication for liraglutide, and will therefore be further investigated in this confirmatory trial. In this trial based on current available data, subjects without pre-diabetes will be treated for a total period of 56 weeks, which should provide ample time to assess the capacity of liraglutide to induce and maintain weight loss and evaluate the long term safety of liraglutide in non-diabetic obese subjects or overweight subjects with co-morbidities. In order to assess the effects of drug cessation on appetite and weight control, possible withdrawal effects and rebound will be ascertained in the 13 week follow-up/12 week off-drug observational re-randomised treatment period (that consist of

28 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 26 of 112 a blinded 12 week on re-randomised treatment [liraglutide treatment arm re-randomised 1:1 to liraglutide or liraglutide placebo and liraglutide placebo arm continues unchanged] and a 2 week off-randomised-treatment period to assess potential liraglutide antibodies). 2.9 Section 6 Trial population 6.1 Number of subjects to be studied Countries planned to participate: Germany, Spain, France, Italy, Belgium, Austria, Switzerland, Hungary, Poland, Serbia, UK, Norway, Finland, Denmark, Netherland, Ireland, Turkey, Israel, Brazil, Mexico, India, Russia, Hong Kong, South Africa, US, Canada, Australia. Planned number of subjects to be screened: 4800 Planned number of subjects to be randomised/started on trial product(s): 3600 Minimum number of subjects randomised to the delayed-onset-of type-2 diabetes part of the trial: Planned number of subjects to complete the main trial (56 weeks): 2160 Planned number of subjects to complete the delayed-onset-of type-2 diabetes part of the trial: Anticipated number of trial sites: 160 Anticipated number of subjects to be randomised/started on trial products(s) at each site: subjects will be screened in order to obtain a minimum of 3600 randomised subjects of which 2400 is are on liraglutide and 1200 on liraglutide placebo. This number is based on an anticipated screening failure rate of 25%. As a minimum of subjects will be randomised stratified to the delayed-onset-of-type-2-diabetes part of the trial (160 weeks of treatment), it is anticipated that the incidence prevalence of pre-diabetes in obese subjects at randomisation is 16-18% will be at least 30% (please refer to section 17). If it turns out that the incidence the prevalence of prediabetes is lower than expected, screening and randomisation will continue with the purpose to identify and enrol subjects with until the target of 600 minimum pre-diabetic subjects pre-diabetes is reached. As a consequence more than 3600 subjects may be randomised. If however, the incidence of pre-diabetes is higher than expected, screening and randomisation will stop when the total number of randomised subjects has reached 3600.There will be no upper limit to the number of subjects with pre-diabetes which can be enrolled within the planned total of Stratification criteria

29 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 27 of 112 In order to To be eligible for stratification to the delayed-onset-of-type-2-diabetes part of the trial, subjects must comply with the following: Pre-diabetes at Screening visit 2, defined as IFG (FPG 110 mg/dl / 6.1 mmol/l, and < 126 mg/dl / 7.0 mmol/l) and/or IGT (2 hr post OGTT plasma glucose 140 mg/dl/7.8 mmol/l and < 200 mg/ dl/ 11.1 mmol/l) Within the main trial (56 weeks) and the delayed-onset-of-type- 2diabetes part of the trial, subjects will furthermore be stratified according the BMI ( 30 kg/m 2 or < 30 kg/m 2 ). FPG 100 mg/dl (5.6 mmol/l) and 125 mg/dl (6.9 mmol/l) AT Screening visit 2, and/or 2 hr post-challenge (OGTT) plasma glucose 140 mg/dl (7.8 mmol/l) and 199 mg/ dl (11.0 mmol/l) at Screening visit 2, and/or HbA 1c % both inclusive, at Screening visit 1 (21) Subjects will furthermore be stratified according to BMI ( 30 kg/m 2 or < 30 kg/m 2 ). 6.4 Exclusion criteria Any clinically significant disease which in the Investigator s opinion could interfere with the safety of trial participants or with the results of the trial HbA 1c % (Screening visit 1) or FPG 126 mg/dl (7.0 mmol/l) (Screening visit 2) or 2 hr post-challenge plasma glucose 200 mg/dl (11.1 mmol/l) (Screening visit 2) HbA 1c 7 % at Visit 1 FPG 126 mg/ dl (7 mmol/l) and/or OGTT 2 hr reading 200 mg/dl (11.1 mmol/l) at Visit 2 Screening calcitonin 50 ng/l Family or personal history of multiple endocrine neoplasia type 2 (MEN2) or familial medullary thyroid carcinoma (FMTC) Personal history of non-familial medullary thyroid carcinoma Simultaneous participation in any other clinical trial of an investigational drug Participation in a clinical trial of weight control within the last 3 months prior to Screening visit 1 Current participation in an organised diet reduction weight reduction program (or within the last 3 months) and /or are currently using or have used within three months before Screening visit 1: pramlintide, sibutramine, orlistat, zonisamide, topiramate, phenteremine, or metformin (either by prescription or as part of a clinical trial) History of major depressive disorder or a PHQ-9 score 15 or history of other severe psychiatric disorders, e.g. schizophrenia or bipolar disorder within the last 2 years History of Major Depressive Disorder within the last 2 years

30 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 28 of 112 History of other severe psychiatric disorders, e.g., schizophrenia, bipolar disorder A PHQ-9 score of 15 Subjects with any lifetime history of a suicide attempt or lifetime history of any suicidal behaviour or any suicidal ideation of type 4 (active suicidal ideation with some intent to act, without specific plan) or type 5 (active suicidal ideation with specific plan and intent) on the C- SSRS questionnaire Any lifetime history of a suicide attempt A history of any suicidal behaviour in the last month prior to randomisation Any suicidal ideation of type 4 or 5 on the C-SSRS in the last month prior to randomisation Subjects who are non-compliant with any of the eligibility criteria, but included in the trial, should be withdrawn immediately. If extraordinary circumstances speak in favour of maintaining the subject in the trial then this is only acceptable if justified and approved by the Independents Ethics Committee (IEC)/Institutional Review Board (IRB), and if the regulatory authorities are notified according to local requirements. Investigators are recommended to refer screening failure subjects having a calcitonin value of 50 ng/l to an endocrinologist for follow-up. 6.5 Rescue criteria The rescue criteria only apply for subjects developing diabetes during the trial. If self measured FPG on three consecutive days/occasions exceeds the limits set below, the subject should contact the Investigator and come in for an unscheduled visit as soon as possible. The next scheduled visit should not be awaited. A FPG should be obtained and analysed by the central laboratory. If this FPG exceeds the limits set below, the background medication dose should initially be escalated to the maximum approved dose, and if not sufficient to achieve glycaemic control, one of the other allowed antidiabetic treatment should be added (please refer to withdrawal criteria no. 4): From baseline to Week 6: FPG > 15 mmol/l (270 mg/dl) From Week 7 to Week 12: FPG > 13.3 mmol/l (240 mg/dl) From Week 13 to Week 56: FPG > 11.1 mmol/l (200 mg/dl) Moreover, if any of the FPG or HbA 1c samples analysed by the central laboratory exceeds the limits set below, the subject should immediately be called in for an unscheduled visit as soon as the result has come to the knowledge of the Investigator. The next scheduled visit should not be awaited. A new FPG should be obtained and analysed by the central laboratory and if this FPG exceeds the limits set below, the background medication dose should initially be escalated to the maximum

31 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 29 of 112 approved dose, and if not sufficient to achieve glycaemic control, one of the other allowed antidiabetic treatment should be added (please refer to withdrawal criteria no. 4): From baseline to Week 6: FPG > 15 mmol/l (270 mg/dl) From Week 7 to Week 12: FPG > 13.3 mmol/l (240 mg/dl) From Week 13 to Week 56: FPG > 11.1 mmol/l (200 mg/dl) or HbA 1c > 8% 6.5 Withdrawal criteria 4. Subjects who develop diabetes during the trial will not be withdrawn but should receive the best standard of care at the discretion of the Investigator. If the Investigator determines that insulin, GLP-1 receptor agonist (e.g., Byetta or Victoza ), or DPP-IV inhibitor, is the best treatment option, the subject must be withdrawn that exenatide a GLP-1 receptor agonist, a DPP-4 inhibitor or insulin to be the best treatment, the subjects must be withdrawn. The medication prescribed by the Investigator will not be provided by In case of withdrawal, the End of Trial form must be filled in and in the IV/WRS the Withdrawal session must be completed. If possible, the subject should be called in for a final visit. Procedures according to Visit 17 should be performed for all subjects who discontinue the trial prematurely before Visit 17, and whereas procedures relating to Visit 21a should be performed for all subjects continuing in the 13 week follow-up/12 week off-drug observational re-randomised treatment period. For subjects stratified to continuing in the delayed-onset-of-type-2-diabetes part of the trial, and past Visit 17, procedures according to Visit 43b should be performed. Subjects that who have discontinued the trial prematurely prior to Visit 17 will be asked to attend a visit (Visit 17x) taking place one year 56 weeks ± 5 days after their randomisation date. The purpose of this visit will be recording of the body weight and reporting of MESIs. 6.6 Subject replacement Withdrawn subjects will not be replaced. However, re-screening is allowed once within the limits of the recruitment period, at the discretion of the Investigator. If a subject is re-screened, a new subject number must be assigned, and new informed consent has to be signed off, and all Screening visit 1 blood sampling and assessments have to be repeated. 6.7 Rationale for trial population Subjects with a presumed high risk of developing diabetes (pre-diabetes defined as either IFG and/or IGT and/or HbA 1c % inclusive (21)[see section 5.1]) have been selected for extended

32 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 30 of 112 investigation (up to 160 weeks) in order to demonstrate not only the long-term safety and efficacy of inducing and maintaining weight loss, but also the ability of liraglutide treatment to delay the development onset of type 2 diabetes. The risk of developing type 2 diabetes over 160 weeks for this population is presumed to be approximately 20% 8, 9, 10 and the trial aims at investigating whether additional factors can be used to identify subjects most at risk., and whether treatment with liraglutide can reduce the risk to 6% Section 7 Trial schedule Planned duration of recruitment period: Planned date for first subject (FPFV): 22 weeks 21-Sep Jun-2011 Planned completion of last subject in the main trial including week FU re-randomised treatment period and two weeks FU period(lplv):jul Mar-2013 Planned completion of clinical trial report: Nov-2011 Aug-2013 Planned completion of last subject in the delayed-onset-of-type-2-diabetes part of the trial (LPLV): Apr-2013 Dec-2014 Planned completion of clinical trial report in the delayed-onset-of-type-2-diabetes part of the trial: Aug 2013 Apr-2015 The end of the clinical trial is defined as the last visit of the last subject globally. After a non-competitive grace period, subject recruitment will be competitive between countries and trial sites in order to ensure that the planned number of randomised subjects is achieved within the timelines. The recruitment will be evaluated by on a continuous basis and country- and regional-specific contingency measures may be applied. A competitive recruitment strategy may be applied if deemed needed. The trial is subject to registration before subject enrolment according to the specifications from the International Committee of Medical Journal Editors (ICMJE) 17. will ensure the trial is registered at ClinicalTrials.gov according to the requirements from the Food and Drug Administration (FDA) 18 and ICMJE 17. For countries outside the US, only the main trial site per country will be disclosed via facility name, city and country on the trial registration. In the US, all trial sites will be registered.

33 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 31 of 112 Protocol information for this trial will be subject to public disclosure at external web sites ( and according to international regulations e.g. the International Committee of Medical Journal Editors (ICMJE) (30), the Food and Drug Administration Amendments Act (FDAAA) (31-33 ) - as reflected in Code of Conduct for Clinical Trial Disclosure Section 8 Methods and assessments 8.1 Visit Procedures Before screening takes place subjects will be provided with written and verbal information about the trial and the procedures involved, in accordance with local requirements. Qualified site staff in accordance with local law will ensure that subjects are fully informed both verbally and in writing about the practical consequences of participating, of their rights and responsibilities while participating in the trial, as well as any possible advantages and disadvantages in being treated with the trial products. Subjects will have the opportunity to ask questions to a medically qualified person and have ample time to consider their participation. Subjects who wish to participate must give signed and dated informed consent. In addition the time for signing the informed consent must be noted by the subject. This must be done prior to any trial related activities, i.e. procedures that would not have been performed during normal management of the subject. The Investigator must keep a subject screening log and a subject enrolment log or an informed consent log according to local practice. These can be combined into one document. It must be stated in the medical record that the subject is participating in the current trial. Subjects enrolled will also be provided with a card stating that they are participating in the trial, besides contact address(es) and telephone number(s). The subjects should be reminded to show the card to other physicians, as applicable. Subjects enrolled in the trial will be provided with a subject ID card, stating that they are participating in a trial and whom to contact (site address, Investigator s name and telephone number) for further information, if necessary. The subjects should be reminded to show the card to other health care providers, as applicable. The subjects should be instructed to return the card to the Investigator at the last visit of the subject or destroy the card after the last visit. For screening failures (subjects who have given informed consent but who do not fulfil all inclusion criteria and/or meet at least one exclusion criterion and hence are not randomised) the screening failure session in the IV/WRS must be completed and the reason for screen failure will be recorded in the IV/WRS ecrf. All screen failure information will be available in the clinical database. Serious and non-serious adverse events from screening failures will be entered by the Investigator into the electronic ecrfs and consequently subsequently transferred to the clinical database. When trial related procedures have been finalised for screening failures, no more adverse events should

34 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 32 of 112 be entered in the ecrf. Follow-up of AEs should be done according to section 12. In addition screening failures should be registered in the IV/WRS. As a minimum, the reason for screening failure must be recorded. Investigators are recommended to refer screening failure subjects having a calcitonin value of 50 ng/l to an endocrinologist for follow-up. For subjects with calcitonin 20 ng/l (from Visit 3 and onwards) a repeat measurement of calcitonin must be performed preferably within four weeks and if confirmed, the event elevated calcitonin should be reported as a MESI (i.e. any confirmed episode of calcitonin concentration value 20 ng/l, if not already reported as an ongoing MESI, see section ). During the 56 week main part of the trial including the 13 week follow-up/12 week off-drug observational re-randomised treatment period subjects should attend the clinic in a fasting condition in the morning (i.e. having consumed only water since midnightat least eight hours overnight fast without food and/or drink intake, except for water) at Visit 2, 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 17x, 18a, 21a and Visit 21a 22a. For the delayed-onset-of-type-2-diabetes part of the trial subjects should attend the clinic in a fasting condition in the morning (i.e. having consumed only water since midnight at least eight hours overnight fast without food and/or drink intake, except for water) at Visit 2, 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 17x, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, 43b and 43b 44b. Trial product should be taken as usual i.e. trial product should not be withheld until blood sampling has been performed at the relevant visit. However, participants in the PK sub-study must take their trial product injection at approximately 10 pm the evening prior to the PK blood sampling at Visit 8. Concomitant medication should be taken as usual during the conduct of the trial i.e. concomitant medication should not be withheld on the day of the visit to the clinic, until blood sampling has been done. All subjects should withheld trial product on Visits 2, 3, 4, 5, 6c, 8, 9, 11, 14, 17, 17x, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, 43b and 44b (fasting visits) until blood sampling has been done. Subjects developing diabetes during the trial should in addition withheld withhold their background diabetes medication and trial product at Visits 2, 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 17x, 17a, 18a, 21a, 22a, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, and Visit 43b and Visit 44b until blood sampling has been done. At all other visits diabetes background medication and trial product should be taken as usual during the conduct of the trial. Weight will be measured at all visits to the clinic except for Screening visit 2. Following randomisation, trial product will be dispensed at all visits to the clinic except for Visit 4, 6, 18a and End of Treatment visits (17 Visits 21a and 43b). Subjects will be asked to bring all empty, partly used and unused trial products for drug accountability.

35 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 33 of 112 If the subject attends a fasting visit in a non-fasting state the subject should preferably be called in for a new visit within the visit window to have the fasting weight measured and the fasting blood samples drawn. If it is not possible to attend the new visit within the visit window, the subject should be called in for an Unscheduled Visit and an Unscheduled Visit Form should be used to obtain information on fasting weight and fasting blood samples. If the subject attends a fasting visit in a non-fasting state the subject should preferably be called in for a new visit within the visit window to have the fasting weight and fasting circumference measured and the fasting blood samples drawn. All other assessments can be performed even though the subject is not fasting. If it is not possible to attend the new visit within the visit window, the subjects should be called in for an Unscheduled Visit and an Unscheduled Visit Form should be used to obtain information on fasting weight, fasting circumference and fasting blood samples. In case a subject is being prematurely withdrawn from the trial, the Investigator must aim to will ensure that undertaking undertake the procedures for the End of Treatment visit (Visit 17, 21a or 43b) as soon as possible, if possible. are undertaken, if possible. The primary reason (adverse event, non-compliance with protocol or other) for discontinuation must be specified in the ecrf and in the IV/WRS a withdrawal session should be completed. Even if the subject is not able to attend a final visit, the End of Trial Form must be completed. The end-of-trial form must be completed, and final drug accountability must be performed even if the subject is not able to attend. Subjects, who have discontinued the trial prematurely before Visit 17, will be asked to attend a visit (Visit 17x) one year 56 weeks ± 5 days after their randomisation date. The purpose of this visit will be to record the body weight and assessment of reporting of MESIs. The assessment will be done by asking the subject if he/she did experienced any MESI (Pancreatitis, Psychiatric disorders, Thyroid disorders, Bile duct disorders, Neoplasms, Cardiovascular Disorders, Immunogenicity - see section and appendix J) since the last contact. The subjects should attend Visit 17x fasting. If the subject is not willing to attend Visit 17x, it should be documented in the patient medical record that effort has been made to contact the subject and that the subject has refused to attend the visit. As liraglutide is not available on prescription for the obesity indication after the end of the trial, subjects will not be able to receive it when the trial ends. This means that will not offer investigational drug after the end of the trial. At the End of Treatment visit (Visit 17, 21a or 43b) the Investigator should provide counselling on future weight management Visit schedule Table 8-1 Visit schedule

36 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 34 of 112 Visit no Time of Visit Type of Visit 56 weeks main trial (all subjects) Visit 1 Week -2 ± 5 days Screening visit 1 Visit 2 Week -1 ± 3 days Screening visit 2 Visit 3 Week 0, baseline Randomisation, start of liraglutide/liraglutide placebo 0.6 mg Visit 4 Week 2 ± 3 days Dose escalation Visit 5 Week 4 ± 5 days Dose escalation Visit 6-16 Week 6-50 ± 5 days Maintenance visits Visit 17 Visit 17x Week 56 ± 5 days Week ± 5 days End of 56 weeks randomised treatment/premature discontinuation visit End of Treatment visit (for subjects in 12 week off-drug observational period)) Body weight/mesi recording visit Visit 17a Week 56 ± 3 days Re-randomisation visit (for subjects in week follow-up period) Rerandomised treatment period (ONLY subjects without prediabetics) Visit 18a Week ± 3 days Re-randomised treatment Follow-up visit (13 week follow-up/12 week off-drug observational period) Visit 19a Week 60 ± 5 days Re-randomised treatment Follow-up visit (13 week follow-up/12 week observational period) Visit 20a Week 64 ± 5 days Follow-up visit (13 14 week follow-up/12 week observational period) Visit 21a Week 68 ± 3 days End of re-randomised Ttreatment /premature discontinuation visit (13 week follow-up period) Visit 22a Week ± 3 days Follow-up visit (13 week follow-up period) ONLY subjects with prediabetes Visit 18b-42b Week ± 5 days Maintenance (delayed-onset-of-type-2-diabetes part of the trial) Visit 43b Week 160 ± 3 days End of 160 week randomised Treatment visit/premature discontinuation (delayed-onset-of-type-2-diabetes part of the trial) Visit 44b Week ± 3 days Follow-up visit (delayed-onset-of-type-2-diabetes part of the trial) The duration of the trial for a subject completing the 56 week main trial including the 13 week follow-up/12 week off-drug observational re-randomised treatment period and the follow-up period will be up to weeks and will comprise a total of 23/22 visits. The total duration for a subject

37 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 35 of 112 completing the delayed-onset-of-type-2-diabetes part of the trial will be up to weeks and will comprise a total of 45 visits. The visits during the 56 week main trial are divided into nine categories [(Screening visit 1, Screening visit 2, Randomisation visit (Visit 3), Non-lab visits, category 1 lab visit, category 2 lab visits, category 3 lab visit, category 4 lab visit and End of Treatment visit (see Table 8-2)]. The 3- month follow-up period is divided into 5 categories (see Table 8-2). The visits during the delayedonset-of-type-2-diabetes part of the trial are divided into eight categories [(Visit 18b, Non-lab visits, category 1 lab visits, category 2 lab visits, category 3 lab visits, category 4 lab visit, End of Treatment visit and Follow-up visit (see Table 8-2)]. In addition a body weight recording visit (Visit 17x) exists for subjects who discontinue the main trial prematurely. This extra visit will take place one year after the subject s randomisation date. Please find description of each visit category in sections, to ). Table 8.2 Visit categories

38 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 36 of 112 Visit category Visit no. 56 week main trial: Screening visit 1(8.1.2) Visit 1 Screening visit 2 (8.1.3) Visit 2 Randomisation (8.1.4) Visit 3 Non-lab visits (8.1.5) Visits 6, 6c, 9, 10, 12, 13, 15 and 16 Category 1 lab visits (8.1.6) Visits 4and 7 Category 2 lab visits (8.1.7) Visits 8 and 14 Category 3 lab visit (8.18) Visit 11 Category 4 lab visit (8.1.19) Visit 5 End of Treatment visit* (8.1.10)) Visit 17 Body weight recording visit (8.1.11)) Visit 17x 13 week follow-up/12 week off-drug observational period : Re-randomisation visit** (8.1.12)) Visit 17a Follow up visit (8.1.13) Visit 18a Follow up visit (8.1.14) Visits 19a and 20a End of Treatment visit (8.1.15) Visit 21a follow-up visit (8.1.16) Visit 22a Delayed-onset-of-type-2-diabetes part: First visit (8.1.17) Visit 18b Non-lab visits (8.1.18) Visits 21b, 24b, 27b, 30b, 33b, 36b, 39b, 40b and 42b Category 1 lab visits (8.1.19) Visits 20b, 26b and 32b Category 2 lab visits (8.1.20) Visits 23b, 29b and 35b Category 3 lab visits (8.1.21) Visits 19b, 22b, 25b, 28b, 31b, 34b, 37b and 38b Category 4 lab visit (8.1.22)) Visit 41b End of Treatment visit (8.1.23) Visit 43b Follow-up visit (8.1.24) Visit 44b * The End of Treatment visit is only applicable for subjects continuing in the 12 week off-drug observational period and the subjects who discontinue the trial prematurely ** The re-randomisation visit is only applicable for the first 900 subjects who continue in the 13 week follow-up period Screening visit 1 Screening visit 1 will take place 2 weeks ± 5 days before randomisation (Visit 3). Subjects must give signed and dated (incl. time) informed consent prior to any trial-related activities. All subjects will be provided with a copy of the subject information and a copy of their own signed and dated Informed Consent Form. The IV/WRS should be called/entered to register the subject as screened (see section 10). The following will be performed and/or recorded in the ecrf:

39 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 37 of 112 Demographics information registered in the IV/WRS and/or recorded in the ecrf (date of birth, sex, race and ethnicity, according to local requirements) Medical history/concomitant illness including family history of type 2 diabetes Signs or symptoms of diabetic complications (see section 8.4.6) History of concomitant cardiovascular disease (see section 8.4.7) History of gallbladder disease (see section 8.4.8) History of psychiatric disorders (see section 8.4.9) Screening visit 2 In order to be eligible for stratification to the 160 week delayed-onset-of-type-2-diabetes part of the trial, the subject must comply with the following definition of pre-diabetes: IFG (FPG mg/dl / mmol/l, and < mg/dl / mmol/l) and/ or IGT (2 hr post OGTT plasma glucose 140 mg/dl / 7.8 mmol/l, and < mg/dl / mmol/l) or HbA 1c % both inclusive (21) If the subjects does not have pre-diabetes is not pre-diabetic, the subject will be stratified to the 56 week main trial (plus the re-randomised treatment period). Remind the subject to bring a sample of their morning urine to Visit Visit 3, randomisation If the subject is found eligible after review of inclusion and exclusion criteria at Screening visit 1 and Screening visit 2 and complies with the pre-diabetes stratification criteria (see section 6.3) the subject will be stratified to the ddelayed-onset-of-type-2-diabetes part of the trial lasting for weeks in total (attending Visit 3 to 17 and Visit 18b to 44b). Subjects found eligible to be stratified to the week trial will be asked to sign an additional informed consent no later than at Visit 3. If the subject is found eligible in accordance with inclusion and exclusion criteria at Screening visit 1, but does not comply with the pre-diabetes stratification criteria, the subject will be stratified to the 56 week main trial including a re-randomised treatment period. trial only (Visit 3-17). At the end of the 56 week main trial subjects who received liraglutide during the 56 week main trial will be re-randomised 1:1 to either continue on liraglutide or be switched to liraglutide placebo during the re-randomised treatment period. approximately 900 subjects without pre-diabetes will be included in a 13 week follow-up period. The subjects selected for participation in the 13 week follow-up period are the subjects first to complete the main trial (56 weeks). Treatment allocation for the 900 subjects will be as follows: subjects receiving placebo will continue liraglutide placebo

40 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 38 of 112 treatment (approximately 300 subjects), whereas subjects receiving active liraglutide treatment (approximately 600 subjects) will be randomised 1:1 to either continue active liraglutide treatment (approximately 300 subjects) or switch from active liraglutide treatment to placebo (approximately 300 subjects). The remaining (approximately 2100) subjects will continue in a 12 week off-drug observational period. The following will be performed and/or recorded in the ecrf: Subject compliance (see section 8.5) Subject s dietary compliance and average daily level of physical activity (see section 8.4.6) Urine sampling for measurement of: Urinary Albumin-to-Creatinine ratio (see section ) Remind the subject to: attend Visit 4 5 fasting (i.e. having consumed only water since midnight i.e., at least eight hours overnight fast without food and/or drink intake, except for water) inject liraglutide/liraglutide placebo before bedtime in the evening prior to Visit 4 keep a dosing diary for the three doses of trial product immediately prior to Visit Non-lab visits during the 56 week main trial Visits 6, 6c, 9, 10, 12, 13, 15, and 16 Non-lab visits are Visit 6, 6c, 9, 10, 12, 13, 15, and 16. Blood sampling is not applicable for the non-lab visits. The following will be performed and/or recorded in the ecrf: Mental health questionnaires, applicable for Visit 6c, 9, 10, 12, 13, 15 and 16 (see section 8.3.9) Blood sampling for measurement of: FPG, only applicable for Visits 6c, 9 and 16 (see section 8.2.3) Remind the subject to: attend category 2, 3 and 4 lab visits visits 6c, 11, 14, 16 and 17 fasting and End of Treatment (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) i.e. having consumed only water since midnight) inject liraglutide/ liraglutide placebo before bedtime in the evening prior to Visit 7 and Visit 11 keep a dosing diary for the three doses of trial product immediately prior to Visit 7 and Visit 11 remind the subject to bring a sample of their morning urine at Visit 11 and Category 1 lab visit during the 56 week main trial Visits 4 and 7

41 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 39 of 112 Category 1 lab visit are Visit 4 and 7. The following will be performed and/or recorded in the ecrf: Mental health questionnaires, only applicable to Visit 7 (see section 8.3.9) Blood sampling for measurement of: FPG (only applicable for Visit 4) (see section 8.2.3) Reminders: Remind the subject to: attend category 2, 3 and 4 lab visits the subsequent visit (Visit 5 and 8) and End of Treatment visit fasting (i.e. having consumed only water since midnight i.e., at least eight hours overnight fast without food and/or drink intake, except for water) Category 2 lab visits during the 56 week main trial Visits 8 and 14 Category 2 lab visits are Visit 8 and 14. Remind the subject to: attend category 2 and 3 lab visits and End of Treatment visit Visit 9 fasting (i.e. having consumed only water since midnight i.e., at least eight hours overnight fast without food and/or drink intake, except for water) Category 3 lab visit during the 56 week main trial Visit 11 Visit 11 is the only category 3 lab visit. The following will be performed and/or recorded in the ecrf: Smoking habits (see section 8.4.3) Subject s dietary compliance and average daily level of physical activity ECG (see section 8.3.4) Urine sampling for measurement of: Urinary Albumin-to-Creatinine ratio (see section ) Remind the subject to: attend category 2 lab visits and End of Treatment visit Visit 14 fasting (i.e. having consumed only water since midnight i.e., at least eight hours overnight fast without food and/or drink intake, except for water) Category 4 lab visits during the 5 week main trial Visit 5 The only category 4 lab visit is Visit 5.

42 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 40 of 112 Remind the subject to: attend category 2, 3 and 4 lab visits Visit 6c and End of Treatment visit fasting (i.e. having consumed only water since midnight i.e., at least eight hours overnight fast without food and/or drink intake, except for water) Visit 17, End of 56 weeks randomised Treatment/premature discontinuation Visit 17 will take place 56 weeks ± 3 5 days after randomisation. This visit is applicable for: Subjects continuing in the 12 week off-drug observational re-randomised treatment period Subjects withdrawing prematurely Subjects continuing in the delayed-onset-of-type-2-diabetes part of the trial Subjects being prematurely withdrawn from the trial prior to Visit 17 should preferably have all procedures for the End of 56 weeks randomised Treatment visit performed. The primary reason (adverse event, non-compliance with protocol or other) for discontinuation must be specified in the ecrf and in the IV/WRS a withdrawal session should be completed. Even if the subject is not able to attend a final visit, the End of Trial Form must be completed. Subjects discontinuing the trial prematurely will be asked to attend a visit 17x 56 weeks ± 5 days after their randomisation date. The following will be performed and/or recorded in the ecrf: Smoking habits (see section 8.4.3) Re-randomisation (only applicable for subjects without pre-diabetes at baseline randomised to liraglutide during the 56 week main trial) Completion of the 56 week main trial Interact with the IV/WRS to: Complete re-randomisation session for all subjects without pre-diabetes at baseline randomised to liraglutide during the 56 week main trial Complete dispensing session for subjects continuing in the re-randomised treatment period Blood sampling for measurement of: For females of childbearing potential: serum pregnancy test (only for subjects completing the 56 week main part of the trial and continuing in the 12 week off-drug observational period not applicable for pre-diabetic subjects continuing in the delayed-onset-of-type-2 diabetes part of the trial) (see section 8.3.5) Urine sampling for measurement of: Urinary Albumin-to-Creatinine ratio (see section ) Reminders:

43 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 41 of 112 Make an appointment for Visit 18a, 58 weeks ± 3 days after randomisation for subjects continuing in the re-randomised treatment period Make an appointment for Visit 18b, 60 weeks ± 5 days after randomisation for subjects continuing in the delayed-onset-of-type-2-diabetes part of the trial Remind the subjects who discontinue the trial prematurely that they will be contacted for measurement of their weight one year 56 weeks ± 5 days after their randomisation date Make an appointment for Visit 18a, 57 weeks ± 3 days after randomisation for subjects randomised to the 56 week main trial and continuing in the 12 week off-drug observational period Remind the subjects to attend Visit 18a fasting Remind the subject to bring a sample of their morning urine at Visit 18a Visit 17x, Body weight/mesi recording visit Visit 17x will take place weeks ± 5 days after randomisation The body weight recording visit is only applicable for subjects who have discontinued the trial prematurely before Visit 17. However, this visit will only be applicable for subjects who have not withdrawn their consent to the trial and accept that they can be contacted 1 year after their randomisation date. Subjects will be asked to attend this visit one year 56 weeks ± 5 days after their randomisation date for the purpose of body weight recording and assessment of MESI. The assessment will be done by asking the subject if he/she has experienced any MESI (see section and Appendix J) since the last contact. In addition the subjects will be asked to complete mental heath questionnaires.the subjects should attend Visit 17x fasting. If the subject is not willing to attend Visit 17x, it should be documented in the patient medical record that effort has been made to contact the subject and that the subject has refused to attend the visit. The following will be performed and recorded in the ecrf: MESI, if any (see section ) Visit 17a, Re-randomisation Visit 17a will take place 56 weeks ± 3 days after randomisation. This visit is only applicable for the first 900 subjects to complete the 56 week main trial. At this visit approximately 900 subjects will be included in a 13 week follow-up period. The subjects selected for participation in the 13 week follow-up period are the subjects first to complete

44 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 42 of 112 the main trial (56 weeks). The 900 subjects will through the IV/WRS system receive treatment allocation as follows: subjects receiving placebo will continue to receive liraglutide placebo treatment for the follow-up period (approximately 300 subjects), whereas subjects receiving active liraglutide treatment (approximately 600 subjects) will be randomised 1:1 to either continue active liraglutide treatment (approximately 300 subjects) or switch from active liraglutide treatment to placebo (approximately 300 subjects). The following will be performed and/or recorded in the CRF: Withdrawal criteria (see section 6.5) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) PRO questionnaire (see section ) Binge eating scale questionnaire (see section ) ECG (see section 8.3.4) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Mental health questionnaires (see section 8.3.9) Diet and physical activity counselling based on the 3-day food diary (see section 5.3.2) Subject s average daily level of physical activity (see section 8.4.6) Interact with the IV/WRS to: Complete dispensing session Complete withdrawal session (if applicable) Interact with the Drug Accountability Module Blood sampling for measurement of: HbA 1c (see section 8.2.2) FPG (see section 8.2.3) Fasting insulin (see section 8.2.4) C-peptide (see section 8.2.5) OGTT (see section 8.2.6) Cardiovascular biomarkers (see section 8.2.9) Lipids (see section ) Haematology and biochemistry (see section 8.3.7) Reminders: Make an appointment for Visit 18a (Follow-up visit), 57 weeks ± 3 days after randomisation for subjects randomised to the 56 week main trial and continuing the 13 week follow-up period Remind the subjects to attend Visit 18a fasting

45 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 43 of Visit 18a, Follow-up/observational period Re-randomised treatment visit Visit 18a will take place weeks ± 3 days after randomisation. This visit is only applicable for subjects continuing in the 13 week follow-up/12 week off-drug observational re-randomised treatment period. The following will be performed and/or recorded in the ecrf: Mental health questionnaire (see section 8.3.9) Urine sampling for measurement of: Urinary Albumin-to-Creatinine ratio (see section ) Blood sampling for measurement of: Liraglutide antibodies, (see section 8.3.8) only applicable for subjects that continue in the 12 week off-drug observational period Fasting insulin (8.2.4) and fasting plasma glucose, only applicable for subjects that continue in the 12 week off-drug observational period C-peptide (see section 8.2.5) Visits 19a + and 20a, follow-up/off-drug observational Re-randomised treatment visits These visits are only applicable for subjects continuing in the 13 /12 week follow-up/off-drug observational re-randomised treatment period. The following will be performed and/or recorded in the ecrf: Dispense 3-day food diary at Visit 19a (see section ) Interact with the IV/WRS to: Complete dispensing session, only applicable for first 900 subjects continuing in the 13 week follow-up period Complete withdrawal session (if applicable) Interact with the Drug Accountability Module, only applicable for first 900 subjects continuing in the 13 week follow-up period Reminders: Make an appointment for Visit 20a 64 weeks ± 5 days after randomisation Make an appointment for the End of the re-randomised Treatment visit (Visit 21a), 68 weeks ± 3 days after randomisation

46 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 44 of 112 Remind the subjects continuing in the 13 week follow up period to attend Visit 21a fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) Remind the subject to bring a sample of their morning urine at Visit 21a Visit 21a, End of re-randomised Treatment visit/ premature discontinuation This visit is only applicable for subjects continuing in the 13 week follow-up/12 week off-drug observational re-randomised treatment period. This visit is the final visit for the subjects included in the 12 week off-drug observational period. Subjects being prematurely withdrawn from the re-randomised treatment period trial during the 13 week follow-up/12 week off-drug observational period should preferably have all procedures for the End of the re-randomised Treatment visit (Visit 21a) performed. The primary reason (adverse event, non-compliance with protocol or other) for discontinuation must be specified in the ecrf and IV/WRS. Even if the subject is not able to attend, the End of Trial Form must be completed. The following will be performed and/or recorded in the ecrf: Smoking habits (see section 8.4.3) only applicable for the first 900 subjects completing the 56 week main part of the trial and continuing in the 13 week follow-up period Physical examination (see section 8.3.1) only applicable for the first 900 subjects completing the 56 week main part of the trial and continuing in the 13 week follow-up period ECG (see section 8.3.4) only applicable for the first 900 subjects completing the 56 week main part of the trial and continuing in the 13 week follow-up period Urine sampling for measurement of: Urinary Albumin-to-Creatinine ratio (see section ) Interact with the IV/WRS to: Interact with the Drug Accountability Module, only applicable for first 900 subjects continuing in the 13 week follow-up period Blood sampling for measurement of: Blood sampling is only applicable for the first 900 subjects continuing in the 13 week follow-up period Liraglutide antibodies (only applicable for subjects who discontinue the main trial after Visit week follow-up period prematurely) (see section 8.3.8) Reminders: Make an appointment for the Visit 22a ( follow-up visit) weeks ± 3 days after randomisation Remind the subjects to attend Visit 22a fasting

47 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 45 of Visit 22a, follow-up Visit 22a will take place weeks ± 3 days after randomisation. This visit is only applicable for the first 900 subjects continuing in the 13 week follow-up period. The following will be performed and/or recorded in the ecrf: Concomitant medication, changes since last visit, if any Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) Adverse events, since last visit ( see section 8.3.6) Mental health questionnaires (see section ) Visit 18b Visit 18b will take place 60 weeks ± 5 days after randomisation. Remind the subject to: attend category 1, 2, 3 and category 4 lab visits 19b s in addition to the End of Treatment visit fasting (i.e. having consumed only water since midnight i.e., at least eight hours overnight fast without food and/or drink intake, except for water) keep a 3-day food diary between Visit 18b and 19b, Visit 20b and 21b, Visit 22b and 23b, Visit 23b and 24b, Visit 25b and 26b, Visit 28b and 29b, Visit 31b and 32b, Visit 34b and 35b, Visit 37b and 38b, Visit 40b and 41b (the diary will be used for diet counselling at Visit 20b, 23b, 26b, 29b, 32b, 35b, 38b and 41b) (see section 5.3.2) Non-lab visits Visits 21b, 24b, 27b, 30b, 33b, 36b, 39b, 40b and 42b during the (delayed-onset-of-type-2-diabetes part of the trial) Non-lab visits are 21b, 24b, 27b, 30b, 33b, 36b, 39b, 40b and 42b. The following will be performed and/or recorded in the ecrf: Smoking habits (see section 8.4.3) Mental health questionnaire, applicable for Visit 21b, 24b, 27b, 30b, 33b, 36b 39b, 40b and 42b (see section 8.3.9) Remind the subject to: attend category 1, 2, 3 and category 4 lab visits 22b, 25b, 28b, 31b, 34b, 37b, 41b and 43b in addition to and the End of Treatment visit fasting (i.e. having consumed only water since midnight i.e., at least eight hours overnight fast without food and/or drink intake, except for water) Remind the subject to bring a sample of their morning urine at Visit 43b

48 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 46 of Category 1 lab Vvisits 20b, 26b and 32b during the (delayed-onset-of type-2- diabetes part of the trial) Category 1 lab visits are 20b, 26b and 32b. The following will be performed and/or recorded in the ecrf: Mental health questionnaire, only applicable for Visit 20b and 32b (see section 8.3.9) Remind the subject to: attend category 2, 3 and 4 lab visit and the End of Treatment visit fasting (i.e. having consumed only water since midnight) Category 2 lab Vvisits 23b, 29b, and 35b in the (delayed-onset-of-type-2-diabetes part of the trial) Category 2 lab visits are 23b, 29b and 35b. Urine sampling for measurement of: Urinary Albumin-to-Creatinine ratio (see section ) Remind the subject to: attend category 1, 2, 3 and 4 lab visits and the End of Treatment visit fasting (i.e. having consumed only water since midnight ) Category 3 lab Vvisits 19b, 22b, 25b, 28b, 31b, 34b, 37b and 38b during the (delayed-onset-of-type-2-diabetes part of the trial) Category 3 lab visits are 19b, 22b, 25b, 28b, 31b, 34b, 37b and 38b. The following will be performed and/or recorded in the ecrf: ECG only at Visit 25b, 31b and 37b (see section 8.3.4) Mental health questionnaires at Visit 19b, 22b, 25b, 28b, 31b, 34b, 37b and 38b (see section 8.3.9) Remind the subject to: attend category 1, 2, 3 and 4 lab visits 20b, 23b, 26b, 29b, 32b, 35b and 38b in addition to the End of Treatment visit fasting (i.e. having consumed only water since midnight i.e., at least eight hours overnight fast without food and/or drink intake, except for water) Remind the subject to bring a sample of their morning urine at Visits 23b, 29b and 35b

49 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 47 of Category 4 lab Vvisit 41b during the (delayed-onset-of-type-2-diabetes part of the trial) The only category 4 lab visit is Visit 41b. Remind the subject to: attend the End of 160 week randomised Treatment visit (Visit 43b) fasting (i.e. having consumed only water since midnight i.e., at least eight hours overnight fast without food and/or drink intake, except for water) Visit 43b, End of 160 weeks randomised Treatment visit/premature discontinuation Subjects being prematurely withdrawn from the trial should preferably have all procedures for the End of 160 week randomised Treatment visit performed. The following will be performed and/or recorded in the ecrf: Smoking habits (see section 8.4.3) PRO questionnaire (see section ) except for BES Urine sampling for measurement of: Urinary Albumin-to-Creatinine ratio (see section ) Interact with the IVRS/WRS to: Make completion call for subjects completing the trial Reminders: Make an appointment for Visit 44b, weeks ± 3 days, after randomisation (Visit 3) Remind the subjects to attend Visit 44b fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) Visit 44b, Follow-up visit /observational period Visit 44b will take place weeks ± 3 days after randomisation. The following will be performed and/or recorded in the ecrf: Mental health questionnaires (see section ) 8.2 Assessments for efficacy Any abnormal, clinically significant result identified at Screening visit 1 and Screening visit 2 will be recorded as concomitant illnesses Weight and height

50 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 48 of 112 Weight will be recorded to the nearest 0.1 kg. Weight should be measured at all visits to the clinic except for Screening visit 2 using calibrated scales. At Visits 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 17x, 17a, 18a, 21a, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, and 43b and 44b weight should be measured in the fasting state Waist circumference Waist circumference will be determined at all visits except for Screening visit 2, according to the procedure specified below. The waist circumference will be determined in a fasting state at Visit 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 17a, 18a, 21a, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, and 43b and 44b HbA 1c Samples will be drawn at Screening visit 1, Visit 3, 5, 8, 11, 14, 17, 17a, 21a, 20b, 23b, 26b, 29b, 32b, 35b, 38b, 41b, and 43b and 44b for measurement of HbA 1c. HbA 1c measured at Screening visit 1 will be used to assess compliance with exclusion criteria no. 2 (see section 6.4). The assay method used will be a National Glycohemoglobin Standardization Program certified method Fasting plasma glucose Samples for determination of FPG will be drawn at Screening visit 2, Visit 3, 5, 6c, 8, 9, 11, 14, 16, 17, 17a, 18a, 21a, 22a, 20b, 23b, 26b, 29b, 32b, 35b, 38b, 41b, 43b and 44b. FPG measured at Screening visit 2 will be used to assess compliance with exclusion criteria no. 4 (see section 6.4). In the event of FPG 126 mg/dl (7.0 mmol/l) at any time during the trial a repeated measurement should be taken within four weeks to confirm/exclude diagnose diagnosis of diabetes. Subjects developing diabetes during the trial should withhold their background diabetes medication and trial product on 2, 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 18a, 21a, 22a, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, 43b and 44b until blood sampling has been done. At all other visits diabetes background medication and trial product should be taken as usual during the conduct of the trial Fasting insulin Blood samples for determination of fasting insulin will be taken at Visit 3, 5, 8, 11, 14, 17, 17a, 18a, 21a, 22a, 20b, 23b, 26b, 29b, 32b, 35b, 38b, 41b, 43b and 44b C-peptide Blood samples for determination of fasting C-peptide will be taken at Visit 3, 5, 8, 11, 14, 17,17a, 21a, 20b, 23b, 26b, 29b, 32b, 35b, 38b, 41b, and 43b. A measure of βbeta-cell function (and insulin resistance) will be derived from FPG, fasting insulin and C-peptide data using the HOMA model 19.

51 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 49 of Oral Glucose Tolerance Test (OGTT) Standardised conditions will minimise lack of reproducibility. The OGTT must be performed after an overnight fast (minimum 8 hours). Subjects should be on an adequate carbohydrate intake before the test, should not be physically active during the test and must not smoke. The 120-min sample must be taken within 5 minutes of 120 minutes. Plasma should be immediately separated, or the sample should be collected into a container with glycolytic inhibitors and placed on ice-water until separated prior to analysis After collection of the baseline blood sample before the glucose load (0h sample) the subject should be given a 75 gram glucose-equivalent oral glucose challenge (to be supplied to the Investigator by A/S via Central lab) over the course of 5 minutes. Timing of the test is from the beginning of the drink. Blood samples will be taken at 0h, 10 min, 20 min, 30 min, 60 min, 90 min and 120 min to evaluate glucose levels. Smoking is not allowed during the test. Insulin and C- peptide concentrations will be assessed as well for estimation of beta-cell function. The OGTT measured at Screening visit 2 will be used to assess the a subject s pre-diabetes status and thereby assess whether the subjects is qualified to be stratified to eligibility for the delayedonset-of-type-2-diabetes part of the trial. In addition, the fasting and 2-hour post-change plasma glucose values will serve to rule out presence of type 2 diabetes at baseline. If FPG 126 mg/dl (7.0 mmol/l) and/or 2-hour post-challenge glucose is 200 mg/dl (11.1 mmol/l) the subject is a screening failure. Throughout the duration of the trial, any change in status based on OGTT-values will be recorded (normal, pre-diabetes, or diabetes). Pre-diabetes or diabetes may also be diagnosed based on HbA1c value(pre-diabetes: HbA 1c % both inclusive, diabetes HbA 1c 6.5 %). Subjects who have pre-diabetes according to their fasting or 2-hour value during an OGTT will be further classified according to the following pre-diabetes categories: isolated IFG, isolated IGT, or combined IFG/IGT (please refer to section for details). Within each of these categories, a subject may in addition fulfil the HbA 1c criteria for pre-diabetes but this information should not be entered in the ecrf. Similar to diagnosis of diabetes by FPG value (see section )Any change in glucose tolerance status (normal, impaired [fasting and/or 120 post glucose load] or diabetic glucose tolerance) as defined according to fasting glucose alone or an OGTT (i.e. both fasting and post load glucose) will be assessed at Visit 11, 17, 17a, 21a, 23b, 29b, 35b, 41b and 43b. In in the event of 2-hour postchallenge plasma glucose OGTT 200 mg/dl (11.1 mmol/l) at any time during the trial, a repeated measurement should be taken within four weeks to confirm in order to diagnose diagnosis of diabetes. The date of diagnosis of diabetes will be recorded in the ecrf (date of the first

52 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 50 of 112 assessment out of range). No subsequent OGTTs will be performed for subjects diagnosed with diabetes. In the event subjects are diagnosed with diabetes they should receive the best standard of care at the discretion of the Investigator. Glycaemic goals should follow recommendations laid out in Standards of Medical Care in Diabetes (21), i.e. HbA 1c <7%. If Investigator determines that insulin, GLP-1 receptor agonist (e.g., Byetta or Victoza ), or DPP-IV inhibitor, is the best treatment option, the subject must be withdrawn (see section 6.6, withdrawal criteria no. 4). Subjects who are diagnosed with diabetes should monitor their blood glucose at home with a glucose meter. The treatment target should be to maintain HbA 1c levels below 7%. If self-measured FPG falls below 6.1 mmol/l (117 mg/dl), or meets the criteria for glycaemic rescue i.e. from baseline to 6 weeks, increases above 15 mmol/l (270 mg/dl) or from 7 weeks to 12 weeks, increases above 13.3 mmol/l (240 mg/dl) or from 13 weeks to 160 weeks, increases above 11.1 mmol/l (200 mg/dl) on three consecutive days/occasions, the subject should contact the Investigator Impaired fasting glucose/ impaired glucose tolerance (IFG/IGT) Pre-diabetes status and classification A subject has pre-diabetes if the following applies: Blood samples for assessment of FPG and OGTT will be drawn at Screening visit 2 to assess prediabetes status (IFG/IGT). In the event of: FPG mg/dl / mmol/l, and < mg/dl / mmol/l) and/or IGT (2 hr post-challenge( OGTT) plasma glucose 140 mg/dl / 7.8 mmol/l, and < mg/dl / mmol/l) and/or HbA 1c % both inclusive (21) In the event of OGTT 200 mg/dl (11.1 mmol/l) and/or FPG 126 mg/dl (7.0 mmol/l) the subject is a screening failure. In the event of FPG 126 mg/dl (7.0 mmol/l) any time during the trial a repeated measurement should be taken within four weeks in order to diagnose diabetes. The date of diagnosis of diabetes will be recorded in the ecrf (date of the first assessment out of range). A subject will be further classified according to their fasting and postprandial glycaemia (irrespective of HbA1c value), as follows:

53 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 51 of 112 Isolated Impaired Fasting Glucose (iifg): FPG 100 mg/dl (5.6 mmol/l) to 125 mg/dl (6.9 mmol/l), both inclusive, and 2-hour post-challenge plasma glucose < 140 mg/dl (7.8 mmol/l) Isolated Impaired Glucose Tolerance (iigt): 2-hour post-challenge plasma glucose 140 mg/dl (7.8 mmol/l) to 199 mg/dl (11.0 mmol/l), both inclusive, and FPG < 100 mg/dl (5.6 mmol/l) Combined IFG/IGT (IFG/IGT): FPG 100 mg/dl (5.6 mmol/l) to 125 mg/dl (6.9 mmol/l), both inclusive, and 2-hour post-challenge plasma glucose 140 mg/dl (7.8 mmol/l) to 199 mg/dl (11.0 mmol/l), both inclusive the A subject who has pre-diabetes at baseline is eligible for stratification to continuing in the delayed-onset-of-type-2-diabetes part of the trial Blood pressure Systolic and diastolic blood pressure will be measured preferably in sitting position at all visits to the clinic except for Screening visit 2. Measurement will be done according to the procedure described in Appendix D. However, re-measurement of blood pressure at Screening visit 2 is allowed if white coat syndrome is suspected at Screening visit 1. Caffeine, smoking and physical activity should be avoided within 30 minutes prior to the blood pressure measurement at all visits to the clinic Cardiovascular biomarkers Blood samples for determination of selected cardiovascular biomarkers will be drawn at Visit 3, 11, 17, 17a, 21a, 23b, 29b, 35b, 41b and 43b. hscrp Adiponectin Fibrinogen PAI Lipids Blood samples for determination of fasting levels of TC, LDL-cholesterol, HDL-cholesterol, VLDL-cholesterol, TG and FFA will be drawn at Screening visit 2, Visit 3, 11, 17, 17a, 21a, 23b, 29b, 35b, 41b and 43b. It is important that the subject has been fasting (i.e. having only consumed water) at least since midnight i.e., at least eight hours overnight fast without food and/or drink intake, except for water) prior to blood sampling for lipids Patient reported outcomes (PRO) questionnaires PRO recordings will be performed in UK, Germany, Russia, Canada, US, Italy, Spain, France, Brazil, Austria, Mexico, Australia, Ireland and Netherlands.

