Supplementary Information s57

Supplementary Information

Effect of liraglutide 3.0 mg in individuals with obesity and moderate or severe obstructive sleep apnea: the SCALE Sleep Apnea randomized clinical trial Adam Blackman, MD, Gary D. Foster, PhD, Gary Zammit, PhD, Russell Rosenberg, PhD, Louis Aronne, MD, Thomas Wadden, PhD, Birgitte Claudius, MD, Christine Bjørn Jensen, MD, PhD, Emmanuel Mignot, MD, PhD; on behalf of the SCALE study group*

Content: 1. Methods Statistical Analysis Participants who withdrew before the end of the trial were requested to come in for the end-of-trial visit (as soon as possible after withdrawal) where a final polysomnographic assessment was performed. Therefore, for participants who withdrew during weeks 0–12 and had an end-of-trial visit, their polysomnographic measurements from the end-of-trial visit were imputed via the LOCF method. If the withdrawn participants (before week 12) did not agree to the end-of-trial visit and had no post-baseline assessment, they were not included in the primary analysis of the primary endpoint (but were included in the baseline observation carried forward sensitivity analysis). For participants who withdrew after week 12 and before week 32 and had an end-of-trial visit, their polysomnographic measurements from the end- of-trial visit were imputed via the LOCF method. If the withdrawn participants (after week 12) did not agree to the end-of-trial visit, their polysomnographic measurements at week 12 were imputed via the LOCF method.

Approximately 7% of participants withdrew from the trial without providing any post-baseline data for the primary endpoint and were therefore excluded from the primary analysis (Supplementary Table E3). Out of 359 total randomized participants (corresponding to those included in the baseline observation carried forward analysis in the Supplementary Table E3), 334 (93%) had post-baseline polysomnographic assessments/data and were therefore included in the primary analysis as shown in the Supplementary Table E3. The remaining 7% withdrew from the trial without providing any post-baseline data for the primary analysis.

2. Results Safety Geometric mean amylase and lipase activity was higher at week 32 (1.03±49.4 U/L and 0.74±75.8 U/L, respectively) compared to baseline (0.92±43.2 U/L and 0.54±44.2 U/L, respectively) with liraglutide, but remained generally unchanged with placebo (Supplementary Table E5). Among the liraglutide-treated participants who experienced elevations in amylase (11.4%) or lipase (37.5%) activity at any time during treatment, most had elevations ≤2 times ULN (amylase: 9.7%, lipase: 29.0%). Geometric mean calcitonin levels after 32 weeks were similar to baseline levels for both genders in both treatment groups (Supplementary Table E6). No persistent increases (ie, normal at baseline and elevated beyond ULN at all post-baseline visits) in calcitonin were observed during the trial. In total, two male subjects with baseline calcitonin levels <5.8 pmol/L (20 ng/L) experienced incidental (1–2 visits) post-baseline elevations in calcitonin levels of ≥5.8 pmol/L (20 ng/L) during treatment with liraglutide. There were no calcitonin elevations of ≥14.6 pmol/L (50 ng/L). 3. Table E1. Complete list of trial inclusion and exclusion criteria Inclusion criteria  Informed consent obtained before any trial-related activities take place  Body mass index ≥30 kg/m²  Stable body weight (<5% self-reported change during the previous 3 months)  Age 18−64 years (both inclusive)  Diagnosis of moderate or severe obstructive sleep apnea (OSA)  Unwilling or unable to use continuous positive airway pressure (CPAP) (or other positive airway pressure) treatment. No CPAP (or other positive airway pressure) treatment for at least 4 weeks prior to screening  Ability and willingness to comply with all protocol procedures e.g. correct handling of trial product, compliance to visit schedule and dietary advice and complete trial-related questionnaires Exclusion criteria  Treatment with glucagon-like peptide-1 (GLP-1) receptor agonists (including liraglutide or exenatide), dipeptidyl peptidase-4 (DPP-4) inhibitors or insulin within the last 3 months prior to screening  Diagnosis of type 1 or type 2 diabetes per judgement of the investigator

