A Trial Comparing the Efficacy and Safety of Liraglutide 1.8 mg/Day to Liraglutide 0.9 mg/Day in Japanese Subjects With Type 2 Diabetes Mellitus.

Change from baseline (week 0) in glycosylated haemoglobin (HbA1c) was evaluated after 26 weeks of treatment. The change from baseline in the response after 26 weeks of treatment is analysed using an analysis of covariance (ANCOVA) model with treatment as a fixed effect and baseline response as a covariate.

Time frame: Week 0, Week 26

Population: FAS, which included all randomised subjects. Missing values were imputed using the last observation carried forward (LOCF) method.

Change from baseline (week 0) in alanine aminotransferase was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in albumin was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in albumin corrected calcium was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in alkaline phosphatase was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in amylase was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in aspartate aminotransferase was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in calcium was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in creatine kinase was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in creatinine was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in estimated glomerular filtration rate (eGFR) was evaluated after 26 weeks and 52 weeks of treatment, respectively. eGFR was evaluated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formula, mL/min/1.73m\^2.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in lipase was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in potassium was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in sodium was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in total bilirubin was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in urea was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in systolic blood pressure (SBP) and diastolic blood pressure (DBP) was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in BMI was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in body weight was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Reported results are number of subjects with low, normal or high calcitonin values at week 0, week 26 and week 52. Number of subjects analyzed = number of subjects contributed to the analysis for individual time point. Calcitonin values were categorised as low, normal or high.

Time frame: Week 0 and Week 26 and Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Reported results are ECG outcomes at week 0, week 26 and week 52. ECG outcomes were evaluated as: 1) normal, 2) abnormal, NCS or 3) abnormal, CS.

Time frame: Week 0 and Week 26 and Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Reported results are eye examination (ophthalmoscopy) outcomes at week 0, week 26 and week 52. Ophthalmoscopy outcomes for both left and right eye were evaluated as: 1) normal, 2) abnormal, NCS or 3) abnormal, CS.

Time frame: Week 0 and Week 26 and Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in FPG was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in basophils was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26 and Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in eosinophils was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26 and Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in erythrocytes was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26 and Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in haematocrit was evaluated after 26 weeks and 52 weeks of treatment, respectively. Haematocrit is the ratio of the volume of red blood cells to the total volume of blood.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in haemoglobin was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in leukocytes was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26 and Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in lymphocytes was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26 and Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in monocytes was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26 and Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in neutrophils was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26 and Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in thrombocytes (platelets) was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26 and Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in HbA1c was evaluated after 52 weeks of treatment.

Time frame: Week 0, Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for this outcome measure, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Reported results are physical examination outcomes at week (wk) 0, wk 26 and wk 52. Physical examination consisted of the following listed examinations and the outcome of each examination was evaluated as: 1) normal, 2) abnormal, not clinically significant (NCS) or 3) abnormal, clinically significant (CS). 1. Cardiovascular system 2. Central and peripheral nervous system (PNS) 3. Gastrointestinal (GI) system including mouth 4. General appearance 5. Head, ears, eyes, nose, throat, neck 6. Lymph node palpation 7. Musculoskeletal system 8. Respiratory system 9. Skin 10. Thyroid gland

Time frame: Week 0 and Week 26 and Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in pulse was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Reported results are 7-point SMBG values at week 0, week 26 and week 52. The 7-point profile blood glucose levels were measured at the following time points always starting with the first: 1. Before breakfast. 2. 90 minutes after start of breakfast. 3. Before lunch. 4. 90 minutes after start of lunch. 5. Before dinner. 6. 90 minutes after start of dinner. 7. At bedtime.

Time frame: Week 0 and Week 26 and Week 52

Population: FAS, which included all randomised subjects. Number analyzed = number of subjects contributed to the analysis. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method

Change from baseline (week 0) in mean of the SMBG 7-point profile was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: FAS, which included all randomised subjects. Number analyzed = number of subjects contributed to the analysis. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in mean of postprandial increments (from before meal to 90 minutes after for breakfast, lunch and dinner) of the SMBG 7-point profile was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: FAS, which included all randomised subjects. Number analyzed = number of subjects contributed to the analysis. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Change from baseline (week 0) in waist circumference was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 0, Week 26, Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Fasting C-peptide was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Fasting glucagon was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Fasting insulin was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Free fatty acids were evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

HDL was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

HOMA-IR was evaluated after 26 weeks and 52 weeks of treatment, respectively. HOMA-IR is an index of insulin resistance and was calculated as: HOMA-IR= fasting insulin (μU/mL) x FPG (mmol/L)/22.5.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

HOMA-B was evaluated after 26 weeks and 52 weeks of treatment, respectively. HOMA-B is an index of beta-cell function and was calculated as: HOMA-B=\[(20 x fasting insulin in µU/mL)/(FPG in mmol/L-3.5)\].

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

LDL was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Treatment emergent adverse events (TEAEs) were evaluated during the 26-week and 52-week treatment period, respectively. TEAE for weeks 0-26: Event that has onset date on or after randomisation (from week 0) and no later than seven days after the last day on liraglutide (maximum till week 26 + 7 days). TEAE for weeks 0-52: Event that has onset date on or after randomisation (from week 0) and no later than seven days after the last day on liraglutide (maximum till week 52 + 7 days). Hence, the following shown 'Time Frame' should be read as 'Weeks 0-26 + 7 days and Weeks 0-52 + 7 days'.