54 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 52 of 112 Patient Reported Outcome (PRO) will be assessed at Visit 3, 11, and 17 for the 56 week main trial, at Visit 17a and 21a for the 56 week main trial including 13 week follow-up/12 week off-drug observational re-randomised treatment period and at Visit 23b, 30b, 36b and 43b for the delayedonset-of-type-2-diabetes part of the trial. Subjects discontinuing the trial prematurely should complete the questionnaire at the end of treatment Visits 17, 21a or Visit 43b, depending on when they discontinue the trial. PRO will be assessed by the questionnaires IWQOL-Lite 20, SF-36 21, 22, 23 and TRIMm- Weight questionnaire 24 and the binge eating scale (BES) Treatment Related Impact Measure (TRIMm) - Weight TRIMm-Weight (41) evaluates subjects overall impact and treatment satisfaction divided in 5 domains: efficacy, impact, side effects, psychological and convenience. TRIMm-Weight takes approximately 10 minutes to complete Binge eating scale questionnaire (BES) Subjects from UK, Germany, Russia, Canada, US, Italy, Spain, France and the Netherlands will complete the Binge Eating Scale questionnaire at Visit 3, 17, 17a and 18a. Subjects discontinuing the trial prematurely should not complete the questionnaire at the end of treatment visit. The Binge Eating Scale questionnaire is a 16-item self-report diagnostic tool designed to capture both the behavioural and emotional characteristics of binge eating. The scale takes approximately 10 minutes to complete 26, 27, Urinary Albumin-to-Creatinine ratio A urine sample will be taken for assessment of Urinary Albumin-to-Creatinine ratio. For the subjects in the56 week main trial including the re-randomised treatment period the urine sample will be collected at visits 3, 11, 17, 18a and Visit 21a. For the pre-diabetic subjects a urine sample will be collected at visits 3, 11, 17, 23b, 29b, 35b, 43b Physical examination Physical examination will be performed at Screening visit 1, and at the End of Treatment visit (Visit 17 for the 56 week main trial, Visit 21a, for the 13 week follow-up period and Visit 43b for the delayed-onset-of-type-2-diabetes part of the trial) according to local procedure. Physical examination should include the cardiovascular system, respiratory system, abdomen, central and peripheral nervous system, musculo-skeletal system and the thyroid gland Hypoglycaemic episodes Hypoglycaemic episodes will be classified in the database as follows:

55 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 53 of 112 Symptoms suggestive of hypoglycaemia are defined as: Adrenergic symptoms (e.g. tachycardia, palpitations, shakiness) Cholinergic symptoms (e.g. sweating) Neurological symptoms (e.g. inability to concentrate, dizziness, hunger, blurred vision, obvious impairment of motor functions, confusion or inappropriate behaviour but still alert enough to seek self-treatment) which reverse after intake of carbohydrates All hypoglycaemic episodes will be reported as AEs or SAEs during the trial. For subjects experiencing a severe hypoglycaemic episode, the event should be recorded as a SAE and as a minimum the following information should be collected/recorded: Whether seizure or coma developed, plasma glucose concentration prior to treatment of hypoglycaemia if available, which type of intervention was provided by another person (food, glucagon or i.v. glucose), and the duration of the symptoms. All episodes categorised as: Documented symptomatic Asymptomatic Probable symptomatic Relative

56 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 54 of 112 Minor hypoglycaemia will be recorded as AEs. An AE fulfilling the Minor hypoglycaemia criterion by definition also meets the Documented symptomatic criterion and should only be reported once (as Minor hypoglycaemia ). Hypoglycaemic episodes will be summarised by both The definitions ( major, minor, and symptoms only ) and ADA definitions (all criteria in figure 8-1, except for Minor hypoglycaemia which is not under ADA definitions). Both definitions will be used for the statistical analysis. If the subject develops diabetes during the trial, they will measure self-monitored glucose using glucose meters (see section 8.3.3) at the discretion of the Investigator. Glucose meters will be supplied by. Subjects developing diabetes should be provided with a diabetes diary and complete the hypo page if they experience a hypo. Hypos experienced by diabetic subjects must be recorded on the hypo page in the ecrf and not on the AE page as for non-diabetics. For subjects developing diabetes: Subjects developing diabetes will be supplied with a glucose meter and a diabetes diary for recording of hypoglycaemia. Plasma glucose should always be measured when there is the suspicion of a hypoglycaemic episode. All plasma glucose (PG) values 3.9 mmol/l (70 mg/dl), as well as values > 3.9 mmol/l (70 mg/dl) when hypoglycaemic symptoms have occurred, should be recorded by the subjects in the diaries. Hypoglycaemic episodes will be recorded by the subject in his/her diary throughout the trial and must be transcribed into the ecrf by the Investigator at each site visit throughout the trial. The recording should include: date of hypoglycaemic episode time of hypoglycaemic episode time of last trial drug prior to episode type of last trial drug prior to episode date and time of diabetes background treatment prior to the episode time of last main meal prior to episode whether the episode was symptomatic whether the episode was in relation to exercise whether the subject was able to treat him-herself (if not answered, the Investigator must provide an explanation in the ecrf) if the subject was not able to treat him-herself, whether he-she recovered with oral administration of carbohydrates the plasma glucose level before treating the episode (if available)

57 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 55 of 112 The answer to the question: Was subject able to treat him/herself? should be answered No if oral carbohydrates, glucagon or intravenous glucose had to be administered to the subject by another person because of severe central nervous system (CNS) dysfunction associated with the hypoglycaemic episode. Oral carbohydrates should not be given in case the subject is unconscious. A hypoglycaemic episode will be defined as treatment emergent if the onset of the episode is on or after the first day of randomised treatment, and no later than 14 days after the last day of randomised treatment. Hypoglycaemic episodes will be defined as nocturnal if the time of onset is between 00:01 and (both included). A hypoglycaemic episode form and an AE form must be filled in for all hypoglycaemic episodes. If the hypoglycaemic episode fulfils the criteria for a serious AE and/or a MESI, a hypoglycaemic episode form, an AE form and a safety information form must be filled in. Severe hypoglycaemic episodes will be recorded as MESIs. Hypoglycaemic episodes will be summarised based on the ADA classification (Figure 8-1), and also according to an additional definition. For subjects without diabetes: Hypoglycaemic episodes during the trial should be recorded on an AE page. A hypoglycaemic episode will be defined as treatment emergent if the onset of the episode is on or after the first day of randomised treatment, and no later than 14 days after the last day of randomised treatment. Hypoglycaemic episodes will be defined as nocturnal if the time of onset is between 00:01 and (both included). If the hypoglycaemic episode fulfils the criteria for a serious AE and/or a MESI, a hypoglycaemic episode form, an AE form and a safety information form must be filled in. Severe hypoglycaemic episodes will be recorded as MESIs. ADA hypoglycaemia classification According to the ADA the definition of a hypoglycaemic episode (Figure 8-1) is categorised as: Severe hypoglycaemia: An episode requiring assistance of another person to actively administer carbohydrate, glucagon, or other resuscitative actions. Documented symptomatic hypoglycaemia: An episode during which typical symptoms of hypoglycaemia are accompanied by a measured plasma glucose concentration 3.9 mmol/l (70 mg/dl). Asymptomatic hypoglycaemia: An episode not accompanied by typical symptoms of hypoglycamia, but with a measured plasma glucose concentration 3.9 mmol/l (70 mg/dl).

58 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 56 of 112 Probable symptomatic hypoglycaemia: An episode during which symptoms of hypoglycaemia are not accompanied by a plasma glucose determination (but that was presumably caused by a plasma glucose concentration 3.9 mmol/l (70 mg/dl)). Relative hypoglycaemia: An episode during which the person with diabetes reports any of the typical symptoms of hypoglycaemia, and interprets those as indicative of hypoglycaemia, but with a measured plasma glucose concentration > 3.9 mmol/l (70 mg/dl). No Severe hypoglycaemia Hypoglycaemia Subject able to treat it him/herself? No PG 3.9 mmol/l Asymptomatic (70 mg/dl) hypoglycaemia Yes Symptoms? Yes PG 3.9 mmol/l Documented (70 mg/dl) symptomatic hypoglycaemia PG>3.9 mmol/l Relative (70 mg/dl) hypoglycaemia No measurement Probable symptomatic hypoglycaemia Figure 8 1 Classification of hypoglycaemia Additional definition In normal physiology, hypoglycaemia symptoms occur at a blood glucose level of approximately < 2.8 mmol/l (50 mg/dl)/plasma glucose level < 3.1 mmol/l (56 mg/dl). Therefore, has used this cut-off value to define minor hypoglycaemia. Minor hypoglycaemic episode is defined as: An episode with symptoms consistent with hypoglycaemia with confirmation by plasma glucose < 3.1 mmol/l (56 mg/dl), or full blood glucose < 2.8 mmol/l (50 mg/dl) and which is handled by the subject himself/herself Or any asymptomatic plasma glucose value < 3.1 mmol/l (56 mg/dl) or full blood glucose value < 2.8 mmol/l (50 mg/dl)

59 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 57 of 112 Minor hypoglycaemic episodes will be summarised according to this definition, and can be subject to additional analysis Self-monitored plasma glucose If the a subject develops diabetes self-monitoring of plasma glucose will be performed using a glucose meter, supplied by. A sufficient amount of test strips, lancets and calibration solutions and a diabetes diary will be supplied together with the glucose meter. Subjects developing diabetes during the trial should withheld withhold their background diabetes medication and trial product on Visit 2, 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 17a, 18a, 21a, 22a, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, and Visit 43b and 44b until blood sampling has been done. At all other visits diabetes background medication and trial product should be taken as usual during the conduct of the trial ECG A 12-lead ECG with a rhythm strip of at least 10 seconds will be performed at Screening visit 1, Visit 11, Visit 17, 21a, 25b, 31b, 37b and 43b and will be evaluated by the Investigator. At Screening visit 1 an ECG taken within the period starting three weeks prior to Screening visit 1 and ending at the date of Screening visit 2 is acceptable. If performed before the subject has signed and dated the informed consent, it must be documented in the subject s notes that the assessment was not related to the trial. At Visit 43b the ECG must be performed on the day of Visit 43b or in the two weeks just prior to the visit. The ECG will be evaluated / recorded as: Normal Abnormal, not clinically significant Abnormal, clinically significant ECG printouts must be dated and signed by the Investigator on the day of evaluation and filed in the subject s record. In order to preserve the quality of the data, a copy of the dated /signed ECG tracings must be filed in the subject s records. Any changes in the ECG at Visit 11, 17, 21a, 25b, 31b, 37b or 43b compared to baseline (Screening visit 1 ECG), which fulfil the criteria for an AE, must be recorded as such (see section 12.2). A 12 lead ECG will be performed at Screening visit 1, Visits 11, 17, 21a, 25b, 31b, 37b and Visit 43b. An ECG taken within the period starting at Screening visit 1 and ending at the date of Visit 3 is acceptable. The ECG will be interpreted, signed and dated by the Investigator. The interpretation must follow the categories: Normal

60 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 58 of 112 Abnormal, not clinically significant Abnormal, clinically significant In case of an abnormal ECG, the Investigator must evaluate if the abnormal ECG fulfils the criteria of an AE or SAE, and if so, should be reported as an AE or SAE. All ECGs will in addition undergo central assessment. Sites will be informed of the central ECG evaluation in case this evaluation reveals an abnormal ECG reading. If the abnormality represent an unreported SAE or MESI, such SAE or MESI must be reported by the Investigator. Section Pregnancy test Females of childbearing potential will have a serum pregnancy test (hcg) performed in connection with Screening visit 1, and Visit 17 and the End of Treatment visit (Visit 17, 21a or 43b). Blood drawn for biochemistry will be used (see section 8.3.7). Urine pregnancy tests will be performed for females of childbearing potential at any time during the trial if a menstrual period is missed or as required by local law. Urine pregnancy kits will be supplied by the central laboratory. The test will be performed at the site. Pregnancy testing will not be required for women who have undergone a hysterectomy or bilateral tubal ligation, or for women above the age of 50, who have been without menstrual period for at least 1 year Haematology and biochemistry Samples will be drawn at Screening visit 1, Visit 3, 5, 8, 11, 14, 17, 17a, 18a, 21a, 19b, 22b, 25b, 28b, 31b, 34b, 37b, 38b, 41b, and 43b. Blood samples for measurement of calcitonin will not be drawn at Screening visit 1. Differential count: Eosinophils Neutrophils Basophils Monocytes Lymphocytes Biochemistry Calcium, albumin corrected Tryptase, if hypersensitivity is suspected

61 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 59 of 112 For subjects with calcitonin 20 ng/l (from Visit 3 and onwards) a repeat measurement of calcitonin must be performed preferably within four weeks and if confirmed, the event elevated calcitonin should be reported as a MESI (i.e. any confirmed episode of calcitonin concentration value 20 ng/l, if not already reported as an ongoing MESI, see section plus Appendix I) Liraglutide antibodies Blood samples to be analysed for serum antibody antibodies against liraglutide will be drawn at Visits 3, and 18a and Visit 22a for subjects completing the 56 week main trial including rerandomised treatment period. and continuing in the 12 week off-drug observational period. For subjects continuing in the blinded 13 week follow-up period antibodies will be drawn at Visit 3 and 22a. For subjects participating in the delayed-onset-of-type-2-diabetes part of the trial ( weeks in total) blood samples will be drawn at Visit 3 and 44b. Blood samples for subjects discontinuing the trial prematurely will be drawn at Visit 17, 21a or 43b depending on when the subject is discontinuing. The sample drawn at baseline (Visit 3) will be analysed either together with the sample drawn during the Follow-up/observational period or with the sample drawn at the End of Treatment visit for subjects discontinuing the trial prematurely Suspicion of acute hypersensitivity (allergic reaction) to trial product If acute hypersensitivity to trial product is suspected, local testing for blood tryptase concentration (total and/or mature tryptase) is recommended. If trial product is discontinued as a consequence of suspicion of acute hypersensitivity, blood sampling for central assessment of liraglutide antibodies and IgE-isotype of liraglutide antibodies should be conducted, at least 14 days after trial product discontinuation. Tryptase concentration (if measured) as well as results of liraglutide antibody and IgE-isotype liraglutide antibodies will be sent to and will be included in the final MESI report Suspicion of immune-complex disease If immune-complex disease is suspected, blood sampling for central assessment of complement levels (C3 and C4) should be conducted. Results should be included when reporting a MESI Mental health questionnaires Mental health will be assessed by the C-SSRS and the PHQ-9 in all subjects at all visits to the clinic except for Screening visit 2, Visit 4, 6,and Visit 17x, 18a, Visits 22a and Visit 44b. Subjects discontinuing the trial prematurely before Visit 17 will complete the mental health questionnaires at the premature discontinuation visit (Visit 17) (end of treatment visit) whereas subjects discontinuing the trial prematurely before after Visit 17 but before Visit 21a will complete

62 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 60 of 112 the mental health questionnaires at the premature discontinuation visit (Visit 21a) (end of treatment visit for subjects continuing in the 14 13/12 week follow-up/off-drug observational period). If subjects randomised to the delayed-onset-of-type-2-diabetes part of the trial discontinue the trial prematurely after Visit 17 but before Visit 43b they will complete the questionnaires at the premature discontinuation visit (Visit 43b). The Investigator or his/her delegate must review the PHQ-9 and C-SSRS questionnaire for completeness and adverse event immediately following administration C-SSRS The C-SSRS is a detailed questionnaire assessing both suicidal behaviour and suicidal ideation. The questionnaire will be administered as an interview by the Investigator 29.The Investigator must assess the C-SSRS score at Screening visit 1 and Visit 3 (randomisation) (see section 6.4) to exclude subjects with any suicidal ideation of type 4 (active suicidal ideation with some intent to act, without specific plan) or type 5 (active suicidal ideation with specific plan and intent) PHQ-9 The PHQ-9 is a 9-item self-reported depression screening tool. The questionnaire takes approximately 10 minutes to complete. The Investigator must assess PHQ-9 score at Screening visit 1 and Visit 3 (randomisation) to exclude subjects with a PHQ-9 score 15 (see section 6.4) Thyroidectomy pathology slides In case a subject undergoes a thyroidectomy (partial or total) for any reason during the duration of the trial, pathology slides of the thyroid tissue will be centrally reviewed in addition to the routine examination at the site level. Both the site pathology report and the central pathology report will be reviewed by an external independent Event Adjudication Committee (EAC). A set of pathology slides routinely made after thyroidectomies by the pathology laboratory of the hospital where the operation was performed will be sent centrally for a second reading by a pathologist with expertise in thyroid and C-cell pathology, who will be blinded to both trial treatment and site diagnosis. Once the samples are re-examined they will be sent back to the site laboratory. The Investigator will be informed of the second pathology report, in order to take appropriate action (e.g. in case of a difference to the diagnosis of the site pathology) Thyroid tissue sample collection in case of thyroidectomy Subjects scheduled for thyroidectomy will be asked to inform the Investigator prior to their operation. These subjects will be asked to consent to have a small sample of the removed thyroid tissue collected for testing of RET Y1062 phosphorylation in the thyroid C-cells. This is only

63 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 61 of 112 applicable if C-cell pathology is confirmed (i.e., hyperplastic or neoplastic thyroid C-cells), and only if allowed by local law. The tissue sample will be destroyed after examination Genetic testing in case of a confirmed C-cell pathology Subject scheduled for thyroidectomy will be asked to consent to be tested (blood sample) to identify germline RET gene mutations associated with MEN2 syndrome. This RET gene mutation detection will be conducted in subjects with pathology reports confirming C-cell abnormality (medullary carcinoma or C-cell hyperplasia). Genetic testing will only be performed if allowed by local law Binge eating scale questionnaire (BES) Subjects from UK, Germany, Russia, Canada, US, Italy, Spain, France, Brazil, Austria, Mexico, Australia, Ireland and the Netherlands will complete the Binge Eating Scale questionnaire at Visit 3, 17, and 18a. Subjects discontinuing the trial prematurely should complete the questionnaire at the premature discontinuation visit (Visit 17). The binge eating scale questionnaire is only applicable for subjects who did not have pre-diabetes at baseline. The Binge Eeating Sscale questionnaire is a 16-item self-report diagnostic tool designed to capture both the behavioural and emotional characteristics of binge eating. The scale takes approximately 10 minutes to complete 26, 27, 28. The questionnaires should preferably be completed by the subject before any trial procedures at the relevant visit are performed. Subjects should be given the opportunity to complete the questionnaires by themselves without interruption (see section ). The Investigator or his/her delegate must review Patient Reported Outcome(s) for completeness and adverse events immediately following administration Liraglutide concentration (population PK) A single blood sample for measurement of plasma liraglutide concentration will be drawn at Visit 4 (dose escalation period), Visit 7 and 11 (maintenance period). At Visit 4 and 7 subjects will be nonfasting for the blood sampling whereas they will be fasting for blood sampling at Visit 4 and 11. Subjects will be instructed to use the same injection site for three consecutive days prior to Visit 4, 7 and 11 and to note the exact date and time of administration, injection site and the dose of liraglutide/liraglutide placebo. Subjects will be instructed to note the exact date and time of administration, injection site and the dose of liraglutide/ liraglutide placebo for the three doses of trial product immediately prior to Visit 4, 7 and 11. A special laboratory will be responsible for the analysis of the liraglutide plasma concentration. Blood samples for liraglutide plasma concentration are collected, treated and shipped according to

64 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 62 of 112 the description in the laboratory manual supplied by the special laboratory. Samples will be analysed on an ongoing basis during the 56 week main trial Liraglutide concentration (PK C max sub-study) Subjects will be instructed to inject liraglutide/ liraglutide placebo at approximately 10 pm in the evening prior to Visit 8. Furthermore, they will be instructed to use the same injection site for three consecutive days prior to Visit 8 and to note the date and time for the injection. Diaries will be used to capture information on the injection date and time as well as injection site and dose. Subjects will be instructed to inject liraglutide/ liraglutide placebo at approximately 10 pm in the evening prior to Visit 8. Furthermore, they will be instructed not to inject liraglutide / liraglutide placebo in the morning of Visit 8 and to use the same injection site for three consecutive days prior to Visit 8. Diaries will be used to capture information on the injection date and time as well as injection site and dose Smoking habits At Screening visit 1, Visit 11, 17, 21a, 30b, 43b it should be recorded whether the subject is a smoker Family history of type 2 diabetes At Screening visit 1 it should be recorded whether a first-degree relative (parent, sibling or child) has been diagnosed with type 2 diabetes day food diary A 3-day food diary will be handed out to the subjects every second month. For subjects participating in the 56 week main trial including the re-randomised treatment period the food diary will be handed out at Screening visit 2, Visit 5, 7, 9, 11, 13, and 16 and Visit 19a. For subjects continuing in the 13 week follow-up period the diary will be handed out at Visit Diabetes diary A diabetes diary will be handed out to the subjects developing diabetes during the trial. The diabetes diary will be used to capture information on hypoglycaemic episodes. Subjects will be instructed by the Investigator or qualified staff how to enter information in the diary. Information on hypos from the diary will be transcribed into the hypo page and the AE page in the ecrf Dietary compliance and physical activity The subject s dietary compliance and the average daily level of physical activity will be recorded every second month. The subject will be questioned on whether they performed less than half an

65 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 63 of 112 hour, between half an hour and one hour or more than 1 hour of physical activity per day. For the main part of the trial including the re-randomised treatment period this applies to Visit 3, 6c, 8, 10, 12, 14, and 17 For the 13 week follow-up period the dietary compliance and physical activity will in addition be recorded at and Visit 20a. For the delayed-onset-of-type-2-diabetes part of the trial the subject s dietary compliance and average daily level of physical activity will be recorded at Visit 3, 6c, 8, 10, 12, 14, 17, 19b, 21b, 23b, 25b, 27b, 29b, 31b, 33b, 35b 37b, 39b, 41b and 43b. An increase in physical activity (recommended minimum150 minutes/week) will be encouraged and reenforced by use of pedometers Signs or symptoms of diabetic complications Information related to signs or symptoms of microvascular complications of diabetes (i.e., diabetic nephropathy (incuding microalbuminuria), retinopathy, neuropathy, and diabetic foot ulcer) will be recorded at Screening visit History of concomitant cardiovascular disease Information related to concomitant cardiovascular disease (i.e., myocardial infarction, disorders of rhythm or conduction, heart failure including NYHA class, ischaemic heart disease including type, Percutaneus coronary intervention (PCI) and coronary artery bypass graft surgery (CABG), left ventricular systolic dysfunction, left ventricular diastolic dysfunction, hypertension, ischemic stroke, transient ischemic attack, haemorrhagic stroke, intracranial artery stenosis, carotid artery stenosis, peripheral arterial disease incl. > 50% stenosis on angiography or other imaging) will be recorded in the ecrf at Screening visit History of gallbladder disease Information related to gallbladder disease will be recorded in the ecrf at Screening visit History of psychiatric disorders Information related to psychiatric disorders (specifically history of depression, suicidal behaviour, anxiety, mood disorders, insomnia, or sleep disorders) will be recorded in the ecrf at Screening visit Subject compliance The Investigator will at each visit remind the subject to follow protocol procedures. At all visits from Visit 5 except for Visit 6 and the Follow-up visits (Visit 22a and 44b) and at any unscheduled visit where drug accountability is performed the subject will return their partly used or unused trial products including all empty packaging material. The Investigator will assess the amount of trial product returned compared to what was dispensed at the last dispensing visit, and ask the subjects if

66 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 64 of 112 the trial product has been used as prescribed. There is no fixed compliance ranges. If a subject is discovered to be non-compliant, the Investigator should counsel the subject on the importance of taking trial products as directed Section 9 Trial supplies 9.1 Trial products The placebo and active drug are visually identical. 9.2 Packaging and labelling of trial product(s) All trial products will be packed and labelled by A/S and will be provided per in non subject specific boxes. Labelling will be in accordance with Annex 13, local law and trial requirements. The boxes will be provided with a label containing the information outlined below. The box label can be translated and modified according to local requirements: Clinical trial use only Trial Identification number Generic product name: see section 9.1 Strength: see section 9.1 Quantity: will be determined at packaging Dosage form Route of administration Code/batch number Expiry date Storage conditions: see section 9.3 Caution statements Unique number Sponsor s name and address Each Investigator site will be supplied with sufficient trial products for the trial on an ongoing basis controlled by the IV/WRS. Dispensing units will be prepared and distributed to the sites according to enrolment. Please refer to the TMM provided by for details regarding trial products standard packages. For further details please see the Trial Materials Manual (TMM). 9.3 Storage and drug accountability of trial product(s)

67 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 65 of 112 After first use of the liraglutide/ liraglutide placebo pen, the product can be stored for 14 days one month at room temperature (not above below +30 C) or in a refrigerator (2 C to 8 C). Keep the pen cap on when liraglutide/ liraglutide placebo pen is not in use in order to protect from light. Liraglutide/ liraglutide placebo should be protected from excessive heat and sunlight. US: 30 days at room temperature (+15 C to +30 C)/(59 F to 86 F) or in a refrigerator (+2 C to +8 C)/(+36 F to +46 F) Section 10 Randomisation, breaking of blinded codes and interactive voice/web response system (IV/WRS) 10.1 Randomisation At randomisation subjects will be stratified according to BMI and baseline pre-diabetes status. Subjects will be randomised to either liraglutide or liraglutide placebo using IV/WRS. Treatment allocation for subjects with pre-diabetes will be double-blind for the first 56 weeks, and single blind for the remaining 104 weeks (i.e. is unblinded, whereas Investigator and subjects remain blinded). At randomisation subjects will be stratified to either 56 weeks (plus 12 weeks re-randomised treatment plus 2 weeks follow-up period) or 160 weeks (plus 2 weeks follow-up period) of treatment based on pre-diabetes status (pre-diabetes Yes/No) at randomisation. Within these two groups subjects are furthermore stratified according to BMI (BMI 30, or BMI < 30). Treatment allocation for subjects with pre-diabetes will be double-blind for the first year (, Investigator and subject are blinded), and single blind for the remaining 2 years (i.e. is unblinded, whereas Investigator and subjects remain blinded). Subjects will be randomised to either liraglutide or liraglutide placebo using IV/WRS. At Visit 17 subjects who have received liraglutide treatment during the first 56 weeks of the trial will be rerandomised 1:1 to receive treatment with either liraglutide or liraglutide placebo for the following 12 weeks. The re-randomisation will be handled by the IV/WRS. Approximately 900 subjects without pre-diabetes participating in the one year main trial will be included in a follow-up period starting at week 56. The subjects selected for participation in the 13 week follow-up period are the subjects first to complete the 56 weeks. Subjects receiving placebo throughout the main part of the trial will continue to receive liraglutide placebo treatment for the follow-up period (approximately 300 subjects), whereas subjects receiving active liraglutide treatment (approximately 600 subjects) will be randomised 1:1 to either continue active liraglutide treatment (approximately 300 subjects) or switch from active liraglutide treatment to placebo (approximately 300 subjects).

68 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 66 of 112 The remaining (approximately 2100) non-pre-diabetic subjects will continue in a 12 week off-drug observational period Breaking of blinded codes The code for a particular subject may be broken in a medical emergency if knowing the identity of the treatment allocation would influence the treatment of the subject. Whenever a code is broken, the person breaking the code must record the time, date and reason as well as his/her initials in the source documents. print the Code Break Confirmation generated by the IV/WRS, record the reason, and sign and date the document. If the trial site needs to break the code, the sponsor should, if possible, be contacted prior to breaking the code. In all cases, the monitor must be notified within 24 hours after the code has been broken. (Monitor and department responsible for global product safety) will be notified immediately after the code break by the IV/WRS. When a subject should be withdrawn following a code break, a withdrawal session should be completed in IV/WRS. When code is broken the treatment allocation will be accessible to the Investigator and the department responsible for global product safety,. In case IV/WRS is not accessible at the time of code break the IV/WRS vendor helpdesk should be contacted. The treatment will be accessible for Investigators, Affiliates and the department responsible for global product safety, or any other relevant party using IV/WRS for code breaking. If the trial site or department responsible for global product safety needs to break the treatment code, they should interact with IV/WRS and obtain the information of the allocated treatment. The IV/WRS will register site number, subject number, caller, date and time, and inform sponsor immediately. If IV/WRS is not available, the IV/WRS supplier Help Desk (24h/7 days) should be contacted. If Affiliate(s), or any other relevant party needs to break the treatment code they should contact the IV/WRS supplier Help Desk (24h/7 days). All documentation on treatment code breaking must be kept throughout the trial period. The monitor will ensure that accountability of all broken or unbroken codes is documented at or after trial closure.

69 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 67 of Interactive voice/web response system (IV/WRS) Subject randomisation will be done by the use of a telephone or web based randomisation system, IV/WRS. The underlying randomisation will be performed by Clinical Supplies Coordination. Details about the use of the IV/WRS will be described in IV/WRS manual which will be available before the trial initiation. Furthermore, the IV/WRS will be used for registration of subjects at screening and completion, and for ordering and accountability of trial products. If additional dispensing of trial product is needed, the IV/WRS must be contacted in order to have correct trial product allocated. All drug accountability including the start day and last day on trial product will be performed in the IV/WRS. All information and procedures are described in the IV/WRS manual. Reports from the IV/WRS will be available for description of drug accountability per subject and to log supplies dispatched to each site. For further information about the use of the IV/WRS system, please see the IV/WRS manual that will be provided to all users. A trial specific IV/WRS will be set up, and can be accessed at any time by the internet. Some sessions may be available through a toll-free telephone number. Accessibility to the IV/WRS must be restricted to and controlled by authorised persons. IV/WRS is used for: screening screening failure randomisation dispensing medication arrival withdrawal completion code break drug accountability live data change An IV/WRS user manual will be provided to the trial site.

70 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 68 of Section 11 Concomitant illnesses/medical history and concomitant medication Definitions: Concomitant illness: any illness that is present at the start of the trial (i.e. at the first visit). Concomitant medication: any medication other than the trial product(s) that is taken during the trial, including the screening and run-in periods. Details of all concomitant illnesses and medication must be recorded at trial entry (i.e. at the first visit). Any changes in concomitant medication must be recorded at each visit. If the change influences the subject s eligibility to continue in the trial, the sponsor must be informed If a change is due to an AE then this must be recorded and reported according to section 12. If the change influences the subject s eligibility to continue in the trial, the sponsor then the Monitor must be informed. The information collected for each concomitant medication includes, (at a minimum), start date, stop date or continuing and indication Section 12 Adverse events and clinical technical complaints pregnancies 12.1 Definitions Adverse Event (AE): Any untoward medical occurrence in a subject or clinical investigation subject administered a pharmaceutical product, and which does not necessarily have to have a causal relationship with this treatment. An adverse event AE can therefore be any unfavourable and unintended sign (including an abnormal laboratory finding, for example), symptom, or disease temporally associated with the use of a medicinal product, whether or not considered related to the medicinal product. Note: This includes events from the first trial related activity after the subject has signed the informed consent and until post treatment follow-up period as defined in the protocol. For the subjects without pre-diabetes the post treatment follow-up period is defined as the period from the end of Visit 21a until the Follow-up visit (Visit 22a). For the pre-diabetic subjects the post treatment follow-up period is defined as the period from the end of the 160 week randomised treatment visit (V43b) until the follow-up visit (V44b). A worsening in concomitant illness must be recorded as an AE. A worsening of an ongoing AE should be reported on a new AE form by making a new assessment for seriousness and/or severity. Serious adverse event (SAE)

71 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 69 of 112 *The term hospitalisation describes a period of at least 24 hours. Over-night stay for observation, treatment at emergency room or treatment on an out-subject-basis does not constitute a hospitalisation. However, medical judgement must always be exercised and when in doubt the case should be considered serious. Hospitalisations for administrative, trial related and social purposes do not constitute hospitalisations as defined by the seriousness criteria for SAEs and should therefore not be reported as such. Likewise, hospital admissions for surgical procedures planned prior to trial inclusion are not considered adverse events. ** The term hospitalisation is the definition of a subject admitted to a hospital/inpatient (irrespective of the duration of physical stay), or not admitted to a hospital/not inpatient, but stays at the hospital for treatment or observation for more than 24 hours. Medical judgement must always be exercised, and when in doubt, the hospital contact should be regarded as a hospitalisation. Hospitalisations for administrative, trial related and social purposes do not constitute AEs and should therefore neither be reported as AEs or SAEs. Likewise, hospital admissions for surgical procedures planned prior to trial inclusion are not considered AEs or SAEs. Non-serious adverse events: Outcome categories and definitions: Recovered with sequelae - As a result of the AE the subject suffered persistent and significant disability/incapacity (e.g. became blind, deaf, paralysed). Any AE recovered with sequelae should be rated as an SAE If the sequelae meet seriousness criteria, the AE must be reported as an SAE Technical complaints A technical complaint is any written, electronic, or oral communication that alleges defects on trial products - listed as trial products in this protocol (9.1). The technical complaint may be associated with an AE, but does not concern the AE itself. A technical complaint may for example concern: the physical or chemical appearance of trial products (eg discolouration, particles or contamination) the packaging material (e.g. leakage, cracks, problems with rubber membrane in the cartridge or errors in labelling text) problems related to devices (e.g. to the injection mechanism, dose setting mechanism, glucose measurement, push button or interface between the pen and the needle)

72 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 70 of Collection, recording and reporting of adverse events All adverse events AEs, SAEs and MESIs must be recorded by the Investigator on an the Adverse Event AE Form in the Electronic Data Capture the (EDC ecrf) application. If more than one sign or symptom is to be reported, create a separate adverse event form for each sign and symptom. For serious adverse events, the safety information form must also be completed. For SAEs and MESIs, the SIF pages also have to be completed for each event in the ecrf. However if several symptoms or diagnosis occur as part of the same clinical picture, only one set of SIF can be used to describe all the SAEs. All concerned AE numbers must be included in the AE number field in the header of the SIF. For MESIs, the specific MESI follow-up questions have also to be completed in the ecrf within 7 calendar days. For MESIs qualifying for adjudication, the Source Data Collection Tool has also to be completed in the ecrf within 7 calendar days. The Investigator must report initial information on all SAEs to within 24 hours of obtaining knowledge about the event. The information must be provided by completion of the Adverse Event Form in the EDC application. If for any reason, the EDC application is unavailable, the information must be provided by fax, telephone or to the local affiliate. For details on the contact person please refer to Attachment II. The Safety Information Form on the paper CRFs must also be completed. The Investigator must forward electronically/fax/courier copies of the safety information form to within 5 calendar days of obtaining knowledge about the SAE. The Investigator must report initial information on all SAEs and MESIs to within 24 hours of obtaining knowledge about the event by completing the following in EDC ecrf: AE form SIF The specific MESI follow-up questions and source data collection tool also have to be completed within 7 calendar days if applicable. If for any reason the EDC application is unavailable, complete the AE form, SIF and if applicable the specific MESI follow-up questions and Source Data Collection Tool on paper CRFs within 7 calendar days and forward a copy electronically in PDF format by , or by fax or courier to within the same timelines Medical events of special interest

73 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 71 of 112 Events such as medication errors (e.g. wrong drug administration or wrong route of administration), suspected transmission of an infectious agent via a trial product, and other events pre-defined as a MESI in this protocol are always to be considered as medical events of special interest irrespective of seriousness and must follow the same reporting requirements and timelines as for SAEs. A MESI (serious or non-serious) is noteworthy event of scientific and medical concern that Novo Nordisk continues to monitor. A MESI does not necessarily have a causal relationship with the IMP. A MESI should be reported following the same reporting requirements and timelines as for SAEs (see section 12.1), irrespective of whether the MESI fulfils a SAE criterion. In this trial certain additional events have been defined as Medical events of special interest: The following are defined as MESIs in this trial (see Appendix J for further definitions) Medication errors concerning trial products administration of wrong drug or use of wrong device wrong route of administration, such as intramuscular instead of subcutaneous administration of a high dose with the intention to cause harm, e.g. suicide attempt administration of an accidental overdose, i.e. dose which may lead to significant health consequences, as judged by the Investigator, irrespective of whether the SAE criteria are fulfilled or not Suspected transmission of an infectious agent via a trial product Death Acute coronary syndrome (myocardial infarction or hospitalisation for unstable angina) Cerebrovascular event (stroke or transient ischemic attack) Heart failure Stent trombosis Revascularisation procedure Hospitalisation for cardiac arrhythmia Pancreatitis or acute, severe, persistent abdominal pain leading to a suspicion of pancreatitis Acute gallstone disease (biliary colic or acute cholecystitis) Elevated lipase or amylase 3 xunr Neoplasms Thyroid disease Any confirmed episode of calcitonin value 20 ng/l (from Visit 3 and onwards) Acute renal failure Severe hypoglycaemic events

74 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 72 of 112 Immunogenicity event ( allergic reactions including allergic reactions at injection sites, or immune-complex disease) AEs leading to withdrawal Psychiatric Disorders (including psychiatric disorders diagnosed by C-SSRS and PHQ-9 questionnaires) For subjects with calcitonin 20 ng/l (from visit 3 and onwards) a repeat measurement of calcitonin must be performed preferably within four weeks and if confirmed, the event elevated calcitonin should be reported as a MESI (i.e., any confirmed episode of calcitonin concentration value 20 ng/l, if not already reported as a MESI). For details on how increased calcitonin levels at screening and follow-up visits should be reported and followed up, please refer to Appendix J for reporting of MESIs and Appendix I describing how the Calcitonin Monitoring Committee works. In addition the following rules for increased lipase and /or amylase apply: If the amylase or lipase baseline (at screening) value is > 3xUNR this information will be recorded as medical history for that subject. If at any post baseline visit the amylase or lipase value is > 3xUNR a MESI should be reported. The information must be provided by completion of the Adverse Event Form in the EDC application. If for any reason the EDC application is unavailable, the information must be provided by fax, telephone or to the local affiliate. For details on the contact person, please refer to Attachment II. The Safety Information Form on the paper CRFs must also be completed and the Investigator must forward electronically/fax/courier copies of the Safety Information Form to the local affiliate. Certain MESIs will be adjudicated by an external independent event adjudication committee as described in section (12.2.2). For further information regarding definitions of MESIs and an overview of which events that should undergo adjudication, please refer to Appendix J. Complete the AE form, Safety Information Form (SIF), specific MESI follow-up questions and if applicable Source Data Collection Tool in the ecrf. If for any reason the electronic data capture (EDC) application is unavailable, complete the AE form, SIF, specific MESI follow-up question and if applicable Source Data Collection Tool on paper CRFs and forward a copy electronically in PDF format by , or by fax or courier to.

75 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 73 of 112 Pancreatitis Cases of acute and chronic pancreatitis should always be reported to the local affiliate as a MESI irrespectively of seriousness. Confirmed cases of pancreatitis should be followed-up with investigations of other potential causes (tests such as gallbladder ultrasound, triglycerides, liver enzymes, detailed history of concomitant medications or alcohol will be suggested for follow up as part of the Investigators consideration relative to local standards of practice). Observation of amylase or lipase > 3xUNR should always be reported to the local Novo Nordisk affiliate as a MESI irrespectively of seriousness.if the enzyme baseline (pre-dose) value is > 3xUNR this information will be recorded as medical history for that subject. If an enzyme post baseline value is > 3xUNR the observation should be reported as a MESI if applicable. Psychiatric Disorders Any psychiatric disorder categorised as an adverse event should always be reported to the local affiliate as a MESI irrespectively of seriousness. Mental health will be assessed using the PHQ-9 and C-SSRS questionnaires (see section 8.3.9). The assessments will be based on the subjects own response which will subsequently be verified by the Investigator. Mental health results categorised as adverse events should always be reported to the department responsible for global product safety on an adverse event form and safety information form irrespectively of seriousness. All questionnaires must be sent to data management for data entry. As a minimum, subjects found to have major depression or having suicidal behaviour/suicidal ideation should be referred to psychiatric follow-up according to local practice, whereas potential referral of subjects with any other mental health disorders to psychiatric follow-up is at the discretion of the investigator. Thyroid disorders All types of adverse events related to the thyroid (including changes in laboratory values considered AEs) should always be reported to the local affiliate as a MESI irrespective of seriousness. If the calcitonin baseline value (pre-dose) is > 2xUNR or if post baseline values > 2xUNR and increase > 50% from baseline, an alert will appear on the lab report and a fax will be sent to the investigator. If the calcitonin baseline value (pre-dose) is > 2xUNR this information should be recorded as medical history for that subject. If a calcitonin post baseline value is > 2xUNR and increase > 50% from baseline the observation should be reported as a MESI. In both cases a followup visit with a clinical endocrinologist is recommended. Bile duct disorders Any bile duct disorder categorised as an adverse event should always be reported as a MESI to the local affiliate irrespectively of seriousness. A potential previous history of cholecystitis, cholelithiasis or biliary colic should be probed by the investigator and be recorded as medical history prior to randomisation.

76 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 74 of 112 Neoplasms All types of neoplasm events should be reported as a MESI to the local affiliate irrespectively of seriousness. Important follow-up data includes information about malignancy versus non-malignancy, preferably histologically verified. Cardiovascular Disorders All cardiovascular disorders regarded as adverse events should always be reported to the local Novo Nordisk affiliate as a MESI irrespectively of seriousness. Immunogenicity Events which in the opinion of the Investigator may be causally related to immune mechanisms to trial product, for example acute IgE mediated reactions (clinical signs may include anaphylactic reactions, angiooedema and urticaria) and delayed type hypersensitivity (clinical signs may include various types of skin rashes), should always be reported as a MESI to the local affiliate irrespectively of seriousness. Adverse Events leading to withdrawal Adverse events leading to withdrawal from the trial, irrespective of organ class classification, should always be reported to the local affiliate as a MESI irrespectively of seriousness Events Adjudication Panel External independent event adjudication committee An independent blinded Events Adjudication Panel (EAP) will be established prior to FPFV with the aim of prospectively ensuring accurate and standardised classification of selected adverse events. Events adjudication will be performed for a subset of MedDRA terms within the MESI categories of Pancreatitis, Thyroid disorders and Psychiatric disorders. The specific MedDRA terms as well as the tasks, responsibilities and composition of the EAP will be described in a trial specific Events Adjudication Panel Charter prior to FPFV. An external independent Event Adjudication Committee (EAC) is constituted for this trial to perform ongoing adjudication, standardisation and assessment of events listed in Appendix J. The following events (also described in under MESIs) will be adjudicated and evaluated by the EAC in an independent and blinded manner: Death Acute coronary syndrome (myocardial infarction or hospitalisation for unstable angina) Cerebrovascular event (stroke or transient ischemic attack) Heart failure requiring hospitalisation Stent trombosis Coronary revascularisation procedure

77 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 75 of 112 Pancreatitis or acute, severe, persistent abdominal pain leading to a suspicion of pancreatitis Neoplasms Thyroid disorders requiring thyroidectomy The EAC is composed of permanent members who cover required medical specialities. The EAC members must disclose any potential conflicts of interest and must be independent of. The EAC works in accordance with written guidelines included in the EAC Charter that describes in detail the composition, tasks, responsibilities, and work processes of the committee. The charter will be finalised prior to first patient first visit. The EAC will perform adjudication based on the criteria and definitions described in the EAC Charter. The cardiovascular events will be classified according to FDA requirements (49). The EAC will review translated copies in English of medical documentation received in the adjudication packages (for example X-ray, ECGs, ultrasound images, discharge summaries, pathology reports, and death certificates). The Investigator will provide them as soon as possible, when they receive the request from. The role of the EAC is solely to adjudicate events in a blinded manner. The EACs will have no authorisation to impact on trial conduct, trial protocol and amendments. The assessments made by the EAC will be included in the CTR as well as assessments made by the Investigator. However, the adjudication made by an EAC, given its independence and in-depth analysis of each event, will be attributed with greater importance of the two. The outcomes of adjudication will be kept in the Global Safety database as well as in the clinical trial database Follow-up of adverse events Follow-up information (corrections, new or additional information) should be reported within 24 hours of obtaining knowledge of the information for SAEs and MESIs, and if previously non-serious AEs become SAEs by updating the AE form and/or SIF in the ecrf. All non-serious AEs classified as severe or possibly/probably related to the trial product must be followed until the subject has recovered or recovered with sequelae, and all queries have been resolved. However, ccases of chronic conditions or cancer or AEs ongoing at time of death (i.e. subject dies from another AE) can be closed with an outcome of recovering or not recovered. Cases can be closed with an outcome of recovering when the subject has completed the post-trial follow-up period and is expected by the Investigator to recover. If subjects die from another event, these cases can be closed with an outcome of recovering or not recovered.

78 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 76 of 112 All other non-serious AEs must be followed until the outcome of the event is recovering (for chronic conditions), or recovered or recovered with sequelae or until the end of the posttreatment follow-up stated in the protocol, whichever comes first, and until all queries related to these AEs have been resolved. If subjects die from another event, these cases can be closed with an outcome of recovering or not recovered. AEs ongoing at time of death (i.e. subjects dies from another AE) can be closed with an outcome of recovering or not recovered. The Investigator must record follow-up information on non-serious adverse events by updating the adverse event form in the EDC application ecrf. The follow-up information should only include new (updated and/or additional) information that reflects the situation at the time of the Investigator s signature. Queries or follow-up requests from should be responded to within 14 calendar days, unless otherwise specified. The Investigator must forward follow-up information on SAEs within 5 calendar days of obtaining the information. The information must be provided by completion of the Adverse Event Form in the EDC application and/or complete the Safety Information Form on the paper CRFs. The Investigator must forward electronically/fax/courier copies of the Safety Information Form to the local Novo Nordisk affiliate. If for any reason the EDC application is unavailable, the information must be provided by fax, telephone or the local affiliate. For details on the contact person, please refer to Attachment II. All serious adverse events must be followed until the outcome of the event is recovered, recovered with sequelae or fatal and until all queries have been resolved. For cases of chronic conditions and cancer or if the subject dies from another event follow-up until the outcome categories are recovered, recovered with sequelae or fatal is not required, as these cases can be closed with an outcome of recovering or not recovered. After access to update the AE form in the EDC application is removed the Investigator must record any SAE follow up information, if required, on the paper CRFs provided at trial closure. Queries or follow-up requests from should be responded to within 7 calendar days, unless otherwise specified. This must be done by updating the AE form and/or SIF in the ecrf. If for any reason EDC/eCRF application is unavailable, then the relevant paper forms have to be filled in, marked as follow-up and forwarded by fax or to within the same timelines. All SAEs and MESIs must be followed up until the outcome of the event is recovered, recovered with sequelae or fatal and until all queries have been resolved. Cases of chronic conditions or

79 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 77 of 112 cancer or AEs ongoing at time of death (i.e. the subject dies form another AE) can be closed with the outcome of recovered or not recovered. Cases can be closed with an outcome of recovering when the subject has completed the trial and is expected by the Investigator to recover. When a MESI of a specific category occurs for the first time in the subject and is selected on the AE form, the predefined follow-up section for that MESI category is automatically activated in the EDC. Each of these sections have follow-up forms related to the reported MESI, and the forms should be considered as a follow-up request from, and should therefore also be responded to within 7 calendar days from the date of awareness of the MESI. After access to update the AE form and SIF in EDC is removed the Investigator must record any SAE and MESI follow-up information, if required, on the paper CRFs provided at trial closure Collection and reporting of technical complaints All technical complaints occurring from the time of first and until the last usage of trial product - must be collected and reported to. The subject must be asked about technical complaints during each contact (site visit or telephone contact) with the Investigator or trial site staff. This may be done by posing a simple question such as have you experienced any problems since the last contact?. The Investigator must assess whether the technical complaint is related to: AE(s), SAE(s) and/or MESI(s) The AE(s), SAE(s) and MESI(s) related to technical complaint(s) must be reported by the Investigator following the same reporting requirements and timelines as for other AEs, SAEs and MESIs (see section12.1). Technical complaints must be reported on the technical complaint form by the Investigator, as described in the following: One technical complaint form must be completed for each trial product, non-investigational medicinal product (NIMP) or auxiliary supply. If the technical complaint involves more than one batch number, a technical complaint form for each batch number must be completed. The Investigator must complete the technical complaint form in the ecrf within the following timelines of the trial site obtaining knowledge of the technical complaint: technical complaint assessed as related to a SAE and/or MESI within 24 hours all other technical complaints within 5 calendar days

80 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 78 of 112 If the ecrf is unavailable, the paper technical complaint form should be completed and faxed to Customer Complaint Centre,, fax: , within the same timelines as for ecrf Collection, storage and shipment of technical complaint samples The Investigator must collect the technical complaint sample from the subject. If the technical complaint sample is unobtainable, the Investigator must specify on the technical complaint form why it is unobtainable. The technical complaint sample and a paper copy of the technical complaint form must be sent to within 5 calendar days of receiving the technical complaint sample at trial site by using the following address: A/S, Att.: Customer Complaint Centre, Nybrovej 80, 2820 Gentofte, Denmark. The Investigator must ensure that the technical complaint sample is labelled with the batch number and, if available, the DUN number. Storage and shipment of the technical complaint sample should be done in accordance with the conditions prescribed for the product (see section 9) Pregnancy Subjects must be instructed to notify the Investigator immediately if they become pregnant during the trial. The information must be provided by completion of the Adverse Event Form in the EDC application. If for any reason the EDC application is unavailable, the information must be provided by fax, telephone or the local affiliate. For details on the contact person, please refer to Attachment II. The Safety Information Form on the paper CRFs must also be completed and the Investigator must forward electronically/fax/courier copies of the Safety Information Form to The AE form and SIF in the ecrf must be completed. If for any reason the EDC/eCRF application is unavailable, then the relevant paper forms have to be completed and forwarded by fax or to. Pregnancies in partners of trial subjects

81 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 79 of 112 In case of an adverse event (with a causal relationship evaluated as possible or probable by the Investigator) in the foetus, newborn infant(s) or infant(s)/toddler(s) of a trial subject s partner, who is potentially exposed to the trial product via the trial subject, the pregnancy and the adverse event should be reported on the same forms as for a subject in the trial Precautions/over-dosage When initiating treatment with liraglutide, the subject may in some cases experience loss of fluids/dehydration, e.g. in cases of vomiting, nausea or diarrhoea. It is important to avoid dehydration by drinking plenty of fluids. Please refer to liraglutide obesity IB, 1 st edition rd edition 2010 or any updates hereof Calcitonin Monitoring Committee Monitoring of calcitonin in regular intervals will be implemented in the trial. Algorithm of further clinical and laboratory evaluation, supervised by an independent committee of thyroid experts (Calcitonin Monitoring Committee) will be recommended to be followed in all subjects with clinically relevant abnormal calcitonin values. This algorithm has been developed in collaboration with leading independent experts in thyroid/c-cell disease. In cases where the follow-up action recommended by the CMC, based on the elevated calcitonin levels results in establishing the presence of a thyroid disease, the thyroid disease should be reported as a new MESI. If the thyroid disease is a neoplasm or result in a thyroidectomy, the event will undergo adjudication by the neoplasm EAC Section 13 Case report forms 13.3 ecrf flow At the end of trial the Investigator must ensure that all remaining data have been entered into the ecrf no later than 3 days after the last subject s last visit at the site in order to ensure the planned lock of the database Questionnaires The PRO questionnaires, binge eating scale questionnaires and mental health questionnaires will be located in a CRF binder.

82 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 80 of 112 The CRF pages will consist of three-part NCR (no carbon required) paper. The original and the first copy are for the Clinical Departments (CRDs) and the second copy is for the Investigator Section 14 Monitoring procedures During the course of the trial, the monitor will visit the trial site to ensure that the protocol has been adhered to, that all issues have been recorded, to perform source data verification and to monitor drug accountability. The visit intervals will depend on the outcome of the remote monitoring of the CRFs, the trial site's recruitment rate and the compliance of the trial site to the protocol and GCP During the course of the trial the Monitor will visit the trial site to ensure that the protocol is adhered to, that all issues have been recorded, to perform source data verification and to monitor drug accountability. The visit intervals will depend on the outcome of the remote monitoring of the ecrfs, the trial site's recruitment rate and the compliance of the trial site to the protocol and GCP. Factors to be considered in this determination may include objective, purpose, design, complexity, blinding, size, and endpoint for the trial. The Monitor must visit the site shortly after the first subject has attended Screening visit 1 to ensure that mistakes are caught early. Hereafter the intervals between visits should not exceed 12 weeks. However, more frequent monitoring visits are required during peak periods such as recruitment period and finalisation of the trial. The intervals must as a minimum be: Shortly after the first subject at a site has attended Screening visit 1 Every 2-6 weeks during the recruitment period Thereafter the interval must preferably not exceed eight weeks For screening failures as a minimum the Screening Failure Form, SAE Form, if any and Affirmation Statement Form will be monitored and source data verified. The Monitor must ensure that all required ecrf forms for screening failures are completed, (e.g. screening failure form and the case book sign off (affirmation statement) is electronically signed by the Investigator). As a minimum requirement the following data must be source data verifiable in source documentation other than the ecrf: In/exclusion criteria Concomitant medication SIF The following data can be recorded directly on the CRFs and will be considered source data:

83 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 81 of 112 PRO questionnaires (IWQoL, SF-36, TRIMm-Weight and BES) For all data recorded the source document must be defined in a source document agreement at each site The existence of subjects must be confirmed either via extensive medical record (not just anamnesis collected at Screening visit 1) e.g. a copy of a passport or ID card could be used Section 15 Data management Data management is the responsibility of Data Management, Headquarters. Data management may be delegated under an agreement of transfer of responsibilities to another data management unit within Section 16 Computerised systems Note: New section has been added, consequently all subsequent headings will be updated in the revised protocol accordingly. However the following headings in this amendment will not be updated in order to keep reference to the original protocol. will capture and process clinical data using computerised systems which are described in Standard Operating Procedures and IT architecture documentation. The use and control of these systems are documented. Investigators working on the trial may use their own electronic systems to capture source data. will collect information on the practical use of these systems within the conduct of this clinical trial Section 16 Evaluability of subjects for analysis The following analysis sets are defined: Full analysis set (FAS) All randomised subjects exposed to at least one dose of the trial product and with at least one postbaseline assessment of any efficacy endpoint will be included. Subjects in the FAS will be analysed according to randomised treatment. The requirement of a post-baseline observation is in alignment with the FDA recommendations [reference: Guidance for Industry developing products for weight management, Draft guidance, February 2007].