 Glycated hemoglobin (HbA1c) ≥6.5% (screening value)  Significant craniofacial abnormalities that may be causing OSA  Respiratory and neuromuscular diseases that could interfere with the results of the trial in the opinion of the investigator  Known diagnosis prior to screening of periodic limb movement disorder  Use of central stimulants, hypnotics, mirtazepine, opioids, trazodone within the previous 3 months prior to screening  Obesity induced by other endocrinologic disorders (eg, Cushing Syndrome)  Treatment with medications within 3 months prior to screening that in the opinion of the investigator may cause significant weight gain  Weight loss attempts using herbal supplements or over-the-counter medications within 3 months prior to screening  Participation in an organized weight reduction program (current or within 3 months prior to screening)  Treatment with pramlintide, sibutramine, orlistat, zonisamide, topiramate or phentermine within 3 month prior to screening  Previous surgical treatment for obesity; e.g., gastric banding procedure (excluding liposuction if performed more than one year before trial entry)  Hypothyroidism/hyperthyroidism defined as thyroid-stimulating hormone (TSH) >6 mIU/L or <0.4 m IU/L  Calcitonin ≥50 ng/L at screening  Familial or personal history of Multiple Endocrine Neoplasia type 2 or familial Medullary Thyroid Carcinoma  Personal history of non-familial Medullary Thyroid Carcinoma  History of chronic pancreatitis or idiopathic acute pancreatitis  History of Major Depressive Disorder within 2 years prior to screening  History of other severe psychiatric disorders (eg, schizophrenia or bipolar disorder)  A patient health questionnaire-9 (PHQ-9) score ≥15  Any lifetime history of a suicide attempt  A history of any suicidal behavior in the 4 weeks prior to screening  Any suicidal ideation of type 4 or 5 on the Columbia Suicide Severity Rating Scale (C-SSRS) in the 4 weeks prior to screening  Surgery scheduled for the trial duration period, except for minor surgical procedures, at the discretion of the investigator  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  History of stroke or Acute Coronary Syndrome within the year prior to screening  History of unstable angina or heart failure corresponding to New York Heart Association (NYHA) functional class III or IV within the year prior to screening  History of arrhythmia  Systolic blood pressure ≥160 mmHg and/or diastolic blood pressure ≥100 mmHg at screening  Known or suspected abuse of alcohol or drugs  Female of childbearing potential who is pregnant, breast-feeding or intends to become pregnant or is not using adequate contraceptive methods (as required by local law or practice)  United States: abstinence and the following methods: diaphragm with spermacide, condom with spermacide (by male partner), intrauterine device, sponge, spermacide, Norplant®, Depo-Provera® or oral contraceptives  Known or suspected hypersensitivity to trial product or related products  Previous participation in this trial. Participation is defined as randomization  Any chronic disorder or severe disease which, in the opinion of the investigator, might jeopardize subject’s safety or compliance with the protocol  Language barrier, mental incapacity, unwillingness or inability to understand and be able to comply with protocol procedures  Subjects from the same household participating in the trial  Participation in another clinical trial within 3 months prior to screening or receipt of any investigational medicinal product within 3 months before the screening visit Table E2. Adverse events of special focus Pre-defined, broad searches based on standard MedDRA queries (SMQ), high-level group terms (HLGT), high-level terms (HLT) and/or preferred terms (PT) for adverse event in the categories below were performed among all adverse events in order to identify relevant events. All adverse events in the below categories (except death) were evaluated based on the above-mentioned search results (marked ‘Pre- defined MedDRA search-based’ below). Some types of events were also evaluated via a blinded adjudication process by an independent, external adjudication committee of medical experts (marked ‘Adjudicated’ below). Based on pre-defined diagnostic criteria, the adjudication committee could either confirm or not confirm the adverse event classification/diagnosis.