Time frame: Weeks 0-26 and Weeks 0-52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for weeks 0-52, as the subjects in this treatment arm received treatment for 26 weeks (main period).

American Diabetes Association (ADA) classification of hypoglycaemia: 1. Severe: Requiring assistance of another person to actively administer carbohydrate/glucagon/take other corrective actions. PG levels may not be available during an event, but neurological recovery following return of PG to normal is considered sufficient evidence that event was induced by a low PG level. 2. Documented symptomatic: PG level ≤3.9 mmol/L with symptoms. 3. Asymptomatic: PG level ≤3.9 mmol/L without symptoms. 4. Probable symptomatic: No measurement with symptoms. 5. Pseudo: PG level \>3.9 mmol/L with symptoms. Treatment emergent hypoglycaemic episode: episode with onset date on or after randomisation (from week 0) and no later than 7 days after the last day on liraglutide (maximum till week 26 and week 52, respectively + 7 days). Hence, the following shown 'Time Frame' should be read as 'Week 0-26 + 7 days and Week 0-52 + 7 days'.

Time frame: Weeks 0-26 and Weeks 0-52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period).

Treatment emergent nocturnal severe or BG confirmed symptomatic hypoglycaemic episodes were evaluated during the26-week and 52-week treatment period, respectively. Nocturnal hypoglycaemic episodes: Those occurring between 00:01 and 05:59 hours, both inclusive. Severe or BG confirmed symptomatic hypoglycaemia: episode that was severe according to the ADA classification or BG confirmed by a PG value \<3.1 mmol/L with symptoms consistent with hypoglycaemia. Treatment emergent: episode with onset date on or after randomisation (from week 0) and no later than 7 days after the last day on liraglutide (maximum till week 26 and week 52, respectively + 7 days). Hence, the following shown 'Time Frame' should be read as 'Week 0-26 + 7 days and Week 0-52 + 7 days'.

Time frame: Weeks 0-26 and Weeks 0-52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period).

Treatment emergent severe or BG confirmed symptomatic hypoglycaemic episodes were evaluated during the 26-week and 52-week treatment period, respectively. Severe or BG confirmed symptomatic hypoglycaemia (hypo):An episode that was severe according to the ADA classification or BG confirmed by a PG value \<3.1 mmol/L with symptoms consistent with hypo. ADA definition of severe hypo:episode requiring assistance of another person to actively administer carbohydrate/glucagon, or take other corrective actions. PG levels may not be available during an event, but neurological recovery following the return of PG to normal is considered sufficient evidence that the event was induced by a low PG level. Treatment emergent: episode with onset date on or after randomisation (from week (wk) 0) and no later than 7 days after the last day on liraglutide (maximum till wk 26 and wk 52, respectively + 7 days). Hence, the following shown 'Time Frame' should be read as 'wk 0-26+7 days and wk 0-52+7 days'.

Time frame: Weeks 0-26 and Weeks 0-52

Population: Safety analysis set, which included all subjects receiving at least one dose of liraglutide. 'Liraglutide 0.9 mg treatment arm' is not applicable for weeks 0-52, as the subjects in this treatment arm received treatment for 26 weeks (main period).

Proinsulin was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Proinsulin/insulin was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Reported results are number of subjects who achieved HbA1c target below 7.0% after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing HbA1c values were imputed using the LOCF method.

Reported results are subjects with HbA1c \<7.0% after 26 weeks and 52 weeks of treatment, respectively without treatment emergent severe or BG confirmed symptomatic hypoglycaemic episodes. Severe or BG confirmed symptomatic hypoglycaemia: severe as per ADA classification or BG confirmed by plasma glucose (PG) value \<3.1 mmol/L with symptoms consistent with hypoglycaemia. Severe hypoglycaemia as per ADA: episode requiring assistance of another person to actively administer carbohydrate/glucagon, or take other corrective actions. PG levels may not be available during an event, but neurological recovery following the return of PG to normal is considered sufficient evidence that the event was induced by a low PG level. Treatment emergent: episode with onset date on or after randomisation (from week (wk)0) and no later than 7 days after the last day on liraglutide (maximum till wk26+7days and wk52+7days). Hence, the following shown 'Time Frame' should be read as 'Wk26+7days and Wk52+7days'

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing HbA1c values were imputed using the LOCF method.

Reported results are number of subjects who achieved HbA1c target below 7.0% without weight gain after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing HbA1c and body weight values were imputed using the LOCF method.

Reported results are number of subjects who achieved HbA1c target below or equal to 6.5% after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing HbA1c values were imputed using the LOCF method.

Total cholesterol was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

Triglycerides were evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.

VLDL was evaluated after 26 weeks and 52 weeks of treatment, respectively.

Time frame: Week 26 and Week 52

Population: FAS, which included all randomised subjects. 'Liraglutide 0.9 mg treatment arm' is not applicable for week 52, as the subjects in this treatment arm received treatment for 26 weeks (main period). Missing values were imputed using the LOCF method.