84 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 82 of 112 Safety analysis set All randomised subjects who have been exposed to at least one dose of trial product. If a subject has received a different treatment than he/she was randomised to, data for the subject will be analysed, tabulated, and/or listed according to the actual treatment he/she has received. Subjects in the safety analysis set will be analysed as treated. Per protocol analysis set (PP) The Per protocol analysis set includes all subjects from the Full Analysis Set (FAS) and without significantly violating the inclusion/exclusion criteria or other aspects of the protocol considered to potentially affect the efficacy results of the primary endpoints All efficacy analyses will be based on the FAS. All safety analyses will be made using the safety analysis set. The PP analysis set will only be used for sensitivity analyses for the primary endpoints. The decision to exclude any subject or observation from the statistical analysis is a joint responsibility of the International Trial Manager, the Medical Writer and the Statistician. The International Medical Officer should be consulted as needed. The subjects or observations to be excluded must be documented and signed by all parties, prior to database release. The documentation must be stored together with the remaining trial documentation Section 17 Statistical considerations It is not acceptable that anyone, neither Investigator nor makes any form of between treatment group comparisons before database release. Biostatistics A/S will be responsible for the statistical analysis Sample size calculation A sample size of 3600 randomised subjects, 2400 randomised to liraglutide treatment and 1200 randomised to liraglutide placebo is considered to provide a reasonable estimation of the safety of liraglutide as a weight-management product. This sample size provides enough power for the primary efficacy endpoints weight change, the proportion of subjects with a weight loss larger than of at least 5%, the proportion of subjects with a weight loss larger greater than 10%, and new onset of diabetes. The hypothesis of equality between liraglutide and liraglutide placebo with respect to the primary endpoints will be tested in a hierarchical manner in the order in which the endpoints are mentioned. The power for the primary endpoint weight change is calculated based on a two sided t-test with a significance level of 5%. The power with regard to the co-primary dichotomous endpoints

85 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 83 of 112 proportion of subjects with a weight loss larger than of at least 5% and or more than 10%, respectively, is calculated based on a two-sided chi-square test. With a sample size of 2400 subjects treated with liraglutide and 1200 subjects treated with liraglutide placebo, the trial will have more than 90% power to detect a difference between liraglutide and liraglutide placebo in the proportion of subjects with a weight loss greater than 10%, given that the probabilities to achieve this weight loss is 10% for liraglutide placebo and 14% for liraglutide. The trial will also have more than 90% power to detect a difference in mean weight change between liraglutide and liraglutide placebo, given that the true difference is 0.70 kg and the standard deviation is 5.9 kg. In the interim analysis of data on weight loss after one year of treatment of obese subjects with either liraglutide, liraglutide placebo or orlistat (liraglutide obesity Investigators Brochure (IB), 1 st 3 rd edition ), the estimated difference in weight change between 3.0 mg liraglutide and liraglutide placebo was 5.82 kg, whereas the SD of weight loss was equal to 5.9 kg. Among subjects treated with 3.0 mg liraglutide the proportion of subjects with a weight loss larger greater than 5% and a weight loss larger greater than 10 % was 75% and 37%, respectively. The corresponding proportions in the placebo group were 28% and 10%. By assuming that the effects of treatment are of this magnitude the power to detect the differences between liraglutide and liraglutide placebo are larger than 99% for each of the three co-primary endpoints weight change, proportion of subjects with a weight loss larger than of at least 5% and proportion of subjects with a weight loss larger than 10%. It is believed that approximately 16-18% of the included obese subjects will have IFG/IGT. This corresponds to 384:192 and 432:216 included pre-diabetic subjects, respectively. This number of pre-diabetic subjects is considered to provide a reasonable estimation of the safety of liraglutide during 160 weeks of treatment of pre-diabetic subjects. The large number of randomised subjects also provides sufficient power for the fourth primary endpoint new onset of diabetes among subjects with pre-diabetes as assessed by IFG/IGT. The endpoint new onset of diabetes will be analysed using methods for analysis of interval censored failure time data. A low conservative estimate of the power may be calculated as if the endpoint diabetes yes/no among completers during the 160 weeks is analysed by use of a two-sided Chisquare test with a significance level of 5%. A high estimate may be calculated by applying ordinary survival analysis (without interval censoring), assuming exponential drop-out. This would correspond to very frequent visits, and the same drop-out rate during the 160 weeks. It is assumed that the annual conversion rate of subjects with IFG/IGT pre-diabetes to diabetes equals 7% among placebo treated subjects, whereas it is 70 60% lower, 2.1%, among liraglutide treated subjects. After 160 weeks of treatment with liraglutide/placebo, the percentage of subjects with diabetes is therefore equal to 1-(1-0.07) 3 = 20% among liraglutide placebo treated subjects,

86 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 84 of 112 and 1-( ) 3 = 6% among liraglutide treated subjects. It is assumed that the drop-out during the 160 weeks may be as large as 65% in both groups. Conversions of 20% and 6% during 160 weeks correspond to hazard rates of 7.4 and 2.1. Furthermore a drop-out equal to 65% during 160 weeks corresponds to a drop-out hazard equal to The power corresponding to these figures for conversion rates for placebo of 5, 7 and 9% and conversion rates 60 and 70% lower in the liraglutide group may be seen in Table Table 18 1 Sample size calculation for the primary endpoint onset of diabetes assessed at week 160 Annual conversion factor for placebo 5% 7% 9% Power for onset of diabetes with: Liraglutide conversion rate 60% lower than placebo Liraglutide conversion rate 70% lower than placebo The power calculations are based on a two-sided chi-square test of equal proportions using a 2:1 distribution and a significance level of 5%. With the minimum planned number of subjects with pre-diabetes and a drop out rate of 65%, 233 and 117 completers are expected in the liraglutide and placebo group respectively. Based on these figures it is apparent that a sample size of 2400 liraglutide treated subjects and 1200 liraglutide placebo treated subjects will provide sufficient power also for the fourth primary endpoint onset of diabetes, provided that 16-18% of the included subjects have IFG/IGT. Therefore a sample size of 3600 subjects is applied, provided the included number of pre-diabetic subjects is at least 600. If the frequency of pre-diabetic subjects is lower, more subjects should be included Statistical methods All statistical tests are two-sided at a 5% significance level. For all efficacy evaluations, only observations on drug (defined as last injection taken the day before or on the day of the visit) will be included in the statistical analyses and summaries. For all weight and glycaemic efficacy endpoints, only observations prior to rescue medication will be included in the statistical analyses and summaries, as rescue medication will confound the subsequent measurement of these parameters. Excluded observations will be listed Primary efficacy endpoints The four primary efficacy endpoints consists of three co-primary endpoints describing weight change and one endpoint describing onset of diabetes: Fasting body weight loss (defined as the change from week 0 to Week 56)

87 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 85 of 112 Change from baseline in fasting body weight at 56 weeks The proportion of subjects losing more thanat least 5% of baseline fasting weight (measured at Week 0) The proportion of subjects losing more than 10% of baseline fasting weight (measured at Week 0) Onset of type 2 diabetes in subjects with pre-diabetes at baseline evaluated at week 160 The first primary endpoint is suggested as a primary endpoint in the FDA guidance 18 as well as the EMEA guidance 32. The second primary endpoint is suggested by FDA, whereas EMEA instead focus on the third of the endpoints Primary analysis of the co-primary endpoints Description of the applied LOCF approach applied in the analysis of the co-primary endpoints describing weight change In the primary analyses of the three first endpoints, LOCF will be applied. Only fasting weight measurements will be used and only measurements performed after randomisation will be carried forward. The follow-up weight measurements at weeks (Visit 17x) after randomisation will not be applied in the primary analyses. Primary analysis of change from baseline in Fasting Body fasting body weight loss after at 56 weeks For the continuous primary endpoint, fasting body weight loss, analysed as the change in fasting body weight from Week 0 to Week 56 (using the last observation carried forward approach as described above) will be compared between liraglutide and liraglutide placebo using an ANCOVA model with treatment, country, pre-diabetes stratification factor, BMI-stratification factor, interaction between stratification factors and gender as fixed effects and with baseline body weight (at Week 0) as a covariate. The analysis will be performed for the FAS. From this model the expected differences in weight change between liraglutide treatment and liraglutide placebo will be estimated together with the associated 95% confidence interval and the p-value corresponding to the test of the hypothesis of no difference between treatments. The proportion of subjects losing more thanat least 5% of baseline fasting weight (measured at Week 0) For this categorical primary endpoint, a logistic regression model with treatment, country, prediabetes stratification factor, BMI-stratification factor, interaction between stratification factors and gender as fixed effects and with baseline fasting body weight (at Week 0) as a covariate, will be used to compare the proportion that after 56 weeks of treatment lose more than at least 5% of their baseline fasting bodyweight in the two groups. The analysis will be performed for the FAS using the LOCF approach described above. From this model the odds ratios between the liraglutide treatment and liraglutide placebo will be estimated together with the associated 95% confidence

88 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 86 of 112 interval and the p-value corresponding to the test of the hypothesis of no difference between treatments. Primary analysis of the proportion of subjects losing more than 10% of baseline fasting weight (measured at Week 0) This endpoint will be analysed in the same manner as the proportion of subjects losing more than at least 5% of baseline fasting weight. Primary analysis of the endpoint Proportion of subjects with onset of type 2 diabetes at 160 weeks onset of diabetes The time of onset of type 2 diabetes in subjects with pre-diabetes at baseline will be registered at specific visits. If presence of diabetes is observed at a visit, the onset must have occurred between this visit and the previous visit where diabetes was examined. Thus the observations can be considered as interval censored data. The time of onset will be set to be in between the first of the two required registrations of elevated FPG or OGTT 2 hr plasma glucose, and the diabetes assessment visit prior to the first registration. The data will be analysed using a Weibull model. The model will include effects of treatment, gender, and BMI stratification factor. Baseline FPG will be included as a covariate. This analysis will be carried out on the FAS. From this model the hazard rate between treatments will be estimated and a test on no difference between treatments will be performed. Description of the testing procedure The hypotheses of equality between liraglutide and liraglutide placebo for each of the four endpoints are tested in a hierarchical manner in the order in which the endpoints are presented. The implications of the testing procedure are as described below: liraglutide will be considered statistically significantly better than liraglutide placebo with respect to the first primary endpoint if the hypothesis of equality between liraglutide and liraglutide placebo is rejected and if the estimated effects of treatment are better in the liraglutide group than in the placebo group for the first primary endpoint. liraglutide will only be considered statistically significantly better than liraglutide placebo with respect to the second primary endpoint if it is considered statistically significantly better with respect to the first primary endpoint, if the hypothesis of equality between liraglutide and liraglutide placebo is rejected for the secondary primary endpoint, and if the estimated effects of treatment are better in the liraglutide group than in the placebo group for the second primary endpoint. liraglutide will only be considered statistically significantly better than liraglutide placebo with respect to the third primary endpoint if it is considered statistically significantly better with respect to the first and the second of the primary endpoints, if the hypothesis of equality between liraglutide and liraglutide placebo is rejected for the third co-primary endpoint, and if

89 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 87 of 112 the estimated effects of treatment are better in the liraglutide group than in the placebo group for the third primary endpoint. liraglutide will only be considered statistically significantly better than liraglutide placebo with respect to the fourth primary endpoint if it is considered statistically significantly better with respect to all the co-primary endpoints describing weight change, if the hypothesis of equality between liraglutide and liraglutide placebo is rejected for the fourth primary endpoint, and if the estimated effects of treatment are better in the liraglutide group than in the placebo group for the fourth primary endpoint Sensitivity analyses of the primary endpoints Sensitivity analyses of change from baseline in Fasting Body weight loss after 56 weeks For supporting evidence, the following sensitivity analyses will be carried out: The same analysis as above will be applied to the completers (week 56)PP analysis set The same analysis as above, applied to all randomised subjects allowing for baseline carried forward for subjects without a post baseline measurements The same analysis as above, applied to the subset of subjects in the PP analysis set with a valid assessment of weight loss at Week 56 The same analysis as above applied to the FAS using the follow-up measurements of fasting body weight at 52 weeks after randomisation (Visit 17x) The same analysis as above, applied to the FAS including the fasting and non-fasting weight measurements, off drug weight measurements, and using the follow-up weight measurements of fasting body weight at 5256 weeks after randomisation (Visit 17x) The same analysis as above, applied to the FAS including the fasting and non-fasting weight measurements, off drug weight measurements, and the follow-up weight measurements 56 weeks after randomisation (Visit 17x) and weight measurements following rescue medication The same analysis as above, applied to FAS, but imputing missing observations with the regression method (52). Five sets of imputations and subsequent analyses will be done The following repeated measures analysis (linear mixed effect model) using all longitudinal fasting weight measurements available for the FAS will be applied. The response variable is the change of body weight from baseline, and the model includes visit, treatment, country, prediabetes stratification factor, BMI-stratification factor, interaction between stratification factors and gender and the interactions between treatment and visit and baseline body weight and visit as fixed effects and with baseline body weight (at Week 0) as a covariate. Subject will be included as a random factor. The model will be used to compare liraglutide and liraglutide placebo at Week 56. Sensitivity analyses of the proportion of subjects losing more than at least 5% of baseline fasting weight (measured at Week 0) For supporting evidence the following sensitivity analyses will be carried out:

90 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 88 of 112 The same analysis as above, applied to the completers (week 56) PP analysis set The same analysis as above, applied to the subset of subjects in the PP analysis set with a valid assessment of weight loss at Week 56 The same analysis as above applied to the FAS using the follow-up measurements of fasting body weight at 52 weeks after randomisation (Visit 17x) The same analysis as above, applied to all randomised subjects allowing for baseline carried forward for subjects without a post baseline measurements The same analysis as above, applied to the FAS including fasting and non-fasting weight measurements, off drug weight measurements and the as well as follow-up weight measurements at 5256 weeks after randomisation (Visit 17x) The same analysis as above, applied to the FAS including fasting and non-fasting weight measurements, off drug weight measurements and the follow-up weight measurements 56 weeks after randomisation (Visit 17x) and weight measurements following rescue medication The same analysis as above, applied to the FAS, but regarding subjects without a valid assessment of weight at 56 weeks as non-responders, i.e. as not having lost 5% of their body weight. (For subjects withdrawing prematurely from the trial, their follow-up weights at 56 weeks after randomisation, if available, will be used) The same analysis as above, applied to FAS, but imputing missing observations with the regression method [Reference: Rubin, D.B. (1987), Multiple Imputation for Nonresponse in Surveys, New York: John Wiley & Sons, Inc.] Five sets of imputations and subsequent analyses will be done. Sensitivity analyses of the proportion of subjects losing more than 10% of baseline fasting weight (measured at Week 0) The sensitivity analyses for this endpoint will be analysed in the same manner as for the proportion of subjects losing more than at least 5% of baseline fasting weight. Sensitivity analyses of the endpoint proportion of subjects with onset of type 2 diabetes at 160 weeks For supporting evidence the following sensitivity analyses will be carried out: The same analysis as above, applied to the completers (week 160 ) the Per Protocol analysis set for the delayed-onset-of-type-2-diabetes part of the trial. A semi-parametric approach assuming proportional hazards applied to the FAS Analysis of secondary efficacy endpoints describing 56 week data All secondary efficacy endpoints will be based on the FAS. All endpoints will be summarized descriptively by visit using observed data. At end of treatment (week 56, 68 and 160) summaries will be presented for both observed and LOCF imputed data.

91 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 89 of 112 Summary statistics for continuous endpoints include number of observations, arithmetic mean, median, standard deviation, minimum and maximum. Summary statistics for categorical endpoints include number of observations, number and percentage of subjects fulfilling the criteria. Table 18.2, 18-3 and 18-4 gives an overview of the statistical analysis for the secondary endpoints at week 56, 68 and 160, respectively. Continuous secondary endpoints will be analysed and presented similarly to the primary analysis of weight change. Baseline values will be included as covariates in the analyses of the corresponding response variables. This analysis is referred to as ANCOVA in table 18-2, 18-3 and18-4. Categorical secondary endpoints will be analysed and presented similarly to the primary analysis of proportion of subjects losing at least 5% of baseline body weight. Continuous baseline values will be included as covariates in the analyses of the corresponding response variables unless otherwise specified. This analysis is referred to as LR in Table 18-2, 18-3 and Table 18-2: Overview of statistical analysis of secondary efficacy endpoints at week 56

92 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 90 of 112 Endpoint Endpoint type Statistical analysis Change from baseline (week 0) to week 56 in body weight (kg) Continuous ANCOVA Change from baseline (week 0) to week 56 in waist circumference Continuous ANCOVA Change from baseline (week 0) to week 56 in HbA1c Continuous ANCOVA Change from baseline (week 0) to week 56 in FPG Continuous ANCOVA Change from baseline (week 0) to week 56 in fasting insulin Continuous ANCOVA Change from baseline (week 0) to week 56 in fasting c-peptide Continuous ANCOVA Change from baseline (week 0) to week 56 in beta cell function (HOMA) Continuous ANCOVA Change from baseline (week 0) to week 56 in insulin resistance (HOMA) Continuous ANCOVA Change from baseline (week 0) to week 56 in OGTT parameters Continuous ANCOVA Change from baseline (week 0) to week 56 in systolic blood pressure Continuous ANCOVA Change from baseline (week 0) to week 56 in diastolic blood pressure Continuous ANCOVA Change from baseline (week 0) to week 56 in hscrp Continuous ANCOVA Change from baseline (week 0) to week 56 in adiponectin Continuous ANCOVA Change from baseline (week 0) to week 56 in fibrinogen Continuous ANCOVA Change from baseline (week 0) to week 56 in PAI-1 Continuous ANCOVA Change from baseline (week 0) to week 56 in total cholesterol Continuous ANCOVA Change from baseline (week 0) to week 56 in LDL cholesterol Continuous ANCOVA Change from baseline (week 0) to week 56 in HDL cholesterol Continuous ANCOVA Change from baseline (week 0) to week 56 in vldl cholesterol Continuous ANCOVA Change from baseline (week 0) to week 56 in triglycerides Continuous ANCOVA Change from baseline (week 0) to week 56 in free fatty acids Continuous ANCOVA Change from baseline (week 0) to week 56 in urinary albumin to creatinine ratio Continuous ANCOVA Scores from PRO questionnaire IWQoL-Lite at week 56 Continuous ANCOVA Scores from PRO questionnaire SF-36 at week 56 Continuous ANCOVA Scores from PRO questionnaire TRIm-Weight at week 56 Continuous ANCOVA Proportion of subjects with change from baseline (week 0) to week 56 in antihypertensives Categorical LR (lowering/ no change or increase) 2 Proportion of subjects change from baseline (week 0) to week 56 in lipid lowering Categorical LR agents (lowering/ no change or increase) 2 Proportion of subjects change from baseline (week 0) to week 56 in oral Categorical LR antidiabetic drugs (lowering/ no change or increase) 2 Proportion of subjects with type 2 diabetes at week 56 (yes/no) 1 Categorical LR Proportion of subjects with pre-diabetes at week 56 (yes/no) 3 Categorical LR 1 FPG at week 0 will be included as covariate 2 Relevant concomitant medication status (present/absent) will be included as covariate 3 Pre-diabetes status at week 0 will be included as covariate

93 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 91 of 112 For assessments in the re-randomised treatment period statistical analysis and summaries will be done for non pre-diabetes subjects in the FAS, who are re-randomised at 56 week. Statistical analysis models are similar to the main treatment period except week 56 is included as baseline covariate. Treatment comparisons will only be made between treatment groups liraglutide/liraglutide and liraglutide/placebo. Descriptive statistics will be done for all treatment groups. Table 18-3: Overview of statistical analysis of secondary efficacy endpoints at week 68 Endpoint Endpoint type Statistical analysis Change from baseline (week 0) to week 68 in fasting body weight (kg and %) Continuous N/A Change from week 56 to 68 in fasting body weight (kg and %) Continuous ANCOVA Change from baseline (week 0) to week 68 in waist circumference Continuous N/A Change from week 56 to 68 in waist circumference Continuous ANCOVA Change from baseline (week 0) to week 68 in FPG Continuous N/A Change from week 56 to 68 in FPG Continuous ANCOVA Change from baseline (week 0) to week 68 in systolic blood pressure Continuous N/A Change from week 56 to 68 in systolic blood pressure Continuous ANCOVA Change from baseline (week 0) to week 68 in albumin to creatinine ratio Continuous N/A Change from week 56 to 68 in albumin to creatinine ratio Continuous ANCOVA Change from baseline (week 0) to week 68 in diastolic blood pressure Continuous N/A Change from week 56 to 68 in diastolic blood pressure Continuous ANCOVA Scores from PRO questionnaire IWQoL-Lite week 0 to 68 Continuous N/A Scores from PRO questionnaire IWQoL-Lite week 56 to 68 Continuous ANCOVA Scores from PRO questionnaire SF-36 week 0 to 68 Continuous N/A Scores from PRO questionnaire SF-36 week 56 to 68 Continuous ANCOVA Scores from PRO questionnaire TRIm-Weight week 0 to 68 Continuous N/A Scores from PRO questionnaire TRIm-Weight week 56 to 68 Continuous ANCOVA N/A Not available For assessments of endpoints related to week 160, statistical analyses and summaries will be done for pre-diabetes subjects in the FAS. Statistical analysis models are similar to the week 56 analysis models, except factors related to the stratification factor pre-diabetes is not included in the model as all subjects are pre-diabetic at baseline. Table 18-4: Overview of statistical analysis of secondary efficacy endpoints at week 160

94 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 92 of 112 Endpoint Endpoint type Statistical analysis Change from baseline (week 0) to week 160 in body weight (% and kg) Continuous ANCOVA Proportion of subjects losing at least 5% of baseline body weight at week 160 Categorical LR Proportion of subjects losing more than 10% of baseline body weight at week 160 Categorical LR Change from baseline (week 0) to week 160 in waist circumference Continuous ANCOVA Change from baseline (week 0) to week 160 in HbA1c Continuous ANCOVA Change from baseline (week 0) to week 160 in FPG Continuous ANCOVA Change from baseline (week 0) to week 160 in fasting insulin Continuous ANCOVA Change from baseline (week 0) to week 160 in fasting c-peptide Continuous ANCOVA Change from baseline (week 0) to week 160 in beta cell function (HOMA) Continuous ANCOVA Change from baseline (week 0) to week 160 in insulin resistance (HOMA) Continuous ANCOVA Change from baseline (week 0) to week 160 in OGTT parameters Continuous ANCOVA Change from baseline (week 0) to week 160 in systolic blood pressure Continuous ANCOVA Change from baseline (week 0) to week 160 in diastolic blood pressure Continuous ANCOVA Change from baseline (week 0) to week 160 in PAI-1 Continuous ANCOVA Change from baseline (week 0) to week 160 in hscrp Continuous ANCOVA Change from baseline (week 0) to week 160 in adiponectin Continuous ANCOVA Change from baseline (week 0) to week 160 in fibrinogen Continuous ANCOVA Change from baseline (week 0) to week 160 in total cholesterol Continuous ANCOVA Change from baseline (week 0) to week 160 in LDL cholesterol Continuous ANCOVA Change from baseline (week 0) to week 160 in HDL cholesterol Continuous ANCOVA Change from baseline (week 0) to week 160 in vldl cholesterol Continuous ANCOVA Change from baseline (week 0) to week 160 in triglycerides Continuous ANCOVA Change from baseline (week 0) to week 160 in free fatty acids Continuous ANCOVA Change from baseline (week 0) to week 160 in albumin to creatinine ratio Continuous ANCOVA Scores from PRO questionnaire IWQoL-Lite at week 160 Continuous ANCOVA Scores from PRO questionnaire SF-36 at week 160 Continuous ANCOVA Scores from PRO questionnaire TRIm-Weight at week 160 Continuous ANCOVA Proportion of subjects with change from baseline (week 0) to week 160 in antihypertensives Categorical LR (lowering/ increase or no change) 1 Proportion of subjects change from baseline (week 0) to week 160 in lipid Categorical LR lowering agents (lowering/ increase or no change) 1 Proportion of subjects change from baseline (week 0) to week 160 in oral antidiabetic drugs (lowering/ increase or no change) 1 Categorical LR 1 Relevant concomitant medication status (present/absent) will be included as covariate Change from baseline to Week 56 in waist circumference, HbA 1c, FPG, insulin, C-peptide, OGTT and HOMA parameters, vital signs (systolic and diastolic blood pressure and pulse), Cardiovascular biomarkers (hscrp, adiponectin, fibrinogen, lipids) These variables will be analysed similarly to the primary analysis of weight change. The baseline values will be included in the models as covariates.

95 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 93 of 112 Metabolic syndrome status (ATP-III) at 56 weeks The metabolic syndrome status will be assessed after 56 weeks using the ATP III criterion introduced by the Adult Treatment Panel III (The ATP III criterion for clinical diagnosis of the metabolic syndrome is the presence of 3 to 5 of factors listed in Table 2 4) under the National Cholesterol Education Program (NCEP) 33. This variable will be analysed and presented similarly as the proportion of subjects losing more than 5% and 10% of initial bodyweight. Baseline status of the metabolic syndrome will be included as a covariate in the model. The analysis will be performed for the FAS and LOCF will be applied. Table 2 4 ATP-III criterion Risk Factor Abdominal Obesity* Waist Circumference Men Women Triglycerides High-density lipoprotein cholesterol Men Women Blood Pressure Fasting Glucose Defining Level >102 cm (>40 in) >88 cm (>35 in) 150 mg/dl <40 mg/dl <50 mg/dl 130 / 85 mmhg 100 mg/dl *Overweight and obesity are associated with insulin resistance and the metabolic syndrome. However, the presence of abdominal obesity is more highly correlated with the metabolic risk factors than is an elevated BMI. Therefore, the simple measure of waist circumference is recommended to identify the body weight component of the metabolic syndrome. Some male subjects can develop multiple metabolic risk factors when the waist circumference is only marginally increased, e.g cm (37-40in). Such subjects may have strong genetic contribution to insulin resistance and they should benefit from changes in life habits, similarly to men with categorical increases in waist circumference. Development of type 2 diabetes (yes/no) The binary endpoint development of type 2 diabetes; yes or no will be analysed by logistic regression. The model will include effects of treatment, gender, and BMI stratification factor. Baseline FPG will be included as a covariate. This analysis will be carried out on the FAS using LOCF. From this model the odds ratios between the liraglutide treatment and liraglutide placebo will be estimated together with the associated 95% confidence interval and the p-value corresponding to the test of the hypothesis of no difference between treatments.

96 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 94 of 112 Pre-diabetes status after 56 weeks. Pre-diabetes status (NGT, pre-diabetes, diabetes) after 56 weeks will be summarised, based on the FAS. LOCF will be applied. Furthermore the dichotomous variable NGT (yes/no) will be analysed in a manner similar to the analysis of metabolic syndrome after 56 weeks. Change in concomitant medication during 56 weeks of treatment For each of the selected medication types anti-hypertensives lipid lowering agents oral antidiabetic drugs The increase of medication during 56 weeks of treatment will be regarded as a dichotomous event. These variables will be analysed similarly as the primary analysis of the proportion of subjects losing more than 5%. The analysis will be performed for the FAS for the one year 56 week main trial. LOCF will be applied. Analysis of secondary efficacy endpoints describing 160 week data These endpoints will be assessed for subjects with pre-diabetes at baseline. Fasting body weight change from baseline to Week 160, Proportion of subjects losing more than 5% and 10% of baseline bodyweight at 160 weeks The endpoints will be analysed similarly to the primary analyses of the corresponding endpoints after 56 weeks of treatment. The analyses will be performed for the FAS. Change from baseline to Week 160 in waist circumference, HbA 1c, FPG, insulin, C-peptide, OGTT and HOMA parameters, vital signs (systolic and diastolic blood pressure and pulse), Cardiovascular biomarkers (hscrp, adiponectin, fibrinogen, lipids) The variables will be analysed similarly to the corresponding endpoints in the 56 weeks main trial. Metabolic syndrome status (ATP-III) after 160 weeks The variable will be analysed similarly to the corresponding endpoints in the 56 weeks main trial. Development of type 2 diabetes The binary endpoint development of type 2 diabetes; yes or no will be analysed by logistic regression. The model will include effects of treatment, gender, and BMI stratification factor. Baseline FPG will be included as a covariate. This analysis will be carried out on the FAS using LOCF. From this model the odds ratios between the liraglutide treatment and liraglutide placebo will be estimated together with the associated 95% confidence interval and the p-value corresponding to the test of the hypothesis of no difference between treatments

97 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 95 of 112 Pre-diabetes status after 160 weeks. Pre-diabetes status (NGT, pre-diabetes, diabetes) after 160 weeks will be summarised, based on the FAS. LOCF will be applied. Change in concomitant medication during 160 weeks The variables will be analysed similarly to the corresponding endpoints in the 56 weeks main trial. Analysis of secondary efficacy endpoints describing the follow-up period for subjects rerandomised to either placebo or liraglutide These endpoints will be assessed for subjects who complete the 56 weeks trial and are rerandomised to the follow-up period. Fasting weight change from baseline (Week 0) to Week 68 The change in fasting body weight from Week 0 to Week 68 (using the last observation carried forward approach as described in 0) will be analysed similarly to the analysis of the primary endpoint weight change. The analysis will be performed for subjects in the FAS. The effect of liraglutide/liraglutide as well as liraglutide/placebo will be tested against placebo/placebo. Fasting weight change from Week 56 to Week 68 The weight assessments in the follow-up period will be summarised by treatment and visit. The change in body weight from Week 56 to Week 68 will be summarised by treatment arm (liraglutide/liraglutide, liraglutide/placebo, placebo/placebo). The analysis will be performed for subjects in the FAS. Change in fasting waist circumference, FPG and Vital signs from baseline (Week 0) to Week 68 These endpoints will be analysed similarly to the weight change in the corresponding period. Change in fasting waist circumference, FPG and vital signs from Week 56 to Week 68 These endpoints will be analysed similarly to the weight change in the corresponding period. Change from baseline (Week 0) in Binge Eating (assessed by BES) to Week 56 and 57 The binge eating score will be summarised by treatment and visit. Change in binge eating score from baseline (Week 0) to Week 52 and 57 will be summarised by treatment. The analysis will be performed for subjects in the FAS Analysis of secondary efficacy endpoints describing the off-drug observational followup period for subjects entering the un-blinded 12 week follow-up period off drug These endpoints will be assessed for subjects who complete the 56 week trial and enter the unblinded 12 week follow-up period off drug.

98 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 96 of 112 Fasting weight change from baseline (Week 0) to Week 68 The change in fasting body weight from Week 0 to Week 68 (using the last observation carried forward approach as described in 0 will be analysed similarly to the analysis of the primary endpoint weight change. The analysis will be performed for subjects in the FAS. The effect of liraglutide will be tested against placebo. Fasting weight change from Week 56 to Week 68 The weight assessments in the observational off-drug period will be summarised by treatment and visit. The change in body weight from Week 56 to Week 68 will be summarised by treatment arm (liraglutide, placebo). The analysis will be performed for subjects in the FAS. Change in fasting waist circumference, FPG and Vital signs from baseline (Week 0) to Week 68 These endpoints will be analysed similarly to the weight change in the corresponding period. Change in fasting waist circumference, FPG and vital signs from Week 56 to Week 68 These endpoints will be analysed similarly to the weight change in the corresponding period. Change from baseline (Week 0) in Binge Eating (assessed by BES) to Week 56 and 57 The binge eating score will be summarised by treatment and visit. Change in binge eating score from baseline (Week 0) to Week 56 and 57 will be summarised by treatment. The analysis will be performed for subjects in the FAS Analysis of the safety endpoints In the 56 week part as well as the 160 week part all conducted analyses will be done with the aim to compare the liraglutide group to the liraglutide placebo group. In the 13 week follow-uprerandomised treatment period where subjects are re-randomised to either liraglutide or liraglutide placebo the treatments liraglutide/liraglutide, liraglutide/placebo and placebo/placebo will be tabulatedcompared. In the 12 week un-blinded observational period where subjects are off drug, the treatment arms liraglutide and liraglutide placebo will be compared. All analyses and tabulations regarding safety endpoints will be done using the safety analysis set. Separate summaries will be made for 56 week data, the 12 week observational period data, the 13 week follow-up period data and 160 week data. Unless otherwise specified, all safety assessments will be presented for each of the periods: Safety analysis set from randomisation (week 0) to week 56 Re-randomised subjects from re-randomisation (week 56) to week 70 Pre-diabetes subjects in the safety analysis set from randomisation (week 0) to week 162 Physical examination

99 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 97 of 112 Physical examinations are recorded at screening and at End of Treatment. Physical examination at screening as well as changes in physical examination will be summarised. Hypoglycaemic episodes Hypoglycaemic episodes follow the same definition as AEs for treatment emergence. All episodes reported by subjects with diabetes will be summarized by both the ADA definition and the definition. The categorisations are severe, asymptomatic, probable symptomatic, relative, documented symptomatic (ADA definitions) and major, minor, or symptoms only (Novo Nordisk definition). Frequencies of subjects experiencing treatment emergent hypoglycaemic episodes will be summarised by severity and treatment. Hypoglycaemic episodes not defined as treatment emergent will be presented in a listing. ECG Summary statistics and the frequencies of shifts from baseline to end of treatment will be tabulated for each treatment group. Adverse events Adverse events will be coded using the current version of MedDRA. A treatment emergent adverse event (TEAE) is defined as an event that either: Occurs before randomisation and increases in severity during the treatment period Has onset date on or after the first day of randomised treatment and no later than 714 days after the last day of randomised treatment Only treatment emergent adverse events (TEAEs) are considered for the tabulation and analysis of adverse events. The incidence of treatment emergent adverse events will be compared between treatment groups by means of descriptive statistics. Further details will be given in the SAP. Adverse events not defined as treatment emergent will be presented in a listing. Treatment emergent adverse events, TEAEs, are summarised descriptively, whereas non-treatment emergent AE s are presented in listings. TEAE data will be displayed in terms of the number of subjects with at least one event (N), the percentage of subjects with at least one event (%), the number of events (E) and the event rate per 100 years (R). Furthermore, TEAE data are summarized by seriousness, severity, relation to trial drug, MESI, withdrawal due to AEs and outcome. Summary tables by system organ class and preferred term are made for all TEAEs, serious TEAEs, possible or probably related to trial drug, severe TEAE, treatment emergent MESI, and TEAE occurring in at least 5% of the subjects in any treatment arm. Mental health questionnaires and binge eating scale

100 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 98 of 112 Results from the questionnaires will be summarised using descriptive statistics. Separate summaries will be made for each questionnaire by treatment and week. Pulse Pulse will be summarised and analysed similar to blood pressure. Analysis of laboratory safety parameters Laboratory safety parameters are measured throughout the trials and comprise haematology and, biochemistry as defined in the flowchart. No formal statistical analyses are planned for the laboratory safety parameters, except for calcitonin. The distribution of each continuous laboratory parameter will be presented using box plots by treatment and week. Continuous Llaboratory values will be compared to the relevant references ranges and results will be presented as follows: Shift tables for each laboratory parameter will be provided by treatment group. The shift tables will include the number of subjects below, within and above the reference ranges at baseline and after end of treatment For each laboratory parameter the proportion of subjects with laboratory values outside the normal ranges will be tabulated per visit and treatment group For each laboratory parameter individual values outside the reference ranges (abnormal values) will be listed by treatment and subject Categorical laboratory parameters will be summarised with shift tables. Amylase and lipase Tables showing shifts from baseline to highest value in treatment period to UNR, 2UNR or 3UNR will be presented. Mean plots by visit will be presented. Number and percentage of subjects with amylase and lipase levels UNR, 2UNR and 3UNR by treatment and visit will be tabulated. At end of treatment (week 56, 68 and 160) and follow-up (week 68) summaries will be presented for both observed and LOCF imputed data. Subjects with values above 2UNR will be presented with spaghetti plots and listings showing medical history of GI AEs, GI AEs and liver lab parameters. Calcitonin

101 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 99 of 112 As a substantial number of the calcitonin measurements are below lower limit of quantification, the data must be considered as partially censored at lower level of quantification (LLOQ). For the analysis of these data a repeated measurement model incorporating gender, visit, treatment and visit by treatment interaction will be used. Subject will be included as a random effect. Calcitonin will be presented as: Shift tables of the changes in levels by category: Below LLOQ, above LLOQ and below lower normal range, above upper normal range but below two times upper normal range, and above two times upper normal range. Done by treatment and gender, safety analysis set Summary tables of results by treatment and gender (data below LLOQ are set to 0.5*LLOQ) Bar chart representing the probabilities of shifting category Analysis results from the repeated measurements analysis All analyses done on the safety analysis set. Number, percentage and incidence of subjects with persistent (all post baseline measurements) and incidental (at least one post baseline measurement) increases in calcitonin for the criteria below will be tabulated for all subjects, males and females. From baseline <UNR to UNR From baseline <UNR to 1.5UNR From baseline <UNR to 20ng/L From baseline <UNR to 50ng/L From baseline <20ng/L to 20ng/L From baseline <50ng/L to 50ng/L Number and percentage of subjects with calcitonin levels UNR, 1.5UNR, 20ng/L and 50ng/L by treatment and visit will be tabulated. At end of treatment (week 56, 68 and 160) and follow-up (week 68) summaries will be presented for both observed and LOCF imputed data. The distribution of all calcitonin measurements across treatment groups and time will be shown with histograms and corresponding cumulative plots by gender and total for actual levels. Sum curves will be done for baseline (pooled) and week 56, 68 and 160 (LOCF) by treatment. In addition histograms of baseline level (pooled) and change at week 56, 68 and 160 (LOCF) will be done by treatment. A summary table showing number and percentage of observations < and LLOQ, minimum, Q25, median, Q75, maximum and geometric mean will be done by treatment group, gender and week. Geometric means will be plotted by treatment and visit in order to assess the pattern of the longitudinal changes.

102 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 100 of 112 In addition, a scatter plot of baseline vs. maximum post baseline calcitonin measurement will be done by treatment. Longitudinal changes for subjects with calcitonin 20ng/L will be evaluated with spaghetti plots. The spaghetti plots will follow the above by treatment group and gender. Selected listings of subjects with at least one post baseline calcitonin measurement above 20ng/L will be done. The listings will include treatment, age, gender, smoking habits at baseline, risk factor information (use of relevant concomitant medication at time of assessment (proton pump inhibitors and H2 blockers) and medical history of thyroid disorder) and calcitonin measurements. Liraglutide antibodies Observations with positive liraglutide antibodies 34 above the upper normal range will be examined for cross-reactivity and neutralizing effect. Frequencies of subjects with liraglutide antibodies will be tabulated by week. Similarly, frequencies of subjects with liraglutide antibodies with neutralizing effect and with cross-reacting liraglutide will be tabulated Interim analysis Not applicable. No interim analysis is planned, but a separate analysis and subsequent report will be made using data collected up to 69 weeks of treatment. This analysis will be made to evaluate the primary objective regarding weight loss, namely to establish efficacy of liraglutide compared to liraglutide placebo in inducing and maintaining weight loss over 56 weeks. Both this analysis and the analysis of the subjects with pre-diabetes in the delayed-onset-of-type-2-diabetes part of the trial based on 160 week data will be described in this protocol Sequential safety analysis/safety monitoring Sequential safety analysis for medical events of special interest is planned to be monitored by an internal Data Monitoring group. No formal statistical analyses are planned, blinded summary statistics and graphical presentation of data will be the basis of the safety group s decisions Explorative statistical analysis for pharmacogenetics and biomarkers Not applicable PK and/or PD modelling The pharmacokinetics of liraglutide 3.0 mg in steady state conditions will be evaluated using population PK analysis methods. The analyses will be based on plasma samples taken from a

103 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 101 of 112 subgroup of subjects, using a rich sampling scheme, and plasma concentrations taken from all subjects, using a sparse sampling scheme. A separate modelling plan will be prepared before Database release outlining details of the analysis Health economics and/or subject reported outcome Please refer to section The PRO scores IWQoL-Lite, SF-36 and Treatment Related Impact Measure - Weight (TRIM- Weight) will be analysed similarly to the primary analysis of weight change based on 56 week data as well as 160 week data. Baseline values will be included as a covariate in the models. The change in PRO-scores from week 56 to 68 and from week 0 to 68 based on data from the follow-up period and the observational period, respectively, will be analysed in a similar manner as the corresponding changes in weight. Baseline values will be included as a covariate in the models Section 18 Ethics The trial will be conducted in compliance with ICH GCP( 53 ), and applicable regulatory requirements, and in accordance with the Declaration of Helsinki (54). All subjects participating in the trial will receive instruction in a hypocaloric diet calculated to induce a moderate weight loss during the trial, and all subjects will receive instruction and encouragement on regular physical activity to aid in weight management and improvement of risk factors. A standard panel of safety laboratory evaluations will be performed regularly during the trial (including vital signs, ECG, haematology, and biochemistry), and side effects will be monitored closely. The adverse events reported by most subjects in the phase 2 study in non-diabetic obese subjects were nausea (between 24 and 48%), diarrhoea (between 8 and 15%) and vomiting (between 5 and 15%), increasing with liraglutide dose; the corresponding percentages for the placebo group were 7%, 10% and 2%, respectively. Most of the gastrointestinal events were transient, occurring within the first 4-6 weeks of treatment, and were of mild or moderate intensity. Headache and dizziness were frequent adverse events, but were not reported more frequently with liraglutide compared to placebo (headache: 12-20% compared to 15% with placebo; dizziness 4-6.5% compared to 7% with placebo). Other common AEs included infections such as upper respiratory tract infections, influenza and nasopharyngitis, and back pain, but in general these events were equally distributed across all treatment groups.

104 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 102 of 112 Patients treated with liraglutide should be advised of the potential risk of dehydration in relation to gastrointestinal side effects and take precautions to avoid fluid depletion. Few cases of acute pancreatitis (inflammation of the pancreas) presenting with persistent severe abdominal pain (usually accompanied by vomiting) have been reported with liraglutide and exenatide (a drug belonging to the same drug-class as liraglutide, and currently approved for the treatment of type 2 diabetes in the US and in Europe). If the investigator suspects acute pancreatitis, all suspected drugs should be discontinued until confirmatory tests have been conducted and appropriate treatment should be initiated. Subjects that are diagnosed with acute pancreatitis (as a minimum 2 of 3: characteristic abdominal pain, amylase and/or lipase >3x UNR or characteristic findings on CT/MRI), should be withdrawn from the study. As of October 1st 2008 a total of 8 pancreatitis adverse events (liraglutide: 7, non-liraglutide: 1) have been reported during the total liraglutide development programme (7 of the events occurred in subjects with type 2 diabetes and 1 event was in a non-diabetic obese subject). Two of the cases of pancreatitis were chronic pancreatitis. All pancreatitis events were reported in the intermediate-term or long-term trials. The rate of total pancreatitis across treatments was 2.2, 0.0, 0.9 and 0.6 events per 1,000 subject years of exposure for total liraglutide, placebo, active comparator and total comparator, respectively. The rate of acute pancreatitis was 1.6, 0.0, 0.9 and 0.6 for total liraglutide, placebo, active comparator and total comparator, respectively. The rate of chronic pancreatitis was 0.6 events per 1,000 subject years of exposure for total liraglutide, no chronic pancreatitis events were reported for placebo or comparator treatments. There was no time correlation between initiation of therapy and onset of acute pancreatitis. Several of the subjects had known risk factors or fully recovered while continuing on liraglutide therapy. Based on this, causality between the development of pancreatitis and treatment with liraglutide cannot be established or excluded. Thyroid C-cell tumours were seen in the 104-week carcinogenicity studies in mice and rats. Based on mechanistic studies and data from the literature, the C-cell proliferative changes seen in rodents are associated with liraglutide. It is possible that chronic activation of the GLP-1 receptor present on thyroid C-cells by liraglutide can lead to hyperplastic and neoplastic changes in the thyroid. This may be associated with an increase in plasma calcitonin levels in rodents. A mode of action analysis is underway to assess the human relevance of the C-cell tumors observed in the rodent carcinogenicity studies. The C-cell tumors induced in mice and rats by dosing of liraglutide are thought to be caused by a non-genotoxic, specific receptor mediated mechanism. The thyroid tumors and C-cell proliferative changes induced by liraglutide in rodents is considered of uncertain relevance to human subjects.

105 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 103 of 112 In the combined liraglutide diabetes and obesity clinical development programs, the overall rates for thyroid adverse events were 33.3, 29.5, 18.8 and 22.0 events per 1000 subject years of exposure for liraglutide, placebo, active comparator and total comparator, respectively. Rates of thyroid neoplasms were comparable for subjects treated with liraglutide and subjects treated with placebo and higher than for subjects treated with active comparator (7.0, 6.3, 0.9 and 2.5 events per 1000 subject years of exposure for liraglutide, placebo, active comparator and total comparator, respectively). The vast majority of thyroid neoplasms were benign. C-cell hyperplasia has been diagnosed in 2 subjects with papillary thyroid cancer treated with liraglutide and in 1 subject with a thyroid cancer treated with glimepiride+metformin. The latter event was a C-cell hyperplasia carcinoma in situ. The distribution of C-cell hyperplasia between the treatment groups was in accordance with the randomisation scheme. Few cases of acute pancreatitis (inflammation of the pancreas) presenting with persistent severe abdominal pain (usually accompanied by vomiting) have been reported with liraglutide and exenatide. Post-marketing surveillance identified at least 30 cases of pancreatitis with exenatide.26 However, a health services registry-based study found no increased frequency of pancreatitis among exenatide users.27 Liraglutide causes dose-dependent and treatment-duration-dependent thyroid C-cell tumors at clinically relevant exposures in both genders of rats and mice. It is unknown whether liraglutide causes thyroid C-cell tumors, including medullary thyroid carcinoma (MTC), in humans, as human relevance could not be ruled out by clinical or nonclinical studies. Liraglutide is contraindicated in patients with a personal or family history of MTC and in patients with Multiple Endocrine Neoplasia syndrome type 2 (MEN 2). Based on the findings in rodents, monitoring with serum calcitonin or thyroid ultrasound was performed during clinical trials, but this may have increased the number of unnecessary thyroid surgeries. It is unknown whether monitoring with serum calcitonin or thyroid ultrasound will mitigate human risk of thyroid C-cell tumors. In a 2-year repeat subcutaneous dose carcinogenicity study of liraglutide injected once a day in CD-1 mice, a treatment-related increase in fibrosarcomas was seen on the dorsal skin and subcutis, the body surface used for drug injection, in males in the 3 mg/kg/day group. These fibrosarcomas were attributed to the high local concentration of drug near the injection site. The liraglutide concentration in the clinical formulation (6 mg/ml) is 10 times higher than the concentration in the formulation used to administer 3 mg/kg/day liraglutide to mice in the carcinogenicity study (0.6 mg/ml). Further information can be obtained in the Liraglutide Investigator's Brochure Obesity, Edition 3, 08 October 2010 or any updates hereof Informed consent form for trial subjects

106 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 104 of 112 In obtaining seeking and documenting informed consent, the Investigator must comply with the applicable regulatory requirement(s) and adhere to the ICH GCP 31 and the requirements in the Declaration of Helsinki 35. A voluntary, signed and personally dated, incl. time, informed consent form will be obtained from the subject prior to any trial-related activity. The responsibility for obtaining seeking informed consent must remain with that of a medically qualified person and cannot be delegated to a non-medically qualified person the Investigator or an adequately medically qualified person delegated by the Investigator. The written informed consent must be signed and personally dated, incl. time, by the person who obtained seeks the informed consent. The information given to the subjects may be.given by e.g. a study nurse, but the Investigator or another physician delegated by the Investigator must always sign and date (i.e. obtain) the informed consent. Any local requirements must be followed. If the subject decides to withdraw consent to trial participation, is entitled to keep and use data which were obtained before the subject withdraws consent. Following subject s trial participation, the Investigator will advise the subject with regards to the best possible treatment options for his/her management of weight control. No additional care will be provided. Liraglutide will not be an available treatment option by the time this trial is planned to be finished, as this product is still under development for the obesity indication and therefore not approved by any Health Authority Data Handling Data collected will be used as part of the per protocol/intention to treat population Safety events will be reported to the department responsible for global product safety, Novo Nordisk /regulatory authorities If data is used, it will always be in accordance with local law and IRB/IEC procedures Data collected will be retained by and entered into the database Safety events will be reported to the department responsible for global product safety, Novo Nordisk/regulatory authorities according to local/national requirements If data is used, it will always be in accordance with local law and Institutional Review Board (IRB)/Independent Ethics Committee (IEC) procedures.

107 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 105 of Section 19 Premature termination of the trial/ trial site, Investigator or a pertinent regulatory authority may decide to stop the trial/trial site or part of the trial at any time, but agreement on procedures to be followed must be obtained. Unplanned discontinuation of the trial could occur if any safety issues arise during the conduct of this or other trials in the obesity development programme. In addition the trial may be discontinued if an overall risk/benefit evaluation is not in favour of continuation. In case of premature discontinuation of the trial, the subjects will be asked to attend a last visit, if possible. At this visit, procedures according to Visit 17/21a/43, including blood sampling should be done. CRF pages for Visit 17/21a/43 and the End of Trial form must be completed. Even if a subject is not able to attend a last visit, the End of Trial form and the Case Book sign off must be completed Section 20 Protocol compliance Deviations from the protocol should not occur be avoided Section 21 Critical documents Before the Investigator starts the trial (ie obtains informed consent from the first subject i.e. the site has green light for screening subjects), the following documents must be available to Novo Nordisk: Signed financial agreement(s). In the US, the following should be stated instead: Verification under disclosures per CFR of Financial Conflict of Interest 36 Signed FDA form 1572 for each Investigator, if applicable Signed FDA financial disclosure form or local equivalent as applicable financial agreement(s) For US: Verification under disclosures per Code of Federal Regulations (CFR) of Financial Conflict of Interest (20) FDA financial disclosure form or local equivalent as applicable. Signed and dated FDA form 1572 for each US Investigator (listing individual US clinical trial staff if directly involved in the treatment or evaluation of research making a direct and significant contribution to the data). Protocol NN (US sites): Intended for US sites Conducted under the Investigational New Drug Application (IND) All US Investigators will sign FDA Form 1572

108 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 106 of 112 Protocol NN (sites outside the US): Intended for participating sites outside the US Not conducted under the IND All Investigators outside the US will not sign FDA Form 1572 As documented in writing by protocol signature, all Investigators will fully comply with ICH GCP (13), applicable regulatory requirements, and in accordance with the Declaration of Helsinki (19). will analyse and report data from all sites together Audits and inspections Any aspect of the clinical trial may be subject to audits conducted by internal Quality Audit System or an inspection from domestic or foreign regulatory authorities. The Investigator and the site staff as well as clinical staff have an obligation to cooperate and assist in such audits and inspections. This includes giving Auditors and Inspectors direct access to all source documents and other documents relevant to the conduct of the clinical trial at the site Section 22 Responsibilities will be responsible for the preparation of the protocol, CRFs, supply of trial products and other trial supplies, monitoring, data management, statistics, and the Clinical Trial report as documented by procedures and internal specific agreements, as well as the current GCP guidelines. A central laboratory will be responsible for all analyses of blood, unless otherwise specified. Liraglutide concentration will be analysed at a special laboratory in Denmark. All staff, site, Central Laboratories, CRO etc. must conduct the trial in compliance with ICH GCP (13), applicable regulatory requirements, and in accordance with the Declaration of Helsinki (19). will be responsible for the preparation of the protocol, ecrf, supply of trial products and stated equipment, monitoring, data management, statistics, and the CTR as documented by procedures and internal specific agreements as well as the current GCP guidelines will provide a system for EDC. This system and support services to the system will be supplied by a clinical services vendor. The activities of the clinical services vendor will be under the direction and supervision of. Furthermore, will be responsible for the IV/WRS.

109 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 107 of 112 A central laboratory will be responsible for providing all lab supplies for the analysis of all blood samples taken during the trial. All results are received as paper copies at the sites as well as electronic transfer to clinical database. The name of the Central Laboratories will appear in the application to the authorities and in protocol Attachment I. The medical care given to, and medical decisions made on behalf of, subjects should always be the responsibility of a qualified physician. A qualified physician, who is an Investigator or a Sub-investigator for the trial, should be responsible for all trial-related medical decisions. In case the Investigator is not able to fulfil the role as Investigator (e.g. retirement), a new Investigator must be appointed in collaboration with. The Investigator will ensure that the last samples are shipped to the central laboratory within 24 hours after the last subject visit at the site Section 23 Reports and publications The signatory Investigator Dr Pi-Sunyer will review and sign the Clinical Trial Report to confirm, to the best of his knowledge, that it accurately describes the conduct and results of the trial. As the Clinical Trial Report contains information that will unblind the treatment allocation to the investgator, these parts (e.g. listings, narratives) will be removed from the CTR before Dr Pi- Sunyer's review of the Clinical Trial Report Communication and publication When the primary results are available, plans to discuss the interpretation of these with the principal Investigator, but reserves the right to release results that may impact Novo Nordisk financial expectations (e.g. a press release directly to the public or similar) without prior consultation with the remaining participating Investigators. reserves the right not to release data until specified milestones, e.g. a clinical trial report is available. This includes the right not to release interim results from clinical trials, because such results may lead to conclusions that are later shown to be incorrect information can invalidate the results of the entire trial. At the end of the trial, one or more manuscripts for publication will be prepared in collaboration collaboratively between Investigator(s) and. will not suppress or veto publications; however reserves the right to postpone publication and/or communication for a short time less than 60 days to protect intellectual property.

110 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 108 of Publication(s) Results from this trial will be reported by A/S on a publicly available database in compliance with the recommendations issued by The International Committee of Medical Journal Editors (ICMJE) 37 and the international pharmaceutical industry association. The results of this trial will be subject to public disclosure at external web sites according to international regulations, which is reflected in Code of Conduct for Clinical Trial Disclosure Investigator access to data and review of results As owners of the trial database, has discretion to determine who will have access to the database. Generally, trial databases are only made available to regulatory authorities. Individual Investigator(s) will have their own research participants' data and will be provided with the randomisation code after results are available. For Investigators participating in the 160 week trial, randomisation codes will not be provided until results for the 160 week trial are available Section 24 Retention of clinical trial documentation The Investigator must agree to archive the documentation (this includes both electronic and paper based records) pertaining to the trial in an archive after completion or discontinuation of the trial if not otherwise notified. The Investigator should not destroy any documents without prior permission from. If the investigator cannot archive the documents at the trial site after trial completion, can refer the Investigator to an independent archiving provider who has a system in place that allows only the Investigator to access the files. The Investigator must be able to get hold of his/her trial documents without involving in any way. will maintain the documentation pertaining to the trial as long as the product is on the market and for a minimum of 20 years after the trial has been completed, or in accordance with national regulations if they require a longer retention period. will maintain s documentation pertaining to the trial as long as the product is on the market plus 20 years. The files from the Investigator site/institution will be retained 15 years after the completion of the trial, or longer if required by national regulations.

111 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 109 of Section 25 Indemnity statements Poland: carries liability for the Study in the scope defined by the applicable laws and in particular by the Civil Code and the Pharmaceutical Law dated 6 September 2001 (uniform version Journal of Laws of 2008 no. 45 item 271 with amendments). In order to support potential claims for liability attributable to the Study, and Investigator are covered by the Insurance Policy issued according to applicable Polish law. Australia: accepts liability in accordance with: Medicines Australia Guidelines for Compensation for Injury Resulting from Participation in a Company-Sponsored Clinical Trial, dated 16 January Spain: Royal Decree 223/04 of 6th February 2004 establishing the requisites concerning clinical trial drugs.spain: accepts liability in accordance with: Spain: Royal Decree 223 of 6 th February 2004 establishing the requisites concerning clinical trial drugs. Germany: accepts liability in accordance with DE law: Drug law dated August 24, 1976, last amended per August 29, last amended per fifteenth law for amendment of the drug law dated July 17, The Netherlands: Wetgeving betreffende geneesmiddelen; geneesmiddelenwet 1 juli 2007 (Medicines Law, 1 July 2007). De Wet Medisch-wetenschappelijk Onderzoek met mensen (WMO), 1 maart 2006 (Medical Research Involving Human Subjects Act, 1 March 2006). Besluit van 23 juni 2003, houdende regels inzake de verplichte verzekering bij medischwetenschappelijk onderzoek met mensen (Decree of 23 June 2003, containing rules for compulsory insurance in medical research involving human subjects (Medical Research (Human Subjects) Compulsory Insurance Decree).