Event type Evaluation Pre-defined MedDRA Adjudicated search based Death No Yes Cardiovascular eventsa Yes Yes Acute coronary syndrome Yes Yes Cerebrovascular Yes Yes Heart failure Yes Yes Stent thrombosis Yes Yes Revascularization procedure Yes Yes Hospitalization for cardiac Yes No arrhythmia Pancreatitis/suspicion of Yes Yes pancreatitisb Gallbladder-related eventsc Yes No Neoplasmsd Yes Yes Thyroid diseasee Yes Events requiring thyroidectomy and thyroid neoplasms only Acute renal failuref Yes No Severe hypoglycemic Not applicable No episodesg Immunogenicity eventsh Allergic reactions Yes No Immune-complex disease Yes No Injection-site reactions Yes No Psychiatric disordersi Yes No The search terms presented below were used to define the preferred terms included in the searches for events of special focus: aCerebrovascular disorders (SMQ), Cardiac failure (SMQ), Embolic and thrombotic events (SMQ), Torsade de pointes/QT prolongation (SMQ), Cardiac arrhythmias (SMQ), Arrhythmia related investigations (signs and symptoms) (SMQ), Bradyarrhythmia terms (nonspecific) (SMQ), Conduction defects (SMQ), Disorders of sinus node function (SMQ), Cardiac arrhythmia terms (nonspecific) (SMQ), Supraventricular tachyarrhythmias (SMQ), Tachyarrhythmia terms (nonspecific) (SMQ), Ventricular tachyarrhythmias (SMQ) bAcute pancreatitis (narrow scope) (SMQ) and Acute and chronic pancreatitis (HLT) cBile duct related disorders (SMQ), Biliary system related disorders and investigations (signs and symptoms) (SMQ), Gallstone related disorders (SMQ), Infectious biliary disorders (SMQ), Site unspecified biliary disorders (SMQ), Gallbladder related disorders (SMQ) dBiliary neoplasms malignant and unspecified (SMQ), Biliary malignant tumors (SMQ), Biliary tumors of unspecified malignancy (SMQ), Breast neoplasms - malignant and unspecified (SMQ), Breast malignant tumors (SMQ), Breast tumors of unspecified malignancy (SMQ), Liver neoplasms - malignant and unspecified (SMQ), Liver malignant tumors (SMQ), Liver tumors of unspecified malignancy (SMQ), Malignant or unspecified tumors (SMQ), Malignant tumors (SMQ), Tumors of unspecified malignancy (SMQ), Ovarian neoplasms - malignant and unspecified (SMQ), Ovarian malignant tumors (SMQ), Ovarian tumors of unspecified malignancy (SMQ), Oropharyngeal neoplasms (SMQ), Premalignant disorders (SMQ), Blood premalignant disorders (SMQ), Gastrointestinal premalignant disorders (SMQ), Premalignant disorders - general conditions and other site specific disorders (SMQ), Reproductive premalignant disorders (SMQ), Skin premalignant disorders (SMQ), Prostate neoplasms - malignant and unspecified (SMQ), Prostate malignant tumors (SMQ), Prostate tumors of unspecified malignancy (SMQ), Skin neoplasms - malignant and unspecified (SMQ), Skin malignant tumors (SMQ), Skin tumors of unspecified malignancy (SMQ), Uterine and fallopian tube neoplasms - malignant and unspecified (SMQ), Uterine and fallopian tube malignant tumors (SMQ), Uterine and fallopian tube tumors of unspecified malignancy (SMQ), Tumor markers (SMQ) eHyperthyroidism (SMQ), Hypothyroidism (SMQ) and Thyroid gland disorders (HLGT) and Calcitonin secretion disorder (PT), Ectopic calcitonin production (PT), Hypercalcitoninemia (PT), Blood calcitonin abnormal (PT), Blood calcitonin increased (PT) fAcute renal failure (SMQ) gADA Workgroup on Hypoglycemia1 h Anaphylactic reaction (narrow