112 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 110 of References The following references have been added: (6) Buteau J, Roduit R, Susini S, Prentki M. Glucagon-like peptide-1 promotes DNA synthesis, activates phosphatidylinositol 3-kinase and increases transcription factor pancreatic and duodenal homeobox gene 1 (PDX-1) DNA binding activity in beta (INS-1)-cells. Diabetologia 1999;42(7): (7) Edvell A, Lindström P. Initiation of increased pancreatic islet growth in young normoglycemic mice (Umeå +/?). Endocrinology 1999;140(2): (8) Xu G, Stoffers DA, Habener JF, Bonner-Weir S. Exendin-4 stimulates both beta-cell replication and neogenesis, resulting in increased beta-cell mass and improved glucose tolerance in diabetic rats. Diabetes 1999;48(12): (9) Flint A, Raben A, Astrup A, Holst JJ. Glucagon-like peptide 1 promotes satiety and suppresses energy intake in humans. J Clin Invest 1998 Feb 1;101(3): (10) Gutzwiller JP, Goke B, Drewe J, Hildebrand P, Ketterer S, Handschin D, et al. Glucagon-like peptide-1: a potent regulator of food intake in humans. Gut 1999;44(1):81-6. (11) Naslund E, Barkeling B, King N, Gutniak M, Blundell JE, Holst JJ, et al. Energy intake and appetite are suppressed by glucagon-like peptide-1 (GLP-1) in obese men. Int J Obes Relat Metab Disord 1999 Mar;23(3): (12) Zander M, Madsbad S, Madsen JL, Holst JJ. Effect of 6-week course of glucagon-like peptide 1 on glycaemic control, insulin sensitivity, and beta-cell function in type 2 diabetes: a parallel-group study. Lancet 2002 Mar 9;359(9309):

113 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 111 of 112 (15) Ahmad SR, Swann J. Exenatide and rare adverse events. N Engl J Med 2008; 358(18): (16) Dore DD, Seeger JD, Chan KA. Use of a claims-based active drug safety surveillance system to assess the risk of acute pancreatitis with exenatide or sitagliptin compared to metformin or glyburide. Current eedical Research and Opinion 2009; 25(4): (21) ADA, Standard of Medical Care in Diabetes Diabetes Care, vol 33 supp 1, Jan 2010) (30) De AC, Drazen JM, Frizelle FA, Haug C, Hoey J, Horton R, et al. Clinical trial registration: a statement from the International Committee of Medical Journal Editors. New Engl J Med 2004;351(12): (31) Food and Drug Administration Amendments Act of Ref Type: Generic (32) International Conference on Harmonisation. ICH Harmonised Tripartite Guideline. Good Clinical Practice. 01 May Ref Type: Generic (33) Food and Drug Administration. Food and Drug Administration Amendments Act of fda gov/regulatoryinformation/legislation/federalfooddrugandcosmetic ActFDCAct/SignificantAmendmentstotheFDCAct/FoodandDrugAdmin istrationamendmentsactof2007/default htm 2007Available from: URL: ugandcosmeticactfdcact/significantamendmentstothefdcact/food anddrugadministrationamendmentsactof2007/default.htm (41) Meryl Brod MHNKSLDMB. Development and validation of the Treatment Related Impact Measure of Weight (TRIM-Weight). Health and Quality of Life Outcomes Ref Type: Generic

114 Substantial Protocol Amendment No. 6 EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 112 of 112 (49) Food and Drug Administration. Standardized Definitions for Cardiovascular Outcomes Trials: Draft Recommendations. Division of Metabolism and Endocrinology Products, Center for Drug Evaluation and Research (CDER), 22-July Rockville, MD; 2009 Jul Ref Type: Generic (52) Rubin DB. Multiple Imputation for Nonresponse in Surveys, New York: John Wiley & Sons, Inc Ref Type: Generic (53) International Conference on Harmonisation. ICH Harmonised Tripartite Guideline. Good Clinical Practice. 01 May Ref Type: Generic (54) World Medical Association. Declaration of Helsinki. Ethical Principles for Medical Research Involving Human Subjects. 59th WMA General Assembly, Seoul Oct 1.

115 Compound or Product Name Substantial Protocol Amendment No. 6 - App A Date: 26 November 2010 Version: 2.0 Page: 1 of 2 Appendix A Approval of Substantial Protocol Amendment no. 6 This confidential document is the property of. No unpublished information contained herein may be disclosed without prior written approval from. Access to this document must be restricted to relevant parties.

116 Compound or Product Name Substantial Protocol Amendment No. 6 - App A Date: 26 November 2010 Version: 2.0 Page: 2 of 2 Approval of Substantial Protocol Amendment novodocs object ID of final substantial protocol amendment: 0900c76e8123ef15 Substantial protocol amendment originator: Trine Nielsen, International Trial Manager Specialist Name & title (printed) Signature Date Head of originating department: Tina Hjorth, Head of ClinOps Lira/Obesity Name & title (printed) Signature Date International Project Statistician: Tu Duyen Le Thi, Statistician Name & title (printed) Signature Date Responsible Clinician: Christine Bjørn Jensen, International Medical Director Name & title (printed) Signature Date

117 A/S Electronic Signature Page 14 Dec :23: substantial-amendment-06-global-approval Documentum ObjectID : 0900c76e8123ef87 This is a representation of an electronic record that was signed electronically and this page is the manifestation of the electronic signature. Document signed by: Initials Full Name Capacity Reason Date and Time of Signature (Server Time) tnel Trine Nielsen Trial Manager Author :57:16 cbjj Christine Bjørn Jensen Medical Responsible Approval :58:09 tmoy Bettina Hjorth Manager Approval :14:22 rtp Rene Tabanera Palacios Biostatistician Approval :59:23

118 Compound or Product Name Substantial Protocol Amendment No. 6 - App B EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 1 of 2 Appendix B Agreement on Substantial Protocol Amendment no. 6 This confidential document is the property of. No unpublished information contained herein may be disclosed without prior written approval from. Access to this document must be restricted to relevant parties.

119 Compound or Product Name Substantial Protocol Amendment No. 6 - App B EudraCT No.: Date: 26 November 2010 Version: 2.0 Page: 2 of 2 Agreement on Substantial Protocol Amendment The Investigator and agree to conduct the trial as outlined in this substantial amendment with reference to national/local regulations, and in accordance with current Good Clinical Practice (GCP) guidelines. Investigator: Name (printed) Signature Date Head of medical/clinical research or designee: Name (printed) Signature Date

120 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 1 of 138 Revised Protocol Revised protocol including amendments: no. 1 dated 28-Aug-2008, no. 3 dated 28-Apr-2009, no. 6 (version 2) dated 26-Nov-2010 SCALE Obesity and Pre-diabetes Satiety and Clinical Adiposity- Liraglutide Evidence in Nondiabetic and Diabetic Subjects Effect of liraglutide on body weight in non-diabetic obese subjects or overweight subjects with comorbidities A randomised, double-blind, placebo controlled, parallel group, multi-centre, multinational trial with stratification of subject to either 56 or 160 weeks of treatment based on pre-diabetes status at randomisation Trial phase: 3a Author: Trine Nielsen, International Trial Manager Specialist Clinical Operations, GLP-1 & Obesity This confidential document is the property of. No unpublished information contained herein may be disclosed without prior written approval from. Access to this document must be restricted to relevant parties.

121 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 2 of 138 Table of contents Page Table of contents...2 Table of tables...6 Table of figures...7 List of abbreviations Summary Flow chart Introduction Basic information Glucagon-like peptide-1 (GLP-1) Liraglutide Rationale for the trial Objectives and endpoints Objective(s) Endpoint(s) Trial design Type of trial Rationale for trial design Treatment of subjects Liraglutide/liraglutide placebo Counselling on diet and physical activity Calculation of estimated TEE Rationale for treatment Trial population Number of subjects to be studied Inclusion criteria Stratification criteria Exclusion criteria Rescue criteria Withdrawal criteria Subject replacement Rationale for trial population Trial schedule Methods and assessments Visit procedures Visit schedule...40

122 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 3 of Screening visit Screening visit Visit 3, randomisation Visits 4 and Visit Visits 6, 6c, 9, 10, 12, 13, 15, and Visits 8 and Visit Visit 17, End of 56 week randomised Treatment/premature discontinuation Visit 17x, Body weight/mesi recording visit Visit 18a, Re-randomised treatment visit Visit 19a and 20a, Re-randomised treatment visits Visit 21a, End of re-randomised Treatment visit / premature discontinuation Visit 22a, follow-up Visit 18b Visits 19b, 22b, 25b, 28b, 31b, 34b, 37b and 38b (delayed-onset-of-type-2- diabetes part of the trial) Visits 20b, 26b and 32b (delayed-onset-of type-2-diabetes part of the trial) Visits 21b, 24b, 27b, 30b, 33b, 36b, 39b, 40b and 42b (delayed-onset-of-type-2- diabetes part of the trial) Visits 23b, 29b and 35b (delayed-onset-of-type-2-diabetes part of the trial) Visit 41b (delayed-onset-of-type-2-diabetes part of the trial) Visit 43b, End of 160 weeks randomised Treatment visit /premature discontinuation Visit 44b, Follow-up visit Assessments for efficacy Body measurements Weight and height Waist circumference HbA 1c Fasting plasma glucose Fasting insulin C-peptide Oral Glucose Tolerance Test (OGTT) Pre-diabetes status and classification Vital signs Blood pressure Pulse Cardiovascular biomarkers Lipids Patient reported outcomes (PRO) questionnaires Impact of Weight on Quality Of Life questionnaire (IWQoL) Lite version SF Treatment Related Impact Measure (TRIm) - Weight Positive predictive value of Finnish Type 2 diabetes (FINDRISC) Urinary Albumin-to-Creatinine ratio...69

123 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 4 of Assessments for safety Physical examination Hypoglycaemic episodes Self-monitored plasma glucose ECG Pregnancy test Adverse events (AEs) Haematology and biochemistry Liraglutide antibodies Suspicion of acute hypersensitivity (allergic reaction) to trial product Suspicion of immune-complex disease Mental health questionnaires C-SSRS PHQ Thyroidectomy pathology slides Thyroid tissue sample collection in case of thyroidectomy Genetic testing in case of a confirmed C-cell pathology Binge eating scale questionnaire (BES) Other assessments Liraglutide concentration (population PK) Liraglutide concentration (PK C max sub-study) Smoking habits Diaries Dosing diary day food diary Diabetes diary Dietary compliance and physical activity Signs or symptoms of diabetic complications History of concomitant cardiovascular disease History of gallbladder disease History of psychiatric disorders Subject compliance Trial supplies Trial product(s) Packaging and labelling of trial product(s) Storage and drug accountability of trial product(s) Auxiliary supply Randomisation, breaking of blinded codes and interactive voice/web response system (IV/WRS) Randomisation Breaking of blinded codes Interactive voice/web response system (IV/WRS) Concomitant illnesses/medical history and concomitant medication Adverse events and pregnancies Definitions...86

124 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 5 of Technical complaints Collection, recording and reporting of adverse events Medical events of special interest External independent event adjudication committee Follow-up of adverse events Collection and reporting of technical complaints Collection, storage and shipment of technical complaint samples Pregnancy Precautions/over-dosage Safety committee Internal safety committee Calcitonin Monitoring Committee Case report forms Rules for completing ecrfs Corrections to ecrfs ecrf flow Monitoring procedures Data management Computerised systems Evaluability of subjects for analysis Statistical considerations Sample size calculation Statistical methods Primary efficacy endpoints The four primary efficacy endpoints consists of three co-primary endpoints describing weight change and one endpoint describing onset of diabetes: The proportion of subjects losing at least 5% of baseline fasting weight (measured at Week 0) The proportion of subjects losing more than 10% of baseline fasting weight (measured at Week 0) Onset of type 2 diabetes in subjects with pre-diabetes at baseline evaluated at week Primary analysis of the co-primary endpoints Sensitivity analyses of the primary endpoints Analysis of secondary efficacy endpoints Analysis of the safety endpoints Interim analysis Sequential safety analysis/safety monitoring Explorative statistical analysis for pharmacogenetics and biomarkers PK and/or PD modelling Health economics and/or subject reported outcome Ethics Informed consent form for trial subjects Data handling...119

125 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 6 of Institutional review boards/independent ethics committee Regulatory authorities Premature termination of the trial/trial site Protocol compliance Audits and inspections Critical documents Responsibilities Reports and publications Communication and publication Authorship Publication(s) Site-specific publication(s) by Investigator(s) Investigator access to data and review of results Retention of clinical trial documentation Indemnity statement References Appendix A Appendix B Appendix C Appendix D Appendix E Appendix F Appendix G Appendix H Appendix I Appendix J Attachment I - global Attachment II - country Approval of final protocol Agreement on the final protocol NYHA criteria for functional capacity Instruction for blood pressure measurement Flow chart: Main trial including a 12 week re-randomised treatment period Flow chart: delayed-onset-of-type-2-diabetes trial (subjects with prediabetes) Patient Reported Outcome Questionnaires Mental Health Questionnaires Calcitonin Monitoring Committee Screening of Calcitonin (CT) Levels Medical Events of Special Interest (MESI) List of key staff and relevant departments List of key staff and relevant departments Table of tables Page Table 5 1 Table 5 2 Table 5 3 Treatment groups: Main trial (56 weeks) incl. 12 week re-randomised treatment period and two weeks FU period*...24 Treatment groups: Delayed-onset-of-type-2-diabetes part of the trial*...25 Equations for estimating BMR in kcal/day*...28

126 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 7 of 138 Table 8 1 Visit schedule...40 Table 18 1 Sample size calculation for the primary endpoint onset of diabetes assessed at week Table 18 2 Overview of statistical analysis of secondary efficacy endpoints at week Table 18 3 : Overview of statistical analysis of secondary efficacy endpoints at week Table 18 4 Overview of statistical analysis of secondary efficacy endpoints at week Table of figures Page Figure 2 1 Trial design diagram...14 Figure 8 1 Clasification of hypoglycaemia...71

127 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 8 of 138 List of abbreviations AE adverse event ALAT alanine aminotransferase ALP alkaline phosphatase ASAT aspartate aminotransferase ATP-III Adult Treatment Panel III BES Binge Eating Scale BMI body mass index BMR basal metabolic rate CABG coronary artery bypass graft surgery C-cell parafollicular cell CPK creatinine phosphokinase (creatinine kinase) ecrf electronic case report form C-SSRS Columbia Suicidality Severity Rating Scale CT Computed Tomography CTA Clinical Trial Application CTR Clinical Trial Report DBP Diastolic blood pressure DPP-4 dipeptidyl peptidase-4 DUN Dispensing Unit Number EAP Events Adjudication panel ECG electrocardiogram EDC electronic data capture EOT End of Trial FAS Full Analysis Set FDA Food and Drug Administration FFA free fatty acids FINDRISC Finnish Type 2 Diabetes Risk Score FMTC familial medullary thyroid carcinoma FPFV first patient first visit FPG fasting plasma glucose (fasting glucose) GCP good clinical practice GLP-1 glucagon-like peptide-1 HbA 1c glycosylated haemoglobin hcg human chorionic gonadotrophin HDL-cholesterol high density lipoprotein cholesterol HOMA homeostasis model assessment hscrp high sensitivity C reactive protein IFG Impaired fasting Glucose

128 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 9 of 138 IGT Impaired Glucose Tolerance IB Investigator Brochure ICH International Conference on Harmonisation ICMJE International Committee of Medical Journal Editors IEC independent ethics committee IRB Institutional Review Board ITT intention- to- treat IUD Intra-Uterine Device IV/WRS interactive voice/web response system (IVRS/IWRS) IWQoL-Lite impact of weight on quality of life questionnaire lite version LDL-cholesterol low density lipoprotein cholesterol LLOQ lower level of quantification LOCF last observation carried forward LPLV last patient last visit MedDRA Medical Dictionary for Regulatory Activities MEN2 multiple endocrine neoplasia type 2 MESI Medical Event of Special Interest MHP Mental Health Professional MRI Magnetic Resonance Imaging NCEP National Cholesterol Education Program NCR No Carbon Required NGT Normal glucose tolerance NYHA New York Heart Association OGTT Oral glucose tolerance test PAI-1 plasminogen activator inhibitor-1 PCI percutaneous coronary intervention PHQ-9 Patient Health Questionnaire-9 PP per protocol PRO Patient reported outcome PK Pharmacokinetics s.c. subcutaneous SAE serious adverse event SBP Systolic blood pressure SD standard deviation SIF safety information form TC total cholesterol (cholesterol) TEAE treatment emergent adverse event TEE total energy expenditure TG triglycerides the time to reach maximum concentration of drug in plasma t max

129 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 10 of 138 TMM TRIm-Weight TSH TVP UNR VLDL-cholesterol Trial Materials Manual Treatment Related Impact measure - Weight thyroid-stimulating hormone Trial Validation Plan Upper normal range very low density lipoprotein cholesterol

130 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 11 of Summary Primary objectives: To establish the efficacy of 3.0 mg liraglutide compared to liraglutide placebo in inducing and maintaining weight loss over 56 weeks in obese subjects and overweight subjects with comorbidities To investigate the long term efficacy of 3.0 mg liraglutide versus liraglutide placebo in delaying the onset of type 2 diabetes in obese subjects and overweight subjects with co-morbidities diagnosed with pre-diabetes Secondary objective: To investigate the long term efficacy of 3.0 mg liraglutide versus liraglutide placebo on cardiovascular risk markers such as blood pressure, lipids, glucose parameters as well as effects on quality of life Safety objective: To evaluate the safety and tolerability of liraglutide Trial design: This is a randomised, double-blind, placebo controlled, parallel group, multi-centre, multinational trial in non-diabetic obese subjects and overweight subjects with co-morbidities. Subjects will be randomised in a 2:1 manner to receive either 3.0 mg liraglutide or liraglutide placebo, and based on body mass index (BMI [kg/m 2 ]) and pre-diabetes status at randomisation subjects will be stratified to either 56 or 160 weeks of treatment (160 week treatment will only applicable to subjects with pre-diabetes at screening). Treatment will be stratified by body mass index (BMI [kg/m 2 ]) (BMI 30 kg/m 2, or BMI < 30 kg/m 2 ) and pre-diabetes status at baseline (Yes/No, see below for definition of pre-diabetes). Treatment allocation for subjects with pre-diabetes will be double-blind for the first 56 weeks (, Investigator, and subject blinded), after which time point sponsor is un-blinded (Investigator and subject remain blinded to treatment allocation). A subject will be classified as having pre-diabetes if falling within the following categories of increased risk for diabetes: Fasting plasma glucose 100 mg/dl ( 5.6 mmol/l) to 125 mg/dl (6.9 mmol/l) (IFG), 2-hour post-challenge (OGTT) glucose 140 mg/dl (7.8 mmol/l) to 199 mg/dl (11.0 mmol/l) (IGT), or HbA 1c % both inclusive. The maximum overall duration of the main part of the trial from screening to follow-up will be 72 weeks including the 12 week re-randomised treatment period. For subjects enrolled in the delayedonset-of-type-2-diabetes part of the trial the maximum total duration of the trial will be 164 weeks.

131 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 12 of 138 Trial population: 4800 subjects are expected to be screened in order to meet the randomisation target of minimum 3600 subjects in total, including a minimum of 1000 randomised subjects with pre-diabetes. If the number of subjects with pre-diabetes equals at least 1000 after randomisation of 3600 subjects, no further randomisation will be done. The screening and randomisation of subjects with pre-diabetes will be monitored closely throughout the recruitment period. If the number of subjects with prediabetes is lower than expected (approximately 30% of screened population), screening will continue with the purpose of identifying and enrolling the minimum number of subjects with prediabetes. If 3600 subjects have been randomised without reaching the minimum number of subjects with pre-diabetes (N = 1000), only subjects with pre-diabetes will be randomised. At week 56 all subjects who did not have pre-diabetes at baseline will continue in a blinded 12 week re-randomised treatment period followed by a 2 week off-treatment period for assessment of liraglutide antibody development. Subjects who have received treatment with liraglutide during the previous 56 weeks will be re-randomised 1:1 to either continue on liraglutide treatment or be switched to liraglutide placebo (for a blinded off active treatment evaluation). Subjects who had been randomised to liraglutide placebo will continue on liraglutide placebo during the 12 week rerandomised treatment period. Key inclusion criteria: Informed consent obtained Body Mass Index (BMI) 30.0 kg/m 2 or 27 kg/m 2 in the presence of co-morbidities (treated or untreated hypertension or dyslipidaemia) Stable body weight Preceding failed dietary effort Key exclusion criteria: Known type 1 or type 2 diabetes HbA 1c 6.5 % (Screening visit 1) or FPG 126 mg/dl (7 mmol/l) (Screening visit 2) or 2 hr post-challenge (OGTT) plasma glucose 200 mg/dl (11.1 mmol/l) (Screening visit 2) Screening calcitonin 50 ng/l Family or personal history of multiple endocrine neoplasia type 2 (MEN2) or familial medullary thyroid carcinoma (FMTC) Personal history of non-familial medullary thyroid carcinoma History of acute or chronic pancreatitis Obesity induced by drug treatment Use of approved weight lowering pharmacotherapy Previous surgical treatment of obesity History of major depressive disorder or suicide attempt

132 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 13 of 138 Uncontrolled hypertension (systolic blood pressure 160 mmhg and/or diastolic blood pressure 100 mmhg) Assessments: Endpoints to support the primary efficacy objectives include change from baseline in body weight at 56 weeks as well as the proportion of subjects losing at least 5% or more than 10% of baseline body weight at 56 weeks. Moreover, the proportion of subjects who have converted to overt diabetes at 160 weeks is included as a co-primary endpoint for subjects who had pre-diabetes at baseline. The secondary efficacy endpoints include change from baseline in body weight at 160 weeks of treatment in subjects with pre-diabetes, the proportion of subjects losing at least 5% or ore than 10% of baseline bodyweight at week 160 and change from baseline in waist circumference, glucose control parameters (HbA 1c, FPG, fasting insulin, C-peptide, OGTT-derived parameters, pre-diabetes status), blood pressure, cardiovascular biomarkers, lipids, urinary albumin-to-creatinine ratio (UACR) and Patient Reported Outcome (PRO) at 56 weeks. Key assessments of safety include physical examination, ECG, pulse, adverse events, haematology, biochemistry (including amylase, lipase and calcitonin), liraglutide antibodies and ratings based on mental health questionnaires. Trial product(s): A/S will supply the following trial products: Liraglutide 6.0 mg/ml, 3 ml FlexPen for subcutaneous injection Liraglutide placebo 3 ml FlexPen for subcutaneous injection

133 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 14 of Flow chart Please refer to Appendices E and F. Figure 2 1 Trial design diagram 3 Introduction In this document Investigator refers to the individual overall responsible for the conduct of the clinical trial at a trial site. 3.1 Basic information The prevalence of both obesity and diabetes continues to increase worldwide. More than one billion adults worldwide are overweight and at least 300 million are obese(1). The prevalence of diabetes is also rising, with worldwide prevalence estimated at 4.0 % in 2007(2), and still rising. Eighty to ninety percent of people with type 2 diabetes are overweight(3) and obesity worsens the metabolic and physiologic abnormalities associated with diabetes, particularly hyperglycemia, hyperlipidemia, and hypertension(4). Weight loss is a cornerstone of diabetes care for overweight persons, as it improves insulin sensitivity and glycemic control(5) and in addition improves blood pressure and lipid profiles by decreasing triglycerides and low-density lipoprotein levels(5) Glucagon-like peptide-1 (GLP-1) GLP-1 is an incretin hormone secreted from the L-cells in the lower gut in response to meal ingestion, which stimulates endogenous insulin secretion in a glucose-dependent manner. GLP-1

134 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 15 of 138 also decreases blood glucagon levels and delays gastric emptying by decreasing gastric motility and increasing satiety and subsequently reducing food intake, and has been shown to promote β-cell growth and proliferation in animal models (6-8). GLP-1 reduces appetite in lean and normal weight individuals, as well as in obese individuals (9-11), and has been shown to reduce body weight in people with type 2 diabetes (12). The underlying mechanism mediating the weight-reducing effects of GLP-1 is most likely a combination of effects on the gastrointestinal tract and the central nervous system, i.e. decreased gastric motility and reduced appetite/increased satiety with a subsequent reduction in food intake. The combination of these mechanisms makes GLP-1 receptor stimulation an attractive mechanism to investigate for weight management and diabetes prevention.(13), (14) Liraglutide Liraglutide is a long-acting GLP-1 analogue developed by A/S and recently approved for the treatment of type 2 diabetes in the US, EU, Japan, and other countries worldwide under the brand name Victoza (1.2 mg or 1.8 mg once-daily). Development of liraglutide 3.0 mg once-daily for weight management is currently ongoing. Compared to human GLP-1, liraglutide has a C16 fatty (palmitic) acid chain attached at position 26 (lysine) of the peptide, and has lysine at position 34 replaced by arginine. When administered subcutaneously, these structural modifications result in a compound with protracted kinetic properties suitable for once daily injection. In vitro receptor studies have shown that liraglutide is a selective, potent and full agonist of the cloned human GLP-1 receptor. The effects of liraglutide include delayed gastric emptying, reduced sensation of hunger and increased satiety leading to decreased food intake and subsequent weight loss. A total of 50 clinical trials with liraglutide have been completed (includes doses up to 3.0 mg). The trials were conducted world-wide, with most being conducted in Europe. Out of more than subjects, more than 7000 subjects were exposed to liraglutide (including 850 subjects treated for 104 weeks in completed trials). A total of 986 obese subjects without type 2 diabetes (< 9% of all subjects) have been included to date in the obesity clinical development programme for liraglutide in the completed phase 2 trial NN and the completed phase 3a trial NN (of which 305 subjects were randomised to liraglutide 3.0 mg). A further 48 obese subjects were randomised in the ongoing phase 1 trial NN Data from finalised trials have shown liraglutide to have a pharmacokinetic profile suitable for one time daily administration, as evidenced by a relatively slow absorption ([tmax] =8-12 hours) with a terminal elimination half-life (t½) of approximately 13 hours. The pharmacokinetic profile is comparable between healthy subjects and subjects with type 2 diabetes.

135 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 16 of 138 Results from a phase 2 trial in obese subjects without type 2 diabetes (NN ) showed a dose dependent weight loss ranging from 3.8 to 7.8 kg with liraglutide doses of mg administered for 52 weeks (20 weeks as double-blind and 32 weeks of open-label treatment (sponsor unblinded at 20 weeks). In addition to weight lowering, there was a decrease in systolic blood pressure and an impact on the number of subjects with pre-diabetes. In addition, of the approximately 30% of subjects who had pre-diabetes at baseline, 85% did not have pre-diabetes after 20 weeks compared to 45% of placebo subjects. Of the 70% of subjects without pre-diabetes at baseline, 20% of placebo subjects developed pre-diabetes, whereas only 2-4% of liraglutide-treated subjects had pre-diabetes after 20 weeks. The safety evaluation was favourable with the main tolerability finding being gastrointestinal side effects (please refer to liraglutide obesity Investigators Brochure (IB), 3 rd Edition, 2010) The first of three confirmatory phase 3 trials within the liraglutide obesity development programme (Trial NN , or SCALE -Maintenance) was recently completed. Reporting is ongoing. The trial was a 56-week randomised, double-blind, placebo-controlled trial investigating treatment of liraglutide 3.0 mg versus placebo as an adjunct to diet and exercise in obese subjects or overweight subjects with co-morbidities who had already lost at least 5% of their body weight during a 4 to 12-week run-in period on a low calorie diet. The mean weight loss for subjects in the run-in period was approximately 6 kg. From a body weight of approximately 100 kg at randomisation, treatment with liraglutide for 56 weeks provided an additional estimated mean weight loss of 6.11% (-5.7 kg), compared to weight-neutrality or maintenance in the placebo group (+0.16 kg vs. baseline). Eighty one (81) percent of liraglutide-treated subjects maintained their runin weight loss compared to 48% in the placebo group. Moreover, 51% of liraglutide-treated subjects lost additional 5% or more of their baseline body weight, compared to 22% in the placebo group. Treatment with liraglutide maintained and in some instances further improved beneficial effects on markers of glycaemic control and cardiovascular risk. Treatment with liraglutide was generally well-tolerated, with high completion rates in groups (75% in liraglutide group, 70% in placebo group). The number of withdrawals due to adverse events was evenly distributed between groups (8.5% in the liraglutide group vs. 8.6% in placebo group). Serious adverse events were relatively uncommon, but were more frequent in liraglutide-treated subjects (4.2%) compared to placebo (2.4%). There were no events of pancreatitis or medullary thyroid cancer, and no treatment-related increases in blood calcitonin levels. Consistent with previous trials with liraglutide, the most commonly reported adverse events were from the gastrointestinal system, with nausea reported by 47% of subjects in the liraglutide group compared to 17% in the placebo groups, and vomiting by 17% vs. 2%, respectively. As in previous trials, the majority of events were reported in the first 6-8 weeks, were mild or moderate in severity and transient in nature. Please refer to liraglutide obesity Investigators Brochure, 3rd Edition, 2010, or any updates hereof.

136 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 17 of 138 Patients treated with liraglutide should be advised of the potential risk of dehydration in relation to gastrointestinal side effects and take precautions to avoid fluid depletion. Few cases of acute pancreatitis (inflammation of the pancreas) presenting with persistent severe abdominal pain (usually accompanied by vomiting) have been reported with liraglutide and exenatide. Post-marketing surveillance identified at least 30 cases of pancreatitis with exenatide.(15).however, a health services registry-based study found no increased frequency of pancreatitis among exenatide users.(16) If the investigator suspects acute pancreatitis, all suspected drugs should be discontinued until confirmatory tests have been conducted, and appropriate treatment should be initiated. Subjects diagnosed with acute pancreatitis (as a minimum 2 of 3: characteristic abdominal pain, amylase and/or lipase >3x UNR or characteristic findings on CT/MRI) should be withdrawn from the study. Liraglutide causes dose-dependent and treatment-duration-dependent thyroid C-cell tumors at clinically relevant exposures in both genders of rats and mice. It is unknown whether liraglutide causes thyroid C-cell tumors, including medullary thyroid carcinoma (MTC), in humans, as human relevance could not be ruled out by clinical or nonclinical studies. Liraglutide is contraindicated in patients with a personal or family history of MTC and in patients with Multiple Endocrine Neoplasia syndrome type 2 (MEN 2). Based on the findings in rodents, monitoring with serum calcitonin or thyroid ultrasound was performed during clinical trials, but this may have increased the number of unnecessary thyroid surgeries. It is unknown whether monitoring with serum calcitonin or thyroid ultrasound will mitigate human risk of thyroid C-cell tumors. In a 2-year repeat subcutaneous dose carcinogenicity study of liraglutide injected once a day in CD- 1 mice, a treatment-related increase in fibrosarcomas was seen on the dorsal skin and subcutis, the body surface used for drug injection, in males in the 3 mg/kg/day group. These fibrosarcomas were attributed to the high local concentration of drug near the injection site. The liraglutide concentration in the clinical formulation (6 mg/ml) is 10 times higher than the concentration in the formulation used to administer 3 mg/kg/day liraglutide to mice in the carcinogenicity study (0.6 mg/ml). Further information can be obtained in the Liraglutide Investigator's Brochure Obesity, Edition 3, 2010 or any updates hereof. 3.2 Rationale for the trial The trial has been designed to further investigate the potential of liraglutide to safely induce long term weight loss in non-diabetic obese subjects, as well as in overweight subjects with comorbidities (treated or untreated hypertension or dyslipidaemia according to ATP-III).

137 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 18 of 138 Treatment duration of 56 weeks is considered to be sufficient to demonstrate weight loss and subsequent weight maintenance. In order to assess the effects of drug cessation on appetite and weight control, possible withdrawal effects and rebound will be ascertained in the 12 week rerandomised treatment period. 12 week re-randomised treatment period: To allow for a blinded assessment of potential rebound and withdrawal effects, all subjects who did not have pre-diabetes at baseline will continue into a 12-week blinded re-randomised treatment period. Subjects who have received active treatment (liraglutide) throughout the 56 week treatment period will be re-randomised 1:1 to either continue active treatment ( on-drug ) or switch to liraglutide placebo ( off-drug ). Subjects who have received liraglutide placebo treatment throughout the 56 week treatment period will continue on liraglutide placebo in the re-randomised treatment period. Subjects will continue to receive guidance on dietary measures and physical activity throughout this period. After the 12 week blinded re-randomised treatment period, randomised treatment will be discontinued and all subjects will be followed for an additional 2 weeks off-treatment for assessment of liraglutide antibody formation. A major health risk associated with being obese is development of type 2 diabetes. This risk is particularly pronounced in subjects with pre-diabetes. In order to investigate whether liraglutide not only safely and efficaciously lowers body weight but also delays onset of type 2 diabetes an additional 104 week treatment period for subjects at high risk of developing diabetes (i.e. subjects with pre-diabetes at baseline) has been included. A treatment period of 160 weeks is predicted to be sufficient to demonstrate a 60% relative risk reduction in the development of type 2 diabetes(17), (18), (19).

138 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 19 of Objectives and endpoints 4.1 Objective(s) Hypotheses: Liraglutide will induce and maintain a significantly greater weight loss compared with liraglutide placebo in non-diabetic obese subjects or overweight subjects with co-morbidities Liraglutide will delay the onset of diabetes in pre-diabetic obese subjects or pre-diabetic overweight subjects with co-morbidities Primary objectives: To establish the efficacy of 3.0 mg liraglutide compared to liraglutide placebo in inducing and maintaining weight loss over 56 weeks To investigate the long term efficacy of 3.0 mg liraglutide in delaying onset of type 2 diabetes in obese subjects and overweight subjects with co-morbidities diagnosed with pre-diabetes Secondary objective: To investigate the long term efficacy of 3.0 mg liraglutide versus liraglutide placebo on cardiovascular risk markers such as blood pressure, lipids, glucose parameters, urinary-albuminto-creatinine-ration (UACR), as well as effects on quality of life, Patient Reported Outcomes (PRO) and Binge Eating Scale (BES) ratings Safety objective: To evaluate the safety and tolerability of 3.0 mg liraglutide 4.2 Endpoint(s) Primary efficacy endpoints: Change from baseline in body weight (fasting) at 56 weeks Proportion of subjects losing at least 5% of baseline body weight at 56 weeks Proportion of subjects losing more than 10% of baseline body weight at 56 weeks Proportion of subjects with onset of type 2 diabetes at 160 weeks (among subjects with prediabetes at baseline) Secondary efficacy endpoints: Change from baseline to Week 56 in: Waist circumference

139 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 20 of 138 Glucose related parameters (HbA 1c, FPG, fasting insulin, C-peptide, endpoints from 2 hr Oral Glucose Tolerance test (OGTT) parameters, proportion of subjects with type 2 diabetes (yes/no), pre-diabetes status, and homeostasis model assessment parameters [HOMA])(20) Urinary-albumin-to-creatinine ratio (UACR) Systolic and diastolic blood pressure (vital signs) Cardiovascular biomarkers (high sensitivity C reactive protein [hscrp], adiponectin, fibrinogen, PAI-1) Lipids (total cholesterol (TC), low density lipoprotein cholesterol (LDL-cholesterol), high density lipoprotein cholesterol ([HDL-cholesterol], very low density lipoprotein cholesterol [VLDL-cholesterol], triglycerides [TG], free fatty acids [FFA]) PRO assessed by IWQoL-Lite, SF-36 and Treatment Related Impact Measure - Weight (TRIm- Weight) Proportion of subjects with change in concomitant medication from baseline to Week 56 in: anti-hypertensive drugs lipid lowering drugs oral antidiabetic drugs Mean change in body weight from baseline to Week 160 (subjects with pre-diabetes at baseline) Proportion of subjects losing at least 5% and more than 10% of baseline body weight at 160 weeks (subjects with pre-diabetes at baseline) Change from baseline to Week 160 in other secondary endpoints (subjects with pre-diabetes at baseline) 12 week re-randomised treatment period efficacy endpoints: Change from Week 56 to Week 68 in: Albumin/ creatinine ratio Fasting body weight Waist circumference FPG Urinary Albumin-to-Creatinine Ration (UACR) Systolic and diastolic blood pressure (vital signs) PRO Change from baseline to Week 68 in: Fasting body weight Waist circumference FPG Urinary Albumin-to-Creatinine Ration (UACR) Systolic and diastolic blood pressure (vital signs) PRO

140 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 21 of 138 Safety endpoints: Physical examination (cardiovascular system, respiratory system, abdomen, central and peripheral nervous system, musculo-skeletal system and the thyroid gland) Pulse (vital sign) ECG Adverse events Haematology and biochemistry including amylase, lipase and calcitonin Formation of liraglutide antibodies Mental health assessed by the Columbia Suicidality Severity Rating Scale (C-SSRS) and Patient Health Questionnaire (PHQ)-9 12 week re-randomised treatment period safety endpoints: Adverse events Pulse (vital sign) ECG Haematology and biochemistry including amylase, lipase, calcitonin Change from baseline to Week 56 and 58 in binge eating (assessed by BES) Mental health assessed by the Columbia Suicidality Severity Rating Scale (C-SSRS) and Patient Health Questionnaire (PHQ-9) Other endpoints: Plasma concentrations of liraglutide (PK sampling) 5 Trial design 5.1 Type of trial This is a randomised, double-blind, placebo controlled, parallel group, multi-centre, multinational trial in non-diabetic obese subjects or overweight subjects with co-morbidities. Subjects will be stratified to either 56 weeks (plus 12 weeks re-randomised treatment period and 2 weeks follow-up period) or 160 weeks plus 2 weeks follow-up period) of treatment based on pre-diabetes status at randomisation. Treatment allocation for subjects with pre-diabetes will be double-blind for the first year (, Investigator and subject are blinded), and single blind for the remaining 2 years (i.e. is unblinded, whereas Investigator and subjects remain blinded).within these groups subjects are furthermore stratified according to BMI (BMI 30 kg/m 2 or BMI < 30 kg/m 2 ). Pre-diabetes is defined as: IFG (FPG 100 mg/dl / 5.6 mmol/l, and 125 mg/dl / 6.9 mmol/l) and/or IGT (2 hr post-challenge OGTT) plasma glucose 140 mg/dl / 7.8 mmol/l, and 199

141 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 22 of 138 mg/dl / 11.0 mmol/l) or HbA 1c % both inclusive (21).Within these groups, subjects are furthermore stratified according to BMI (BMI 30, or BMI < 30). Subjects will be randomised in a 2:1 manner to receive either 3.0 mg liraglutide or liraglutide placebo subjects are expected to be screened in order to meet the randomisation target of a minimum of 3600 subjects in total, including a minimum of 1000 subjects with pre-diabetes. These numbers are based on an anticipated screen failure rate of 25% and an expected incidence of prediabetes in the trial population of 30% (please refer to section 18). The duration of the main part of the trial from screening to end follow-up will be a maximum of 72 weeks including a 12 week re-randomised treatment period and a 2 weeks follow-up period, while for subjects enrolled in the delayed-onset-of-type-2-diabetes part of the trial the total duration of the trial will be 164 weeks. Subjects will be randomised (Visit 3) to either liraglutide or liraglutide placebo using an Interactive Voice/ Web Response System (IV/WRS). Subjects will follow a fixed dose escalation. The dose will be gradually escalated to 3.0 mg starting with 0.6 mg and with a dose level increment of 0.6 mg every 7 days. After successful escalation, subjects will attend the clinic at four-week intervals (Visit 6-17). At week 56 (Visit 17) all subjects who did not have pre-diabetes at baseline and had been randomised to liraglutide during the first 56 weeks will be re-randomised. They will attend Visit 18a two weeks after re-randomisation, attend Visit 19a two weeks after Visit 18a and then afterwards attend the clinic at 4 week intervals. In addition they will attend a final follow-up visit (Visit 22a) at week 70. The primary purpose of this visit is to collect blood samples for analysis of liraglutide antibodies. After completion of dose escalation, subjects with pre-diabetes at baseline will attend the clinic at four week intervals (Visits 6 to 17 + Visits 18b to 43b). A safety follow-up visit (Visit 44b) is included two weeks after Visit 43b (End of 160 week randomised Treatment ). The primary purpose of this visit is to collect blood samples for liraglutide antibody analysis. Subjects who have discontinued the trial prematurely will be asked to attend a visit (17x) 56 weeks ± 5 days after their randomisation date. The purpose of this visit will be recording of the body weight and assessment of MESIs. In addition the subjects will be asked to complete mental health questionnaires. The assessment will be done by asking the subject if he/she has experienced any MESI (see section and appendix J) since the last contact. The subjects should attend Visit 17x fasting. If the subject is not willing to attend Visit 17x, it should be documented in the patient medical record that the subject has refused to attend the visit.

142 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 23 of Rationale for trial design Obesity is a serious medical condition which is increasing in incidence worldwide. Obesity induces metabolic abnormalities which contribute to the development of diabetes mellitus and cardiovascular disease. In addition obesity is associated with increased risk of morbidity and mortality. The treatment modalities currently available for the treatment of obesity (i.e. lifestyle interventions, pharmacotherapy and surgery) have limited long-lasting success in producing major and sustained weight loss and are often associated with undesirable side effects. Therefore a need exists for new and effective therapeutic strategies to prevent and reduce obesity. Within the last decade, the ability of GLP-1 to decrease appetite and energy intake has been demonstrated in both lean and obese subjects. Furthermore, studies have shown that obese subjects have attenuated GLP-1 release in response to meals. Results from a phase 2 trial in obese subjects without type 2 diabetes showed a dose-dependent weight loss ranging from 3.8 to 7.8 kg with liraglutide doses of mg administered for 52 weeks (20 weeks as double blind and 32 weeks of open-label treatment (i.e. unblinded at week 20). In addition to weight lowering, there was a decrease in systolic blood pressure and a reduction of the number of subjects with pre-diabetes. The safety evaluation was favourable with the main tolerability finding being gastro-intestinal side effects. The present trial is a confirmatory trial and a double-blind, randomised, parallel group, placebo controlled trial design will be employed. Treatment allocation for subjects with pre-diabetes will be double-blind for the first 56 weeks (i.e., Investigator and subjects blinded), and single blind (i.e. unblinded, whereas Investigator and subjects remain blinded) for the remaining 104 weeks. The trial has been designed to further investigate the potential of liraglutide to safely induce long term weight loss in non-diabetic obese subjects, as well as in overweight subjects with co-morbidities. A treatment duration of 56 weeks should be sufficient to demonstrate weight loss and subsequent weight maintenance. In order to assess the effects of drug cessation on appetite and weight control, possible withdrawal effects and rebound will be ascertained in a blinded 12 week re-randomised treatment period (12 weeks on liraglutide or liraglutide placebo treatment plus two weeks off treatment). The re-randomised treatment period is only applicable to subjects who did not have pre-diabetes at baseline as subjects with pre-diabetes at baseline will continue in the104 week extension. Subjects who have been on active treatment during the first 56 weeks of the trial will be re-randomised in a 1:1 manner to either continue on active treatment or switch to liraglutide placebo during the re-randomised treatment period. Subjects who have received placebo treatment during the first 56 weeks of the trial will continue on liraglutide placebo treatment during the re-randomised treatment period. During this period treatment allocation will remain blinded to, Investigator and subjects. After the blinded 12 week re-randomised treatment period all subjects will discontinue randomised treatment and will be followed for an additional 2 weeks in an off-treatment period to assess potential antibodies to

143 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 24 of 138 liraglutide. A major health risk associated with being obese is development of type 2 diabetes. This risk is particularly pronounced in subjects with pre-diabetes (IFG and/or IGT and HbA 1c % both inclusive (22). In order to demonstrate that liraglutide not only safely and efficaciously lowers body weight, but also delays the onset of type 2 diabetes, an additional blinded (subjects and Investigator are blinded to treatment allocation) 104 week extension period for subjects at high risk of developing diabetes has been included. For this population a total treatment period of 160 weeks is predicted to be sufficient to demonstrate a 60% relative risk reduction in the proportion of subjects with new onset of type 2 diabetes. 5.3 Treatment of subjects Subjects will attend Screening visit 1 to assess their eligibility. If found eligible the subjects will return at Screening visit 2, where they will have laboratory samples taken for assessment of prediabetes/diabetes status (FPG and OGTT) and fasting lipids. At Visit 3 subjects will be randomised to one of the two treatment groups as shown in Table 5 1 and Table 5 2. If subjects based on either FPG, 2-hr post-challenge plasma glucose or HbA 1c are classified as having pre-diabetes (see section 5.1 for definition) at Screening visit 2, subjects will be asked to participate in the delayed-onset-oftype-2-diabetes part of the trial lasting for 162 weeks. Furthermore, subjects will be asked to sign an additional informed consent form no later than at Visit 3 confirming this. All other subjects (not classified as having pre-diabetes) will be randomised to the 56 week main trial plus a 12 week rerandomised treatment period and a two weeks follow-up period. In addition, a subset population of approximately 60 subjects from US and Europe (UK and the Netherlands) will be selected to participate in the PK sub-study (Visit 8) as well. A separate informed consent form will be signed for the participation in the sub-study (see section 8.4.2). Table 5 1 Treatment groups: Main trial (56 weeks) incl. 12 week re-randomised treatment period and two weeks FU period* *(please note that this table does not include subjects with pre-diabetes at baseline. For treatment of subjects with pre-diabetes at baseline, please refer to Table 5-2)

144 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 25 of 138 Trial periods Visits in each period Duration of each period Treatment N arm Screening V1-V2 Liraglutide 867 Screening Liraglutide 867 Screening Liraglutide placebo Randomisati on/ Dose escalation V3-V5 Maintenance V6, V6c- V17 Rerandomised treatment period V18a-21a FU V22a 2 weeks 4 weeks 52 weeks 12 weeks 2 week 867 Screening Liraglutide mg* Liraglutide mg* Liraglutide placebo Liraglutide 3.0 mg Liraglutide 3.0 mg Liraglutide placebo Liraglutide 3.0 mg Liraglutide placebo Liraglutide placebo FU FU FU Table 5 2 Treatment groups: Delayed-onset-of-type-2-diabetes part of the trial* *(applies to subjects with pre-diabetes at baseline) Trial periods Screening Randomisation/ Dose escalation Maintenance End of 160 week randomised treatment Follow-up Visits in each period V1-V2 V3-V5 V6, V6c-V43b V44b Duration of each period 2 weeks 4 weeks 156 weeks 2weeks Treatment arm N Liraglutide 666 Screening Liraglutide Liraglutide mg 3.0 mg Follow-Up Liraglutide placebo 333 Screening Liraglutide placebo Liraglutide placebo Follow-Up

145 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 26 of Liraglutide/liraglutide placebo Liraglutide will be available in a concentration of 6.0 mg/ml and supplied in 3 ml FlexPen. Treatment allocation will be blinded to subjects (irrespective of pre-diabetes status at baseline), Investigator and throughout the main part of the trial (56 weeks). However, after database lock of the 70-week-data (56-week-data plus12 week re-randomised treatment data and two weeks follow-up data), will be unblinded. As a consequence, only subjects and Investigator remain blinded to the treatment allocation in the 104-week extension (plus 2-week follow up) period. Investigator will be notified of the treatment allocation when the 164 week trial has ended and the results have been released. Dosing with liraglutide or liraglutide placebo FlexPen is controlled by turning the dose selector until the dose indicator shows the relevant dose (10, 20, 30, 40, 50 clicks). 1 click equals 0.06 mg. Liraglutide is administered once daily by subcutaneous injections with the FlexPen either in the abdomen, thigh or upper arm. Injections can be done at any time of day irrespective of meals. However, it is preferable that liraglutide be injected during the same overall time period on a day to day basis. The injection site does not have to be kept consistent throughout the trial. Subjects will be instructed to perform an air shot before the first injection with the FlexPen. For further information please see the handling instruction for liraglutide FlexPen. Handling instructions will be provided in local language together with the trial product. The Investigator must instruct subjects how to inject liraglutide or liraglutide placebo and must ensure that subjects are familiar with the instructions. It must be documented in the subject s medical record that the subject has been instructed in the use of the FlexPen. The treatment duration (for a subject without pre-diabetes at baseline) will be up to 56 weeks in the main trial plus 12 weeks in the re-randomised treatment period and 2 weeks follow-up (a total of 70 weeks). The treatment duration for a subject with pre-diabetes at baseline who completes the delayed-onset-of-type-2-diabetes part of the trial will be up to 162 weeks. Subjects will follow a fixed dose escalation. The dose will be gradually escalated to 3.0 mg starting with 0.6 mg and with a dose level increment of 0.6 mg every 7 days. In order to reduce the level of side effects, liraglutide is gradually escalated up to the maintenance dose. If subjects do not tolerate an increase in dose during dose escalation, the Investigator has the option to individualise the dose escalation with a total delay of up to 7 days. All subjects must be at the target dose of 3.0 mg 35 days after randomisation. After reaching the target dose of 50 clicks per day (3.0 mg liraglutide or liraglutide placebo), dose and dosing frequency must not be changed at any time during the treatment period. If subjects do not tolerate the treatment dose they must be withdrawn from the trial. If the investigator suspects

146 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 27 of 138 acute pancreatitis, all suspected drugs should be discontinued until confirmatory tests have been conducted. If tests reveal that a subject does not have acute pancreatitis, the subject can remain in the trial with re-initiation of titration until the target dose is reached. The exact date, time, dose and site of injection of the 3 doses of trial product immediately prior to Visit 4, 7, 8 (sub-study) and Visit 11, must be noted in the case report form (ecrf) for evaluation of PK. Subjects should be encouraged to preferably use the same injection site three days prior to blood sampling for PK measurement on Visit 8 (sub-study) Counselling on diet and physical activity At visit 3 subjects will receive diet counselling (either in a group or individually) by a qualified dietician according to local standard and will be put on a hypocaloric diet containing max. 30% of energy from fat, approximately 20% of energy from protein, approximately 50% of energy from carbohydrates and with an energy deficit of approximately 500 kcal/day compared to the subject s estimated total energy expenditure (TEE) (see ). The hypocaloric diet is continued after randomisation and throughout the treatment period. If after 28 weeks of treatment the subjects are unable to lose additional weight despite having a BMI 25, recalculation of recommended energy intake with no kcal deficit (maintenance diet) is accepted. If a BMI 22 is reached the recommended energy intake should be recalculated with no kcal deficit (maintenance diet) for the remainder of the trial. All subjects will receive diet and physical activity counselling (either in a group or individually) at Visit 3 (randomisation), Visit 5, Visit 6c and at all subsequent visits with the exception of Visit 18a. All subjects are instructed by dieticians to keep a 3-day food diary every second month between Visits 2 and 3, Visits 5 and 6c, Visits 7 and 8, Visits 9 and 10, Visits 11 and 12, Visit 13 and 14, Visits 16 and 17 for the main part of the trial. During the re-randomised treatment period, the 3-day food diary will be used to register food intake between Visits 16 and 17 and between Visit 19a and 20a. In the delayed-onset-of-type-2-diabetes part of the trial, registration of food intake using the 3- day food diary will be done between Visits 18b and 19b, Visits 20b and 21b, Visits 22b and 23b, Visits 24b and 25b, Visits 26b and 27b, Visits 28b and 29b, Visits 30b and 31b, Visits 32b and 33b, Visits 34b and 35b, Visits 36b and 37b, Visits 38b and 39b, Visits 40b and 41b, Visits 42b and 43b. The 3-day food diary will be used for diet counselling. The subject s dietary compliance and the average daily level of physical activity will be recorded every second month. The subject will be asked if they performed less than half an hour, between half an hour and one hour or more than 1 hour of physical activity per day. For the main part of the trial this applies to Visit 3 (only average level of physical activity), 6c, 8, 10, 12, 14 and 17. For the re-randomised treatment period, the dietary compliance and physical activity will also be recorded at Visit 20a. For the delayed-onset-of-type-2-diabetes part of the trial, the subject s dietary

147 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 28 of 138 compliance and average daily level of physical activity will be recorded at Visit 3, 6c, 8, 10, 12, 14, 17, 19b, 21b, 23b, 25b, 27b, 29b, 31b, 33b, 35b 37b, 39b, 41b and 43b. An increase in physical activity (recommended minimum150 minutes/week) will be encouraged and re-enforced by use of pedometers. Whether or not the subject is in compliance with the prescribed diet is at the discretion of the dietician after review of the food diary. At Visit 21a and Visit 43b a dietary session on future weight management will be done for subjects ending treatment Calculation of estimated TEE The TEE is calculated by multiplying the estimated Basal Metabolic Rate (BMR) (see Table 5 3) with a Physical Activity Level (PAL) value of 1.3(23): The TEE (kcal/day) = BMR x 1.3 Table 5 3 Equations for estimating BMR in kcal/day* Sex Age BMR (kcal/day) Men years x actual weight in kg years x actual weight in kg > 60 years x actual weight in kg Women years x actual weight in kg years x actual weight in kg > 60 years x actual weight in kg * Revised WHO equations(23) 5.4 Rationale for treatment Based on a balanced evaluation of the efficacy and safety/tolerability results from the doubleblinded phase 2 dose-range finding trial (NN ) and subsequent 32 week open label extension (Investigator and subjects remained blinded), a dose of 3.0 mg has been chosen as the optimal dose for the weight management indication for liraglutide, and will therefore be further investigated in this confirmatory trial. Liraglutide 3.0 mg induced a clinically relevant weight loss from baseline, while improving weight-related risk factors, such as waist circumference, systolic blood pressure and glucose control. The safety of 3.0 mg liraglutide was satisfactory with the main tolerability issue identified being gastro-intestinal side-effects (nausea, diarrhoea and vomiting). The gastro-intestinal side effects were predominantly of mild to moderate in severity, were mostly transient in nature and were mostly present during the first 6-8 weeks of treatment. Only few

148 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 29 of 138 symptoms of hypoglycaemia have been reported in subjects without diabetes and in the phase 2 trial no subjects had confirmed low fasting blood glucose (< 3.1 mmol/l). In this trial all subjects will at randomisation be instructed in symptom recognition and handling of hypoglycaemia. This trial will investigate the efficacy and safety of liraglutide at a dose level of 50 clicks per day (3.0 mg liraglutide or liraglutide placebo). The dose will be gradually escalated to 3.0 mg/day with a dose level increment of 0.6 mg every 7 days. Based on data from an ongoing clinical development program of liraglutide, the current trial is anticipated to confirm the safety and efficacy of liraglutide on body weight loss in obese or overweight subjects without type 2 diabetes. In this trial based on current available data, subjects without pre-diabetes will be treated for a total period of 56 weeks, which should provide ample time to assess the capacity of liraglutide to induce and maintain weight loss and evaluate the long term safety of liraglutide in non-diabetic obese subjects or overweight subjects with co-morbidities. In order to assess the effects of drug cessation on appetite and weight control, possible withdrawal effects and rebound will be ascertained in the re-randomised treatment period (that consist of a blinded 12 week on re-randomised treatment [liraglutide treatment arm re-randomised 1:1 to liraglutide or liraglutide placebo and liraglutide placebo arm continues unchanged] and a 2 week off-randomised-treatment period to assess potential liraglutide antibodies). A treatment period of 160 weeks for subjects with pre-diabetes at randomisation is considered to be a sufficient duration to demonstrate delayed-onset-of-type-2-diabetes in subjects at high risk, while confirming the long-term safety of liraglutide. For further information please refer to the liraglutide obesity IB 3 rd edition, 2010 and any subsequent updates thereof.

149 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 30 of Trial population 6.1 Number of subjects to be studied Countries planned to participate: Germany, Spain, France, Italy, Belgium, Austria, Switzerland, Hungary, Poland, Serbia, UK, Norway, Finland, Denmark, Netherland, Ireland, Turkey, Israel, Brazil, Mexico, India, Russia, Hong Kong, South Africa, US, Canada, Australia. Planned number of subjects to be screened: 4800 Planned number of subjects to be randomised/started on trial product(s): 3600 Minimum number of subjects randomised to the delayed-onset-of type-2 diabetes part of the trial: 1000 Planned number of subjects to complete the main trial (56 weeks): 2160 Planned number of subjects to complete the delayed-onset-of type-2 diabetes part of the trial: 350 Anticipated number of trial sites: subjects will be screened in order to obtain a minimum of 3600 randomised subjects of which 2400 are on liraglutide and 1200 on liraglutide placebo. This number is based on an anticipated screening failure rate of 25%. As a minimum of 1000 subjects will be stratified to the delayedonset-of-type-2-diabetes part of the trial (160 weeks of treatment), it is anticipated that the prevalence of pre-diabetes in obese subjects at randomisation will be at least 30% (please refer to section 18). If the prevalence of pre-diabetes is lower than expected, screening and randomisation will continue with the purpose to identify and enrol subjects with pre-diabetes. There is no upper limit to the number of subjects with pre-diabetes which can be enrolled within the planned total of The anticipated screening failure rate of 25% and the anticipated drop-out rate at 56 weeks of 40% are based on reported rates from obesity trials. A drop-out rate of 65% over 160 weeks is anticipated based on previous experience from diabetes prevention trials(24).

150 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 31 of Inclusion criteria 1. Informed consent obtained before any trial-related activities (Trial related activities are any procedure that would not have been performed during the normal management of the subject) 2. Body Mass Index (BMI) 30.0 kg/m2 or 27 kg/m2 with presence of co-morbidities of treated or untreated dyslipidemia and/or hypertension. (Untreated dyslipidemia is defined as LDL 160 mg/dl, or Triglycerides 150 mg/dl, or HDL < 40 mg/dl for male and < 50 mg/dl for female(25),(26), untreated hypertension is defined as Systolic Blood Pressure 140 or Diastolic Blood Pressure 90 mmhg(27) 3. Stable body weight (less than 5 kg self-reported change during the previous 3 months) 4. Preceding failed dietary effort 5. Age 18 years 6.3 Stratification criteria To be eligible for stratification to the delayed-onset-of-type-2-diabetes part of the trial, subjects must comply with the following: FPG 100 mg/dl (5.6 mmol/l) and 125 mg/dl (6.9 mmol/l) AT Screening visit 2, and/or 2 hr post-challenge (OGTT) plasma glucose 140 mg/dl (7.8 mmol/l) and 199 mg/ dl (11.0 mmol/l) at Screening visit 2, and/or HbA 1c % both inclusive, at Screening visit 1 (28) Subjects will furthermore be stratified according to BMI ( 30 kg/m 2 or < 30 kg/m 2 ). 6.4 Exclusion criteria 1. Diagnosis of type 1 or type 2 diabetes per the judgement of the Investigator 2. HbA 1c 6.5 % (Screening visit 1) or FPG 126 mg/dl (7.0 mmol/l) (Screening visit 2) or 2 hr port-challenge plasma glucose 200 mg/dl (11.1 mmol/l) (Screening visit 2) 3. Previous treatment with GLP-1 receptor agonists (including liraglutide or exenatide) within the last 3 months 4. Untreated or uncontrolled hypothyroidism/ hyperthyroidism defined as TSH > 6 miu/l or < 0.4 miu/l 5. Screening calcitonin 50 ng/l 6. Family or personal history of multiple endocrine neoplasia type 2 (MEN2) or familial medullary thyroid carcinoma (FMTC) 7. Personal history of non-familial medullary thyroid carcinoma 8. History of chronic pancreatitis or idiopathic acute pancreatitis

151 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 32 of Obesity induced by other endocrinologic disorders (e.g. Cushing Syndrome) 10. Current or history of treatment with medications that may cause significant weight gain, within 3 months prior to Screening visit 1, including systemic corticosteroids (except for a short course of treatment, i.e., 7-10 days), tri-cyclic antidepressants, atypical antipsychotic and mood stabilizers (e.g., imipramine, amitryptiline, mirtazapin, paroxetine, phenelzine, clorpromazine, thioridazine, clozapine, olanzapine, valproic acid and its derivatives, and lithium) 11. Diet attempts using herbal supplements or over-the-counter medications within 3 months before Screening visit Current participation in an organised weight reduction program (or within the last 3 months) and /or are currently using or have used within three months before Screening visit 1: pramlintide, sibutramine, orlistat, zonisamide, topiramate, phenteremine, or metformin (either by prescription or as part of a clinical trial) 13. Participation in a clinical trial within the last 3 months prior to Screening visit Simultaneous participation in any other clinical trial of an investigational drug 15. Previous surgical treatment for obesity (excluding liposuction if performed > one year before trial entry) 16. History of Major Depressive Disorder within the last 2 years 17. History of other severe psychiatric disorders, e.g., schizophrenia, bipolar disorder 18. A PHQ-9 score of Any lifetime history of a suicidal attempt 20. A history of any suicidal behaviour in the last month prior to randomisation 21. Any suicidal ideation of type 4 or 5 on the C-SSRS in the last month prior to randomisation 22. Surgery scheduled for the trial duration period, except for minor surgical procedures, at the discretion of the Investigator 23. Uncontrolled treated/untreated hypertension (systolic blood pressure 160 mmhg and/or diastolic blood pressure 100 mmhg). If white-coat hypertension is suspected at the Screening visit 1 a repeated measurement at Screening visit 2 prior to other trial related activities is allowed 24. Cancer (past or present, except basal cell skin cancer or squamous cell skin cancer), which in the Investigator s opinion could interfere with the results of the trial 25. Known or suspected hypersensitivity to trial product(s) or related product(s) 26. Previous participation in the randomised phase of this trial. Re-screening is allowed once within the limit of the recruitment period 27. Known or suspected abuse of alcohol or narcotics 28. Language barrier, mental incapacity, unwillingness or inability to understand and be able to complete the mental health questionnaire in the provided language 29. Subjects from the same house hold participating in the trial 30. Females of child bearing potential who are pregnant, breast-feeding or intend to become pregnant or are not using adequate contraceptive methods (adequate contraceptive measures as required by local law or practice). US: abstinence and the following methods: diaphragm with

152 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 33 of 138 spermacide, condom with spermacide (by male partner), intrauterine device, sponge, spermacide, Norplant, Depo-Provera or oral contraceptives. Germany: adequate contraceptive measures are implants, injectables, combined oral contraceptives, hormonal Intra-Uterine Device (IUD), sexual abstinence or vasectomised partner. UK: adequate contraceptive measures are defined as sterilisation, intra-uterine device, oral contraceptives, consistent use of barrier methods, male sterilisation or true abstinence. 31. The receipt of any investigational drug within four weeks prior to screening for this trial (Screening visit 1) Subjects who are non-compliant with any of the eligibility criteria, but included in the trial, should be withdrawn immediately. If extraordinary circumstances speak in favour of maintaining the subject in the trial then this is only acceptable if justified and approved by the Independents Ethics Committee (IEC)/Institutional Review Board (IRB), and if the regulatory authorities are notified according to local requirements. Investigators are recommended to refer screening failure subjects having a calcitonin value of 50 ng/l to an endocrinologist for follow-up. 6.5 Rescue criteria The rescue criteria only apply for subjects developing diabetes during the trial. If self measured FPG on three consecutive days/occasions exceeds the limits set below, the subject should contact the Investigator and come in for an unscheduled visit as soon as possible. The next scheduled visit should not be awaited. A FPG should be obtained and analysed by the central laboratory. If this FPG exceeds the limits set below, the background medication dose should initially be escalated to the maximum approved dose, and if not sufficient to achieve glycaemic control, one of the other allowed antidiabetic treatment should be added (please refer to withdrawal criteria no. 4): From baseline to Week 6: FPG > 15 mmol/l (270 mg/dl) From Week 7 to Week 12: FPG > 13.3 mmol/l (240 mg/dl) From Week 13 to Week 56: FPG > 11.1 mmol/l (200 mg/dl) Moreover, if any of the FPG or HbA 1c samples analysed by the central laboratory exceeds the limits set below, the subject should immediately be called in for an unscheduled visit as soon as the result has come to the knowledge of the Investigator. The next scheduled visit should not be awaited. A new FPG should be obtained and analysed by the central laboratory and if this FPG exceeds the limits set below, the background medication dose should initially be escalated to the maximum approved dose, and if not sufficient to achieve glycaemic control, one of the other allowed antidiabetic treatment should be added (please refer to withdrawal criteria no. 4):

153 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 34 of 138 From baseline to Week 6: FPG > 15 mmol/l (270 mg/dl) From Week 7 to Week 12: FPG > 13.3 mmol/l (240 mg/dl) From Week 13 to Week 56: FPG > 11.1 mmol/l (200 mg/dl) or HbA 1c > 8% 6.6 Withdrawal criteria The subject may be withdrawn from the trial at the discretion of the Investigator or due to a safety concern or if judged non-compliant with trial procedures. A subject must be withdrawn if the following applies: 1. The subject may withdraw from the trial at will at any time 2. If the target treatment dose of the randomised trial product is not tolerated by the subjects 3. Pregnancy or intention of becoming pregnant 4. Subjects who develop diabetes during the trial will not be withdrawn but should receive the best standard of care at the discretion of the Investigator. If the Investigator determines that insulin, GLP-1 receptor agonist (e.g., Byetta or Victoza ), or DPP-IV inhibitor, is the best treatment option, the subject must be withdrawn. The medication prescribed by the Investigator will not be provided by 5. If the investigator suspects acute pancreatitis, all suspected drugs should be discontinued until confirmatory tests have been conducted and appropriate treatment should be initiated. Subjects that are diagnosed with acute pancreatitis (as a minimum 2 of 3: characteristic abdominal pain, amylase and/or lipase >3x UNR or characteristic findings on CT/MRI), must be withdrawn from the trial. 6. A subject should be referred to a Mental Health Professional (MHP) if he/she has a PHQ-9 score 10, OR, any suicidal behaviour, OR, any suicidal ideation of type 4 (active suicidal ideation with some intent to act, without specific plan) or type 5 (active suicidal ideation with specific plan and intent) on any C-SSRS assessment A referral to a Mental Health Professional (MHP) should also be made if in the opinion of the Investigator it is necessary for the safety of the subject. If a subject s psychiatric disorder can be adequately treated with psycho- and/or pharmacotherapy, then the subject, at the discretion of the Investigator (in agreement with the MHP), may be continued in the trial on randomised therapy, otherwise, the subject must be withdrawn In case of withdrawal, the End of Trial form must be filled in and in the IV/WRS the Withdrawal session must be completed. If possible, the subject should be called in for a final visit. Procedures according to Visit 17 should be performed for all subjects who discontinue the trial prematurely

154 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 35 of 138 before Visit 17, and procedures relating to Visit 21a should be performed for all subjects continuing in the re-randomised treatment period. For subjects continuing in the delayed-onset-of-type-2- diabetes part of the trial and past Visit 17, procedures according to Visit 43b should be performed. Subjects who have discontinued the trial prematurely prior to Visit 17 will be asked to attend a visit (Visit 17x) taking place 56 weeks ± 5 days after their randomisation date. The purpose of this visit will be recording of the body weight and reporting of MESIs. 6.7 Subject replacement Withdrawn subjects will not be replaced. However, re-screening is allowed once within the recruitment period at the discretion of the Investigator. If a subject is re-screened, a new subject number must be assigned, and a new informed consent has to be signed off, and all Screening visit 1 blood sampling and assessments have to be repeated. 6.8 Rationale for trial population Obese subjects or overweight subjects with co-morbidities have been included in the initial 56 weeks, since this population is likely to benefit from weight reduction. Subjects with diabetes have been excluded from the trial as there may be a confounding effect of changes in glycaemic control on weight loss and subjects with type 2 diabetes may have specific safety concerns which will be addressed in a dedicated trial. Subjects with a presumed high risk of developing diabetes (pre-diabetes defined as either IFG and/or IGT and/or HbA 1c % inclusive (29) (see section 5.1 )) have been selected for extended investigation (up to 160 weeks) in order to demonstrate not only the long-term safety and efficacy of inducing and maintaining weight loss, but also the ability of liraglutide treatment to delay the onset of type 2 diabetes. The risk of developing type 2 diabetes over 160 weeks for this population is presumed to be approximately 20% (17), (18), (19) and the trial aims at investigating whether additional factors can be used to identify subjects most at risk.

155 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 36 of Trial schedule Planned duration of recruitment period: Planned date for first subject (FPFV): 22 weeks 01-Jun-2011 Planned completion of last subject in the main trial including 12 week re-randomised treatment period and two weeks FU period(lplv): Mar Planned completion of clinical trial report: Planned completion of last subject in the delayed-onset-of-type-2-diabetes part of the trial (LPLV): Aug-2013 Dec-2014 Planned completion of clinical trial report in the delayed-onset-of-type-2-diabetes part of the trial: Apr The end of the clinical trial is defined as the last visit of the last subject globally. The recruitment will be evaluated by on a continuous basis and country- and regional- specific contingency measures may be applied. A competitive recruitment may be applied if deemed needed. To ensure that only the required number of subjects is randomised, screened subjects will be monitored closely via IV/WRS. All Investigators will be notified immediately when the enrolment period comes to an end, after which no subjects must be screened, and the IV/WRS will be closed for further screening. Protocol information for this trial will be subject to public disclosure at external web sites ( and according to international regulations e.g. the International Committee of Medical Journal Editors (ICMJE) (30), the Food and Drug Administration Amendments Act (FDAAA) (31-33) - as reflected in Code of Conduct for Clinical Trial Disclosure.

156 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 37 of Methods and assessments 8.1 Visit procedures Before screening takes place subjects will be provided with written and verbal information about the trial and the procedures involved. Qualified site staff in accordance with local law will ensure that subjects are fully informed both verbally and in writing about the practical consequences of participating, of their rights and responsibilities while participating in the trial, as well as any possible advantages and disadvantages in being treated with the trial products. Subjects will have the opportunity to ask questions to a medically qualified person and have ample time to consider their participation. Subjects who wish to participate must give signed and dated informed consent. This must be done prior to any trial related activities, i.e. procedures that would not have been performed during normal management of the subject. The Investigator must keep a subject screening log and a subject enrolment log. These can be combined into one document. It must be stated in the medical record that the subject is participating in the current trial. Subjects enrolled in the trial will be provided with a subject ID card, stating that they are participating in a trial and whom to contact (site address, Investigator s name and telephone number) for further information, if necessary. The subjects should be reminded to show the card to other health care providers, as applicable. The subjects should be instructed to return the card to the Investigator at the last visit of the subject or destroy the card after the last visit. For screening failures (subjects who have given informed consent but who do not fulfil all inclusion criteria and/or meet at least one exclusion criterion and hence are not randomised) the screening failure session in the IV/WRS must be completed and the reason for screen failure will be recorded in the ecrf. All screen failure information will be available in the clinical database. Serious and non-serious adverse events from screening failures will be entered by the Investigator into the electronic ecrfs and subsequently transferred to the clinical database. When trial related procedures have been finalised for screening failures, no more adverse events should be entered in the ecrf. Follow-up of AEs should be made according to section 12. Investigators are recommended to refer screening failure subjects having a calcitonin value of 50 ng/l to an endocrinologist for follow-up. For subjects with calcitonin 20 ng/l (from Visit 3 and onwards) a repeat measurement of calcitonin must be performed preferably within four weeks and if confirmed, the event elevated calcitonin should be reported as a MESI (i.e. any confirmed episode of calcitonin concentration value 20 ng/l, if not already reported as an ongoing MESI, see section

157 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 38 of 138 During the 56 week main part of the trial including the re-randomised treatment period subjects should attend the clinic in a fasting condition in the morning (at least eight hours overnight fast without food and/or drink intake, except for water) at Visit 2, 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 17x, 18a, 21a and Visit 22a. For the delayed-onset-of-type-2-diabetes part of the trial subjects should attend the clinic in a fasting condition in the morning (at least eight hours overnight fast without food and/or drink intake, except for water) at Visit 2, 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 17x, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, 43b and 44b. Concomitant medication should be taken as usual during the conduct of the trial i.e. concomitant medication should not be withheld on the day of the visit to the clinic, until blood sampling has been done. All subjects should withheld trial product on Visits 2, 3, 4, 5, 6c, 8, 9, 11, 14, 17, 17x, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, 43b and 44b (fasting visits) until blood sampling has been done. Subjects developing diabetes during the trial should in addition withhold their background diabetes medication at Visits 2, 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 17x, 17a, 18a, 21a, 22a, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, 43b and Visit 44b until blood sampling has been done. At all other visits diabetes background medication and trial product should be taken as usual during the conduct of the trial. Weight will be measured at all visits to the clinic except for Screening visit 2. Following randomisation, trial product will be dispensed at all visits to the clinic except for Visit 4, 6, 18a and End of Treatment visits (Visits 21a and 43b). Subjects will be asked to bring all empty, partly used and unused trial products for drug accountability. If the subject attends the clinic for a visit not described in the protocol, an Unscheduled Visit Form must be completed. This only applies if any assessments are carried out, e.g. not for the purpose of dispensing trial products. If an unscheduled visit is made for the purpose of dispensing trial products to the subjects then in the IV/WRS the unscheduled dispensing session must be completed. If the subject attends a fasting visit in a non-fasting state the subject should preferably be called in for a new visit within the visit window to have the fasting weight and fasting circumference measured and the fasting blood samples drawn. All other assessments can be performed even though the subject is not fasting. If it is not possible to attend the new visit within the visit window, the subjects should be called in for an Unscheduled Visit and an Unscheduled Visit Form should be used to obtain information on fasting weight, fasting circumference and fasting blood samples. In case a subject is being prematurely withdrawn from the trial, the Investigator must aim to undertake procedures for the End of Treatment visit (Visit 17, 21a or 43b) as soon as possible, if possible. The primary reason (adverse event, non-compliance with protocol or other) for discontinuation must be specified in the ecrf and in the IV/WRS a withdrawal session should be completed. The end-of-trial from must be completed, and final drug accountability must be performed even if the subject is not able to attend.

158 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 39 of 138 Subjects who have discontinued the trial prematurely before Visit 17 will be asked to attend a visit (Visit 17x) 56 weeks ± 5 days after their randomisation date. The purpose of this visit will be to record the body weight and reporting of MESIs. The assessment will be done by asking the subject if he/she experienced any MESI (see section and appendix J) since the last contact. The subjects should attend Visit 17x fasting. If the subject is not willing to attend Visit 17x, it should be documented in the patient medical record that the subject has refused to attend the visit. As liraglutide is not available on prescription for the obesity indication after the end of the trial, subjects will not be able to receive it when the trial ends. This means that will not offer investigational drug after the end of the trial. At the End of Treatment visit (Visit 21a or 43b) the Investigator should provide counselling on future weight management.

159 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 40 of Visit schedule Subjects will attend the clinic as stated in Table 8 1. Table 8 1 Visit schedule Visit no Time of Visit Type of Visit 56 weeks main trial (all subjects) Visit 1 Week -2 ± 5 days Screening visit 1 Visit 2 Week -1 ± 3 days Screening visit 2 Visit 3 Week 0, baseline Randomisation, start of liraglutide/liraglutide placebo 0.6 mg Visit 4 Week 2 ± 3 days Dose escalation Visit 5 Week 4 ± 5 days Dose escalation Visit 6-16 Week 6-50 ± 5 days Maintenance visits Visit 17 Week 56 ± 5 days End of 56 weeks randomised treatment/premature discontinuation visit Visit 17x Week 56 ± 5 days Body weight/ MESI recording visit Rerandomised treatment period (ONLY subjects without prediabetes) Visit 18a Week 58 ± 3 days Re-randomised treatment Visit 19a Week 60 ± 5 days Re-randomised treatment Visit 20a Week 64 ± 5 days Re-randomised treatment Visit 21a Week 68 ± 3 days End of re-randomised treatment/ premature discontinuation visit Visit 22a Week 70 ± 3 days Follow-up visit ONLY subjects with pre-diabetes Visit 18b-42b Week ± 5 days Maintenance (delayed-onset-of-type-2-diabetes part of the trial) Visit 43b Week 160 ± 3 days End of 160 week randomised treatment visit/ premature discontinuation (delayed-onset-of-type-2-diabetes part of the trial) Visit 44b Week 162 ± 3 days Follow-up visit (delayed-onset-of-type-2-diabetes part of the trial) The duration of the trial for a subject completing the 56 week main trial including the rerandomised treatment period and the follow-up period will be up to 72 weeks and will comprise a total of 23 visits. The total duration for a subject completing the delayed-onset-of-type-2-diabetes part of the trial will be up to 164 weeks and will comprise a total of 45 visits.

160 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 41 of Screening visit 1 Screening visit 1 will take place 2 weeks ± 5 days before randomisation (Visit 3). Subjects must give signed and dated informed consent prior to any trial-related activities. All subjects will be provided with a copy of the subject information and a copy of their own signed and dated Informed Consent Form. The IV/WRS should be called/entered to register the subject as screened (see section 10). The following will be performed and/or recorded in the ecrf: Informed consent, signed and dated (see section 8.1) Inclusion and exclusion criteria (see section 6.2, 6.4) Demographics information registered in the IV/WRS and/or recorded in the ecrf (date of birth, sex, race and ethnicity, according to local requirements) Medical history/concomitant illness Concomitant medication Smoking habits (see section 8.4.3) Body measurement (see section 8.2.1) Vital signs (se section 8.2.8) Physical examination (see section 8.3.1) ECG (see section 8.3.4) Mental health questionnaires (see section ) Signs or symptoms of diabetic complications (see section 8.4.6) History of concomitant cardiovascular disease (see section 8.4.7) History of gallbladder disease (see section 8.4.8) History of psychiatric disorders (see section 8.4.9) Record the subject on the subject screening log or the informed consent log according to local practice Blood sampling for measurement of: HbA 1c (see section 8.2.2) Haematology and biochemistry (see section 8.3.7) For females of childbearing potential: serum pregnancy test (see section 8.3.5) Once all data relating to Screening visit 1 have been obtained, they must be reviewed by the Investigator to ensure that the subject is eligible to continue in the trial. If the subject is not eligible to continue, the subject is a screening failure, and the IV/WRS must be contacted to register the subject as a screening failure, and the reason for screen failure must be recorded. Reminders: Make an appointment for Screening visit 2, 1 week ± 3 days after Screening visit 1

161 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 42 of 138 Remind the subject to attend Screening visit 2 fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) Remind the subject to avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at Screening visit Screening visit 2 Screening visit 2 will take place 1 week ± 3 days before randomisation (Visit 3). If the subject is found eligible to continue in the trial based on Screening visit 1 data, the subject will have blood samples drawn for assessment of FPG and OGTT to assess the pre-diabetes status. In order to be eligible for stratification to the 160 week delayed-onset-of-type-2-diabetes part of the trial, the subject must comply with the following definition of pre-diabetes: IFG (FPG 100 mg/dl / 5.6 mmol/l, and < 125 mg/dl / 6.9 mmol/l) and/or IGT (2 hr post OGTT plasma glucose 140 mg/dl /7.8 mmol/l and < 199 mg/dl / 11.0 mmol/l) or HbA 1c % both inclusive (34) If the subject does not have pre-diabetes, the subject will be stratified to the 56 week main trial (plus the re-randomised treatment period). The following will be performed and/or recorded in the ecrf: Inclusion and exclusion criteria (see section 6.2,6.4) Concomitant medication, changes since last visit, if any Re-measurement of blood pressure is allowed if white coat hypertension is suspected at Screening visit 1 For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Dispense 3-day food diary (see section ) Blood sampling for measurement of: FPG (see section 8.2.3) OGTT (see section 8.2.6) Lipids (see section ) Reminders: Make an appointment for Visit 3, 1 week after Screening visit 2 Remind the subject to attend Visit 3 fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water)

162 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 43 of 138 Remind the subject to avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at Visit 3 Remind the subject to keep a 3-day food diary between Screening visit 2 and Visit 3 (the diary will be used for diet counselling at Visit 3 (see section 5.3.2) Remind the subject to bring a sample of their morning urine to Visit Visit 3, randomisation Visit 3 will take place one week after Screening visit 2. If the subject is found eligible after review of inclusion and exclusion criteria at Screening visit 1 and Screening visit 2 and complies with the pre-diabetes stratification criteria (see section 6.3) the subject will be stratified to the delayed-onset-of-type-2-diabetes part of the trial lasting for 162 weeks in total (attending Visit 3 to 17 and Visit 18b to 44b). Subjects found eligible to be stratified to the 162 week trial will be asked to sign an additional informed consent no later than at Visit 3. If the subject is found eligible in accordance with inclusion and exclusion criteria at Screening visit 1, but does not comply with the pre-diabetes stratification criteria, the subject will be stratified to the 56 week main trial including a re-randomised treatment period. At the end of the 56 week main trial subjects who received liraglutide during the 56 week main trial will be re-randomised 1:1 to either continue on liraglutide or be switched to liraglutide placebo during the re-randomised treatment period. All subjects will be randomised using IV/WRS (see section 10.3). The IV/WRS will allocate the dispensing unit numbers (DUN) of medication to be dispensed to the subject. Subjects will be supplied with trial product according to randomisation and instructed in daily injections of liraglutide or liraglutide placebo. The trial products dispensed will cover the dose escalation period. Injections can be done at anytime of day irrespective of meals. However, it is preferable that liraglutide/liraglutide placebo be injected during the same overall time period on a day to day basis. The injection site does not have to be kept consistent throughout the trial. Subjects will follow a fixed dose escalation. The dose will be gradually escalated to 3.0 mg starting with 0.6 mg and with a dose level increment of 0.6 mg every 7 days. If subjects do not tolerate an increase in dose during dose-escalation, the Investigator has the option to individualise the dose escalation with a total delay of up to 7 days. All subjects must be at the target dose of 3.0 mg 35 days after randomisation. The following will be performed and/or recorded in the ecrf: Informed Consent, signed and dated, for the delayed-onset-of-type-2-diabetes part of the trial (see section 8.1)Inclusion and exclusion criteria (see section 6.2,6.4) Stratification criteria (see section 6.3) Concomitant medication, changes since last visit, if any Body measurements (see section 8.2.1)

163 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 44 of 138 Vital signs (see section 8.2.8) PRO questionnaires (see section ) Binge eating scale questionnaire (see section ), not applicable for the subjects randomised to the delayed-onset-of-type-2-diabetes part of the trial For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Mental health questionnaire (see section ) Dispense study card FINDRISC score (see section ) Dispense pedometer Diet and physical activity counselling based on 3-day food diary (see section 5.3.2) Subject s average daily level of physical activity (see section 8.4.5) Train the subject in the use of the FlexPen Train the subject in symptom recognition and handling of hypoglycaemic events Interact with the IV/WRS to: Complete randomisation session Complete dispensing session Urine sampling for the measurement of: Urinary Albumin-to-Creatinine ratio (see section ) Blood sampling for measurement of: HbA 1c (see section 8.2.2) FPG (see section 8.2.3) Fasting insulin (see section 8.2.4) C-peptide (see section 8.2.5) Cardiovascular biomarkers (see section 8.2.9) Lipids (see section ) Haematology and biochemistry (see section 8.3.7) Liraglutide antibodies (see section 8.3.8) Reminders: Make an appointment for the subsequent visit Remind the subject to: attend visit 4 fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) keep a dosing diary for the three doses of trial product immediately prior to Visit 4

164 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 45 of 138 avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at all visits keep a 3-day food diary between Visit 5 and 6c (see section 5.3.2) return all empty, partly used and unused trial product drug accountability Visits 4 and 7 The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Mental health questionnaires (see section ) Dispense 3-day food diary at Visit 7 (see section ) Diet and physical activity counselling only applicable to Visit 7 (see section 5.3.2) Interact with the IV/WRS to: Complete dispensing session, only applicable for Visit 7 Complete withdrawal session (if applicable) Interact with the Drug Accountability Module, only applicable for Visit 7 Blood sampling for measurement of: Liraglutide plasma concentration (see section 8.4.1) FPG (only applicable for Visit 4) (se section 8.2.3) Reminders: Make an appointment for the subsequent visit Remind the subject to: attend the subsequent visit (Visit 5 and 8) fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) inject liraglutide/ liraglutide placebo at approximately 10 pm in the evening prior to Visit 8. Subjects should be instructed to use the same injection site for three consecutive days prior to Visit 8 and to note the date and time for the injection (see section 8.4.2) avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at all visits return all empty, partly used and unused trial product at all visits for drug accountability

165 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 46 of Visit 5 The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Mental health questionnaires (see section ) Dispense 3-day food diary (see section ) Diet and physical activity counselling (see section 5.3.2) Interact with the IV/WRS to: Complete dispensing session Complete withdrawal session (if applicable) Interact with the Drug Accountability Module Blood sampling for measurement of: HbA 1c (see section 8.2.2) FPG (see section 8.2.3) Fasting insulin (see section 8.2.4) C-peptide (see section 8.2.5) Haematology and biochemistry (see section 8.3.7) Reminders: Make an appointment for the subsequent visit Remind the subject to: attend Visit 6c fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at all visits keep a 3-day food diary between Visit 5 and 6c (the diary will be used for diet counselling at Visit 6c) (see section 5.3.2) return all empty, partly used and unused trial product at all visits for drug accountability

166 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 47 of Visits 6, 6c, 9, 10, 12, 13, 15, and 16. The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Mental health questionnaires (see section ) Dispense 3-day food diary at Visit 9, 13 and Visit 16 (see section ) Diet and physical activity counselling based on 3-day food diary at Visit 6c, 10 and Visit 12 (see section 5.3.2) Diet and physical activity counselling at Visit 9, 13, 15 and Visit 16 (see section 5.3.2) Subject s dietary compliance and average daily level of physical activity at Visit 6c, 10, and Visit 12 (see section 8.4.5) Interact with the IV/WRS to: Complete dispensing session (not applicable for Visit 6) Complete withdrawal session (if applicable) Interact with Drug Accountability Module (not applicable for Visit 6) Blood sampling for measurement of: FPG, only applicable for Visits 6c, 9 and 16 (see section 8.2.3) Reminders: Make an appointment for the subsequent visit Remind the subject to: attend visits 6c, 11, 14, 16 and 17 fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at all visits keep a 3-day food diary between Visit 9 and 10, Visit 13 and 14 and Visit 16 and 17, 16 and 17a (see section 5.3.2) keep a dosing diary for the three doses of trial product immediately prior to Visit 7 and Visit 11 return all empty, partly used and unused trial product at all visits for drug accountability remind the subject to bring a sample of their morning urine at Visit 11 and 17

167 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 48 of Visits 8 and 14 The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Mental health questionnaires ( see section ) Diet and physical activity counselling based on the 3-day food diary (see section 5.3.2) Subject s dietary compliance and average daily level of physical activity (see section 8.4.5) Interact with the IV/WRS to: Complete dispensing session Complete withdrawal session (if applicable) Interact with the Drug Accountability Module Blood sampling for measurement of: HbA 1c (see section 8.2.2) FPG (see section 8.2.3) Fasting insulin (see section 8.2.4) C-peptide (see section 8.2.5) Haematology and biochemistry (see section 8.3.7) Liraglutide plasma concentration (sub-study), only applicable for Visit 8 (see section 8.4.2) Reminders: Make an appointment for the subsequent visit Remind the subject to: attend Visit 9 fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at all visits keep a 3-day food diary between Visit 8 and 9 (the diary will be used for diet counselling at Visit 9) (see section 5.3.2) return all empty, partly used and unused trial product at all visits for drug accountability

168 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 49 of Visit 11 The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) PRO questionnaires (see section ) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Mental health questionnaires (see section ) Dispense 3-day food diary (see section ) Diet and physical activity counselling (see section 5.3.2) ECG (see section 8.3.4) Interact with the IV/WRS to: Complete dispensing session Complete withdrawal session (if applicable) Interact with the Drug Accountability Module Blood sampling for measurement of: HbA 1c (see section 8.2.2) FPG (see section 8.2.3) Fasting insulin (see section 8.2.4) C-peptide (see section 8.2.5) OGTT (see section 8.2.6) Cardiovascular biomarkers (see section 8.2.9) Lipids (see section ) Liraglutide plasma concentration (see section 8.4.1) Haematology and biochemistry (see section 8.3.7) Urine sampling for measurement of: Urinary Albumin-to-Creatinine ratio (see section ) Reminders: Make an appointment for the subsequent visit Remind the subject to: attend the subsequent visit 14 fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water)

169 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 50 of 138 avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at all visits return all empty, partly used and unused trial product at all visits for drug accountability Visit 17, End of 56 week randomised Treatment/premature discontinuation Visit 17 will take place 56 weeks ± 5 days after randomisation. This visit is applicable for: Subjects continuing in the re-randomised treatment period Subjects withdrawing prematurely Subjects continuing in the delayed-onset-of-type-2-diabetes part of the trial Subjects being prematurely withdrawn from the trial prior to Visit 17 should preferably have all procedures for the End of 56 weeks randomised Treatment visit performed. The primary reason (adverse event, non-compliance with protocol or other) for discontinuation must be specified in the ecrf and in the IV/WRS a withdrawal session should be completed. Even if the subject is not able to attend a final visit, the End of Trial Form must be completed. Subjects discontinuing the trial prematurely will be asked to attend a visit 17x 56 weeks ± 5 days after their randomisation date. The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Re-randomisation (only applicable for subjects without pre-diabetes at baseline randomised to liraglutide during the 56 week main trial) Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) PRO questionnaire (see section ) Binge eating scale questionnaire (see section ) not applicable for the subjects randomised to the delayed onset-of-type-2 diabetes part of the trial Physical examination (see section 8.3.1) ECG (see section 8.3.4) For females of childbearing potential: urine pregnancy test, if applicable (only applicable for subjects continuing in the delayed-onset-of-type-2-diabetes part of the trial) (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Diet and physical activity counselling based on the 3-day food diary (see section 5.3.2) Subject s average daily level of physical activity (see section 8.4.5) Mental health questionnaires (see section ) Completion of the 56 week main trial

170 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 51 of 138 Interact with the IV/WRS to: Complete dispensing session for subjects continuing in the delayed-onset-of-type-2-diabetes part of the trial Complete re-randomisation session for all subjects without pre-diabetes at baseline randomised to liraglutide during the 56 week main trial Complete dispensing session for subjects continuing in the re-randomised treatment period Complete withdrawal session (if applicable) Interact with the Drug Accountability Module Blood sampling for measurement of: HbA 1c (see section 8.2.2) FPG (see section 8.2.3) Fasting insulin (see section 8.2.4) C-peptide (see section 8.2.5) OGTT (see section 8.2.6) Cardiovascular biomarkers (see section 8.2.9) Lipids (see section ) For females of childbearing potential: serum pregnancy test (not applicable for pre-diabetic subjects continuing in the delayed-onset-of-type-2 diabetes part of the trial) (see section 8.3.5) Haematology and biochemistry (see section 8.3.7) Liraglutide antibodies (only applicable for subjects who discontinue the trial prematurely)(see section 8.3.8) Urine sampling for measurement of: Urinary Albumin-to-Creatinine ratio (see section ) Reminders: Make an appointment for Visit 18a, 58 weeks ± 3 days after randomisation for subjects continuing in the re-randomised treatment period Make an appointment for Visit 18b, 60 weeks ± 5 days after randomisation for subjects continuing in the delayed-onset-of-type-2-diabetes part of the trial Remind the subjects who discontinue the trial prematurely that they will be contacted for measurement of their weight 56 weeks ± 5 days after their randomisation date Remind subjects to attend Visit 18a fasting Remind the subject to bring a sample of their morning urine at Visit 18a

171 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 52 of Visit 17x, Body weight/mesi recording visit Visit 17x will take place 56 weeks ± 5 days after randomisation. The body weight recording visit is only applicable for subjects who have discontinued the trial prematurely before Visit 17. Subjects will be asked to attend this visit 56 weeks ± 5 days after their randomisation date for the purpose of body weight recording and assessment of MESI. The assessment will be done by asking the subject if he/she has experienced any MESIs (see section ) and Appendix J since the last contact. In addition the subjects will be asked to complete mental heath questionnaires. The subjects should attend Visit 17x fasting. If the subject is not willing to attend Visit 17x, it should be documented in the patient medical record that the subject has refused to attend the visit. The following will be performed and recorded in the ecrf: Body weight (see section ) MESI, if any (see section ) Visit 18a, Re-randomised treatment visit Visit 18a will take place 58 weeks ± 3 days after randomisation. This visit is only applicable for subjects continuing in the re-randomised treatment period. The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Binge eating scale questionnaire (see section ) Mental health questionnaire (see section ) Interact with the IV/WRS to: Complete withdrawal session (if applicable) Urine sampling for measurement of: Urinary Albumin-to-Creatinine ratio (see section )

172 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 53 of 138 Blood sampling for measurement of: FPG (see section 8.2.3) Haematology and biochemistry (see section 8.3.7) Liraglutide antibodies (see section 8.3.8) Fasting insulin (see section 8.2.4) C-peptide (see section 8.2.5) Reminders: Make an appointment for Visit 19a, 60 weeks ± 5 days after randomisation and for Visit 20a, 64 weeks ± 5 days after randomisation Visit 19a and 20a, Re-randomised treatment visits Visit 19a will take place 60 weeks ± 5 days after randomisation and Visit 20a will take place 64 weeks ± 5 days after randomisation. These visits are only applicable for subjects continuing in the re-randomised treatment period. The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Mental health questionnaires (see section ) Diet and physical activity counselling based on the 3-day food diary at visit 20a (see section 5.3.2) Diet and physical activity counselling at visit 19a (see section 5.3.2) Subject s dietary compliance and average daily level of physical activity (only applicable for Visit 20a) (see section 8.4.5) Dispense 3-day food diary at Visit 19a (see section ) Interact with the IV/WRS to: Complete dispensing session Complete withdrawal session (if applicable) Reminders: Make an appointment for Visit 20a 64 weeks ± 5 days after randomisation

173 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 54 of 138 Make an appointment for the End of the re-randomised Treatment visit (Visit 21a), 68 weeks ± 3 days after randomisation Remind the subjects to attend Visit 21a fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) Remind the subject to bring a sample of their morning urine at Visit 21a Visit 21a, End of re-randomised Treatment visit / premature discontinuation Visit 21a will take place 68 weeks ± 3 days after randomisation. This visit is only applicable for subjects continuing in the re-randomised treatment period. Subjects being prematurely withdrawn from the re-randomised treatment period should preferably have all procedures for the End of the re-randomised Treatment visit (Visit 21a) performed. The primary reason (adverse event, non-compliance with protocol or other) for discontinuation must be specified in the ecrf and IV/WRS. Even if the subject is not able to attend, the End of Trial Form must be completed. The following will be performed and/or recorded in the ecrf: Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) PRO questionnaires (see section ) Physical examination (see section 8.3.1) ECG (see section 8.3.4) Adverse events since last visit (see section 8.3.6) Counselling on future weight management (see section 5.3.2) Mental health questionnaire (see section ) Complete the End of Trial Form Urine sampling for measurement of: Urinary Albumin-to-Creatinine ratio (see section ) Interact with the IV/WRS to: Make completion call for subjects completing the trial Interact with the Drug Accountability Module Blood sampling for measurement of: HbA 1c (see section 8.2.2)

174 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 55 of 138 FPG (see section 8.2.3) Fasting insulin (see section 8.2.4) C-peptide (see section 8.2.5) OGTT (see section 8.2.6) Cardiovascular biomarkers (see section 8.2.9) Lipids (see section ) For females of childbearing potential: serum pregnancy test (see section 8.3.5) Haematology and biochemistry (see section 8.3.7) Liraglutide antibodies (only applicable for subjects who discontinue the main trial after Visit 17 (see section 8.3.8) Reminders: Make an appointment for the Visit 22a ( follow up visit) 70 weeks ± 3 days after randomisation Remind the subjects to attend Visit 22a fasting Visit 22a, follow-up Visit 22a will take place 70 weeks ± 3 days after randomisation. The following will be performed and/or recorded in the ecrf: Concomitant medication, changes since last visit, if any Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) Adverse events, since last visit (see section 8.3.6) Mental health questionnaires (see section ) Blood sampling for measurement of: Liraglutide antibodies (see section 8.3.8) Fasting insulin (see section 8.2.4) and fasting plasma glucose (see section 8.2.3) Visit 18b Visit 18b will take place 60 weeks ± 5 days after randomisation. The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5)

175 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 56 of 138 Adverse events since last visit (see section 8.3.6) Mental health questionnaires (see section ) Dispense 3-day food diary (see section ) Diet and physical activity counselling (see section 5.3.2) Interact with the IV/WRS to: Complete dispensing session Complete withdrawal session (if applicable) Interact with Drug Accountability Module Reminders: Make an appointment for the subsequent visit Remind the subject to: attend visit 19b fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at all visits keep a 3-day food diary between Visit 18b and 19b (see section 5.3.2) return all empty, partly used and unused trial product at all visits for drug accountability Visits 19b, 22b, 25b, 28b, 31b, 34b, 37b and 38b (delayed-onset-of-type-2-diabetes part of the trial) The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) ECG only at Visit 25b, 31b and 37b (see section 8.3.4) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Mental health questionnaires (see section ) Dispense 3-day food diary at Visit 19b, 22b, 28b, 34b (see section ) Diet and physical activity counselling based on the 3-day food diary at Visit 19b, 25b, 31b, and Visit 37b (see section 5.3.2) Diet and physical activity counselling at Visit 22b, 28b, 34b, and Visit 38b (see section 5.3.2) Subject s dietary compliance average daily level of physical activity at Visit 19b, 25b, 31b, and Visit 37b (see section 8.4.5)

176 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 57 of 138 Interact with the IV/WRS to: Complete dispensing session Complete withdrawal session (if applicable) Interact with Drug Accountability Module Blood sampling for measurement of: HbA 1c only at Visit 38b (see section 8.2.2) FPG (see section 8.2.3) Fasting insulin only at Visit 38b (see section 8.2.4) C-peptide only at Visit 38b (see section 8.2.5) Haematology and biochemistry (see section 8.3.7) Reminders: Make an appointment for the subsequent visit Remind the subject to: attend visit 20b, 23b, 26b, 29b, 32b, 35b, and 38b fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at all visits keep a 3-day food diary between Visit 22b and 23b and Visit 37b and 38b (the diary will be used for diet counselling at Visit 38b) (see section 5.3.2) return all empty, partly used and unused trial product at all visits for drug accountability Remind the subject to bring a sample of their morning urine at Visits 23b, 29b and 35b Visits 20b, 26b and 32b (delayed-onset-of type-2-diabetes part of the trial) The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Mental health questionnaire (see section ) Dispense 3-day food diary at Visit 20b, 26b and Visit 32b (see section ) Diet and physical activity counselling based on the 3-day food diary (see section 5.3.2) Interact with the IV/WRS to: Complete dispensing session

177 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 58 of 138 Complete withdrawal session (if applicable) Interact with the Drug Accountability module Blood sampling for measurement of: HbA 1c (see section 8.2.2) FPG (see section 8.2.3) Fasting insulin (see section 8.2.4) C-peptide (see section 8.2.5) Reminders: Make an appointment for the subsequent visit Remind the subject to: avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at all visits return all empty, partly used and unused trial product at all visits for drug accountability Visits 21b, 24b, 27b, 30b, 33b, 36b, 39b, 40b and 42b (delayed-onset-of-type-2- diabetes part of the trial) The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) PRO questionnaires, only applicable for Visit 30b and 36b (see section ) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Mental health questionnaire (see section ) Dispense 3-day food diary at 24b, 30b, 36b, 40b, and Visit 42b (see section ) Diet and physical activity counselling based on the 3-day food diary ay Visit 21b, 27b, 33b, 39b, and Visit 41b (see section 5.3.2) Diet and physical activity counselling at Visit 24b, 30b, 36b, 40b, and Visit 42b (see section 5.3.2) Subject s dietary compliance and average daily level of physical activity at Visit 21b, 27b, 33b and Visit 39b (see section 8.4.5) Interact with the IV/WRS to: Complete dispensing session Complete withdrawal session (if applicable)

178 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 59 of 138 Interact with Drug Accountability Module Reminders: Make an appointment for the subsequent visit Remind the subject to: attend visits fasting 22b, 25b, 28b, 31b, 34b, 37b, 41b and 43b (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at all visits Remind the subject to bring a sample of their morning urine at Visit 43b return all empty, partly used and unused trial product at all visits for drug accountability Visits 23b, 29b and 35b (delayed-onset-of-type-2-diabetes part of the trial) The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) PRO questionnaires, only applicable for Visit 23b (see section ) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Mental health questionnaires (see section ) Diet and physical activity counselling based on the 3-day food diary (see section 5.3.2) Subject s dietary compliance average daily level of physical activity (see section 8.4.5) Interact with the IV/WRS to: Complete dispensing session Complete withdrawal session (if applicable) Interact with Drug Accountability Module Urine sampling for measurement of: Urinary Albumin-to-Creatinine ratio (see section ) Blood sampling for measurement of: HbA 1c (see section 8.2.2) FPG (see section 8.2.3) Fasting insulin (see section 8.2.4) C-peptide (see section 8.2.5)

179 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 60 of 138 OGTT (see section 8.2.6) Cardiovascular biomarkers (see section 8.2.9) Lipids (see section ) Reminders: Make an appointment for the subsequent visit Remind the subject to: avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at all visits return all empty, partly used and unused trial product at all visits for drug accountability Visit 41b (delayed-onset-of-type-2-diabetes part of the trial) The following will be performed and/or recorded in the ecrf: Withdrawal criteria (see section 6.6) Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) For females of childbearing potential: urine pregnancy test, if applicable (see section 8.3.5) Adverse events since last visit (see section 8.3.6) Mental health questionnaires (see section ) Diet and physical activity counselling based on the 3-day food diary (see section 5.3.2) Subject s dietary compliance and average daily level of physical activity (see section 8.4.5) Interact with the IV/WRS to: Complete dispensing session Complete withdrawal session (if applicable) Interact with the Drug Accountability module Blood sampling for measurement of: HbA 1c (see section 8.2.2) FPG (see section 8.2.3) Fasting insulin (see section 8.2.4) C-peptide (see section 8.2.5) OGTT (see section 8.2.6) Cardiovascular biomarkers (see section 8.2.9) Lipids (see section ) Haematology and biochemistry (see section 8.3.7) Reminders:

180 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 61 of 138 Make an appointment for the subsequent visit Remind the subject to: attend the End of 160 week randomised Treatment visit fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) avoid caffeine, smoking and physical activity within 30 minutes prior to the blood pressure measurement at all visits return all empty, partly used and unused trial product at all visits for drug accountability Visit 43b, End of 160 weeks randomised Treatment visit /premature discontinuation Visit 43b will take place 160 weeks ± 3 days after randomisation. Subjects being prematurely withdrawn from the trial should preferably have all procedures for the End of 160 week randomised Treatment visit performed. The primary reason (adverse event, noncompliance with protocol or other [see section 6.6]) for discontinuation must be specified in the ecrf and IV/WRS. Even if the subject is not able to attend, the End of Trial Form must be completed. The following will be performed and/or recorded in the ecrf: Concomitant medication, changes since last visit, if any Subject compliance (see section 8.5) Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) PRO questionnaire (see section ) Physical examination (see section 8.3.1) ECG (see section 8.3.4) Adverse events since last visit (see section 8.3.6) Mental health questionnaires (see section ) Counselling on future weight management (see section 5.3.2) Subject s dietary compliance and average daily level of physical activity (see section 8.4.5) Complete the End of Trial Form Interact with the Drug Accountability module Urine sampling for measurement of: Urinary Albumin-to-Creatinine ratio (see section ) Interact with the IV/WRS to: Make completion call for subjects completing the trial Blood sampling for measurement of:

181 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 62 of 138 HbA 1c (see section 8.2.2) FPG (see section 8.2.3) Fasting insulin (see section 8.2.4) C-peptide (see section 8.2.5) OGTT (see section 8.2.6) Cardiovascular biomarkers (see section 8.2.9) Lipids (see section ) For females of childbearing potential: serum pregnancy test (see section 8.3.5) Haematology and biochemistry (see section 8.3.7) Liraglutide antibodies (only applicable for subjects who discontinue the trial prematurely)(see section 8.3.8) Reminders: Make an appointment for Visit 44b, 162 weeks ± 3 days, after randomisation (Visit 3) Remind the subjects to attend Visit 44b fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) Visit 44b, Follow-up visit Visit 44b will take place 162 weeks ± 3 days after randomisation. The following will be performed and/or recorded in the ecrf: Concomitant medication, changes since last visit, if any Body measurements (see section 8.2.1) Vital signs (see section 8.2.8) Adverse events since last visit (see section 8.3.6) Mental health questionnaires (see section ) Blood sampling for measurement of: Liraglutide antibodies for subjects completing the trial (see section 8.3.8) Fasting insulin and fasting plasma glucose 8.2 Assessments for efficacy Any abnormal, clinically significant result identified at Screening visit 1 and Screening visit 2 will be recorded as concomitant illnesses. The laboratory analyses for efficacy and safety will be outsourced to a Central Laboratory unless otherwise specified. Descriptions of assay methods, instrumentation and procedures for obtaining samples, handling and storage of samples will be described in a trial specific laboratory manual provided by the Central Laboratory.

182 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 63 of 138 Blood samples that are not drawn on the day of the scheduled visit may be drawn on another day at an additional visit (e.g. if the subject is not fasting), preferably within the visit window. Samples will be coded with the intention that the subject s identity will remain encrypted but information such as age, sex, race, health information and response to liraglutide will be correlated. The samples will only be used in relation to the present trial. and its representatives and/or regulatory authorities, may have access to this information. However, the subject s identity will not be revealed. Laboratory analysis results will be sent to the Investigator on an ongoing basis except for the results of liraglutide antibody analyses and liraglutide plasma concentration. All laboratory reports must be dated and signed by the Investigator on the day of evaluation. If a result is outside the normal range, the Investigator has to judge whether the abnormality is clinically significant or not (see section 12.1). The signed laboratory report is retained at the Investigator site as source documentation Body measurements Body measurements include weight, height and waist circumference Weight and height Weight will be recorded to the nearest 0.1 kg. Weight should be measured at all visits to the clinic except for Screening visit 2 using calibrated scales. At Visits 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 17x, 18a, 21a, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, 43b and 44b weight should be measured in the fasting state. If the weight is not measured in a fasting state within the visit window, the subject should be called in for an Unscheduled Visit. The same pair of scales should preferably be used throughout the trial. Weight should be measured with an empty bladder, without shoes and only wearing light clothing. Weight measured at Screening visit 1 will only be used for the Investigator s calculation of BMI, whereas weight measured at Visit 3 will be used as baseline for assessment of change in body weight. Height without shoes will be recorded at Screening visit 1. BMI will be calculated as follows: BMI = weight (kg)/height 2 (m 2 ) Waist circumference Waist circumference will be determined at all visits except for Screening visit 2, according to the procedure specified below. The waist circumference will be determined in a fasting state at Visit 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 18a, 21a, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, 43b and 44b. Three consecutive measurements will be performed at each visit and will be recorded in the ecrf. The waist circumference will be measured in the horizontal plane to the nearest 0.5 cm using a non-stretchable measuring tape.

183 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 64 of 138 Subjects should be measured in the standing position with an empty bladder and wearing only light clothing. The measuring tape should lie flat against the skin without compressing the soft tissue and need to be removed between each measurement. The observer should be sitting in front of the subject during the measurement. The subject should be standing with arms at their side and feet together. Subjects should be asked to breathe normally and the measurement should be performed when the subject is breathing out gently. The waist circumference is defined as the abdominal circumference located midway between the lower rib margin and the iliac crest HbA 1c Samples will be drawn at Screening visit 1, Visit 3, 5, 8, 11, 14, 17, 21a, 20b, 23b, 26b, 29b, 32b, 35b, 38b, 41b, 43b and 44b for measurement of HbA 1c. HbA 1c measured at Screening visit 1 will be used to assess compliance with exclusion criteria no. 2 (see section 6.4). The assay method used will be a National Glycohemoglobin Standardization Program certified method Fasting plasma glucose Samples for determination of FPG will be drawn at Screening visit 2, Visit 3, 5, 6c, 8, 9, 11, 14, 16, 17, 18a, 21a, 22a, 20b, 23b, 26b, 29b, 32b, 35b, 38b, 41b, 43b and 44b. FPG measured at Screening visit 2 will be used to assess compliance with exclusion criteria no. 4 (see section 6.4). In the event of FPG 126 mg/dl (7.0 mmol/l) at any time during the trial a repeated measurement should be taken within four weeks to confirm/exclude diagnosis of diabetes. Subjects developing diabetes during the trial should withhold their background diabetes medication and trial product on 2, 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 18a, 21a, 22a, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, 43b and 44b until blood sampling has been done. At all other visits diabetes background medication and trial product should be taken as usual during the conduct of the trial Fasting insulin Blood samples for determination of fasting insulin will be taken at Visit 3, 5, 8, 11, 14, 17, 18a, 21a, 22a, 20b, 23b, 26b, 29b, 32b, 35b, 38b, 41b, 43b and 44b C-peptide Blood samples for determination of fasting C-peptide will be taken at Visit 3, 5, 8, 11, 14, 17, 21a, 20b, 23b, 26b, 29b, 32b, 35b, 38b, 41b, and 43b. A measure of beta-cell function (and insulin resistance) will be derived from FPG, fasting insulin and C-peptide data using the HOMA model(35)

184 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 65 of Oral Glucose Tolerance Test (OGTT) A 75 gram OGTT will be performed at Screening visit 2, Visit 11, 17, 21a, 23b, 29b, 35b, 41b 43b. Standardised conditions will minimise lack of reproducibility. The OGTT must be performed after an overnight fast (minimum 8 hours). Subjects should be on an adequate carbohydrate intake before the test, should not be physically active during the test and must not smoke. The 120-min sample must be taken within 5 minutes of 120 minutes. Plasma should be immediately separated, or the sample should be collected into a container with glycolytic inhibitors and placed on ice-water until separated prior to analysis After collection of the baseline blood sample (0h sample) the subject should be given a 75 gram glucose-equivalent oral glucose challenge (to be supplied to the Investigator by A/S via Central lab) over the course of 5 minutes. Timing of the test is from the beginning of the drink. Blood samples will be taken at 0h, 10 min, 20 min, 30 min, 60 min, 90 min and 120 min to evaluate glucose levels. Insulin and C-peptide concentrations will be assessed as well for estimation of betacell function. The OGTT at Screening visit 2 will be used to assess a subject s pre-diabetes status and thereby eligibility for the delayed-onset-of-type-2-diabetes part of the trial. In addition, the fasting and 2- hour post-change plasma glucose values will serve to rule out presence of type 2 diabetes at baseline. If FPG 126 mg/dl (7.0 mmol/l) and/or 2-hour post-challenge glucose is 200 mg/dl (11.1 mmol/l) the subject is a screening failure. Throughout the duration of the trial, any change in status based on OGTT-values will be recorded (normal, pre-diabetes, or diabetes). Pre-diabetes or diabetes may also be diagnosed based on HbA1c value(pre-diabetes: HbA 1c % both inclusive, diabetes HbA 1c 6.5 %). Subjects who have pre-diabetes according to their fasting or 2-hour value during an OGTT will be further classified according to the following pre-diabetes categories: isolated IFG, isolated IGT, or combined IFG/IGT (please refer to section for details). Within each of these categories, a subject may in addition fulfil the HbA 1c criteria for pre-diabetes but this information should not be entered in the ecrf. Similar to diagnosis of diabetes by FPG value (see section 8.2.3), in the event of 2-hour postchallenge plasma glucose 200 mg/dl (11.1 mmol/l) at any time during the trial a repeated measurement should be taken within four weeks to confirm diagnosis of diabetes. The date of diagnosis of diabetes will be recorded in the ecrf (date of the first assessment out of range). No subsequent OGTTs will be performed for subjects diagnosed with diabetes.

185 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 66 of 138 In the event that subjects are diagnosed with diabetes they should receive the best standard of care at the discretion of the Investigator. Glycaemic goals should follow recommendations laid out in Standards of Medical Care in Diabetes (36), i.e. HbA 1c <7%. If Investigator determines that insulin, GLP-1 receptor agonist (e.g., Byetta or Victoza ), or DPP-IV inhibitor, is the best treatment option, the subject must be withdrawn (see section 6.6, withdrawal criteria no. 4). Subjects who are diagnosed with diabetes should monitor their blood glucose at home with a glucose meter. The treatment target should be to maintain HbA 1c levels below 7%. If self-measured FPG falls below 6.1 mmol/l (117 mg/dl), or meets the criteria for glycaemic rescue i.e. from baseline to 6 weeks, increases above 15 mmol/l (270 mg/dl) or from 7 weeks to 12 weeks, increases above 13.3 mmol/l (240 mg/dl) or from 13 weeks to 160 weeks, increases above 11.1 mmol/l (200 mg/dl) on three consecutive days/occasions, the subject should contact the Investigator Pre-diabetes status and classification A subject has pre-diabetes if the following applies: FPG 100 mg/dl (5.6 mmol/l) and 125 mg/dl (6.9 mmol/l) and/or 2 hr post-challenge (OGTT) plasma glucose 140 mg/dl (7.8 mmol/l) and 199 mg/dl (11.0 mmol/l) and/or HbA 1c % both inclusive (37) A subject will be further classified according to their fasting and postprandial glycaemia (irrespective of HbA1c value), as follows: Isolated Impaired Fasting Glucose (iifg): FPG 100 mg/dl (5.6 mmol/l) to 125 mg/dl (6.9 mmol/l), both inclusive, and 2-hour post-challenge plasma glucose < 140 mg/dl (7.8 mmol/l) Isolated Impaired Glucose Tolerance (iigt): 2-hour post-challenge plasma glucose 140 mg/dl (7.8 mmol/l) to 199 mg/dl (11.0 mmol/l), both inclusive, and FPG < 100 mg/dl (5.6 mmol/l) Combined IFG/IGT (IFG/IGT): FPG 100 mg/dl (5.6 mmol/l) to 125 mg/dl (6.9 mmol/l), both inclusive, and 2-hour post-challenge plasma glucose 140 mg/dl (7.8 mmol/l) to 199 mg/dl (11.0 mmol/l), both inclusive A subject who has pre-diabetes at baseline is eligible for continuing in the delayed-onset-of-type-2- diabetes part of the trial. For subjects diagnosed with diabetes please refer to section

186 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 67 of Vital signs Blood pressure Systolic and diastolic blood pressure will be measured preferably in sitting position at all visits to the clinic except for Screening visit 2. Measurement will be done according to the procedure described in Appendix D. However, re-measurement of blood pressure at Screening visit 2 is allowed if white coat syndrome is suspected at Screening visit 1. Caffeine, smoking and physical activity should be avoided within 30 minutes prior to the blood pressure measurement at all visits to the clinic Pulse Pulse will be recorded after resting for five minutes in a sitting position at all visits to the clinic except for Screening visit Cardiovascular biomarkers Blood samples for determination of selected cardiovascular biomarkers will be drawn at Visit 3, 11, 17, 21a, 23b, 29b, 35b, 41b and 43b. hscrp Adiponectin Fibrinogen PAI Lipids Blood samples for determination of fasting levels of TC, LDL-cholesterol, HDL-cholesterol, VLDL-cholesterol, TG and FFA will be drawn at Screening visit 2, Visit 3, 11, 17, 21a, 23b, 29b, 35b, 41b and 43b. It is important that the subject has been fasting (i.e., at least eight hours overnight fast without food and/or drink intake, except for water) prior to blood sampling for lipids Patient reported outcomes (PRO) questionnaires PRO recording will be performed in UK, Germany, Russia, Canada, US, Italy, Spain, France, Brazil, Austria, Mexico, Australia, Ireland and the Netherlands.

187 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 68 of 138 Patient Reported Outcome (PRO) will be assessed at Visit 3, 11, 17 and 21a for the 56 week main trial including the re-randomised treatment period and at Visit 23b, 30b, 36b and 43b for the delayed-onset-of-type-2-diabetes part of the trial. Subjects discontinuing the trial prematurely should complete the questionnaire Visits Visit 17, 21a or Visit 43b, depending on when they discontinue the trial. PRO will be assessed by the questionnaires IWQOL-Lite(38), SF-36(39), (40), (41) and TRIm - Weight questionnaire(42).the questionnaires are validated and will be translated into local language before being handed out to the subjects participating in the trial. The questionnaires should preferably be completed by the subject before any trial procedures at the relevant visit are performed. Subjects should be given the opportunity to complete the questionnaires by themselves without interruption. The Investigator or his/her delegate must review Patient Reported Outcome(s) for completeness and adverse events immediately following administration Impact of Weight on Quality Of Life questionnaire (IWQoL) Lite version IWQOL-Lite provides an individual perception of impact of weight on physical function, selfesteem, sexual life, public distress, and work. The IWQOL-Lite contains 31 items and it takes approximately 10 minutes to complete SF-36 SF-36 (version 2) measures the individual overall health related quality of life on 8 domains; Physical functioning, Role functioning, Bodily pain, General health, Vitality, Social functioning, Role emotional and mental health. SF-36 contains 36 items and it takes approximately 10 minutes to complete Treatment Related Impact Measure (TRIm) - Weight TRIm-Weight(43) evaluates subjects overall impact and treatment satisfaction divided in 5 domains: efficacy, impact, side effects, psychological and convenience. TRIm-Weight takes approximately 10 minutes to complete Positive predictive value of Finnish Type 2 diabetes (FINDRISC) The FINDRISC(44) score is a composite of data collected in the evaluation of obese subjects suspected for being at risk of development of diabetes. Attributes include, BMI, waist circumference, age, family history of diabetes, history of high glucose, use of antihypertensive medication, daily physical activity level and daily intake of fruit and vegetables. The positive predictive value of Finnish Type 2 Diabetes Risk Score (FINDRISC) at baseline (Visit 3) on development of type 2 diabetes will be assessed. The predictive value of having a FINDRISC score of > 15 will be evaluated at trial completion. Questions used to assess the FINDRISC score will be implemented in the ecrf.

188 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 69 of Urinary Albumin-to-Creatinine ratio A urine sample will be taken for assessment of Urinary Albumin-to-Creatinine ratio. For the subjects in the 56 week main trial including the re-randomised treatment period the urine sample will be collected at visits 3, 11, 17, 18a and Visit 21a For the pre-diabetic subjects a urine sample will be collect at visits 3, 11, 17, 23b, 29b, 35b, 43b. 8.3 Assessments for safety Physical examination Physical examination will be performed at Screening visit 1,Visit 17, Visit 21a and Visit 43b according to local procedure. Physical examination should include the cardiovascular system, respiratory system, abdomen, central and peripheral nervous system, musculo-skeletal system and the thyroid gland. Any abnormal, clinical significant findings at Screening visit 1 must be recorded as a concomitant illness (see section 11). Any changes in subsequent visits as compared to Screening visit 1 which fulfil the criteria for an AE must be recorded as an AE (see section 12) Hypoglycaemic episodes For subjects developing diabetes: Subjects developing diabetes will be supplied with a glucose meter and a diabetes diary for recording of hypoglycaemia. Plasma glucose should always be measured when there is the suspicion of a hypoglycaemic episode. All plasma glucose (PG) values 3.9 mmol/l (70 mg/dl), as well as values > 3.9 mmol/l (70 mg/dl) when hypoglycaemic symptoms have occurred, should be recorded by the subjects in the diaries. Hypoglycaemic episodes will be recorded by the subject in his/her diary throughout the trial and must be transcribed into the ecrf by the Investigator at each site visit throughout the trial. The recording should include: date of hypoglycaemic episode time of hypoglycaemic episode time of last trial drug prior to episode type of last trial drug prior to episode date and time of diabetes background treatment prior to the episode time of last main meal prior to episode whether the episode was symptomatic whether the episode was in relation to exercise

189 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 70 of 138 whether the subject was able to treat him-herself (if not answered, the Investigator must provide an explanation in the ecrf) if the subject was not able to treat him-herself, whether he-she recovered with oral administration of carbohydrates the plasma glucose level before treating the episode (if available) The answer to the question: Was subject able to treat him/herself? should be answered No if oral carbohydrates, glucagon or intravenous glucose had to be administered to the subject by another person because of severe central nervous system (CNS) dysfunction associated with the hypoglycaemic episode. Oral carbohydrates should not be given in case the subject is unconscious. A hypoglycaemic episode will be defined as treatment emergent if the onset of the episode is on or after the first day of randomised treatment, and no later than 14 days after the last day of randomised treatment. Hypoglycaemic episodes will be defined as nocturnal if the time of onset is between 00:01 and (both included). A hypoglycaemic episode form and an AE form must be filled in for all hypoglycaemic episodes. If the hypoglycaemic episode fulfils the criteria for a serious AE and/or a MESI, a hypoglycaemic episode form, an AE form and a safety information form must be filled in. Severe hypoglycaemic episodes will be recorded as MESIs. Hypoglycaemic episodes will be summarised based on the ADA classification (45), and also according to an additional definition. For subjects without diabetes: Hypoglycaemic episodes during the trial should be recorded on an AE page. A hypoglycaemic episode will be defined as treatment emergent if the onset of the episode is on or after the first day of randomised treatment, and no later than 14 days after the last day of randomised treatment. Hypoglycaemic episodes will be defined as nocturnal if the time of onset is between 00:01 and (both included). If the hypoglycaemic episode fulfils the criteria for a serious AE and/or a MESI, a hypoglycaemic episode form, an AE form and a safety information form must be filled in. Severe hypoglycaemic episodes will be recorded as MESIs. ADA hypoglycaemia classification According to the ADA the definition of a hypoglycaemic episode Figure 8 1 is categorised as: Severe hypoglycaemia: An episode requiring assistance of another person to actively administer carbohydrate, glucagon, or other resuscitative actions.

190 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 71 of 138 Documented symptomatic hypoglycaemia: An episode during which typical symptoms of hypoglycaemia are accompanied by a measured plasma glucose concentration 3.9 mmol/l (70 mg/dl). Asymptomatic hypoglycaemia: An episode not accompanied by typical symptoms of hypoglycamia, but with a measured plasma glucose concentration 3.9 mmol/l (70 mg/dl). Probable symptomatic hypoglycaemia: An episode during which symptoms of hypoglycaemia are not accompanied by a plasma glucose determination (but that was presumably caused by a plasma glucose concentration 3.9 mmol/l (70 mg/dl)). Relative hypoglycaemia: An episode during which the person with diabetes reports any of the typical symptoms of hypoglycaemia, and interprets those as indicative of hypoglycaemia, but with a measured plasma glucose concentration > 3.9 mmol/l (70 mg/dl). No Severe hypoglycaemia Hypoglycaemia Subject able to treat it him/herself? No PG 3.9 mmol/l Asymptomatic (70 mg/dl) hypoglycaemia Yes Symptoms? Yes PG 3.9 mmol/l Documented (70 mg/dl) symptomatic hypoglycaemia PG>3.9 mmol/l Relative (70 mg/dl) hypoglycaemia No measurement Probable symptomatic hypoglycaemia Figure 8 1 Clasification of hypoglycaemia Additional definition In normal physiology, hypoglycaemia symptoms occur at a blood glucose level of approximately < 2.8 mmol/l (50 mg/dl)/plasma glucose level < 3.1 mmol/l (56 mg/dl). Therefore, Novo Nordisk has used this cut-off value to define minor hypoglycaemia.

191 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 72 of 138 Minor hypoglycaemic episode is defined as: An episode with symptoms consistent with hypoglycaemia with confirmation by plasma glucose < 3.1 mmol/l (56 mg/dl), or full blood glucose < 2.8 mmol/l (50 mg/dl) and which is handled by the subject himself/herself Or any asymptomatic plasma glucose value < 3.1 mmol/l (56 mg/dl) or full blood glucose value < 2.8 mmol/l (50 mg/dl) Minor hypoglycaemic episodes will be summarised according to this definition, and can be subject to additional analysis Self-monitored plasma glucose If a subject develops diabetes self-monitoring of plasma glucose will be performed using a glucose meter, supplied by. A sufficient amount of test strips, lancets and calibration solutions and a diabetes diary will be supplied together with the glucose meter. The subjects will be instructed in the use of the device, and will also be provided with written instructions. The Investigator must ensure that the subject uses the glucose meter correctly. The instructions should include: Advice on regular calibration according to the manufacturer s instructions Instruction in performing self-measurement of plasma glucose concentration The glucose meters use test strips calibrated to plasma values. This means that all capillary blood glucose measurements performed with these glucose meters are automatically converted into plasma equivalent glucose values, which will be shown in the display. These values are to be used for evaluation purposes. Subjects who develop diabetes are encouraged to measure their FPG concentrations on a regular basis at the discretion of the Investigator. Also subjects should measure their plasma glucose at least every time the subject has symptoms of hypoglycaemia or hyperglycaemia. The Investigator may ask the subject to perform additional plasma self-measurements if needed for any safety reason. If self-measured FPG falls below 6.1 mmol/l (117 mg/dl), or from baseline to 6 weeks, increases above 15 mmol/l (270 mg/dl) or from 7 weeks to 12 weeks, increases above 13.3 mmol/l (240 mg/dl) or from 13 weeks to 160 weeks, increases above 11.1 mmol/l (200 mg/dl) on three consecutive days/occasions, the subject should contact the Investigator. The outcome of the contact will be recorded on an Unscheduled Visit Form in the ecrf. Subjects developing diabetes during the trial should withhold their background diabetes medication and trial product on 2, 3, 4, 5, 6c, 8, 9, 11, 14, 16, 17, 18a, 21a, 22a, 19b, 20b, 22b, 23b, 25b, 26b, 28b, 29b, 31b, 32b, 34b, 35b, 37b, 38b, 41b, 43b and 44b until blood sampling has been done. At all

192 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 73 of 138 other visits diabetes background medication and trial product should be taken as usual during the conduct of the trial ECG A 12 lead ECG will be performed at Screening visit 1, Visits 11, 17, 21a, 25b, 31b, 37b and Visit 43b. An ECG taken within the period starting at Screening visit 1 and ending at the date of Visit 3 is acceptable. The ECG will be interpreted, signed and dated by the Investigator. The interpretation must follow the categories: Normal Abnormal, not clinically significant Abnormal, clinically significant In case of an abnormal ECG, the Investigator must evaluate if the abnormal ECG fulfils the criteria of an AE or SAE, and if so, should be reported as an AE or SAE. All ECGs will in addition undergo central assessment. Sites will be informed of the central ECG evaluation in case this evaluation reveals an abnormal ECG reading. If the abnormality represent an unreported SAE or MESI, such SAE or MESI must be reported by the Investigator Pregnancy test Females of childbearing potential will have a serum pregnancy test (hcg) performed in connection with Screening visit 1, Visit 17 and the End of Treatment visit (Visit 21a or 43b). Blood drawn for biochemistry will be used (see section 8.3.7). Urine pregnancy tests will be performed for females of childbearing potential at any time during the trial if a menstrual period is missed or as required by local law. Urine pregnancy kits will be supplied by the central laboratory. The test will be performed at the site. Pregnancy testing will not be required for women who have undergone a hysterectomy or bilateral tubal ligation, or for women above the age of 50, who have been without menstrual period for at least 1 year Adverse events (AEs) AEs will be recorded at each visit after Screening visit 1 according to procedures described in section 12.

193 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 74 of Haematology and biochemistry Samples will be drawn at Screening visit 1, Visit 3, 5, 8, 11, 14, 17, 18a, 21a, 19b, 22b, 25b, 28b, 31b, 34b, 37b, 38b, 41b, and 43b. Haematology Haemoglobin Haematocrit Thrombocytes Erythrocytes Leucocytes Differential count Eosinophils Neutrophils Basophils Monocytes Lymphocytes Biochemistry Creatinine CPK Urea Albumin Bilirubin (total) ALAT ASAT ALP Amylase Lipase Sodium Potassium Calcium, total Calcitonin TSH Tryptase, if hypersensitivity is suspected For subjects with calcitonin 20 ng/l (from Visit 3 and onwards) a repeat measurement of calcitonin must be performed preferably within four weeks and if confirmed, the event elevated

194 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 75 of 138 calcitonin should be reported as a MESI (i.e. any confirmed episode of calcitonin concentration value 20 ng/l, if not already reported as a MESI, see section ) and Appendix J Liraglutide antibodies Blood samples to be analysed for serum antibodies against liraglutide will be drawn at Visits 3,18a and Visit 22a for subjects completing the 56 week main trial including the re-randomised treatment period. For subjects participating in the delayed-onset-of-type-2-diabetes part of the trial (164 weeks in total) blood samples will be drawn at Visit 3 and 44b. Blood samples for subjects discontinuing the trial prematurely will be drawn at Visit 17, 21a or 43b depending on when the subject is discontinuing. Antibody positive samples will be further characterised for neutralising effect and cross-reactivity. It is not required that subjects are fasting for antibody sampling. A special lab will be responsible for the analysis of liraglutide antibodies. Blood samples are collected, treated and shipped according to the description in the laboratory manual supplied by the central laboratory Suspicion of acute hypersensitivity (allergic reaction) to trial product If acute hypersensitivity to trial product is suspected, local testing for blood tryptase concentration (total and/or mature tryptase) is recommended. If trial product is discontinued as a consequence of suspicion of acute hypersensitivity, blood sampling for central assessment of iraglutide antibodies and IgE-isotype of liraglutide antibodies should be conducted, at least 14 days after trial product discontinuation. Tryptase concentration (if measured) as well as results of liraglutide antibody and IgE-isotype liraglutide antibodies will be sent to and will be included in the final MESI report Suspicion of immune-complex disease If immune-complex disease is suspected, blood sampling for central assessment of complement levels (C3 and C4) should be conducted. Results should be included when reporting a MESI Mental health questionnaires In recent years there has been increasing attention on the influence of seemingly innocuous changes in body chemistry on mental health, and even drugs thought to be largely free of mental effects are now seen as having the potential to affect mental health. Makers of drugs to treat obesity, urinary incontinence, epilepsy, smoking cessation, depression and many other conditions are being asked by regulators to put comprehensive mental health assessments into clinical trials. Therefore, even though liraglutide has not been associated with depression or suicidality the two mental health questionnaires; C-SSRS and PHQ-9 have been included in the present trial.

195 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 76 of 138 Mental health will be assessed by the C-SSRS and the PHQ-9 in all subjects at all visits to the clinic except for Screening visit 2 and Visit 17x In addition, mental health assessed by the PHQ-9 will be done at Screening Visit 1 to exclude subjects with a major depression disorder (PHQ-9 score 15). Subjects discontinuing the trial prematurely before Visit 17 will complete the mental health questionnaires at the premature discontinuation visit (Visit 17) whereas subjects discontinuing the trial prematurely after Visit 17 but before Visit 21a will complete the mental health questionnaires at the premature discontinuation visit (Visit 21a). If subjects randomised to the delayed-onset-oftype-2-diabetes part of the trial discontinue the trial prematurely after Visit 17 but before Visit 43b they will complete the questionnaires at the premature discontinuation visit (Visit 43b). The questionnaires should be completed by the subject after adverse event reporting has been completed. The subject should be given the opportunity to complete the PHQ-9 questionnaire by themselves without interruption. The Investigator or his/her delegate must review the PHQ-9 and C-SSRS questionnaire for completeness and adverse event immediately following administration C-SSRS The C-SSRS is a detailed questionnaire assessing both suicidal behaviour and suicidal ideation. The questionnaire will be administered as an interview by the Investigator(46).The Investigator must assess the C-SSRS score at Screening visit 1 and Visit 3 (randomisation) (see section 6.4) to exclude subjects with any suicidal ideation of type 4 (active suicidal ideation with some intent to act, without specific plan) or type 5 (active suicidal ideation with specific plan and intent). The questionnaire takes approximately 10 minutes to complete PHQ-9 The PHQ-9 is a 9-item self-reported depression screening tool. The questionnaire takes approximately 10 minutes to complete. The Investigator must assess PHQ-9 score at Screening visit 1 and Visit 3 (randomisation) to exclude subjects with a PHQ-9 score 15 (see section 6.4)(47) Thyroidectomy pathology slides In case a subjects undergoes a thyroidectomy (partial or total) for any reason during the duration of the trial, pathology slides of the thyroid tissue will be centrally reviewed in addition to the routine examination at the site level. Both the site pathology report and the central pathology report will be reviewed by an external independent Event Adjudication Committee (EAC). A set of pathology slides routinely made after thyroidectomies by the pathology laboratory of the hospital where the

196 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 77 of 138 operation was performed will be sent centrally for a second reading by a pathologist with expertise in thyroid and C-cell pathology, who will be blinded to both trial treatment and site diagnosis. Once the samples are re-examined they will be sent back to the site laboratory. The Investigator will be informed of the second pathology report, in order to take appropriate action (e.g. in case of a difference to the diagnosis of the site pathology) Thyroid tissue sample collection in case of thyroidectomy Subjects scheduled for thyroidectomy will be asked to inform the Investigator prior to their operation. These subjects will be asked to consent to have a small sample of the removed thyroid tissue collected for testing of RET Y1062 phosphorylation in the thyroid C-cells. This is only applicable if C-cell pathology is confirmed (i.e., hyperplastic or neoplastic thyroid C-cells), and only if allowed by local law. The tissue sample will be destroyed after examination Genetic testing in case of a confirmed C-cell pathology Subject scheduled for thyroidectomy will be asked to consent to be tested (blood sample) to identify germline RET gene mutations associated with MEN2 syndrome. This RET gene mutation detection will be conducted in subjects with pathology reports confirming C-cell abnormality (medullary carcinoma or C-cell hyperplasia). Genetic testing will only be performed if allowed by local law Binge eating scale questionnaire (BES) Subjects from UK, Germany, Russia, Canada, US, Italy, Spain, France, Brazil, Austria, Mexico, Australia, Ieland and the Netherlands will complete the Binge Eating Scale questionnaire at Visit 3, 17,and 18a. Subjects discontinuing the trial prematurely should complete the questionnaire at the premature discontinuation visit (Visit 17). The binge eating scale questionnaire is only applicable for subjects who did not have pre-diabetes at baseline. The Binge Eating Scale questionnaire is a 16-item self-report diagnostic tool designed to capture both the behavioural and emotional characteristics of binge eating. The scale takes approximately 10 minutes to complete(48), (49), (50).

197 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 78 of 138 The questionnaires should preferably be completed by the subject before any trial procedures at the relevant visit are performed. Subjects should be given the opportunity to complete the questionnaires by themselves without interruption (see section ). The Investigator or his/her delegate must review Patient Reported Outcome(s) for completeness and adverse events immediately following administration. 8.4 Other assessments Liraglutide concentration (population PK) A single blood sample for measurement of plasma liraglutide concentration will be drawn at Visit 4 (dose escalation period), Visit 7 and 11 (maintenance period). At Visit 7 subjects will be nonfasting for the blood sampling whereas they will be fasting for blood sampling at Visit 4 and 11. Subjects will be instructed to note the exact date and time of administration, injection site and the dose of liraglutide/liraglutide placebo for the three doses of trial product immediately prior to Visit 4, 7 and 11. Diaries will be used to capture information on the injection site, date and time of administration and the dose injected. Exact date, time and injection site of liraglutide/ liraglutide placebo will be transcribed from the diaries into the ecrfs. A special laboratory will be responsible for the analysis of the liraglutide plasma concentration. Blood samples for liraglutide plasma concentration are collected, treated and shipped according to the description in the laboratory manual supplied by the special laboratory. Codes with the randomisation number containing information about treatment for the particular subject will be available to the special laboratory., Investigator and subject will remain blinded to the treatment until after database release Liraglutide concentration (PK C max sub-study) The sub-study will be performed in order to relate exposure to effect in a representative trial population. A small population of approximately 60 subjects will participate in the sub-study. All participants will be enrolled on a few selected sites in EU (UK and Netherlands) and US. An informed consent to participate in the sub-study will be obtained separately from the consent given to the main trial. The aim of the sub-study is to characterise liraglutide disposition in steady state in the obese population. Subjects will be instructed to inject liraglutide/ liraglutide placebo at approximately 10 pm in the evening prior to Visit 8. Furthermore, they will be instructed not to inject liraglutide / liraglutide placebo in the morning of Visit 8 and to use the same injection site for three consecutive days prior to Visit 8. Diaries will be used to capture information on the injection date and time as well as injection site and dose. Exact date, time, injection site and the dose of liraglutide/ liraglutide

198 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 79 of 138 placebo and time for sampling of the blood will be transcribed from the diaries into the ecrfs. Blood samples for bioanalysis of liraglutide/ liraglutide placebo plasma concentration will be drawn at approximately 8 am on the day of Visit 8 (Week 16). Samples will be drawn at time 0, 1, 2, 3 and 4 hours. Time 0 is defined as the start time of the PK blood sampling. A special laboratory will be responsible for the analysis of the liraglutide plasma concentration. Blood samples for liraglutide plasma concentration are collected, treated and shipped according to the description in the laboratory manual supplied by the special laboratory Smoking habits At Screening visit 1 it should be recorded whether the subject is a smoker Diaries Dosing diary A dosing diary will be handed out to the subjects prior to Visit 4, 7 and Visit 11. For subjects participating in the PK sub-study an additional dosing diary will be handed out prior to Visit 8. Subjects will be instructed by the Investigator or qualified staff to register information on administration of liraglutide. Information from the diary will be transcribed into the ecrf day food diary A 3-day food diary will be handed out to the subjects every second month. For subjects participating in the 56 week main trial including the re-randomised treatment period the food diary will be handed out at Screening visit 2, Visit 5, 7, 9, 11, 13, 16 and Visit 19a. For subjects participating in the delayed-onset-of-type-2-diabetes part of the trial the 3-day food diary will be handed out at Screening visit 2, Visit 5, 7, 9, 11, 13, 16, 18b, 20b, 22b, 24b, 26b, 28b, 30b, 32b, 34b, 36b, 38b, 40b and 42b. Subjects will be instructed by the dietician to register food intake. The dietician should collect and review the diaries. The diary will be used for diet counselling Diabetes diary A diabetes diary will be handed out to the subjects developing diabetes during the trial. The diabetes diary will be used to capture information on hypoglycaemic episodes. Subjects will be instructed by the Investigator or qualified staff how to enter information in the diary. Information on hypos from the diary will be transcribed into the hypo page and the AE page in the ecrf Dietary compliance and physical activity The subject s dietary compliance and the average daily level of physical activity will be recorded every second month. The subject will be questioned whether they performed less than half an hour, between half an hour and one hour or more than 1 hour of physical activity per day. For the main part of the trial including the re-randomised treatment period this applies to Visit 3, 6c, 8, 10, 12,

199 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 80 of ,17, and Visit 20a. For the delayed-onset-of-type-2-diabetes part of the trial the subject s dietary compliance and average daily level of physical activity will be recorded at Visit 3, 6c, 8, 10, 12, 14, 17, 19b, 21b, 23b, 25b, 27b, 29b, 31b, 33b, 35b 37b, 39b, 41b and 43b. An increase in physical activity (recommended minimum150 minutes/week) will be encouraged and re-enforced by use of pedometers. Whether or not the subject is in compliance with the prescribed diet is at the discretion of the dietician after review of the food diary Signs or symptoms of diabetic complications Information related to signs or symptoms of microvascular complications of diabetes (i.e., diabetic nephropathy (incl. microalbuminuria), retinopathy, neuropathy, and diabetic foot ulcer) will be recorded at Screening visit History of concomitant cardiovascular disease Information related to concomitant cardiovascular disease (i.e., myocardial infarction, disorders of rhythm or conduction, heart failure incl. NYHA class, ischemic heart disease incl. type, PCI and CABG, left ventricular systolic dysfunction, left ventricular diastolic dysfunction, hypertension, ischemic stroke, transient ischemic attack, hemorrhagic stroke, intracranial artery stenosis, carotid artery stenosis, peripheral arterial disease incl. > 50% stenosis on angiography or other imaging) will be recorded in the ecrf at Screening visit History of gallbladder disease Information related to gallbladder disease will be recorded in the ecrf at Screening visit History of psychiatric disorders Information related to psychiatric disorders (specifically history of depression, suicidal behaviour, anxiety, mood disorders, insomnia, or sleep disorders) will be recorded in the ecrf at Screening visit Subject compliance The Investigator will at each visit remind the subject to follow protocol procedures. At all visits from Visit 5 except for Visit 6 and the Follow up visits (Visit 22a and 44b) and at any unscheduled visit where drug accountability is performed the subject will return their partly used or unused trial products including all empty packaging material. The Investigator will assess the amount of trial product returned compared to what was dispensed at the last dispensing visit, and ask the subjects if the trial product has been used as prescribed. If a subject is discovered to be non-compliant, the Investigator should counsel the subject on the importance of taking trial products as directed.

200 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 81 of 138 Withdrawal of subjects due to non-compliance should be discussed with in advance. 9 Trial supplies 9.1 Trial product(s) The listed trial products will be supplied by A/S, Denmark Liraglutide 6.0 mg/ml, 3 ml FlexPen for subcutaneous (s.c.) injection Liraglutide placebo 3 ml FlexPen for subcutaneous (s.c.) injection The administration of liraglutide/liraglutide placebo will be as outlined in section The placebo and active drug are visually identical. 9.2 Packaging and labelling of trial product(s) All trial products will be packed and labelled by A/S and provided in non subject specific boxes. Labelling will be in accordance with Annex 13, local law and trial requirements. Sponsor s name and address Each Investigator site will be supplied with sufficient trial products for the trial on an ongoing basis controlled by the IV/WRS. Dispensing units will be prepared and distributed to the sites according to enrolment. Please refer to the TMM provided by for details regarding trial products standard packages. The Investigator will provide each subject with a handling instruction for liraglutide FlexPen at each drug dispensing visit. 9.3 Storage and drug accountability of trial product(s) The Investigator must keep track of all received, used, partly used and unused trial products by the use of the drug accountability module in the IV/WRS. Store in a refrigerator (2 C to 8 C). Do not store in the freezer or directly adjacent to the refrigerator cooling element. Do not freeze liraglutide/ liraglutide placebo and do not use liraglutide/ liraglutide placebo if it has been frozen.

201 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 82 of 138 Liraglutide/ liraglutide placebo should not be used if it does not appear clear and colourless. After first use of the liraglutide/ liraglutide placebo pen, the product can be stored for one month at room temperature (below +30 C) or in a refrigerator (2 C to 8 C). Keep the pen cap on when liraglutide/ liraglutide placebo pen is not in use in order to protect from light. Liraglutide/ liraglutide placebo should be protected from excessive heat and sunlight. US: 30 days at room temperature (+15 C to +30 C)/(59 F to 86 F) or in a refrigerator (+2 C to +8 C)/(+36 F to +46 F). Always remove the injection needle after each injection and store the liraglutide/ liraglutide placebo pen without an injection needle attached. This prevents contamination, infection, and leakage. It also ensures that the dosing is accurate. No trial product which has exceeded the expiry date must be used. The Investigator must ensure the availability of proper storage conditions and record and evaluate the temperature. The temperature must be recorded and evaluated on a daily basis (working days) using a calibrated min/max thermometer. A log to document the temperature must be kept Storage facilities should be checked frequently (at least every working day). In case of incorrect storage the Investigator or site staff must contact the monitor without delay. Trial product must be set on hold and not dispensed to subjects until notified by the monitor. No trial product(s) should be dispensed to any person not enrolled in the trial. Returned trial product(s) (partly used or unused including empty packaging material) must be stored separately from non-allocated trial product(s) until the monitor has performed drug accountability. The monitor will be responsible for coordination of used trial products. Destruction of trial products will be done according to local laws and will be recorded on a Destruction Form, which must be signed by the person responsible for destruction, as agreed by the monitor. 9.4 Auxiliary supply A/S will provide the following auxiliary supplies: NovoFine needles, blood glucose meters (Abbott Diabetes Care), lancets, test strips, control solution and pedometers. For further details please see the TMM.

202 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 83 of Randomisation, breaking of blinded codes and interactive voice/web response system (IV/WRS) 10.1 Randomisation In order to avoid bias the trial is conducted as a double-blinded trial with regards to treatment allocation to liraglutide or liraglutide placebo. At randomisation subjects will be stratified to either 56 weeks (plus 12 weeks re-randomised treatment plus 2 weeks follow-up period) or 160 weeks (plus 2 weeks follow-up period) of treatment based on pre-diabetes status (pre-diabetes Yes/No) at randomisation. Within these two groups subjects are furthermore stratified according to BMI (BMI 30, or BMI < 30). Treatment allocation for subjects with pre-diabetes will be double-blind for the first year (, Investigator and subject are blinded), and single blind for the remaining 2 years (i.e. is unblinded, whereas Investigator and subjects remain blinded). Subjects will be randomised to either liraglutide or liraglutide placebo using IV/WRS. At Visit 17 subjects without pre-diabetes at baseline who have received liraglutide treatment during the first 56 weeks of the trial will be re-randomised 1:1 to receive treatment with either liraglutide or liraglutide placebo for the following 12 weeks. The re-randomisation will be handled by the IV/WRS. Subjects randomised in the trial will continue with the subject number allocated at screening Breaking of blinded codes The code for a particular subject may be broken in a medical emergency if knowing the identity of the treatment allocation would influence the treatment of the subject. Whenever a code is broken, the person breaking the code must print the Code Break Confirmation generated by the IV/WRS, record reason, and sign and date the document. If the trial site needs to break the code, should, if possible, be contacted prior to breaking the code. (Monitor and department responsible for global product safety) will be notified immediately after the code break by the IV/WRS. If the subject should be withdrawn following a code break, a withdrawal session should be completed in IV/WRS. When code is broken the treatment allocation will be accessible to the Investigator and the department responsible for global product safety,.

203 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 84 of 138 In case IV/WRS is not accessible at the time of code break the IV/WRS vendor helpdesk should be contacted Interactive voice/web response system (IV/WRS) A trial specific IV/WRS will be set-up, and can be accessed at any time by the internet. Some sessions may be available through a toll-free telephone number. Accessibility to the IV/WRS must be restricted to and controlled by authorised persons. IV/WRS is used for: screening screening failure randomisation dispensing medication arrival withdrawal completion code break drug accountability live data change An IV/WRS user manual will be provided to the trial site.

204 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 85 of Concomitant illnesses/medical history and concomitant medication Definitions: Concomitant illness: any illness that is present at the start of the trial (i.e. at the first visit). Concomitant medication: any medication other than the trial product(s) that is taken during the trial, including the screening period. Details of all concomitant illnesses and medication must be recorded at trial entry (i.e. at the first visit). Any changes in concomitant medication must be recorded at each visit. If a change is due to an AE then this must be recorded and reported according to section 12. If the change influences the subject s eligibility to continue in the trial, then the Monitor must be informed. The information collected for each concomitant medication includes (at a minimum) start date, stop date or continuing and indication.

205 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 86 of Adverse events and pregnancies 12.1 Definitions Adverse event (AE): Any untoward medical occurrence in a subject or clinical investigation subject administered a pharmaceutical product and which does not necessarily have a causal relationship with this treatment. An AE can therefore be any unfavourable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of a medicinal product, whether or not considered related to the medicinal product. Note: This includes events from the first trial related activity after the subject has signed the informed consent and until post treatment follow-up period as defined in the protocol. For the subjects without pre-diabetes the post treatment follow-up period is defined as the period from the end of Visit 21a until the Follow-up visit (Visit 22a). For the pre-diabetic subjects the post treatment follow-up period is defined as the period from the end of the 160 week randomised treatment visit (Visit 43b) until the follow-up visit (Visit 44b). A worsening in concomitant illness must be recorded as an AE. A worsening of an ongoing AE should be reported on a new AE form by making a new assessment for seriousness and/or severity. The following should not be recorded as AEs: Pre-planned procedure unless the condition for which the procedure was planned has worsened from the first trial related activity after the subject has signed the informed consent. Pre-existing conditions found as a result of screening procedures. These should be recorded as medical history/concomitant illness. An AE can also be a clinical laboratory abnormality regarded as clinically significant, i.e. an abnormality that suggests a disease and/or organ toxicity and is of a severity which requires active management (i.e. change of dose, discontinuation of trial product, more frequent follow-up or diagnostic investigation). Serious adverse event (SAE): A SAE is an experience that at any dose results in any of the following: Death A life-threatening* experience In-patient hospitalisation** or prolongation of existing hospitalisation A persistent or significant disability/incapacity***

206 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 87 of 138 A congenital anomaly/birth defect Important medical events**** that may not result in death, be life-threatening*, or require hospitalisation may be considered a SAE when, based upon appropriate medical judgement, they may jeopardise the subject and may require medical or surgical intervention to prevent one of the outcomes listed in this definition * The term life-threatening in the definition of SAE refers to an event in which the subject was at risk of death at the time of the event. It does not refer to an event which hypothetically might have caused death if it were more severe. ** The term hospitalisation is the definition of a subject admitted to a hospital/inpatient (irrespective of the duration of physical stay), or not admitted to a hospital/not inpatient, but stays at the hospital for treatment or observation for more than 24 hours. Medical judgement must always be exercised, and when in doubt, the hospital contact should be regarded as a hospitalisation. Hospitalisations for administrative, trial related and social purposes do not constitute AEs and should therefore neither be reported as AEs or SAEs. Likewise, hospital admissions for surgical procedures planned prior to trial inclusion are not considered AEs or SAEs.. ***The term disability/incapacity means that following the event the subject or clinical investigation subject has significant, persistent or permanent change, impairment, damage or disruption in his body function or structure, physical activity and/or quality of life. ****The term important medical events means events which may jeopardise the subject or require intervention to prevent a seriousness criterion. It can be adverse events which suggest a significant hazard or puts the subjects or clinical investigation subject at risk, such as druginteraction, contra-indications or precautions, occurrence of malignancies or development of drug dependency or drug abuse. Non-serious adverse event: A non-serious AE is any AE which does not fulfil the definition of a serious AE. Severity assessment definitions: Mild - No or transient symptoms, no interference with the subject s daily activities. Moderate - Marked symptoms, moderate interference with the subject s daily activities. Severe - Considerable interference with the subject s daily activities, unacceptable. Relationship to trial product assessment definitions: Probable: good reasons and sufficient documentation to assume a causal relationship

207 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 88 of 138 Possible: a causal relationship is conceivable and cannot be dismissed Unlikely: the event is most likely related to an aetiology other than the trial product Outcome categories and definitions: Recovered - Fully recovered, or by medical or surgical treatment the condition has returned to the level observed at the first trial related activity after the subject signed the informed consent. Recovering - The condition is improving and the subject is expected to recover from the event. This term should only be used when the subject has completed the trial. Recovered with sequelae - As a result of the AE the subject suffered persistent and significant disability/incapacity (e.g. became blind, deaf, paralysed). If the sequelae meet seriousness criteria, the AE must be reported as an SAE Not recovered. Fatal. Unknown - This term should only be used in cases where the subject is lost to follow-up Technical complaints A technical complaint is any written, electronic, or oral communication that alleges defects on trial products - listed as trial products in this protocol (9.1). The technical complaint may be associated with an AE, but does not concern the AE itself. A technical complaint may for example concern: the physical or chemical appearance of trial products (eg discoloration, particles or contamination) the packaging material (e.g. leakage, cracks, problems with rubber membrane in the cartridge or errors in labelling text) problems related to devices (e.g. to the injection mechanism, dose setting mechanism, glucose measurement, push button or interface between the pen and the needle) 12.2 Collection, recording and reporting of adverse events All events meeting the definition of an AE must be collected and reported. At each contact with the trial site (visit or telephone, excluding safety visits, where the subject is not seeing the Investigator or his staff (e.g. visits to the laboratory)), the subject must be asked about adverse events. The subject will be asked about AEs in the following manner: Have you experienced any problems since the last contact?. All AEs, either observed by the Investigator or reported by the subject, must be recorded by the Investigator and evaluated. 's assessment of expectedness is done according to the reference documents:

208 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 89 of 138 Liraglutide obesity: IB, 3 rd edition, 2010 or any updates hereof The Investigator should record the diagnosis, if available. If no diagnosis is available the Investigator should record each sign and symptom as individual adverse events. All AEs, SAEs and MESIs must be recorded by the Investigator on the AE Form in the EDC ecrf. If more than one sign or symptom is to be reported, create a separate adverse event form for each sign and symptom. For SAEs and MESIs, the SIF pages also have to be completed for each event in the ecrf. However if several symptoms or diagnosis occur as part of the same clinical picture, only one set of SIF can be used to describe all the SAEs. All concerned AE numbers must be included in the AE number field in the header of the SIF. For MESIs, the specific MESI follow-up questions have also to be completed in the ecrf within 7 calendar days. For MESIs qualifying for adjudication, the Source Data Collection Tool has also to be completed in the ecrf within 7 calendar days. The Investigator must report initial information on all SAEs and MESIs to within 24 hours of obtaining knowledge about the event by completing the following in EDC: AE form SIF The specific MESI follow-up questions and source data collection tool also have to be completed within 7 calendar days if applicable. If for any reason the EDC application is unavailable, complete the AE form, SIF and if applicable the specific MESI follow-up questions and Source Data Collection Tool on paper CRFs within 7 calendar days and forward a copy electronically in PDF format by , or by fax or courier to within the same timelines. must inform the regulatory authorities and IECs/IRBs in accordance with the local requirements in force and (ICH GCP)31. will notify the Investigator of trial product related suspected unexpected serious adverse reactions in accordance with the local requirements. (e.g. European Directive 2001/20/EC and International Conference on Harmonisation/Good Clinical Practice (ICH GCP31). In addition, the Investigator will be informed of any trial related procedure SAE which may warrant a change of any trial procedure.

209 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 90 of 138 Investigators will be notified of trial-related SAEs in accordance with the local requirements in force and International Conference on Harmonisation (ICH) GCP31. The monitor must be informed accordingly Medical events of special interest A MESI (serious or non-serious) is noteworthy event of scientific and medical concern that Novo Nordisk continues to monitor. A MESI does not necessarily have a causal relationship with the IMP. A MESI should be reported following the same reporting requirements and timelines as for SAEs (see section 12.1 ), irrespective of whether the MESI fulfils a SAE criterion. The following are defined as MESIs in this trial (see Appendix J for further definitions) Medication errors concerning trial products administration of wrong drug or use of wrong device wrong route of administration, such as intramuscular instead of subcutaneous administration of a high dose with the intention to cause harm, e.g. suicide attempt administration of an accidental overdose, i.e. dose which may lead to significant health consequences, as judged by the Investigator, irrespective of whether the SAE criteria are fulfilled or not Suspected transmission of an infectious agent via a trial product Death Acute coronary syndrome (myocardial infarction or hospitalisation for unstable angina) Cerebrovascular event (stroke or transient ischemic attack) Heart failure Stent trombosis Revascularisation procedure Hospitalisation for cardiac arrhythmia Pancreatitis or acute, severe, persistent abdominal pain leading to a suspicion of pancreatitis Acute gallstone disease (biliary colic or acute cholecystitis) Elevated lipase or amylase 3x UNR Neoplasms Thyroid disease Any confirmed episode of calcitonin value 20 ng/l (from Visit 3 and onwards) Acute renal failure Severe hypoglycaemic events Immunogenicity event ( allergic reactions including allergic reactions at injection sites, or immune-complex disease)

210 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 91 of 138 AEs leading to withdrawal Psychiatric Disorders (including psychiatric disorders diagnosed by C-SSRS and PHQ-9 questionnaires) For subjects with calcitonin 20 ng/l (from visit 3 and onwards) a repeat measurement of calcitonin must be performed preferably within four weeks and if confirmed, the event elevated calcitonin should be reported as a MESI (i.e., any confirmed episode of calcitonin concentration value 20 ng/l, if not already reported as a MESI). For details on how increased calcitonin levels at screening and follow-up visits should be reported and followed up, please refer to Appendix J for reporting of MESIs and Appendix I describing how the Calcitonin Monitoring Committee works. In addition the following rules for increased lipase and /or amylase apply: If the amylase or lipase baseline (at screening) value is > 3xUNR this information will be recorded as medical history for that subject. If at any post baseline visit the amylase or lipase value is > 3xUNR a MESI should be reported. Certain MESIs will be adjudicated by an external independent event adjudication committee as described in section (12.2.2). For further information regarding definitions of MESIs and an overview of which events that should undergo adjudication, please refer to Appendix J. Complete the AE form, Safety Information Form (SIF), specific MESI follow-up questions and if applicable Source Data Collection Tool in the ecrf. If for any reason the electronic data capture (EDC) application is unavailable, complete the AE form, SIF, specific MESI follow-up question and if applicable Source Data Collection Tool on paper CRFs and forward a copy electronically in PDF format by , or by fax or courier to External independent event adjudication committee An external independent Event Adjudication Committee (EAC) is constituted for this trial to perform ongoing adjudication, standardisation and assessment of events listed in Appendix J. The following events (also described in under MESIs) will be adjudicated and evaluated by the EAC in an independent and blinded manner: Death Acute coronary syndrome (myocardial infarction or hospitalisation for unstable angina) Cerebrovascular event (stroke or transient ischemic attack) Heart failure requiring hospitalisation

211 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 92 of 138 Stent trombosis Coronary revascularisation procedure Pancreatitis or acute, severe, persistent abdominal pain leading to a suspicion of pancreatitis Neoplasms Thyroid disorders requiring thyroidectomy The EAC is composed of permanent members who cover required medical specialties. The EAC members must disclose any potential conflicts of interest and must be independent of Novo Nordisk. The EAC works in accordance with written guidelines included in the EAC Charter that describes in detail the composition, tasks, responsibilities, and work processes of the committee. The charter will be finalised prior to first patient first visit. The EAC will perform adjudication based on the criteria and definitions described in the EAC Charter. The cardiovascular events will be classified according to FDA requirements (51). The EAC will review translated copies in English of medical documentation received in the adjudication packages (for example X-ray, ECGs, ultrasound images, discharge summaries, pathology reports, and death certificates). The Investigator will provide them as soon as possible, when they receive the request from. The role of the EAC is solely to adjudicate events in a blinded manner. The EACs will have no authorisation to impact on trial conduct, trial protocol and amendments. The assessments made by the EAC will be included in the CTR as well as assessments made by the Investigator. However, the adjudication made by an EAC, given its independence and in-depth analysis of each event, will be attributed with greater importance of the two. The outcomes of adjudication will be kept in the Global Safety database as well as in the clinical trial database Follow-up of adverse events During and following a subject s participation in a clinical trial, the Investigator/institution should ensure that adequate medical care is provided to the subject for any adverse events, including clinically significant laboratory values related to the trial. The Investigator/institution should inform the subject when medical care is needed for adverse event(s) of which the Investigator becomes aware. The follow up information should only include new (updated and/or additional) information that reflects the situation at the time of the Investigator s signature.

212 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 93 of 138 Follow-up information (corrections, new or additional information) should be reported within 24 hours of obtaining knowledge of the information for SAEs and MESIs, and if previously nonserious AEs become SAEs by updating the AE form and/or SIF in the ecrf. All non-serious AEs classified as severe or possibly/probably related to the trial product must be followed until the subject has recovered or recovered with sequelae, and all queries have been resolved. Cases of chronic conditions or cancer or AEs ongoing at time of death (i.e.subject dies from another AE) can be closed with an outcome of recovering or not recovered. Cases can be closed with an outcome of recovering when the subject has completed the post-trial follow-up period and is expected by the Investigator to recover. All other non-serious AEs must be followed until the outcome of the event is recovering, recovered or recovered with sequelae or until the end of the post-treatment follow-up stated in the protocol, whichever comes first, and until all queries related to these AEs have been resolved. AEs ongoing at time of death (i.e. subjects dies from another AE) can be closed with an outcome of recovering or not recovered. The Investigator must ensure that the worst case severity and seriousness is kept consistent. The Investigator must record follow-up information on non-serious adverse events by updating the adverse event form in the ecrf. The follow-up information should only include new (updated and/or additional) information that reflects the situation at the time of the Investigator s signature. Queries or follow-up requests from should be responded to within 7 calendar days, unless otherwise specified. This must be done by updating the AE form and/or SIF in the ecrf. If for any reason EDC/eCRF application is unavailable, then the relevant paper forms have to be filled in, marked as follow-up and forwarded by fax or to within the same timelines. All SAEs and MESIs must be followed up until the outcome of the event is recovered, recovered with sequelae or fatal and until all queries have been resolved. Cases of chronic conditions or cancer or AEs ongoing at time of death (i.e. the subject dies form another AE) can be closed with the outcome of recovered or not recovered. Cases can be closed with an outcome of recovering when the subject has completed the trial and is expected by the Investigator to recover. When a MESI of a specific category occurs for the first time in the subject and is selected on the AE form, the predefined follow-up section for that MESI category is automatically activated in the EDC. Each of these sections have follow-up forms related to the reported MESI, and the forms should be considered as a follow-up request from, and should therefore also be responded to within 7 calendar days from the date of awareness of the MESI.

213 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 94 of 138 After access to update the AE form and SIF in EDC is removed the Investigator must record any SAE and MESI follow-up information, if required, on the paper ecrfs provided at trial closure Collection and reporting of technical complaints All technical complaints occurring from the time of first and until the last usage of trial product - must be collected and reported to. The subject must be asked about technical complaints during each contact (site visit or telephone contact) with the Investigator or trial site staff. This may be done by posing a simple question such as have you experienced any problems since the last contact?. The Investigator must assess whether the technical complaint is related to: AE(s), SAE(s) and/or MESI(s) The AE(s), SAE(s) and MESI(s) related to technical complaint(s) must be reported by the Investigator following the same reporting requirements and timelines as for other AEs, SAEs and MESIs (see section12.1). Technical complaints must be reported on the technical complaint form by the Investigator, as described in the following: One technical complaint form must be completed for each trial product, non-investigational medicinal product (NIMP) or auxiliary supply. If the technical complaint involves more than one batch number, a technical complaint form for each batch number must be completed. The Investigator must complete the technical complaint form in the ecrf within the following timelines of the trial site obtaining knowledge of the technical complaint: technical complaint assessed as related to a SAE and/or MESI within 24 hours all other technical complaints within 5 calendar days If the ecrf is unavailable, the paper technical complaint form should be completed and faxed to Customer Complaint Centre,, fax: , within the same timelines as for ecrf Collection, storage and shipment of technical complaint samples The Investigator must collect the technical complaint sample from the subject. If the technical complaint sample is unobtainable, the Investigator must specify on the technical complaint form why it is unobtainable. The technical complaint sample and a paper copy of the technical complaint form must be sent to within 5 calendar days of receiving the technical complaint sample at trial site by using the following address:

214 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 95 of 138 A/S, Att.: Customer Complaint Centre, Nybrovej 80, 2820 Gentofte, Denmark. The Investigator must ensure that the technical complaint sample is labelled with the batch number and, if available, the DUN number. Storage and shipment of the technical complaint sample should be done in accordance with the conditions prescribed for the product (see section 9) Pregnancy Subjects must be instructed to notify the Investigator immediately if they become pregnant during the trial. The Investigator must report any pregnancy reported during the trial to except for pregnancies occurring in the screening period. Trial subjects will give consent on enrolment that the Investigator will report any pregnancy during the trial to and they will be asked to provide information about the pregnancy, delivery and the health of her infant until one month of age. The Investigator must report information on pregnancy and follow-up within 14 calendar days of obtaining the information using the pregnancy form part A and the pregnancy form part B respectively. If the pregnancy results in an abnormal outcome, such as congenital anomalities, foetal death, spontaneous abortion or SAEs in the neonate, this should be regarded as an SAE with the same reporting requirements and timelines as for SAEs. If an SAE occurs in relation to a pregnancy, either to the mother or the newborn, then follow the same reporting requirements and timelines as for SAEs. The AE form and SIF in the ecrf must be Completed. If for any reason the EDC/eCRF application is unavailable, then the relevant paper forms have to be completed and forwarded by fax or to Precautions/over-dosage When initiating treatment with liraglutide, the subject may in some cases experience loss of fluids/dehydration, e.g. in cases of vomiting, nausea or diarrhoea. It is important to avoid dehydration by drinking plenty of fluids. There is one overdose report for liraglutide. A male subject accidentally administered 17.4 mg of liraglutide instead of the prescribed 0.6 mg. The subject recovered with no intervention and no lasting sequelae was seen. Please refer to liraglutide obesity IB, 3 rd edition 2010 or any updates hereof.

215 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 96 of Safety committee Internal safety committee will constitute an internal safety committee to perform ongoing safety surveillance. The safety committee will conduct ongoing monitoring of blinded safety data (all haematology and biochemistry results including calcitonin, amylase and lipase). The safety committee may recommend unblinding of any data for further analysis. If so, an independent ad hoc group will be established to maintain the blinding Calcitonin Monitoring Committee Monitoring of calcitonin in regular intervals will be implemented in the trial. Algorithm of further clinical and laboratory evaluation, supervised by an independent committee of thyroid experts (Calcitonin Monitoring Committee) will be recommended to be followed in all subjects with clinically relevant abnormal calcitonin values. This algorithm has been developed in collaboration with leading independent experts in thyroid/c-cell disease. In cases where the follow-up action recommended by the CMC, based on the elevated calcitonin levels results in establishing the presence of a thyroid disease, the thyroid disease should be reported as a new MESI. If the thyroid disease is a neoplasm or result in a thyroidectomy, the event will undergo adjudication by the neoplasm EAC.

216 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 97 of Case report forms will provide a system for electronic data capture. This system and support services to the system will be supplied by a clinical services vendor Phase Forward. The activities of this vendor will be under the direction and supervision of Rules for completing ecrfs Ensure that all relevant questions are answered and that no empty data blocks exist. If a test/assessment has not been done and will not be available, or if the question is irrelevant (e.g. is not applicable) indicate this according to the data entry instructions. The Investigator or Investigator s authorised staff must ensure that all information derived from source documentation is consistent with the source information. By signing the Case Book sign off electronically, the Investigator confirms that the information is complete and correct Corrections to ecrfs Corrections to the ecrf data will be made by the Investigator or the Investigator s authorised staff. An audit trail will be maintained in the EDC application containing as a minimum: identification of the person entering the data, date and time of the entry and reason for the correction. If corrections are made by the Investigator s authorised staff after the date of the Investigator s signature on the Case Book sign off, the form must be signed again by the Investigator ecrf flow The Investigator must ensure that data is recorded in the ecrfs as soon as possible after the visit preferably within 3 working days. When data is entered it will be available to for data verification activities. Site specific ecrf data (in an electronic readable format) will be provided to the Investigator site after the trial database is released and access to update the trial data on the EDC application is removed. This data will be retained by the site. The Investigator will receive the laboratory reports directly from the central laboratory. The Investigator must review, sign and date the laboratory report on the day of evaluation. The signed laboratory report is retained by the site as source documentation. When the final clinical trial report is available the data will be archived by.

217 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 98 of 138 At the end of trial the Investigator must ensure that all remaining data have been entered into the ecrf no later than 3 days after the last subject s last visit at the site in order to ensure the planned lock of the database. 14 Monitoring procedures During the course of the trial the Monitor will visit the trial site to ensure that the protocol is adhered to, that all issues have been recorded, to perform source data verification and to monitor drug accountability. The visit intervals will depend on the outcome of the remote monitoring of the ecrfs, the trial site's recruitment rate and the compliance of the trial site to the protocol and GCP. Factors to be considered in this determination may include objective, purpose, design, complexity, blinding, size, and endpoint for the trial. The Monitor must visit the site shortly after the first subjects attended Screening visit 1 to ensure that mistakes are caught early. Hereafter the intervals between visits should not exceed 12 weeks. However, more frequent monitoring visits are required during peak periods such as recruitment period and finalisation of the trial. The monitor must be given direct access to source documents (original documents, data and records). Direct access includes permission to examine, analyse, verify and reproduce any record(s) and report(s) that are important to evaluation of the clinical trial. In addition, the monitor should be available for discussions by telephone. The Monitor must ensure that all required ecrf forms for screening failures are completed, (e.g. screening failure form and the case book sign off (affirmation statement) is electronically signed by the Investigator). As a minimum requirement the following data must be source data verifiable in source documentation other than the ecrf: Existence of subject (subject identifier; subject number and date of birth) Confirmation of participation in the trial (subject identification number (ID), trial ID and signed and dated informed consent form) Diagnosis/indication under investigation Visit dates Data from: adverse event form(s) safety information form(s) pregnancy form(s) Relevant medical history, concomitant illness Reason for exclusion or withdrawal SIF Body weight OGTT

218 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 99 of 138 The source data must reflect/document the dose regimen, but not the full drug accountability. The following data can be recorded directly on the CRFs and will be considered source data: PRO questionnaires (IWQoL, SF-36, TRIm-Weight and BES) Mental health questionnaires (C-SSRS and PHQ-9) For all other data in the ecrfs, it must be possible to verify these against source documents. The monitor will ensure that the ecrfs are completed. For all data recorded the source document must be defined in a source document agreement at each site. The existence of subjects must be confirmed either via extensive medical record (not just anamnesis collected at Screening visit 1) e.g. a copy of a passport or ID card could be used.

219 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 100 of Data management Data management is the responsibility of Headquarters. Data management may be delegated under an agreement of transfer of responsibilities to another data management unit within. The subject and the biological material obtained from the subject will be identified by subject number, trial site and trial identification number. Appropriate measures such as encryption or deletion will be enforced to protect the identity of human subjects in all presentations and publications as required by local/regional/national requirements. Appropriate measures such as encryption of data files will be used to assure confidentiality of subject data when it is transmitted over open networks. Laboratory data will be transferred electronically from the central laboratory performing clinical analyses. The electronic laboratory data will be considered source data. In cases where laboratory data is transferred via non-secure electronic networks, data will be encrypted during transfer.

220 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 101 of Computerised systems will capture and process clinical data using computerised systems which are described in Standard Operating Procedures and IT architecture documentation. The use and control of these systems are documented. Investigators working on the trial may use their own electronic systems to capture source data. will collect information on the practical use of these systems within the conduct of this clinical trial.

221 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 102 of Evaluability of subjects for analysis The following analysis sets are defined: Full analysis set (FAS) All randomised subjects exposed to at least one dose of the trial product and with at least one postbaseline assessment of any efficacy endpoint will be included. Subjects in the FAS will be analysed according to randomised treatment. The requirement of a post-baseline observation is in alignment with the FDA recommendations [reference: Guidance for Industry developing products for weight management, Draft guidance, February 2007]. Safety analysis set All randomised subjects who have been exposed to at least one dose of trial product. Subjects in the safety analysis set will be analysed as treated.

222 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 103 of Statistical considerations Biostatistics A/S will be responsible for the statistical analysis Sample size calculation A sample size of 3600 randomised subjects, 2400 randomised to liraglutide treatment and 1200 randomised to liraglutide placebo is considered to provide a reasonable estimation of the safety of liraglutide as a weight-management product. This sample size provides enough power for the primary efficacy endpoints weight change, the proportion of subjects with a weight loss of at least 5%, the proportion of subjects with a weight loss greater than 10%, and new onset of diabetes. The hypothesis of equality between liraglutide and liraglutide placebo with respect to the primary endpoints will be tested in a hierarchical manner in the order in which the endpoints are mentioned. The power for the primary endpoint weight change is calculated based on a two sided t-test with a significance level of 5%. The power with regard to the co-primary dichotomous endpoints proportion of subjects with a weight loss of at least 5% or more than 10%, respectively, is calculated based on a two-sided chi-square test. With a sample size of 2400 subjects treated with liraglutide and 1200 subjects treated with liraglutide placebo, the trial will have more than 90% power to detect a difference between liraglutide and liraglutide placebo in the proportion of subjects with a weight loss greater than 10%, given that the probabilities to achieve this weight loss is 10% for liraglutide placebo and 14% for liraglutide. The trial will also have more than 90% power to detect a difference in mean weight change between liraglutide and liraglutide placebo, given that the true difference is 0.70 kg and the standard deviation is 5.9 kg. In the interim analysis of data on weight loss after one year of treatment of obese subjects with either liraglutide, liraglutide placebo or orlistat (liraglutide obesity Investigators Brochure (IB), 3 rd edition 2010), the estimated difference in weight change between 3.0 mg liraglutide and liraglutide placebo was 5.82 kg, whereas the SD of weight loss was equal to 5.9 kg. Among subjects treated with 3.0 mg liraglutide the proportion of subjects with a weight loss greater than 5% and a weight loss greater than 10 % was 75% and 37%, respectively. The corresponding proportions in the placebo group were 28% and 10%. By assuming that the effects of treatment are of this magnitude the power to detect the differences between liraglutide and liraglutide placebo are larger than 99% for each of the three co-primary endpoints weight change, proportion of subjects with a weight loss of at least 5% and proportion of subjects with a weight loss larger than 10%.

223 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 104 of 138 The large number of randomised subjects also provides sufficient power for the fourth primary endpoint new onset of diabetes among subjects with pre-diabetes. The endpoint new onset of diabetes will be analysed using methods for analysis of interval censored failure time data. A conservative estimate of the power may be calculated as if the endpoint diabetes yes/no among completers during the 160 weeks is analysed by use of a two-sided Chi-square test with a significance level of 5%. It is assumed that the annual conversion rate of subjects with pre-diabetes to diabetes equals 7% among placebo treated subjects, whereas it is 60% lower, 2.1%, among liraglutide treated subjects. After 160 weeks of treatment with liraglutide/placebo, the percentage of subjects with diabetes is therefore equal to 1-(1-0.07) 3 = 20% among liraglutide placebo treated subjects, and 1-( ) 3 = 6% among liraglutide treated subjects. It is assumed that the drop-out during the 160 weeks may be as large as 65% in both groups. The power for conversion rates for placebo of 5, 7 and 9% and conversion rates 60 and 70% lower in the liraglutide group may be seen in Table Table 18 1 Sample size calculation for the primary endpoint onset of diabetes assessed at week 160 Annual conversion factor for placebo 5% 7% 9% Power for onset of diabetes with: Liraglutide conversion rate 60% lower than placebo Liraglutide conversion rate 70% lower than placebo The power calculations are based on a two-sided chi-square test of equal proportions using a 2:1 distribution and a significance level of 5%. With the minimum planned number of subjects with pre-diabetes and a drop out rate of 65%, 233 and 117 completers are expected in the liraglutide and placebo group respectively. Based on these figures it is apparent that a sample size of 2400 liraglutide treated subjects and 1200 liraglutide placebo treated subjects will provide sufficient power also for the fourth primary endpoint onset of diabetes Statistical methods All statistical tests are two-sided at a 5% significance level. For all efficacy evaluations, only observations on drug (defined as last injection taken the day before or on the day of the visit) will be included in the statistical analyses and summaries. For all weight and glycaemic efficacy endpoints, only observations prior to rescue medication will be included in the statistical analyses and summaries, as rescue medication will confound the subsequent measurement of these parameters. Excluded observations will be listed.

224 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 105 of Primary efficacy endpoints The four primary efficacy endpoints consists of three co-primary endpoints describing weight change and one endpoint describing onset of diabetes: Change from baseline in fasting body weight at 56 weeks The proportion of subjects losing at least 5% of baseline fasting weight (measured at Week 0) The proportion of subjects losing more than 10% of baseline fasting weight (measured at Week 0) Onset of type 2 diabetes in subjects with pre-diabetes at baseline evaluated at week 160 The first primary endpoint is suggested as a primary endpoint in the FDA guidance(52) as well as the EMEA guidance(53). The second primary endpoint is suggested by FDA, whereas EMEA instead focus on the third of the endpoints Primary analysis of the co-primary endpoints Description of the applied LOCF approach applied in the analysis of the co-primary endpoints describing weight change In the primary analyses of the three first endpoints, LOCF will be applied. Only fasting weight measurements will be used and only measurements performed after randomisation will be carried forward. The follow-up weight measurements at 56 weeks (Visit 17x) after randomisation will not be applied in the primary analyses. Primary analysis of change from baseline in fasting body weight loss at 56 weeks For the continuous primary endpoint, fasting body weight loss, analysed as change in fasting body weight from Week 0 to Week 56 (using the last observation carried forward approach as described above) will be compared between liraglutide and liraglutide placebo using an ANCOVA model with treatment, country, pre-diabetes stratification factor, BMI-stratification factor, interaction between stratification factors and gender as fixed effects and with baseline body weight (at Week 0) as a covariate. The analysis will be performed for the FAS. From this model the expected differences in weight change between liraglutide treatment and liraglutide placebo will be estimated together with the associated 95% confidence interval and the p-value corresponding to the test of the hypothesis of no difference between treatments. The proportion of subjects losing at least 5% of baseline fasting weight (measured at Week 0) For this categorical primary endpoint, a logistic regression model with treatment, country, prediabetes stratification factor, BMI-stratification factor, interaction between stratification factors and gender as fixed effects and with baseline fasting body weight (at Week 0) as a covariate, will be used to compare the proportion that after 56 weeks of treatment lose at least 5% of their baseline fasting bodyweight in the two groups. The analysis will be performed for the FAS using the LOCF approach described above. From this model the odds ratios between the liraglutide treatment and

225 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 106 of 138 liraglutide placebo will be estimated together with the associated 95% confidence interval and the p-value corresponding to the test of the hypothesis of no difference between treatments. Primary analysis of the proportion of subjects losing more than 10% of baseline fasting weight (measured at Week 0) This endpoint will be analysed in the same manner as the proportion of subjects losing at least 5% of baseline fasting weight. Primary analysis of the endpoint Proportion of subjects with onset of type 2 diabetes at 160 weeks The time of onset of type 2 diabetes in subjects with pre-diabetes at baseline will be registered at specific visits. If presence of diabetes is observed at a visit, the onset must have occurred between this visit and the previous visit where diabetes was examined. Thus the observations can be considered as interval censored data. The time of onset will be set to be in between the first of the two required registrations of elevated FPG or OGTT 2 hr plasma glucose, and the diabetes assessment visit prior to the first registration. The data will be analysed using a Weibull model. The model will include effects of treatment, gender, and BMI stratification factor. Baseline FPG will be included as a covariate. This analysis will be carried out on the FAS. From this model the hazard rate between treatments will be estimated and a test on no difference between treatments will be performed. Description of the testing procedure The hypotheses of equality between liraglutide and liraglutide placebo for each of the four endpoints are tested in a hierarchical manner in the order in which the endpoints are presented. The implications of the testing procedure are as described below: liraglutide will be considered statistically significantly better than liraglutide placebo with respect to the first primary endpoint if the hypothesis of equality between liraglutide and liraglutide placebo is rejected and if the estimated effects of treatment are better in the liraglutide group than in the placebo group for the first primary endpoint. liraglutide will only be considered statistically significantly better than liraglutide placebo with respect to the second primary endpoint if it is considered statistically significantly better with respect to the first primary endpoint, if the hypothesis of equality between liraglutide and liraglutide placebo is rejected for the secondary primary endpoint, and if the estimated effects of treatment are better in the liraglutide group than in the placebo group for the second primary endpoint. liraglutide will only be considered statistically significantly better than liraglutide placebo with respect to the third primary endpoint if it is considered statistically significantly better with respect to the first and the second of the primary endpoints, if the hypothesis of equality between liraglutide and liraglutide placebo is rejected for the third co-primary endpoint, and if

226 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 107 of 138 the estimated effects of treatment are better in the liraglutide group than in the placebo group for the third primary endpoint. liraglutide will only be considered statistically significantly better than liraglutide placebo with respect to the fourth primary endpoint if it is considered statistically significantly better with respect to all the co-primary endpoints describing weight change, if the hypothesis of equality between liraglutide and liraglutide placebo is rejected for the fourth primary endpoint, and if the estimated effects of treatment are better in the liraglutide group than in the placebo group for the fourth primary endpoint Sensitivity analyses of the primary endpoints Sensitivity analyses of change from baseline in Fasting Body weight after 56 weeks For supporting evidence, the following sensitivity analyses will be carried out: The same analysis as above will be applied to the completers (week 56) The same analysis as above, applied to all randomised subjects allowing for baseline carried forward for subjects without a post baseline measurements The same analysis as above, applied to the FAS including the fasting and non-fasting weight measurements, off drug weight measurements, and the follow-up weight measurements 56 weeks after randomisation (Visit 17x) The same analysis as above, applied to the FAS including the fasting and non-fasting weight measurements, off drug weight measurements, and the follow-up weight measurements 56 weeks after randomisation (Visit 17x) and weight measurements following rescue medication The same analysis as above, applied to FAS, but imputing missing observations with the regression method (54). Five sets of imputations and subsequent analyses will be done. The following repeated measures analysis (linear mixed effect model) using all longitudinal fasting weight measurements available for the FAS will be applied. The response variable is the change of body weight from baseline, and the model includes visit, treatment, country, prediabetes stratification factor, BMI-stratification factor, interaction between stratification factors and gender and the interactions between treatment and visit and baseline body weight and visit as fixed effects and with baseline body weight (at Week 0) as a covariate. Subject will be included as a random factor. The model will be used to compare liraglutide and liraglutide placebo at Week 56. Sensitivity analyses of the proportion of subjects losing at least 5% of baseline fasting weight (measured at Week 0) For supporting evidence the following sensitivity analyses will be carried out: The same analysis as above, applied to the completers (week 56) The same analysis as above, applied to all randomised subjects allowing for baseline carried forward for subjects without a post baseline measurements

227 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 108 of 138 The same analysis as above, applied to the FAS including fasting and non-fasting weight measurements, off drug weight measurements and the follow-up weight measurements 56 weeks after randomisation (Visit 17x) The same analysis as above, applied to the FAS including fasting and non-fasting weight measurements, off drug weight measurements and the follow-up weight measurements 56 weeks after randomisation (Visit 17x) and weight measurements following rescue medication The same analysis as above, applied to the FAS, but regarding subjects without a valid assessment of weight at 56 weeks as non-responders, i.e. as not having lost 5% of their body weight. (For subjects withdrawing prematurely from the trial, their follow-up weights at 56 weeks after randomisation, if available, will be used) The same analysis as above, applied to FAS, imputing missing observations with the regression method [Reference: Rubin, D.B. (1987), Multiple Imputation for Nonresponse in Surveys, New York: John Wiley & Sons, Inc.] Five sets of imputations and subsequent analyses will be done. Sensitivity analyses of the proportion of subjects losing more than 10% of baseline fasting weight (measured at Week 0) The sensitivity analyses for this endpoint will be analysed in the same manner as for the proportion of subjects losing at least 5% of baseline fasting weight. Sensitivity analyses of the endpoint proportion of subjects with onset of type 2 diabetes at 160 weeks For supporting evidence the following sensitivity analyses will be carried out: The same analysis as above, applied to the completers (week 160) A semi-parametric approach assuming proportional hazards applied to the FAS Analysis of secondary efficacy endpoints All secondary efficacy endpoints will be based on the FAS. All endpoints will be summarized descriptively by visit using observed data. At end of treatment (week 56, 68 and 160) summaries will be presented for both observed and LOCF imputed data. Summary statistics for continuous endpoints include number of observations, arithmetic mean, median, standard deviation, minimum and maximum. Summary statistics for categorical endpoints include number of observations, number and percentage of subjects fulfilling the criteria. Table 18 2,Table 18 3 and Table 18 4 gives an overview of the statistical analysis for the secondary endpoints at week 56, 68 and 160, respectively. Continuous secondary endpoints will be analysed and presented similarly to the primary analysis of weight change. Baseline values will be included as covariates in the analyses of the corresponding

228 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 109 of 138 response variables. This analysis is referred to as ANCOVA in Table 18 2,Table 18 3 and Table Categorical secondary endpoints will be analysed and presented similarly to the primary analysis of proportion of subjects losing at least 5% of baseline body weight. Continuous baseline values will be included as covariates in the analyses of the corresponding response variables unless otherwise specified. This analysis is referred to as LR in Table 18 2,Table 18 3,Table 18 4.

229 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 110 of 138 Table 18 2 Overview of statistical analysis of secondary efficacy endpoints at week 56 Endpoint Endpoint type Statistical analysis Change from baseline (week 0) to week 56 in body weight (kg) Continuous ANCOVA Change from baseline (week 0) to week 56 in waist circumference Continuous ANCOVA Change from baseline (week 0) to week 56 in HbA1c Continuous ANCOVA Change from baseline (week 0) to week 56 in FPG Continuous ANCOVA Change from baseline (week 0) to week 56 in fasting insulin Continuous ANCOVA Change from baseline (week 0) to week 56 in fasting c-peptide Continuous ANCOVA Change from baseline (week 0) to week 56 in beta cell function (HOMA) Continuous ANCOVA Change from baseline (week 0) to week 56 in insulin resistance (HOMA) Continuous ANCOVA Change from baseline (week 0) to week 56 in OGTT parameters Continuous ANCOVA Change from baseline (week 0) to week 56 in systolic blood pressure Continuous ANCOVA Change from baseline (week 0) to week 56 in diastolic blood pressure Continuous ANCOVA Change from baseline (week 0) to week 56 in hscrp Continuous ANCOVA Change from baseline (week 0) to week 56 in adiponectin Continuous ANCOVA Change from baseline (week 0) to week 56 in fibrinogen Continuous ANCOVA Change from baseline (week 0) to week 56 in PAI-1 Continuous ANCOVA Change from baseline (week 0) to week 56 in total cholesterol Continuous ANCOVA Change from baseline (week 0) to week 56 in LDL cholesterol Continuous ANCOVA Change from baseline (week 0) to week 56 in HDL cholesterol Continuous ANCOVA Change from baseline (week 0) to week 56 in vldl cholesterol Continuous ANCOVA Change from baseline (week 0) to week 56 in triglycerides Continuous ANCOVA Change from baseline (week 0) to week 56 in free fatty acids Continuous ANCOVA Change from baseline (week 0) to week 56 in urinary albumin to creatinine ratio Continuous ANCOVA Scores from PRO questionnaire IWQoL-Lite at week 56 Continuous ANCOVA Scores from PRO questionnaire SF-36 at week 56 Continuous ANCOVA Scores from PRO questionnaire TRIm-Weight at week 56 Continuous ANCOVA Proportion of subjects with change from baseline (week 0) to week 56 in antihypertensives Categorical LR (lowering/ no change or increase) 2 Proportion of subjects change from baseline (week 0) to week 56 in lipid lowering Categorical LR agents (lowering/ no change or increase) 2 Proportion of subjects change from baseline (week 0) to week 56 in oral Categorical LR antidiabetic drugs (lowering/ no change or increase) 2 Proportion of subjects with type 2 diabetes at week 56 (yes/no) 1 Categorical LR Proportion of subjects with pre-diabetes at week 56 (yes/no) 3 Categorical LR 1 FPG at week 0 will be included as covariate 2 Relevant concomitant medication status (present/absent) will be included as covariate 3 Pre-diabetes status at week 0 will be included as covariate For assessments in the re-randomised treatment period statistical analysis and summaries will be done for non pre-diabetes subjects in the FAS, who are re-randomised at 56 week. Statistical

230 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 111 of 138 analysis models are similar to the main treatment period except week 56 is included as baseline covariate. Treatment comparisons will only be made between treatment groups liraglutide/liraglutide and liraglutide/placebo. Descriptive statistics will be done for all treatment groups. Table 18 3 : Overview of statistical analysis of secondary efficacy endpoints at week 68 Endpoint Endpoint type Statistical analysis Change from baseline (week 0) to week 68 in fasting body weight (kg and %) Continuous N/A Change from week 56 to 68 in fasting body weight (kg and %) Continuous ANCOVA Change from baseline (week 0) to week 68 in waist circumference Continuous N/A Change from week 56 to 68 in waist circumference Continuous ANCOVA Change from baseline (week 0) to week 68 in FPG Continuous N/A Change from week 56 to 68 in FPG Continuous ANCOVA Change from baseline (week 0) to week 68 in systolic blood pressure Continuous N/A Change from week 56 to 68 in systolic blood pressure Continuous ANCOVA Change from baseline (week 0) to week 68 in albumin to creatinine ratio Continuous N/A Change from week 56 to 68 in albumin to creatinine ratio Continuous ANCOVA Change from baseline (week 0) to week 68 in diastolic blood pressure Continuous N/A Change from week 56 to 68 in diastolic blood pressure Continuous ANCOVA Scores from PRO questionnaire IWQoL-Lite week 0 to 68 Continuous N/A Scores from PRO questionnaire IWQoL-Lite week 56 to 68 Continuous ANCOVA Scores from PRO questionnaire SF-36 week 0 to 68 Continuous N/A Scores from PRO questionnaire SF-36 week 56 to 68 Continuous ANCOVA Scores from PRO questionnaire TRIm-Weight week 0 to 68 Continuous N/A Scores from PRO questionnaire TRIm-Weight week 56 to 68 Continuous ANCOVA N/A Not available For assessments of endpoints related to week 160, statistical analyses and summaries will be done for pre-diabetes subjects in the FAS. Statistical analysis models are similar to the week 56 analysis models, except factors related to the stratification factor pre-diabetes is not included in the model as all subjects are pre-diabetic at baseline.

231 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 112 of 138 Table 18 4 Overview of statistical analysis of secondary efficacy endpoints at week 160 Endpoint Endpoint type Statistical analysis Change from baseline (week 0) to week 160 in body weight (% and kg) Continuous ANCOVA Proportion of subjects losing at least 5% of baseline body weight at week 160 Categorical LR Proportion of subjects losing more than 10% of baseline body weight at week 160 Categorical LR Change from baseline (week 0) to week 160 in waist circumference Continuous ANCOVA Change from baseline (week 0) to week 160 in HbA1c Continuous ANCOVA Change from baseline (week 0) to week 160 in FPG Continuous ANCOVA Change from baseline (week 0) to week 160 in fasting insulin Continuous ANCOVA Change from baseline (week 0) to week 160 in fasting c-peptide Continuous ANCOVA Change from baseline (week 0) to week 160 in beta cell function (HOMA) Continuous ANCOVA Change from baseline (week 0) to week 160 in insulin resistance (HOMA) Continuous ANCOVA Change from baseline (week 0) to week 160 in OGTT parameters Continuous ANCOVA Change from baseline (week 0) to week 160 in systolic blood pressure Continuous ANCOVA Change from baseline (week 0) to week 160 in diastolic blood pressure Continuous ANCOVA Change from baseline (week 0) to week 160 in PAI-1 Continuous ANCOVA Change from baseline (week 0) to week 160 in hscrp Continuous ANCOVA Change from baseline (week 0) to week 160 in adiponectin Continuous ANCOVA Change from baseline (week 0) to week 160 in fibrinogen Continuous ANCOVA Change from baseline (week 0) to week 160 in total cholesterol Continuous ANCOVA Change from baseline (week 0) to week 160 in LDL cholesterol Continuous ANCOVA Change from baseline (week 0) to week 160 in HDL cholesterol Continuous ANCOVA Change from baseline (week 0) to week 160 in vldl cholesterol Continuous ANCOVA Change from baseline (week 0) to week 160 in triglycerides Continuous ANCOVA Change from baseline (week 0) to week 160 in free fatty acids Continuous ANCOVA Change from baseline (week 0) to week 160 in albumin to creatinine ratio Continuous ANCOVA Scores from PRO questionnaire IWQoL-Lite at week 160 Continuous ANCOVA Scores from PRO questionnaire SF-36 at week 160 Continuous ANCOVA Scores from PRO questionnaire TRIm-Weight at week 160 Continuous ANCOVA Proportion of subjects with change from baseline (week 0) to week 160 in antihypertensives Categorical LR (lowering/ increase or no change) 1 Proportion of subjects change from baseline (week 0) to week 160 in lipid lowering Categorical LR agents (lowering/ increase or no change) 1 Proportion of subjects change from baseline (week 0) to week 160 in oral antidiabetic drugs (lowering/ increase or no change) 1 Categorical LR 1 Relevant concomitant medication status (present/absent) will be included as covariate

232 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 113 of Analysis of the safety endpoints In the 56 week part as well as the 160 week part all conducted analyses will be done with the aim to compare the liraglutide group to the liraglutide placebo group. In the re-randomised treatment period where subjects are re-randomised to either liraglutide or liraglutide placebo the treatments liraglutide/liraglutide, liraglutide/placebo and placebo/placebo will be tabulated. All analyses and tabulations regarding safety endpoints will be done using the safety analysis set. Unless otherwise specified, all safety assessments will be presented for each of the periods: Safety analysis set from randomisation (week 0) to week 56 Re-randomised subjects from re-randomisation (week 56) to week 70 Pre-diabetes subjects in the safety analysis set from randomisation (week 0) to week 162 Physical examination Physical examination at screening as well as changes in physical examination will be summarised. Hypoglycaemic episodes Hypoglycaemic episodes follow the same definition as AEs for treatment emergence. All episodes reported by subjects with diabetes will be summarized by both the ADA definition and the definition. The categorisations are severe, asymptomatic, probable symptomatic, relative, documented symptomatic (ADA definitions) and major, minor, or symptoms only (Novo Nordisk definition). Frequencies of subjects experiencing treatment emergent hypoglycaemic episodes will be summarised by severity and treatment. Hypoglycaemic episodes not defined as treatment emergent will be presented in a listing. ECG Summary statistics and the frequencies of shifts from baseline to end of treatment will be tabulated for each treatment group. Adverse events Adverse events will be coded using the current version of MedDRA. A treatment emergent adverse event (TEAE) is defined as an event that either: Occurs before randomisation and increases in severity during the treatment period Has onset date on or after the first day of randomised treatment and no later than 14 days after the last day of randomised treatment Treatment emergent adverse events, TEAEs, are summarised descriptively, whereas non-treatment emergent AE s are presented in listings. TEAE data will be displayed in terms of the number of subjects with at least one event (N), the percentage of subjects with at least one event (%), the

233 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 114 of 138 number of events (E) and the event rate per 100 years (R). Furthermore, TEAE data are summarized by seriousness, severity, relation to trial drug, MESI, withdrawal due to AEs and outcome. Summary tables by system organ class and preferred term are made for all TEAEs, serious TEAEs, possible or probably related to trial drug, severe TEAE, treatment emergent MESI, and TEAE occurring in at least 5% of the subjects in any treatment arm. Mental health questionnaires and binge eating scale Results from the questionnaires will be by treatment and week. Pulse Pulse will be summarised and analysed similar to blood pressure. Analysis of laboratory safety parameters Laboratory safety parameters are measured throughout the trials and comprise haematology and, biochemistry as defined in the flowchart. No formal statistical analyses are planned for the laboratory safety parameters. The distribution of each continuous laboratory parameter will be presented using box plots by treatment and week. Continuous laboratory values will be compared to the relevant references ranges and results will be presented as follows: Shift tables for each laboratory parameter will be provided by treatment group. The shift tables will include the number of subjects below, within and above the reference ranges at baseline and after end of treatment For each laboratory parameter the proportion of subjects with laboratory values outside the normal ranges will be tabulated per visit and treatment group For each laboratory parameter individual values outside the reference ranges (abnormal values) will be listed by treatment and subject Categorical laboratory parameters will be summarised with shift tables. Amylase and lipase Tables showing shifts from baseline to highest value in treatment period to UNR, 2UNR or 3UNR will be presented. Mean plots by visit will be presented.

234 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 115 of 138 Number and percentage of subjects with amylase and lipase levels UNR, 2UNR and 3UNR by treatment and visit will be tabulated. At end of treatment (week 56, 68 and 160) and follow-up (week 68) summaries will be presented for both observed and LOCF imputed data. Subjects with values above 2UNR will be presented with spaghetti plots and listings showing medical history of GI AEs, GI AEs and liver lab parameters. Calcitonin Number, percentage and incidence of subjects with persistent (all post baseline measurements) and incidental (at least one post baseline measurement) increases in calcitonin for the criteria below will be tabulated for all subjects, males and females. From baseline <UNR to UNR From baseline <UNR to 1.5UNR From baseline <UNR to 20ng/L From baseline <UNR to 50ng/L From baseline <20ng/L to 20ng/L From baseline <50ng/L to 50ng/L Number and percentage of subjects with calcitonin levels UNR, 1.5UNR, 20ng/L and 50ng/L by treatment and visit will be tabulated. At end of treatment (week 56, 68 and 160) and follow-up (week 68) summaries will be presented for both observed and LOCF imputed data. The distribution of all calcitonin measurements across treatment groups and time will be shown with histograms and corresponding cumulative plots by gender and total for actual levels. Sum curves will be done for baseline (pooled) and week 56, 68 and 160 (LOCF) by treatment. In addition histograms of baseline level (pooled) and change at week 56, 68 and 160 (LOCF) will be done by treatment. A summary table showing number and percentage of observations < and LLOQ, minimum, Q25, median, Q75, maximum and geometric mean will be done by treatment group, gender and week. Geometric means will be plotted by treatment and visit in order to assess the pattern of the longitudinal changes. In addition, a scatter plot of baseline vs. maximum post baseline calcitonin measurement will be done by treatment. Longitudinal changes for subjects with calcitonin 20ng/L will be evaluated with spaghetti plots. The spaghetti plots will follow the above by treatment group and gender.

235 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 116 of 138 Selected listings of subjects with at least one post baseline calcitonin measurement above 20ng/L will be done. The listings will include treatment, age, gender, smoking habits at baseline, risk factor information (use of relevant concomitant medication at time of assessment (proton pump inhibitors and H2 blockers) and medical history of thyroid disorder) and calcitonin measurements. Liraglutide antibodies Frequencies of subjects with liraglutide antibodies will be tabulated by week. Similarly, frequencies of subjects with liraglutide antibodies with neutralizing effect and with cross-reacting liraglutide will be tabulated Interim analysis Not applicable Sequential safety analysis/safety monitoring Sequential safety analysis for medical events of special interest is planned to be monitored by an internal Data Monitoring group. No formal statistical analyses are planned, blinded summary statistics and graphical presentation of data will be the basis of the safety group s decisions Explorative statistical analysis for pharmacogenetics and biomarkers Not applicable PK and/or PD modelling The pharmacokinetics of liraglutide 3.0 mg in steady state conditions will be evaluated using population PK analysis methods. The analyses will be based on plasma samples taken from a subgroup of subjects, using a rich sampling scheme, and plasma concentrations taken from all subjects, using a sparse sampling scheme. A separate modelling plan will be prepared before Database release outlining details of the analysis Health economics and/or subject reported outcome Please refer to section

236 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 117 of Ethics The trial will be conducted in compliance with ICH GCP (32), and applicable regulatory requirements, and in accordance with the Declaration of Helsinki(55). All subjects participating in the trial will receive instruction in a hypocaloric diet calculated to induce a moderate weight loss during the trial, and all subjects will receive instruction and encouragement on regular physical activity to aid in weight management and improvement of risk factors. A standard panel of safety laboratory evaluations will be performed regularly during the trial (including vital signs, ECG, haematology, and biochemistry), and side effects will be monitored closely. In this trial all subjects will be instructed in symptom recognition and handling of hypoglycaemia as well as regular clinic measurement of fasting plasma glucose as precautionary measures. Patients treated with liraglutide should be advised of the potential risk of dehydration in relation to gastrointestinal side effects and take precautions to avoid fluid depletion. The trial drugs may be associated with AEs, but relevant precautions have been implemented in the design and planned conduct of the trial, in order to minimise the risks and inconveniences of participating in the trial. These precautions include thorough information regarding the correct administration of the trial drugs and gradual dose escalation. Furthermore, subjects are fully informed about possible AEs and inconveniences, and will be instructed to contact the Investigator in case of any concerns regarding the trial participation. In order to reduce the level of side effects, liraglutide is gradually escalated up to maximum dose (3.0 mg). If subjects do not tolerate an increase in dose during dose-escalation, the Investigator has the option to individualise the dose escalation with a total delay of up to 7 days. All subjects must be at the target dose of 50 clicks per day (3.0 mg liraglutide or liraglutide placebo) 35 days after randomisation. Few cases of acute pancreatitis (inflammation of the pancreas) presenting with persistent severe abdominal pain (usually accompanied by vomiting) have been reported with liraglutide and exenatide. Post-marketing surveillance identified at least 30 cases of pancreatitis with exenatide.(15). However, a health services registry-based study found no increased frequency of pancreatitis among exenatide users (16). The subjects will have the right to withdraw from the trial at any time, without giving a specific reason. will be entitled to keep the data collected until withdrawal of the subject.

237 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 118 of 138 Liraglutide causes dose-dependent and treatment-duration-dependent thyroid C-cell tumors at clinically relevant exposures in both genders of rats and mice. It is unknown whether liraglutide causes thyroid C-cell tumors, including medullary thyroid carcinoma (MTC), in humans, as human relevance could not be ruled out by clinical or nonclinical studies. Liraglutide is contraindicated in patients with a personal or family history of MTC and in patients with Multiple Endocrine Neoplasia syndrome type 2 (MEN 2). Based on the findings in rodents, monitoring with serum calcitonin or thyroid ultrasound was performed during clinical trials, but this may have increased the number of unnecessary thyroid surgeries. It is unknown whether monitoring with serum calcitonin or thyroid ultrasound will mitigate human risk of thyroid C-cell tumors. In a 2-year repeat subcutaneous dose carcinogenicity study of liraglutide injected once a day in CD- 1 mice, a treatment-related increase in fibrosarcomas was seen on the dorsal skin and subcutis, the body surface used for drug injection, in males in the 3 mg/kg/day group. These fibrosarcomas were attributed to the high local concentration of drug near the injection site. The liraglutide concentration in the clinical formulation (6 mg/ml) is 10 times higher than the concentration in the formulation used to administer 3 mg/kg/day liraglutide to mice in the carcinogenicity study (0.6 mg/ml). Further information can be obtained in the Liraglutide Investigator's Brochure Obesity, Edition 3, 08 October 2010 or any updates hereof Informed consent form for trial subjects In seeking and documenting informed consent, the Investigator must comply with the applicable regulatory requirement(s) and adhere to the ICH GCP(32) and the requirements in the Declaration of Helsinki(55). Prior to any trial-related activity, the Investigator must give the subject oral and written information about the trial in a form that the subject can read and understand. A voluntary, signed and personally dated, informed consent form will be obtained from the subject prior to any trial-related activity. The responsibility for seeking informed consent must remain with the Investigator or an adequately medically qualified person delegated by the Investigator. The written informed consent must be signed and personally dated by the person who seeks the informed consent. If information becomes available that may be relevant to the subject s willingness to continue participating in the trial, the Investigator must inform the subject in a timely manner, and a revised written informed consent must be obtained.

238 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 119 of Data handling If the subject withdraws the previously given informed consent the subject s data will be handled as follows: Data collected will be retained by and entered into the database Safety events will be reported to the department responsible for global product safety, Novo Nordisk/regulatory authorities according to local/national requirements If data is used, it will always be in accordance with local law and Institutional Review Board (IRB)/Independent Ethics Committee (IEC) procedures Institutional review boards/independent ethics committee Prior to commencement of the trial, the protocol, any amendments, subject information/informed consent form, any other written information to be provided to the subject, subject recruitment procedures, if any, IB, information about payments and compensation available to subjects if not mentioned in the subject information, the Investigator s current CV and/or other documentation evidencing qualifications, and other documents as required by the local IRB/IEC should be submitted. The submission letter should clearly identify (by trial identification number, including version, EudraCT no., if applicable, title and/or date of the document) which documents have been submitted to the IRB/IEC. Written approval/favourable opinion must be obtained from IRB/IEC prior to commencement of the trial. During the trial, the Investigator must promptly in accordance with local requirements report the following to the IRB/IEC: updates to IB, unexpected SAEs where a causal relationship cannot be ruled out, substantial amendments to the protocol, non-substantial amendments according to local requirements, deviations to the protocol implemented to eliminate immediate hazards to the subjects, new information that may affect adversely the safety of the subjects or the conduct of the trial (including new risk/benefit analysis in case it will have an impact on the planned follow-up of the subjects), annually written summaries of the trial status and other documents as required by the local IRB/IEC. Substantial amendments must not be implemented before approval/favourable opinion, unless necessary to eliminate hazards to the subjects. The Investigator must maintain an accurate and complete record of all submissions made to the IRB/IEC. The records should be filed in the investigator s trial file and copies must be sent to Novo Nordisk.

239 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 120 of Regulatory authorities Regulatory authorities will receive the Clinical Trial Application (CTA), substantial/non-substantial amendments to the protocol, reports on SAEs, and the clinical trial report (CTR) according to national requirements.

240 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 121 of Premature termination of the trial/trial site, Investigator or a pertinent regulatory authority may decide to stop the trial/trial site or part of the trial at any time but agreement on procedures to be followed must be obtained. If a trial is prematurely terminated or suspended, the Investigator should promptly inform the subjects and assure appropriate therapy and follow-up. Furthermore, the Investigator and/or Novo Nordisk should promptly inform the IEC/IRB and provide a detailed written explanation. The pertinent regulatory authorities should be informed according to national regulations. If after the termination of the trial the risk/benefit analysis has changed, the new evaluation should be provided to the IEC/IRB in case it will have an impact on the planned follow-up of the subjects who have participated in the trial. If so, the actions needed to protect the subjects should be described.

241 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 122 of Protocol compliance Deviations from the protocol should be avoided. If deviations occur, the Investigator must inform the monitor and the implications of the deviation must be reviewed and discussed. Protocol deviations must be documented stating the reason, date, the action(s) taken, and the impact for the subjects and/or the trial except for protocol deviations where no corrections are required as described in the trial specific validation checks in the approved Trial Validation Plan (TVP). The Investigator must approve these as outlined in the TVP. The documentation for the protocol deviations must be kept in the Investigator s trial file and Novo Nordisk s trial master file Audits and inspections Any aspect of the clinical trial may be subject to audits conducted by internal Quality Audit System or an inspection from domestic or foreign regulatory authorities. The Investigator and the site staff as well as clinical staff have an obligation to cooperate and assist in such audits and inspections. This includes giving Auditors and Inspectors direct access to all source documents and other documents relevant to the conduct of the clinical trial at the site.

242 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 123 of Critical documents Before the Investigator starts the trial (ie the site has green light for screening subjects), the following documents must be available to : Regulatory approval and/or notification as required Curricula vitae of Investigator and Sub-investigator(s) (current, dated and signed and/or supported by an official regulatory document) Signed and dated agreement on the final protocol Signed and dated agreement on any substantial amendment(s), if applicable Approval/favourable opinion from IEC/IRB clearly identifying the documents reviewed: the protocol, any substantial amendments, subject information/informed consent form and any other written information to be provided to the subject, subject recruitment procedures Copy of IEC/IRB approved subject information/informed consent form/any other written information/advertisement List of IEC/IRB members/constitution Signed receipt of IB by Investigator Other critical documents as required by local regulations Financial agreement(s) For US: Verification under disclosures per Code of Federal Regulations (CFR) of Financial Conflict of Interest (56) FDA financial disclosure form or local equivalent as applicable. Signed and dated FDA form 1572 for each US Investigator (listing individual US clinical trial staff if directly involved in the treatment or evaluation of research making a direct and significant contribution to the data). Protocol NN (US sites): Intended for US sites Conducted under the Investigational New Drug Application (IND) All US Investigators will sign FDA Form 1572 Protocol NN (sites outside the US): Intended for participating sites outside the US Not conducted under the IND All Investigators outside the US will not sign FDA Form 1572 As documented in writing by protocol signature, all Investigators will fully comply with ICH GCP (32), applicable regulatory requirements, and in accordance with the Declaration of Helsinki (57). will analyse and report data from all sites together.

243 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 124 of Responsibilities All staff, site, Central Laboratories, CRO etc. must conduct the trial in compliance with ICH GCP (32), applicable regulatory requirements, and in accordance with the Declaration of Helsinki (57). will be responsible for the preparation of the protocol, ecrf, supply of trial products and stated equipment, monitoring, data management, statistics, and the CTR as documented by procedures and internal specific agreements as well as the current GCP guidelines will provide a system for EDC. This system and support services to the system will be supplied by a clinical services vendor. The activities of the clinical services vendor will be under the direction and supervision of. Furthermore, will be responsible for the IV/WRS. A central laboratory will be responsible for providing all lab supplies for the analysis of all blood samples taken during the trial. All results are received as paper copies at the sites as well as electronic transfer to clinical database. The name of the Central Laboratories will appear in the application to the authorities and in protocol Attachment I. A qualified physician, who is an Investigator or a Sub-investigator for the trial, should be responsible for all trial-related medical decisions. The Investigator is accountable for the conduct of the trial. If any tasks are delegated, the Investigator should maintain a list of appropriately qualified persons to whom he/she has delegated specified significant trial-related duties. In case the Investigator is not able to fulfil the role as Investigator (e.g. retirement), a new Investigator must be appointed in collaboration with. The Investigator will ensure that the last samples are shipped to the central laboratory within 24 hours after the last subject visit at the site. The Investigator will follow the instructions from when processing data. The Investigator will take all necessary technical and organisational safety measures to prevent accidental or wrongful destruction, loss or deterioration of data. The Investigator will prevent any unauthorised access to data or any other processing of data against applicable law.

244 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 125 of 138 Upon request from, the Investigator will provide with the necessary information to enable to ensure that such technical and organisational safety measures have been taken. 24 Reports and publications The information obtained during the conduct of this trial is considered confidential and can be used by for regulatory purposes and for the general development of the trial product. All information supplied by in connection with this trial shall remain the sole property of and is to be considered confidential information. No confidential information shall be disclosed to others without prior written consent from. Such information shall not be used except in the performance of this trial. The information obtained during this trial may be made available to other physicians who are conducting other clinical trials with the trial product, if deemed necessary by. The signatory Investigator Dr Pi-Sunyer will review and sign the Clinical Trial Report to confirm, to the best of his knowledge, that it accurately describes the conduct and results of the trial. As the Clinical Trial Report contains information that will unblind the treatment allocation to the investgator, these parts (e.g. listings, narratives) will be removed from the CTR before Dr Pi- Sunyer's review of the Clinical Trial Report Communication and publication No permission to publish shall be granted to any clinical research organisation involved in the trial described in this protocol. commits to communicating or otherwise making available for public disclosure results of trials regardless of outcome. Public disclosure includes publication of a paper in a scientific journal, abstract submission with a poster or oral presentation at a scientific meeting, or by other means. When the primary results are available, plans to discuss the interpretation of these with the principal Investigator, but reserves the right to release results that may impact Novo Nordisk financial expectations (e.g. a press release directly to the public or similar) without prior consultation with the remaining participating Investigators. reserves the right not to release data until specified milestones, e.g. a clinical trial report is available. This includes the right not to release interim results from clinical trials, because such information can invalidate the results of the entire trial.

245 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 126 of 138 At the end of the trial, one or more manuscripts for publication will be prepared collaboratively between Investigator(s) and. reserves the right to postpone publication and/or communication for less than 60 days to protect intellectual property Authorship Authorship of publications should be in accordance with guidelines from The International Committee of Medical Journal Editors Uniform Requirements (sometimes referred to as the Vancouver Criteria(58)). The signatory Investigator will together with establish and appoint members to a publication writing group consisting of minimum 3 Investigators that have been involved in the design of the trial or the interpretation of the results and minimum 3 Investigators considered to have contributed substantially to the acquisition of data. The remaining participating Investigators will be acknowledged as the NN trial group in the author section of publications Publication(s) The results of this trial will be subject to public disclosure at external web sites according to international regulations, which is reflected in Code of Conduct for Clinical Trial Disclosure. In all cases, the trial results shall be reported in an objective, accurate, balanced and complete manner, with a discussion of the strengths and limitations of the trial. All authors will be given the relevant statistical tables, figures, and reports needed to support the planned publication. In the event of any disagreement about the content of any publication, both the Investigators and Novo Nordisk s opinions shall be fairly and sufficiently represented in the publication. maintains the right to be informed of any Investigator plans for publication and to review any scientific paper, presentation, communication or other information concerning the investigation described in this protocol. Any such communication must be submitted in writing to the trial manager prior to submission for comments. Comments will be given within four weeks from receipt of the planned communication Site-specific publication(s) by Investigator(s) For a multi-centre clinical trial, analyses based on single-site data usually have significant statistical limitations and frequently do not provide meaningful information for healthcare professionals or subjects, and therefore may not be supported by. It is s policy that such individual reports do not precede the primary manuscript and should always reference the primary manuscript of the trial.

246 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 127 of 138 reserves the right to prior review of such publication and to ask for deferment of publication of individual site results until after the primary manuscript is accepted for publication Investigator access to data and review of results As owners of the trial database, has discretion to determine who will have access to the database. Generally, trial databases are only made available to regulatory authorities. Individual Investigator(s) will have their own research participants' data and will be provided with the randomisation code after results are available. For Investigators participating in the 160 week trial, randomisation codes will not be provided until results for the 160 week trial are available.

247 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 128 of Retention of clinical trial documentation Subject notes must be kept for the maximum period permitted by the hospital, institution or private practice. The Investigator must agree to archive the documentation (this includes both electronic and paper based records) pertaining to the trial in an archive after completion or discontinuation of the trial if not otherwise notified. The Investigator should not destroy any documents without prior permission from. If the investigator cannot archive the documents at the trial site after trial completion, can refer the Investigator to an independent archiving provider who has a system in place that allows only the Investigator to access the files. The Investigator must be able to get hold of his/her trial documents without involving Novo Nordisk in any way Clinical trial documentation must be retained until at least 2 years after the last approval of a marketing application in an ICH region and until there are no pending or contemplated marketing applications in an ICH region or at least 2 years have elapsed since the formal discontinuation of clinical development of the investigational product. will maintain s documentation pertaining to the trial as long as the product is on the market plus 20 years. The files from the Investigator site/institution will be retained 15 years after the completion of the trial, or longer if required by national regulations. For Spain the following applies: Any record of the participants will be kept confidential, in accordance with Organic Act. 15/1999 of 13 December of Personal Data/Records Protection..

248 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 129 of Indemnity statement carries product liability for its products and liability assumed under the special laws, acts and/or guidelines for conducting clinical trials in any country, unless others have shown negligence. assumes no liability in the event of negligence or any other liability by the clinics or doctors conducting experiments or by persons for whom the said clinic or doctors are responsible. accepts liability in accordance with: Germany: accepts liability in accordance with DE law: Drug law dated August 24, 1976, last amended per fifteenth law for amendment of the drug law dated July 17, France: In accordance with the French Public Health Code article L (law n of 9 August 2004 art. 88 I,IX Journal Officiel of 11 August 2004). "The sponsor is responsible for identification of the harmful consequences of the biomedical-cal research for the person lending himself thereto and for indemnification of his beneficiaries, except in case of proof, incumbent on it, that the prejudice is not attributable to his fault or to the fault of any intervening party, without the sponsor's being entitled to call on acts by a third party or the voluntary withdrawal of the person who had initially consented to cooperating in the research." Austria: Arzneimittelgesetz (AMG 1983), Änderung idf: BGBl. 1 Nr. 153/2005 The Netherlands: Wetgeving betreffende geneesmiddelen; geneesmiddelenwet 1 juli 2007 (Medicines Law, 1 July 2007). De Wet Medisch-wetenschappelijk Onderzoek met mensen (WMO), 1 maart 2006 (Medical Research Involving Human Subjects Act, 1 March 2006). Besluit van 23 juni 2003, houdende regels inzake de verplichte verzekering bij medischwetenschappelijk onderzoek met mensen (Decree of 23 June 2003, containing rules for compulsory insurance in medical research involving human subjects (Medical Research (Human Subjects) Compulsory Insurance Decree). Spain: accepts liability in accordance with: Spain: Royal Decree 223 of 6 th February 2004 establishing the requisites concerning clinical trial drugs. Switzerland: Federal Law on Therapeutic Products of 15 December 2000 (HMG/LPTh) and Ordinance on clinical trials with therapeutic products (Oclin/Vklin) of 17 October Belgium: Law concerning experiments on the human person of 7 May 2004 Poland: carries liability for the Study in the scope defined by the applicable laws and in particular by the Civil Code and the Pharmaceutical Law dated 6 September 2001 (uniform

249 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 130 of 138 version Journal of Laws of 2008 no. 45 item 271 with amendments). In order to support potential claims for liability attributable to the Study, and Investigator are covered by the Insurance Policy issued according to applicable Polish law. Australia: accepts liability in accordance with: Medicines Australia Guidelines for Compensation for Injury Resulting from Participation in a Company-Sponsored Clinical Trial, dated 16 January 2004.

250 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 131 of 138 References Reference List (1) Haslam DW, James WPT. Obesity. Lancet 2005;366(9492): (2) International O. Global Prevalence of Adult Obesity esityfeb2009v2.pdf. Ref Type: Generic (3) Wing RR. Weight loss in the management of type 2 diabetes. In: Gerstein HC, Haynes RB, editors. Evidence-based Diabetes Care.Ontario, Canada: B.C. Decker, Inc.; p (4) Maggio CA, Pi-Sunyer FX. The prevention and treatment of obesity. Application to type 2 diabetes. Diabetes Care 1997;20(11): (5) Pi-Sunyer FX. Weight loss and mortality in type 2 diabetes. Diabetes Care 2000;23(10): (6) Buteau J, Roduit R, Susini S, Prentki M. Glucagon-like peptide-1 promotes DNA synthesis, activates phosphatidylinositol 3-kinase and increases transcription factor pancreatic and duodenal homeobox gene 1 (PDX-1) DNA binding activity in beta (INS-1)-cells. Diabetologia 1999;42(7): (7) Edvell A, Lindström P. Initiation of increased pancreatic islet growth in young normoglycemic mice (Umeå +/?). Endocrinology 1999;140(2): (8) Xu G, Stoffers DA, Habener JF, Bonner-Weir S. Exendin-4 stimulates both beta-cell replication and neogenesis, resulting in

251 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 132 of 138 increased beta-cell mass and improved glucose tolerance in diabetic rats. Diabetes 1999;48(12): (9) Flint A, Raben A, Astrup A, Holst JJ. Glucagon-like peptide 1 promotes satiety and suppresses energy intake in humans. J Clin Invest 1998 Feb 1;101(3): (10) Gutzwiller JP, Goke B, Drewe J, Hildebrand P, Ketterer S, Handschin D, et al. Glucagon-like peptide-1: a potent regulator of food intake in humans. Gut 1999;44(1):81-6. (11) Naslund E, Barkeling B, King N, Gutniak M, Blundell JE, Holst JJ, et al. Energy intake and appetite are suppressed by glucagon-like peptide-1 (GLP-1) in obese men. Int J Obes Relat Metab Disord 1999 Mar;23(3): (12) Zander M, Madsbad S, Madsen JL, Holst JJ. Effect of 6-week course of glucagon-like peptide 1 on glycaemic control, insulin sensitivity, and beta-cell function in type 2 diabetes: a parallel-group study. Lancet 2002 Mar 9;359(9309): (13) Holst JJ. Enteroglucagon. Annu Rev Physiol 1997;59: (14) Knudsen LB, Nielsen PF, Huusfeldt PO, Johansen NL, Madsen K, Pedersen FZ, et al. Potent derivatives of glucagon-like peptide-1 with pharmacokinetic properties suitable for once daily administration. J Med Chem 2000 May 4;43(9): (15) Ahmad SR, Swann J. Exenatide and rare adverse events. N Engl J Med 2008 May 1;358(18): (16) Dore DD, Seeger JD, Chan KA. Use of a claims-based active drug safety surveillance system to assess the risk of acute pancreatitis with exenatide or sitagliptin compared to metformin or glyburide. Current Medical Research and Opinion 2009 Apr;25(4):

252 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 133 of 138 (17) DPP Research Group. Reduction in the Incidence of Type 2 Diabetes with Lifestyle Intervention or Metformin. 346, NEJM. Ref Type: Generic (18) Gerstein HC et al. Effect of rosiglitazone on the frequency of diabetes in patients with impaired glucose tolerance or impaired fasting glucose: a randomised controlled trial. 23, Lancet. Ref Type: Generic (19) Thorgersson JS et al. XENical in the prevention of diabetes in obese subjects (XENDOS) study: a randomized study of orlistat as an adjunct to lifestyle changes for the prevention of type 2 diabetes in obese patients. 27, Diabetes Care. Ref Type: Generic (20) Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28(7): (21) ADA. Standard of Medical Care in Diabetes Diabetes Care 33, supp Ref Type: Generic (22) ADA. Standard of Medical Care in Diabetes Diabetes Care 33, supp Ref Type: Generic (23) FAO/WHO/UNU. Human energy requirements. Report of a joint FAO/WHO/UNU expert consultation. FAO: food and nutrition technical report series 1. Rome: FAO/WHO/UNU; (24) Torgerson JS, Hauptman J, Boldrin MN, Sjostrom L. XENical in the prevention of diabetes in obese subjects (XENDOS) study: a

253 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 134 of 138 randomized study of orlistat as an adjunct to lifestyle changes for the prevention of type 2 diabetes in obese patients. Diabetes Care 2004;27(1): (25) NCEP report. Implications of recent clinical trials for the national cholesterol education program Adult Treatment Panel III guidelines Endorsed by the National Heart, Lung, and Blood Institute, American College of Cardiology Foundation, and American Heart Association, Circulation. 110, Ref Type: Generic (26) NCEP report. Definition of Metabolic Syndrome, report of the National Heart, Lung, and Blood Institute/American Heart Association conference on scientific issues related to definition Circulation. 109, Ref Type: Generic (27) National Heart LaBINHBPEP. The seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. 289(19), JAMA. Ref Type: Generic (28) ADA. Standard of Medical Care in Diabetes Diabetes Care 33, supp Ref Type: Generic (29) ADA. Standard of Medical Care in Diabetes Diabetes Care 33, supp Ref Type: Generic (30) De AC, Drazen JM, Frizelle FA, Haug C, Hoey J, Horton R, et al. Clinical trial registration: a statement from the International Committee of Medical Journal Editors. New Engl J Med 2004;351(12):

254 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 135 of 138 (31) Food and Drug Administration Amendments Act of Ref Type: Generic (32) International Conference on Harmonisation. ICH Harmonised Tripartite Guideline. Good Clinical Practice. 01 May Ref Type: Generic (33) Food and Drug Administration. Food and Drug Administration Amendments Act of fda gov/regulatoryinformation/legislation/federalfooddrugandcosmet icactfdcact/significantamendmentstothefdcact/foodanddruga dministrationamendmentsactof2007/default htm 2007Available from: URL: /FederalFoodDrugandCosmeticActFDCAct/SignificantA mendmentstothefdcact/foodanddrugadministrationa mendmentsactof2007/default.htm (34) ADA. Standard of Medical Care in Diabetes Diabetes Care 33, supp Ref Type: Generic (35) Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985 Jul;28(7): (36) ADA. Standard of Medical Care in Diabetes Diabetes Care 33, supp Ref Type: Generic (37) ADA. Standard of Medical Care in Diabetes Diabetes Care 33, supp Ref Type: Generic

255 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 136 of 138 (38) Kolotkin RL, Crosby RD, Kosloski KD, Williams GR. Development of a brief measure to assess quality of life in obesity. Obes Res 2001;9(2): (39) Ware JE, Jr., Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992 Jun;30(6): (40) Turner-Bowker DM BPWJJ. SF-36 Health Survey & "SF" Bibliography Lincoln, RI: QualityMetric Incorporated. Ref Type: Generic (41) Ware JE KMDJ. How to Score Version Two of the SF-36 Health Survey Ref Type: Generic (42) Brod et al. Understanding and assessing the impact of prescription weight loss medication; Conceptual, gender and cultural issues. Poster at ISPOR 2008" and binge eating scale (BES) Ref Type: Generic (43) Meryl Brod MHNKSLDMB. Development and validation of the Treatment Related Impact Measure of Weight (TRIM-Weight). Health and Quality of Life Outcomes Ref Type: Generic (44) Lindström J TJ. The Diabetes Risk Score: A practical tool to predict type 2 diabetes risk. Diabetes Care. 26, Ref Type: Generic (45) ADA - American Diabetes Association. Defining and reporting hypoglycemia in diabetes: a report from the American Diabetes Association Workgroup on Hypoglycemia. Diabetes Care 2005 May;28(5):

256 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 137 of 138 (46) Posner K, Oquendo MA, Gould M, Stanley B, Davies M. Columbia Classification Algorithm of Suicide Assessment (C-CASA): classification of suicidal events in the FDA's pediatric suicidal risk analysis of antidepressants. Am J Psychiatry 2007 Jul;164(7): (47) Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 2001 Sep;16(9): (48) Gormally J, Black S, Daston S, Rardin D. The assessment of binge eating severity among obese persons. Addict Behav 1982;7(1): (49) Gladis MM, Wadden TA, Foster GD, Vogt RA, Wingate BJ. A comparison of two approaches to the assessment of binge eating in obesity. Int J Eat Disord 1998 Jan;23(1): (50) d'amore A, Massignan C, Montera P, Moles A, De LA, Scucchi S. Relationship between dietary restraint, binge eating, and leptin in obese women. Int J Obes Relat Metab Disord 2001 Mar;25(3): (51) Food and Drug Administration. Standardized Definitions for Cardiovascular Outcomes Trials: Draft Recommendations. Division of Metabolism and Endocrinology Products, Center for Drug Evaluation and Research (CDER), 22-July Rockville, MD; 2009 Jul Ref Type: Generic (52) Food and Drug Administration. Guidance for Industry: Information Program on Clinical Trials for Serious or Life-Threatening Diseases and Conditions Ref Type: Generic (53) EMEA, Committee for medicinal products for human use (CHMP). Guideline on clinical evaluation of medicinal products used in weight control Nov 15.

257 Revised Protocol-supersedes version 3 Date: 10 December 2010 Version: 4.0 Page: 138 of 138 (54) Rubin DB. Multiple Imputation for Nonresponse in Surveys, New York: John Wiley & Sons, Inc Ref Type: Generic (55) World Medical Association. Declaration of Helsinki. Ethical Principles for Medical Research Involving Human Subjects. 59th WMA General Assembly, Seoul Oct 1. (56) Food and Drug Administration. Food and Drug Administration: Code of Federal Regulations, 21 CFR Part 54, Financial Disclosure by Clinical Investigators. accessdata fda gov/scripts/cdrh/cfdocs/cfcfr/cfrsearch cfm?cfrpart=54&showfr=1 2006Available from: URL: h.cfm?cfrpart=54&showfr=1 (57) World Medical Association. Declaration of Helsinki. Ethical Principles for Medical Research Involving Human Subjects. 52nd WMA General Assembly, Edinburgh, Scotland, October Last amended with Note of Clarification on Paragraph 29 by the WMA General Assembly, Washington 2002, and Note of Clarification on Paragraph 30 by the WMA General assembly, Tokyo Ref Type: Generic (58) International Committee of Medicinal Journal Editors. Uniform Requirements for Manuscripts submitted to Biomedical Journals. 336, N Engl J Med. Ref Type: Generic

258 Revised Protocol - Appendix A EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 1 of 2 Appendix A Approval of final revised protocol This confidential document is the property of. No unpublished information contained herein may be disclosed without prior written approval from. Access to this document must be restricted to relevant parties.

259 Revised Protocol - Appendix A EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 2 of 2 Approval of final revised protocol novodocs object ID of final revised protocol: 0900c76e8123f07e Protocol originator: Trine Nielsen, International Trial Manager Specialist Name & title (printed) Signature Date Head of originating department: Tina Hjorth, Head of Clin Ops Lira and Obesity Name & title (printed) Signature Date Responsible statistician: Tu Duyen Le Thi Statistician Name & title (printed) Signature Date Responsible clinician: Christine Bjørn Jensen International Medical Director Name & title (printed) Signature Date

260 A/S 1839-revised appendix A Electronic Signature Page 14 Dec :26:38 Documentum ObjectID : 0900c76e8123f10f This is a representation of an electronic record that was signed electronically and this page is the manifestation of the electronic signature. Document signed by: Initials Full Name Capacity Reason Date and Time of Signature (Server Time) tnel Trine Nielsen Trial Manager Author :20:21 tmoy Bettina Hjorth Manager Approval :12:36 cbjj Christine Bjørn Jensen Medical Responsible Approval :36:08 rtp Rene Tabanera Palacios Biostatistician Approval :57:19

261 Revised Protocol - Appendix B EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 1 of 2 Appendix B Agreement on the final revised protocol This confidential document is the property of. No unpublished information contained herein may be disclosed without prior written approval from. Access to this document must be restricted to relevant parties.

262 Revised Protocol - Appendix B EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 2 of 2 Agreement on the final revised protocol Trial ID:NN The Investigator and agree to conduct the trial as outlined in this protocol with reference to national/local regulations and in accordance with current Good Clinical Practice (GCP) guidelines. Any modification to the protocol must be agreed upon by both the Investigator and and documented in writing. By written agreement to this protocol, the Investigator agrees to allow direct access to all documentation, including source data, to authorised individuals representing (including monitoring staff and auditors), to institutional review boards/independent ethics committees (IEC/IRB) and/or to regulatory authorities. Investigator: Name (printed) Signature Date Head of medical/clinical research or designee: Name (printed) Signature Date

263 Revised Protocol - Appendix C EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 1 of 2 Appendix C New York Heart Association (NYHA) Criteria for Functional Capacity This confidential document is the property of. No unpublished information contained herein may be disclosed without prior written approval from. Access to this document must be restricted to relevant parties.

264 Revised Protocol - Appendix C EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 2 of 2 Criteria for Functional Capacity Functional Capacity Class I. Patients with cardiac disease but without resulting limitation of physical activity. Ordinary physical activity does not cause undue fatigue, palpitation, dyspnea, or anginal pain. Objective Assessment A. No objective evidence of cardiovascular disease. Class II. Patients with cardiac disease resulting in slight limitation of physical activity. They are comfortable at rest. Ordinary physical activity results in fatigue, palpitation, dyspnea, or anginal pain. B. Objective evidence of minimal cardiovascular disease. Class III. Patients with cardiac disease resulting in marked limitation of physical activity. They are comfortable at rest. Less than ordinary activity causes fatigue, palpitation, dyspnea, or anginal pain. C. Objective evidence of moderately severe cardiovascular disease. Class IV. Patients with cardiac disease resulting in inability to carry on any physical activity without discomfort. Symptoms of heart failure or the anginal syndrome may be present even at rest. If any physical activity is undertaken, discomfort is increased. D. Objective evidence of severe cardiovascular disease. *The Criteria Committee of the New York Heart Association. Nomenclature and Criteria for Diagnosis of Diseases of the Heart and Great Vessels. 9th ed. Boston, Mass: Little, Brown & Co; 1994:

265 Revised Appendix D EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 1 of 3 Appendix D Instruction for Blood Pressure Measurement This confidential document is the property of. No unpublished information contained herein may be disclosed without prior written approval from. Access to this document must be restricted to relevant parties.

266 Revised Appendix D EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 2 of 3 Instructions for Blood Pressure Measurement For the purpose of standardisation, blood pressure measurements must be taken according to recent recommendations 1,2 : The auscultatory method should be used to measure blood pressure. Caffeine, smoking and physical activity should be avoided at least 30 minutes before measurement. Before performing a blood pressure reading, the patient should be asked to remove all clothing that covers the location of cuff placement. The sleeve should not be rolled up such that it has a tourniquet effect above the blood pressure cuff. The same type of sphygmomanometer should be used throughout the trial. The measurement should be taken in a sitting position, with the legs uncrossed, the back and arm supported. Subject should be sitting for at least 5 minutes before the first reading is taken. Location for the measurement should be the upper arm, with the stethoscope at the elbow crease over the brachial artery. The middle of the cuff on the upper arm should be at the level of right atrium (the mid-point of the sternum). The same arm should be used for blood pressure measurements at all visits. The size of the cuff should be selected so that the bladder of the cuff encircles at least 80% of the arm circumference, and the width of the cuff is at least 40% of the arm circumference. Cuff placement must be preceded with palpation of the brachial artery in the antecubital fossa. The midline of the cuff bladder must be placed over the location of the arterial pulsation. The lower edge of the cuff should be 2-3 cm above the antecubital fossa to allow for stethoscope placement. The cuff should be inflated to at least 30 mmhg above the point at which the radial pulse disappears. The pressure should then be reduced at 2 to 3 mm/second. Korotkoff sounds should be used to measure blood pressure: the onset of phase I (appearance of clear tapping sound corresponding to the appearance of palpable pulse) should indicate the

267 Revised Appendix D EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 3 of 3 systolic blood pressure, while phase V (the disappearance of sounds) should be used for recording diastolic pressure. The measurement must be taken with the precision to the nearest 2 mmhg. Neither the patient nor the observer should talk during the measurement. At least two measurements at intervals of at least 2 minutes should be performed. In case of >5 mmhg difference between the first and the second reading of diastolic blood pressure, one additional reading should be obtained. References: 1. Recommendations for Blood Pressure Measurement in Humans and Experimental Animals. Part 1: Blood Pressure Measurement in Humans. A Statement for Professionals From the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research. TG Pickering, JE Hall, LJ Appel, BE Falkner, J Graves, MN Hill, DW Jones, T Kurtz, SG Sheps, EJ Roccella. Hypertension 2005; 45: The Seventh Report of the Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure. US Department of Health and Human Services, National Institutes of Health, National Heart, Lung, and Blood Institute. December 2003.

268 Revised Appendix E EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 1 of 11 Protocol Appendix E NN Flow chart Main trial including a 12 week re-randomised treatment period (assessments for subjects with pre-diabetes are addressed in a separate flow chart) This confidential document is the property of. No unpublished information contained herein may be disclosed without prior written approval from. Access to this document must be restricted to relevant parties.

269 Revised Appendix E Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 2 of 11 Screen 1 Screen 2 Randomisation Dose escalation Maintenance period Visit Number c x Weeks in relation to Visit Visit window, days ± 5 ± 3 ± 3 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 SUBJECTS Informed consent X In/exclusion criteria X X X Stratification criteria X 3 Withdrawal criteria X X X X X X X X X X X X X X X Rescue criteria 4 (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) Re-randomisation X Demography X Medical history /Concomitant illness X Signs or symptoms of diabetic complications X History of concomitant X cardiovascular disease History of gallbladder disease X History of psychiatric disorders X Concomitant medication X X X X X X X X X X X X X X X X X X Smoking habits X EFFICACY Body measurements: Body weight X X X X X X X X X X X X X X X X X X 5 Height X Waist X X X X X X X X X X X X X X X X X EOT 1 2

270 Revised Appendix E Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 3 of 11 Screen 1 Screen 2 Randomisation Dose escalation Maintenance period Visit Number c x Weeks in relation to Visit Visit window, days ± 5 ± 3 ± 3 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 circumference Glucose related parameters: HbA 1C X X X X X X X Fasting plasma glucose X 6 X X X X X X X X X X OGTT X 6 X X Fasting insulin X X X X X X C-peptide X X X X X X Urinary Albumin-to- Creatinine Ratio X X X Vital signs: Systolic blood pressure, sitting X X 7 X X X X X X X X X X X X X X X X Diastolic blood pressure, sitting X X 7 X X X X X X X X X X X X X X X X Pulse, sitting X X X X X X X X X X X X X X X X X Cardiovascular biomarkers: hscrp X X X Adiponectin X X X Fibrinogen X X X PAI-1 X X X Lipids: Cholesterol X X X X LDL cholesterol X X X X HDL cholesterol X X X X EOT 1 2

271 Revised Appendix E Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 4 of 11 Screen 1 Screen 2 Randomisation Dose escalation Maintenance period Visit Number c x Weeks in relation to Visit Visit window, days ± 5 ± 3 ± 3 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 VLDL cholesterol X X X X Triglycerides X X X X Free fatty acids X X X X PRO questionnaires 8, 9 X X X 10 OTHER ASSESSEMENTS Liraglutide plasma concentration X X X 11 X FINDRISC score X SAFETY Physical examination X X Binge eating Scale 12 X X 13 ECG 14 X X X Pregnancy test 15 X (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) X Adverse Events X X X X X X X X X X X X X X X X X X 16 Haematology: Haemoglobin X X X X X X X Haematocrit X X X X X X X Thrombocytes X X X X X X X Erythrocytes X X X X X X X Leucocytes X X X X X X X Differential count: X X X X X X X Eosinophils X X X X X X X Neutrophils X X X X X X X Basophils X X X X X X X Monocytes X X X X X X X EOT 1 2

272 Revised Appendix E Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 5 of 11 Screen 1 Screen 2 Randomisation Dose escalation Maintenance period Visit Number c x Weeks in relation to Visit Visit window, days ± 5 ± 3 ± 3 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 Lymphocytes X X X X X X X Biochemistry: Creatinine X X X X X X X Creatine kinase X X X X X X X Urea X X X X X X X Albumin X X X X X X X Bilirubin, total X X X X X X X ALAT X X X X X X X ASAT X X X X X X X Alkaline phosphatase X X X X X X X Amylase X X X X X X X Lipase X X X X X X X Sodium X X X X X X X Potassium X X X X X X X Calcium, total X X X X X X X Calcitonin X X X X X X X TSH X X X X X X X Liraglutide antibodies X X 17 X Mental health 18 X 19 X 19 X X X X X X X X X X X X X X X 20 TRIAL MATERIAL Dispense study card X Dispense pedometer X Dispense 3-day food diary X X X X X X X Diet and physical X 21 X X 21 X X 21 X X 21 X X 21 X X 21 X X X 21 EOT 1 2

273 Revised Appendix E Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 6 of 11 Screen 1 Screen 2 Randomisation Dose escalation Maintenance period Visit Number c x Weeks in relation to Visit Visit window, days ± 5 ± 3 ± 3 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 activity counselling Recording of dietary compliance and X 22 X X X X X X 22 physical activity Dispense dosing diary X X X 23 X Dispense diabetes diary 24 (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) Dispense trial drug X X X X X X X X X X X X X X Drug accountability X X X X X X X X X X X X X IVRS/WRS session X X X X X X X X X X X X X X X X REMINDERS Attend visit fasting X X X X X X X X X X X X EOT 1 2

274 Revised Appendix E Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 7 of 11 Re-rand. treatment period EOT 25 FU Visit Number 18a 19a 20a 21a 22a Weeks in relation to Visit Visit window, days ± 3 ± 5 ± 5 ± 3 ± 3 SUBJECTS Withdrawal criteria X X X Rescue criteria 4 (X) (X) (X) Concomitant medication X X X X X Smoking habits EFFICACY Body measurements: Body weight X X X X X Waist circumference X X X X X Glucose related parameters: HbA 1C X Fasting plasma glucose X X X OGTT X Fasting Insulin X X X C-peptide X X Urinary Albumin-to-Creatinine Ratio X X Vital signs: Systolic blood pressure, sitting X X X X X Diastolic blood pressure, sitting X X X X X Pulse, sitting X X X X X Cardiovascular biomarkers: hscrp X Adiponectin X Fibrinogen X PAI-1 X Lipids: Cholesterol X

275 Revised Appendix E Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 8 of 11 Re-rand. treatment period EOT 25 FU Visit Number 18a 19a 20a 21a 22a Weeks in relation to Visit Visit window, days ± 3 ± 5 ± 5 ± 3 ± 3 LDL-cholesterol X HDL-cholesterol X VLDL-cholesterol X Triglycerides X Free fatty acids X PRO questionnaires 8, 9 X 10 SAFETY Physical examination X Binge eating Scale X ECG X Pregnancy test 15 (X) (X) (X) X Adverse Events X X X X X Haematology: Haemoglobin X X Haematocrit X X Thrombocytes X X Erythrocytes X X Leucocytes X X Differential count: X X Eosinophils X X Neutrophils X X Basophils X X Monocytes X X Lymphocytes X X Biochemistry: Creatinine X X Creatine kinase X X

276 Revised Appendix E Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 9 of 11 Re-rand. treatment period EOT 25 FU Visit Number 18a 19a 20a 21a 22a Weeks in relation to Visit Visit window, days ± 3 ± 5 ± 5 ± 3 ± 3 Urea X X Albumin X X Bilirubin, total X X ALAT X X ASAT X X Alkaline phosphatase X X Amylase X X Lipase X X Sodium X X Potassium X X Calcium, total X X Calcitonin X X TSH X X Liraglutide antibodies X X 17 X Mental health 18 X X X X X TRIAL MATERIAL Dispense 3-day food diary X Diet and physical activity counselling X X 21 X 26 Recording of dietary compliance and physical activity X Dispense diabetes diary 27 (X) (X) (X) Dispense trial drug X X Drug accountability X X X IVRS/WRS session X X X X 28 End of treatment X REMINDERS Attend visit fasting X X X

277 Revised Appendix E Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 10 of 11 1 End of 56 week randomised treatment/premature discontinuation 2 Body weight/mesi recording visit 3 A subject is eligible for stratification to the delayed onset of type 2 diabetes part of the trial if having pre-diabetes by any one or more of the following criteria: Impaired Fasting Glucose, Impaired Glucose Tolerance, or HbA 1c % (both inclusive) 4 Rescue criteria is only relevant for subjects developing diabetes 5 Only for subjects who have discontinued the trial prematurely prior to Visit 17 6 Fasting plasma glucose and OGTT glucose values are used to determine pre-diabetes status and subcategory (Y/N; iifg, iigt, IFG/IGT), and to exclude presence of diabetes 7 Re-measurements are allowed if white coat hypertension is suspected at Screening visit 1 8 PRO assessment as evaluated by IWQoL-Lite, SF-36 and TRIm-Weight. The Investigator or his delegate must review Patient Reported Outcome(s) for completeness and adverse event(s) immediately following administration 9 The assessment is only applicable for subjects in Canada, France, Germany, Italy, Netherlands, Russia, Spain, United Kingdom, Brazil, Austria, Mexico, Australia, Ireland and USA 10 For subjects discontinuing the trial prematurely the assessment should be completed 11 The PK Substudy is only applicable for subjects in Netherlands, United Kingdom and USA 12 The assessment is only applicable for subjects in Canada, France, Germany, Italy, Netherlands, Russia, Spain, United Kingdom, Brazil, Austria, Mexico, Australia, Ireland and USA 13 Only applicable to subjects discontinuing the trial prematurely. All other subjects will complete the binge eating questionnaire at Visit 18a 14 An ECG taken within the period starting at Screening visit 1 and ending at the date of Visit 3 is acceptable 15 Urine pregnancy test will be performed for females of childbearing potential at any time during the trial if a menstrual period is missed or as required by local law, except for screening visit 1, end of 56 week randomised treatment (V17) and end of re-randomised treatment visit (V21a) where the test will be a serum pregnancy test 16 Only recording of MESIs

278 Revised Appendix E Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 11 of Only applicable to subjects discontinuing the trial prematurely. If a subject discontinues the trial within the 56 week main trial liraglutide antibodies will be measured at the end of the 56 week randomised treatment visit (Visit 17) and for subjects discontinuing prematurely after Visit 17 antibodies will be drawn at Visit 21a 18 Mental Health assessed by PHQ-9 and C-SSRS. The Investigator or his delegate must review PHQ-9 for completeness and adverse event(s) immediately following administration 19 The Investigator must assess the PHQ-9 and C-SSRS score at Screening Visit 1 and Visit 3 (randomisation) to exclude subjects with major depression (PHQ-9 15) or any suicidal ideation (of type 4 or type 5) 20 Mental Health questionnaires should be completed for subjects discontinuing the trial prematurely 21 Diet counselling based on a 3-day food diary 22 Only physical activity is recorded at the visit 23 Only for subjects participating in the PK sub-study 24 The diabetes diary will be handed out to subjects developing diabetes 25 Visit 21a is the end of the re-randomised treatment visit 26 Dietary session on future weight management for subjects completing the trial The diabetes diary will only be handed out to subjects developing diabetes Make an IVRS call for subjects completing the trial

279 Revised Appendix F EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 1 of 13 Protocol Appendix F NN Flow chart Delayed-onset-of-type-2-diabetes part of the trial (subjects with pre-diabetes) This confidential document is the property of. No unpublished information contained herein may be disclosed without prior written approval from. Access to this document must be restricted to relevant parties.

280 Revised Appendix F Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 2 of 13 Screening 1 Screening 2 Randomisation Dose escalation Maintenance period Body Weight Visit Number c x Weeks in relation to Visit Visit window, days ± 5 ± 3 ± 3 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 SUBJECTS Informed consent X X In/exclusion criteria X X X Stratification criteria X 1 Withdrawal criteria X X X X X X X X X X X X X X X Rescue criteria 2 (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) Demography X Medical history / Concomitant illness X Signs or symptoms of diabetic complications X History of concomitant cardiovascular disease X History of gallbladder disease X History of psychiatric disorders X Concomitant medication X X X X X X X X X X X X X X X X X X Smoking habits X EFFICACY Body measurements: Body weight X X X X X X X X X X X X X X X X X X 3 Height X Waist circumference X X X X X X X X X X X X X X X X X Glucose related parameters: HbA 1C X X X X X X X Fasting plasma glucose X 4 X X X X X X X X X X

281 Revised Appendix F Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 3 of 13 Screening 1 Screening 2 Randomisation Dose escalation Maintenance period Body Weight Visit Number c x Weeks in relation to Visit Visit window, days ± 5 ± 3 ± 3 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 OGTT X 4 X X Fasting insulin X X X X X X C-peptide X X X X X X Urinary Albumin-to- Creatinine Ratio X X X Vital signs: Systolic blood pressure, sitting X X 5 X X X X X X X X X X X X X X X X Diastolic blood pressure, sitting X X 5 X X X X X X X X X X X X X X X X Pulse, sitting X X X X X X X X X X X X X X X X X Cardiovascular biomarkers: hscrp X X X Adiponectin X X X Fibrinogen X X X PAI-1 X X X Lipids: Cholesterol X X X X LDL cholesterol X X X X HDL cholesterol X X X X VLDL cholesterol X X X X Triglycerides X X X X Free fatty acids X X X X PRO questionnaires 6, 7 X X X OTHER ASSESSEMENTS Liraglutide plasma concentration X X X 8 X

282 Revised Appendix F Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 4 of 13 Screening 1 Screening 2 Randomisation Dose escalation Maintenance period Body Weight Visit Number c x Weeks in relation to Visit Visit window, days ± 5 ± 3 ± 3 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 FINDRISC score X SAFETY Physical examination X X ECG 9 X X X Pregnancy test 10 X (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) Adverse Events X X X X X X X X X X X X X X X X X X 11 Haematology: Haemoglobin X X X X X X X Haematocrit X X X X X X X Thrombocytes X X X X X X X Erythrocytes X X X X X X X Leucocytes X X X X X X X Differential count: X X X X X X X Eosinophils X X X X X X X Neutrophils X X X X X X X Basophils X X X X X X X Monocytes X X X X X X X Lymphocytes X X X X X X X Biochemistry: Creatinine X X X X X X X Creatine kinase X X X X X X X Urea X X X X X X X Albumin X X X X X X X Bilirubin, total X X X X X X X ALAT X X X X X X X ASAT X X X X X X X Alkaline phosphatase X X X X X X X

283 Revised Appendix F Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 5 of 13 Screening 1 Screening 2 Randomisation Dose escalation Maintenance period Body Weight Visit Number c x Weeks in relation to Visit Visit window, days ± 5 ± 3 ± 3 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 Amylase X X X X X X X Lipase X X X X X X X Sodium X X X X X X X Potassium X X X X X X X Calcium, total X X X X X X X Calcitonin X X X X X X X TSH X X X X X X X Liraglutide antibodies X Mental health 12 X 13 X 13 X X X X X X X X X X X X X X X TRIAL MATERIAL Dispense study card X Dispense pedometer X Dispense 3-day food diary X X X X X X X Diet and physical activity counselling X 14 X X 14 X X 14 X X 14 X X 14 X X 14 X X X 14 Recording of dietary compliance and physical X 15 X X X X X X activity Dispense dosing diary X X X 16 X Dispense trial drug X X X X X X X X X X X X X X Drug accountability X X X X X X X X X X X X X IVRS/WRS session X X X X X X X X X X X X X X X X End of trial REMINDERS Attend visit fasting X X X X X X X X X X X X

284 Revised Appendix F Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 6 of 13 Maintenance period Visit Number 18b 19b 20b 21b 22b 23b 24b 25b 26b 27b 28b 29b 30b 31b 32b 33b 34b 35b 36b 37b 38b Weeks in relation to Visit Visit window, days ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 SUBJECTS Withdrawal criteria X X X X X X X X X X X X X X X X X X X X X Rescue criteria 2 (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) Concomitant medication X X X X X X X X X X X X X X X X X X X X X Smoking habits EFFICACY Body measurements: Body weight X X X X X X X X X X X X X X X X X X X X X Waist circumference X X X X X X X X X X X X X X X X X X X X X Glucose related parameters: HbA 1C X X X X X X X Fasting plasma glucose X X X X X X X OGTT X X X Fasting Insulin X X X X X X X C-peptide X X X X X X X Urinary Albumin-to-Creatinine Ratio X X X Vital signs: Systolic blood pressure, sitting X X X X X X X X X X X X X X X X X X X X X Diastolic blood pressure, sitting X X X X X X X X X X X X X X X X X X X X X Pulse, sitting X X X X X X X X X X X X X X X X X X X X X Cardiovascular biomarkers: hscrp X X X Adiponectin X X X Fibrinogen X X X PAI-1 X X X Lipids: Cholesterol X X X LDL-cholesterol X X X

285 Revised Appendix F Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 7 of 13 Maintenance period Visit Number 18b 19b 20b 21b 22b 23b 24b 25b 26b 27b 28b 29b 30b 31b 32b 33b 34b 35b 36b 37b 38b Weeks in relation to Visit Visit window, days ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 HDL-cholesterol X X X VLDL-cholesterol X X X Triglycerides X X X Free fatty acids X X X PRO questionnaires 6, 7 X X X SAFETY Physical examination ECG X X X Pregnancy test 10 (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) Adverse Events X X X X X X X X X X X X X X X X X X X X X Haematology: Haemoglobin X X X X X X X X Haematocrit X X X X X X X X Thrombocytes X X X X X X X X Erythrocytes X X X X X X X X Leucocytes X X X X X X X X Differential count: X X X X X X X X Eosinophils X X X X X X X X Neutrophils X X X X X X X X Basophils X X X X X X X X Monocytes X X X X X X X X Lymphocytes X X X X X X X X Biochemistry: Creatinine X X X X X X X X Creatine kinase X X X X X X X X Urea X X X X X X X X Albumin X X X X X X X X Bilirubin, total X X X X X X X X

286 Revised Appendix F Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 8 of 13 Maintenance period Visit Number 18b 19b 20b 21b 22b 23b 24b 25b 26b 27b 28b 29b 30b 31b 32b 33b 34b 35b 36b 37b 38b Weeks in relation to Visit Visit window, days ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ± 5 ALAT X X X X X X X X ASAT X X X X X X X X Alkaline phosphatase X X X X X X X X Amylase X X X X X X X X Lipase X X X X X X X X Sodium X X X X X X X X Potassium X X X X X X X X Calcium, total X X X X X X X X Calcitonin X X X X X X X X TSH X X X X X X X X Liraglutide antibodies Mental health 12 X X X X X X X X X X X X X X X X X X X X X TRIAL MATERIAL Dispense 3-day food diary X X X X X X X X X X X Diet and physical activity counselling X X 14 X X 14 X X 14 X X 14 X X 14 X X 14 X X 14 X X 14 X X 14 X X 14 X Recording of dietary compliance and physical activity X X X X X X X X X X Dispense diabetes diary 17 (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) (X) Dispense trial drug X X X X X X X X X X X X X X X X X X X X X Drug accountability X X X X X X X X X X X X X X X X X X X X X IVRS/WRS session X X X X X X X X X X X X X X X X X X X X X End of trial REMINDERS Attend visit fasting X X X X X X X X X X X X X X

287 Revised Appendix F Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 9 of 13 Maintenance period EOT FU Visit Number 39b 40b 41b 42b 43b 44b Weeks in relation to Visit Visit window, days ± 5 ± 5 ± 5 ± 5 ± 3 ± 3 SUBJECTS Withdrawal criteria X X X X Rescue criteria 2 (X) (X) (X) (X) Concomitant medication X X X X X X Smoking habits EFFICACY Body measurements: Body weight X X X X X X Waist circumference X X X X X X Glucose related parameters: HbA 1C X X Fasting plasma glucose X X X OGTT X X Fasting Insulin X X X C-peptide X X Urinary Albumin-to-Creatinine ratio X Vital signs: Systolic blood pressure, sitting X X X X X X Diastolic blood pressure, sitting X X X X X X Pulse, sitting X X X X X X Cardiovascular biomarkers: hscrp X X Adiponectin X X Fibrinogen X X PAI-1 X X Lipids: Cholesterol X X

288 Revised Appendix F Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 10 of 13 Maintenance period EOT FU Visit Number 39b 40b 41b 42b 43b 44b Weeks in relation to Visit Visit window, days ± 5 ± 5 ± 5 ± 5 ± 3 ± 3 LDL-cholesterol X X HDL-cholesterol X X VLDL-cholesterol X X Triglycerides X X Free fatty acids X X PRO questionnaires 6, 7 X,18 SAFETY Physical examination X ECG X Pregnancy test 10 (X) (X) (X) (X) X Adverse Events X X X X X X Haematology: Haemoglobin X X Haematocrit X X Thrombocytes X X Erythrocytes X X Leucocytes X X Differential count: X X Eosinophils X X Neutrophils X X Basophils X X Monocytes X X Lymphocytes X X Biochemistry: Creatinine X X Creatine kinase X X Urea X X

289 Revised Appendix F Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 11 of 13 Maintenance period EOT FU Visit Number 39b 40b 41b 42b 43b 44b Weeks in relation to Visit Visit window, days ± 5 ± 5 ± 5 ± 5 ± 3 ± 3 Albumin X X Bilirubin, total X X ALAT X X ASAT X X Alkaline phosphatase X X Amylase X X Lipase X X Sodium X X Potassium X X Calcium, total X X Calcitonin X X TSH X X Liraglutide antibodies X 19 X Mental health 12 X X X X X X TRIAL MATERIAL Dispense 3-day food diary X X Diet and physical activity counselling X 14 X X 14 X X 20 Recording of dietary compliance and physical activity X X X Dispense trial drug X X X X Drug accountability X X X X X IVRS/WRS session X X X X X 21 End of treatment X REMINDERS Attend visit fasting X X X

290 Revised Appendix F Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 12 of 13 1 A subject is eligible for stratification to the delayed onset of type 2 diabetes part of the trial if having pre-diabetes by any one or more of the following criteria: Impaired Fasting Glucose, Impaired Glucose Tolerance, or HbA 1c % (both inclusive) 2 Rescue criteria is only relevant for subjects developing diabetes 3 Only for subjects who have discontinued the trial prematurely prior to Visit 17 4 Fasting plasma glucose and OGTT are used to assess the pre-diabetic (IFG/IGT) status 5 Re-measurements are allowed if white coat hypertension is suspected at Screening visit 1 6 PRO assessment as evaluated by IWQoL-Lite, SF-36 and TRIm-Weight. The Investigator or his delegate must review Patient Reported Outcome(s) for completeness and adverse event(s) immediately following administration 7 The assessment is only applicable for subjects in Canada, France, Germany, Italy, Netherlands, Russia, Spain, United Kingdom, Brazil, Austria, Mexico, Australia, Ireland and USA 8 The PK Sub-study is only applicable for subjects in Netherlands, United Kingdom and USA 9 An ECG taken within the period starting at Screening visit 1 and ending at the date of Visit 3 is acceptable 10 Urine pregnancy test will be performed for females of childbearing potential at any time during the trial if a menstrual period is missed or as required by local law, except for screening visit 1 end of the 160 week randomised treatment (Visit 43b) where the test will be a serum pregnancy test 11 Only recording of MESIs 12 Mental Health assessed by PHQ-9 and C-SSRS. The Investigator or his delegate must review PHQ-9 for completeness and adverse event(s) immediately following administration 13 The Investigator must assess the PHQ-9 and C-SSRS score at Screening Visit 1 and Visit 3 (randomisation) to exclude subjects with major depression (PHQ-9 15) or any suicidal ideation (of type 4 or type 5) 14 Diet counselling based on a 3-day food diary 15 Only physical activity is recorded at the visit 16 Only for subjects participating in the PK sub-study 17 The diabetes diary will only be handed out to subjects developing diabetes

291 Revised Appendix F Date: 10 December 2010 EudraCT No.: Version: 4.0 Page: 13 of For subjects discontinuing the trial prematurely the assessment should be completed 19 Only applicable for subjects discontinuing the trial prematurely (i.e. prior to Visit 43b). For all other subjects liraglutide antibodies will be measured at the Visit 44b 20 Dietary session on future weight management 21 Make IV/WRS completion call for subjects completing the trial

292 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 1 of 16 Appendix G NN Patient Reported Outcome This confidential document is the property of. No unpublished information contained herein may be disclosed without prior written approval from. Access to this document must be restricted to relevant parties.

293 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 2 of 16 SF-36 version 2 Your Health and Well-Being This survey asks for your views about your health. This information will help keep track of how you feel and how well you are able to do your usual activities. Thank you for completing this survey! For each of the following questions, please tick the one box that best describes your answer. 1. In general, would you say your health is: Excellent Very good Good Fair Poor Compared to one year ago, how would you rate your health in general now? Much better now than one year ago Somewhat better now than one year ago About the same as one year ago Somewhat worse now than one year ago Much worse now than one year ago

294 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 3 of The following questions are about activities you might do during a typical day. Does your health now limit you in these activities? If so, how much? Yes, limited a lot Yes, limited a little No, not limited at all a b Vigorous activities, such as running, lifting heavy objects, participating in strenuous sports Moderate activities, such as moving a table, pushing a vacuum cleaner, bowling, or playing golf c Lifting or carrying groceries d Climbing several flights of stairs e Climbing one flight of stairs f Bending, kneeling, or stooping g Walking more than a mile h Walking several hundred yards i Walking one hundred yards j Bathing or dressing yourself

295 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 4 of During the past 4 weeks, how much of the time have you had any of the following problems with your work or other regular daily activities as a result of your physical health? All of the time Most of the time Some of the time A little of the time None of the time a b C d Cut down on the amount of time you spent on work or other activities Accomplished less than you would like Were limited in the kind of work or other activities Had difficulty performing the the work or other activities (for example, it took extra effort) During the past 4 weeks, how much of the time have you had any of the following problems with your work or other regular daily activities as a result of any emotional problems (such as feeling depressed or anxious)? All of the time Most of the time Some of the time A little of the time None of the time a b C Cut down on the amount of time you spent on work or other activities Accomplished less than you would like Did work or other activities less carefully than usual

296 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 5 of During the past 4 weeks, to what extent has your physical health or emotional problems interfered with your normal social activities with family, friends, neighbours, or groups? Not at all Slightly Moderately Quite a bit Extremely How much bodily pain have you had during the past 4 weeks? None Very mild Mild Moderate Severe Very severe During the past 4 weeks, how much did pain interfere with your normal work (including both work outside the home and housework)? Not at all A little bit Moderately Quite a bit Extremely

297 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 6 of These questions are about how you feel and how things have been with you during the past 4 weeks. For each question, please give the one answer that comes closest to the way you have been feeling. How much of the time during the past 4 weeks All of the time Most of the time Some of the time A little of the time None of the time a Did you feel full of life? b Have you been very nervous? c d Have you felt so down in the dumps that nothing could cheer you up? Have you felt calm and peaceful? e Did you have a lot of energy? f Have you felt downhearted and low? g Did you feel worn out? h Have you been happy? i Did you feel tired?

298 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 7 of During the past 4 weeks, how much of the time has your physical health or emotional problems interfered with your social activities (like visiting with friends, relatives, etc.)? All of the time Most of the time Some of the time A little of the time None of the time How TRUE or FALSE is each of the following statements for you?

299 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 8 of 16 IWQoL-Lite Impact of Weight on Quality of Life Please answer the following statements by circling the number that best applies to you in the past week. Be as honest as possible. There are no right or wrong answers. Physical Function ALWAYS TRUE USUALLY TRUE SOMETIMES TRUE RARELY TRUE NEVER TRUE 1. Because of my weight I have trouble picking up objects. 2. Because of my weight I have trouble tying my shoelaces. 3. Because of my weight I have difficulty getting up from chairs. 4. Because of my weight I have trouble using stairs. 5. Because of my weight I have difficulty putting on or taking off my clothes. 6. Because of my weight I have trouble with mobility (getting around). 7. Because of my weight I have trouble crossing my legs. 8. I feel short of breath with only mild exertion (e.g. climbing a single flight of stairs) I am troubled by painful or stiff joints My ankles and lower legs are swollen at the end of the day I am worried about my health

300 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 9 of 16 Self-esteem ALWAYS TRUE USUALLY TRUE SOMETIMES TRUE RARELY TRUE NEVER TRUE 1. Because of my weight I am selfconscious. 2. Because of my weight my self-esteem is not what it could be. 3. Because of my weight I feel unsure of myself. 4. Because of my weight I don t like myself. 5. Because of my weight I am afraid of being rejected. 6. Because of my weight I avoid looking in mirrors or seeing myself in photographs. 7. Because of my weight I am embarrassed to be seen in public places Sexual Life ALWAYS TRUE USUALLY TRUE SOMETIMES TRUE RARELY TRUE NEVER TRUE 1. Because of my weight I do not enjoy sexual activity. 2. Because of my weight I have little or no sexual desire. 3. Because of my weight I have difficulty with sexual performance. 4. Because of my weight I avoid sexual encounters whenever possible

301 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 10 of 16 Public Distress ALWAYS TRUE USUALLY TRUE SOMETIMES TRUE RARELY TRUE NEVER TRUE 1. Because of my weight I experience ridicule, teasing, or unwanted attention. 2. Because of my weight I worry about fitting into seats in public places (e.g. theatres, cinemas, restaurants, cars, or aeroplanes). 3. Because of my weight I worry about fitting through aisles or turnstiles. 4. Because of my weight I worry about finding chairs that are strong enough to hold my weight. 5. Because of my weight I experience discrimination by others Work (Note: For those not in paid employment, answer with respect to your daily activities.) ALWAYS TRUE USUALLY TRUE SOMETIMES TRUE RARELY TRUE NEVER TRUE 1. Because of my weight I have trouble getting things done or carrying out my responsibilities. 2. Because of my weight I am less productive than I could be. 3. Because of my weight I don t receive appropriate pay rises, promotions or recognition at work. 4. Because of my weight I am afraid to go for job interviews

302 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 11 of 16 Treatment Related Impact Measure Weight (TRIM-Weight) The following questions are concerned with your PRESCRIPTION WEIGHT LOSS MEDICATION. If you take medication for other conditions, please think only about the prescription medication you take for weight loss when answering the questions. Please circle the response that most closely represents how you have felt about your weight loss medication over the PAST TWO WEEKS. Please circle only one number for each question. Remember there are no right or wrong answers to these questions. 1. How satisfied or dissatisfied are you with: Not at all satisfied A little satisfied Somewhat satisfied Very satisfied Extremely satisfied a. How well your medication helps you lose weight... b. The ease and convenience of your medication How much of a problem or not a problem is it for you to: Never/ Almost never a problem Rarely a problem Sometimes a problem Often a problem Almost always/ Always a problem a. Remember to take your medication... b. Be as active as you would like (because of your medication)... c. Have side effects from your medication How satisfied or dissatisfied are you with your medication s ability to: a. Control your appetite Not at all satisfied A little satisfied Somewhat satisfied Very satisfied Extremely satisfied How bothered or not bothered are you by: Not at all bothered A little bothered Somewhat bothered Very bothered Extremely bothered a. Weight loss plateaus (periods of no weight loss)... b. Mood swings due to your medication (emotional ups and downs)... c. Being tired or drowsy during the day because of your medication

303 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 12 of How convenient or inconvenient is: Not at all convenient A little convenient Somewhat convenient Very convenient Extremely convenient a. Planning when to take your medication b. The number of times a day you need to take your medication Because of your weight loss medication, how often do you feel: a. Physical discomfort (for example, sleep problems, insomnia, jitteriness, diarrhea, gas, bloating, dry mouth)... b. Emotional distress (for example, anxiety, agitation, irritability, short-tempered)... c. You have problems with mental functioning (for example, concentrating, focusing, being distracted)... Because of your weight loss medication, how often do you feel: Never/ Almost never Rarely Sometimes Often Almost always/ Always Never/ Almost never Rarely Sometimes Often Almost always/ Always d. Worried about long-term side effects... e. Depressed... f. Frustrated... g. Embarrassed... h. Stressed How often does your medication interfere or not interfere with your: Never/ Almost never interferes Rarely interferes Sometimes interferes Often interferes Almost always/ Always interferes a. Social activities (meeting with friends, going out)... b. Being as productive as you would like (either at home or at work)... c. Relationships with friends, family or at work Thank You!

304 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 13 of 16 Binge Eating Scale (BES) Code no. Eating Habits Checklist Instructions: Below are groups of numbered statements. Read all of the statements in each group and check the one that best describes the way you feel about the problems you have controlling your eating behavior. #1 1. I don t feel self-conscious about my weight or body size when I m with others. 2. I feel concerned about how I look to others, but it normally does not make me feel disappointed with myself. 3. I do get self-conscious about my appearance and weight which makes me feel disappointed in myself. 4. I feel very self-conscious about my weight and frequently, I feel intense shame and disgust for myself. I try to avoid social contacts because of my self-consciousness. #2 1. I don t have any difficulty eating slowly in the proper manner. 2. Although I seem to gobble down foods, I don t end up feeling stuffed because of eating too much. 3. At times, I tend to eat quickly and then, I feel uncomfortably full afterwards. 4. I have the habit of bolting down my food, without really chewing it. When this happens I usually feel uncomfortably stuffed because I ve eaten too much. #3 1. I feel capable to control my eating urges when I want to. 2. I feel like I have failed to control my eating more than the average person. 3. I feel utterly helpless when it comes to feeling in control of my eating urges. 4. Because I feel so helpless about controlling my eating I have become very desperate about trying to get in control.

305 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 14 of 16 #4 1. I don t have the habit of eating when I m bored. 2. I sometimes eat when I m bored, but often I m able to get busy and get my mind off food. 3. I have a regular habit of eating when I m bored, but occasionally, I can use some other activity to get my mind off eating. 4. I have a strong habit of eating when I m bored. Nothing seems to help me break the habit. #5 1. I m usually physically hungry when I eat something. 2. Occasionally, I eat something on impulse even though I really am not hungry. 3. I have the regular habit of eating foods that I might not really enjoy, to satisfy a hungry feeling even though physically, I don t need the food. 4. Even though I m not physically hungry, I get a hungry feeling in my mouth that only seems to be satisfied when I eat a food, like a sandwich, that fills my mouth. Sometimes, when I eat the food to satisfy my mouth hunger, I then spit the food out so I won t gain weight. #6 1. I don t feel any guilt or self-hate after I overeat. 2. After I overeat, occasionally I feel guilt or self-hate. 3. Almost all the time I experience strong guilt or self hate after I overeat. #7 1. I don t loose total control of my eating when dieting even after periods when I overeat. 2. Sometimes when I eat a forbidden food on a diet, I feel like I blew it and eat even more. 3. Frequently, I have the habit of saying to myself, I ve blown it now, why not go all the way when I overeat on a diet. When that happens I eat even more. 4. I have a regular habit of starting strict diets for myself, but I break the diets by going on an eating binge. My life seems to be either a feast or famine. #8 1. I rarely eat so much food that I feel uncomfortable stuffed afterwards. 2. Usually about once a month, I eat such a quantity of food; I end up feeling very stuffed. 3. I have regular periods during the month when I eat large amounts of food, either at mealtime or at snacks. 4. I eat so much food that I regularly feel quite uncomfortable after eating and sometimes at bit nauseous.

306 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 15 of 16 #9 1. My level of calorie intake does not go up very high or go down very low on a regular basis. 2. Sometimes after I overeat, I will try to reduce my caloric intake to almost nothing to compensate for the excess calories I ve eaten. 3. I have a regular habit of overeating during the night. It seems that my routine is not to be hungry but overeat in the evening. 4. In my adult years, I have had week long periods when I overeat. It seems I live a life of either feast or famine. #10 1. I usually am able to stop eating when I want to. I know when enough is enough. 2. Every so often, I experience a compulsion to eat which I can t seem to control. 3. Frequently, I experience strong urges to eat which I seem unable to control, but at other times I can control my eating urges. 4. I feel incapable of controlling urges to eat. I have to fear of not being able to stop eating voluntarily. #11 1. I don t have any problem stopping eating when I feel full. 2. I usually can stop eating when I feel full but occasionally overeat leaving me feeling uncomfortably stuffed. 3. I have a problem stopping eating once I start and usually I feel uncomfortable stuffed after I eat a meal. 4. Because I have a problem not being able to stop eating when I want, I sometimes have to induce vomiting to relieve my stuffed feeling. #12 1. I seem to eat just as much when I m with others (family, social gatherings) as when I m by myself. 2. Sometimes, when I m other persons, I don t eat as much as I want to eat because I m selfconscious about my eating. 3. Frequently, I eat only a small amount of food when others are present, because I m very embarrassed about my eating. 4. I feel so ashamed about overeating that I pick times to overeat when I know no on will see me. I feel like a closet eater.

307 Revised Protocol - Appendix G EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 16 of 16 #13 1. I eat three meals a day with only an occasional between meal snack. 2. I eat 3 meals a day, but I also normally snack between meals. 3. When I am snacking heavily, I get in the habit of skipping regular meals. 4. There are regular periods when I seem to be continually eating, with no planned meals. #14 1. I don t think much about trying to control unwanted eating urges. 2. At least some of the time, I feel my thoughts are pre-occupied with trying to control my eating urges. 3. I feel that frequently I spend much time thinking about how much I ate or about trying not to eat more. 4. It seems to me that most of my waking hours are pre-occupied with thoughts about eating or not eating. I feel like I m constantly struggling not to eat. #15 1. I don t think about food a great deal. 2. I have strong cravings for food but they last only for brief periods of time. 3. I have days when I can t seem to think about anything else but food. 4. Most of my days seem to be pre-occupied with thoughts about food. I feel like I live to eat. #16 1. I usually know whether or not I m physically hungry. I take the right portion of food to satisfy me. 2. Occasionally, I feel uncertain about knowing whether or nor I m physically hungry. At these times, its hard to know how much food I should take to satisfy me. 3. Even though I might know how many calories I should eat, I don t have any idea what is a normal amount of food for me.

308 Revised Protocol - Appendix H EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 1 of 10 Appendix H NN Mental Health Questionnaires This confidential document is the property of. No unpublished information contained herein may be disclosed without prior written approval from. Access to this document must be restricted to relevant parties.

309 Revised Protocol - Appendix H EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 2 of 10 Patient Health Questionnaire 9 (PHQ-9) PATIENT HEALTH QUESTIONNAIRE-9 Only the patient (subject) should enter information onto this questionnaire. Over the last 2 weeks, how often have you been bothered by any of the following problems? Not at all Several days More than half the days 1. Little interest or pleasure in doing things Feeling down, depressed, or hopeless Trouble falling or staying asleep, or sleeping too much Feeling tired or having little energy Poor appetite or overeating Feeling bad about yourself or that you are a failure or have let yourself or your family down 7. Trouble concentrating on things, such as reading the newspaper or watching television 8. Moving or speaking so slowly that other people could have noticed? Or the opposite being so fidgety or restless that you have been moving around a lot more than usual 9. Thoughts that you would be better off dead or of hurting yourself in some way Nearly every day SCORING FOR USE BY STUDY PERSONNEL ONLY =Total Score: If you checked off any problems, how difficult have these problems made it for you to do your work, take care of things at home, or get along with other people? Not difficult at all Somewhat difficult Very difficult Extremely difficult Copyright 2005 Pfizer, Inc. All rights reserved. Reproduced with permission. Investigator signature: Date:

310 Revised Protocol - Appendix H EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 3 of 10 Colombia Suicide Severity Rating Scale (C-SSRS), Baseline

311 Revised Protocol - Appendix H EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 4 of 10

312 Revised Protocol - Appendix H EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 5 of 10

313 Revised Protocol - Appendix H EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 6 of 10

314 Revised Protocol - Appendix H EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 7 of 10 Colombia Suicide Severity Rating Scale (C-SSRS), Since last visit

315 Revised Protocol - Appendix H EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 8 of 10

316 Revised Protocol - Appendix H EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 9 of 10

317 Revised Protocol - Appendix H EudraCT No.: Date: 10 December 2010 Version: 4.0 Page: 10 of 10