scope) (SMQ), Anaphylactic/anaphylactoid shock conditions (narrow scope) (SMQ), Angioedema (narrow scope) (SMQ), Severe cutaneous adverse reactions (narrow scope) (SMQ), Asthma/bronchospasm (narrow scope) (SMQ), Documented hypersensitivity to administered drug (PT), Type II hypersensitivity (PT), Type IV hypersensitivity reaction (PT), Systemic lupus erythematous (narrow scope) (SMQ), Vasculitis (narrow scope) (SMQ), Guillain-Barre syndrome (narrow scope) (SMQ) and Serum sickness (PT), Serum sickness-like reaction (PT), Cryoglobulin urine present (PT), Cryoglobulins (PT), Cryoglobulinuria (PT), Acute interstitial pneumonitis (PT), Granulomatous pneumonitis (PT), Pneumonitis (PT), Fibrillary glomerulonephritis (PT), Glomerulonephritis (PT), Glomerulonephritis acute (PT), Glomerulonephritis chronic (PT), Glomerulonephritis membranoproliferative (PT), Glomerulonephritis membranous (PT), Glomerulonephritis minimal lesion (PT), Glomerulonephritis proliferative (PT), Glomerulonephritis rapidly progressive (PT), Immunotactoid glomerulonephritis (PT), Mesangioproliferative glomerulonephritis (PT), Immune complex level increased (PT), Type III immune complex mediated reaction (PT), Administration site reactions (HLT), Application and instillation site reactions (HLT), Infusion site reactions (HLT), Lipodystrophies (HLT), Injection site reactions (HLT) iSearch results were all reported adverse events included in the system organ class of ‘psychiatric disorders’ (including primary and secondary preferred terms) MedDRA, Medical Dictionary for Regulatory Activities 4. Table E3. Sensitivity analyses for the primary endpoint (change from baseline in apnea–hypopnea index [events/h] after 32 weeks). Type of analysis Liraglutide Placebo Estimated treatment P value 3.0 mg difference for liraglutide vs placebo (95% CI)* Primary analysis N=168 N=166 -12.17 -6.07 -6.10 (-11.0 to -1.19) 0.015 Completer populationa N=134 N=142 -13.35 -7.42 -5.94 (-11.4 to -0.48) 0.033 Repeated measuresb N=168 N=166 -12.65 -6.31 -6.34 (-11.6 to -1.06) 0.0187 Baseline value carried N=180 N=179 forwardc -11.25 -5.79 -5.46 (-10.1 to -0.85) 0.0203 Multiple imputationd N=180 N=179 -11.27 -5.99 -5.28 (-10.7 to 0.09) 0.05 aSame analysis as the primary applied to participants in the FAS who completed week 32 and had a valid non-imputed measurement at that week. bAnalysis was conducted using a linear mixed-effect model. cParticipants without a valid post-baseline measurement were included in this analysis. dSame analysis as the primary but imputing missing observations with a regression method. Participants without any post-baseline polysomnographic assessments were excluded from the primary and repeated measures analyses. Data for liraglutide and placebo are estimated means. CI, confidence interval; N, number of participants. 5. Table E4. Model-estimated changes in polysomnographic, quality of life and cardiometabolic endpoints after 32 weeks of treatment. Endpoint Baseline values Least square mean changes Estimated treatment P value from baseline to week 32 difference/odds ratio for Liraglutide 3.0 mg Placebo Liraglutide Placebo liraglutide vs placebo N=180 N=179 3.0 mg (95% CI)a Primary endpoint Apnea-hypopnea 49.0±27.5 49.3±27.5 -12.2 -6.1 -6.1 (-11.0 to -1.2) 0.015 index (events/h) PSG-derived secondary endpoints Lowest oxygen 74.2±10.5 74.7±10.4 1.5 0.8 0.8 (−1.0 to 2.5) 0.40 saturation (%) Time with 14.9±18.2 14.4±18.9 -2.2 -1.3 -0.9 (-3.7 to 1.8) 0.51 oxygen saturation <90% (%) Oxygen 43.7±26.1 44.1±26.1 -9.5 -5.1 -4.37 (-8.94 to 0.06 desaturation ≥4% 0.20) index (events/h) Total sleep time 356.3±62.2 348.4±63.6 23.2 15.5 7.7 (−3.1 to 18.6) 0.16 (min) Wake time after 20.4±11.7 22.3±12.3 -4.6 -2.9 −1.7 (−3.6 to 0.3) 0.10 sleep onset (%) Body weight- related endpoints Body weight (%) -5.7 -1.6 -4.2 (-5.2 to -3.1) <0.0001 Body weight (kg) 116.5±23.0 118.7±25.4 -6.8 -1.8 -4.9 (-6.2 to -3.7) <0.0001 ≥5% body weight 46.4 18.1 3.9 (2.4 to 6.4) <0.0001 loss (%) >10% body 22.4 1.5 19.0 (5.7 to 63.1) <0.0001 weight loss (%) Body-mass 38.9±6.4 39.4±7.4 -2.2 -0.6 -1.6 (-2.0 to -1.2) <0.0001 index (kg/m2)b Waist 122.3±14.5 122.7±15.0 -6.4 -3.1 -3.2 (-4.5 to -2.0) <0.0001 circumference (cm) Neck 44.5±4.5 44.2±4.6 -2.1 -1.3 -0.8 (-1.2 to -0.3) 0.0014 circumference (cm) Glycemic endpoints Glycated 5.7±0.4 5.6±0.4 -0.4 -0.2 -0.2 (-0.3 to -0.1) <0.0001 hemoglobin (%) Fasting glucose 5.4±0.6 5.4±0.9 -0.14 0.16 -0.3 (-0.4 to <0.0001 (mmol/L) -0.16) Blood pressure Systolic (mmHg) 125.8±11.5 127.1±12.3 -3.7 0.4 -4.1 (-6.3 to -1.9) 0.0003 Diastolic 81.2±7.6 82.2±8.8 -1.0 -0.1 -1.0 (-2. 5 to 0.5) 0.20 (mmHg) Cardiovascular biomarkers hsCRP (nmol/L) 35.1±103.9 34.2±112.9 -21.2d -7.7d 0.9 (0.8 to 1.0)d 0.05 Quality of life ESS total score 9.2±5.1 10.3±5.4 -2.7 -2.2 -0.5 (-1.3 to 0.2) 0.15 FOSQ total score 17.1±2.4 17.2±2.5 1.3 1.1 0.2 (−0.1 to 0.5) 0.16 SF-36 overall 46.5±9.2 47.0±8.7 2.8 2.0 0.9 (−0.5 to 2.2) 0.21 physical health score SF-36 overall 53.0±8.1 52.8±7.9 1.5 0.9 0.6 (−0.9 to 2.0) 0.43 mental health score Baseline values are mean±SD for all parameters except hsCRP (geometric mean±CV). aEstimated treatment differences are from an analysis of covariance using the full analysis set (FAS) with last-observation-carried-forward (LOCF) imputation. Loss of ≥5% and >10% of body weight are analyzed by logistic regression using the FAS with LOCF and are presented as odds ratios. bBody-mass index is the weight in kilograms divided by the square of the height in meters. dData were log transformed and are presented as relative differences from baseline (%) (raw changes) and estimated ratio between treatments. ESS, Epworth Sleepiness Scale; FOSQ, Functional Outcomes of Sleep Questionnaire; hsCRP, high sensitivity C-reactive protein; PSG, polysomnography; SF- 36, Short Form 36 health survey 6. Table E5. Amylase and lipase activity at baseline and end-of-treatment. Liraglutide 3.0 mg Placebo N=176 N=179 Amylase (μkat/L) Baseline 0.92±43.2 0.92±38.9 Week 32* 1.03±49.4 0.92±42.2 Lipase (μkat/L) Baseline 0.54±44.2 0.55±62.9 Week 32* 0.74±75.8 0.56±62.8 Numbers are geometric mean plus/minus coefficient of variation. *Last observation carried forward method was used Normal range for amylase activity is: 0.3–1.9 μkat/L. Normal range for lipase activity is: 0–1.0 μkat/L.

7. Table E6. Calcitonin levels at baseline and end-of-treatment. Liraglutide 3.0 mg Placebo Males N=128 N=129 Baseline (pmol/L) 0.70±112.8 0.82±116.6 Week 32* (pmol/L) 0.70±112.5 0.76±110.7 Females N=48 N=50 Baseline (pmol/L) 0.29±22.2 0.41±111.5 Week 32* (pmol/L) 0.32±98.7 0.38±90.5 Values are geometric mean plus/minus coefficient of variation. *Last observation carried forward method was used. Upper limit of normal for calcitonin is 2.5 pmol/L for males and 1.5 pmol/L for females. 8. Figure E1. OSA severity category at baseline and after 32 weeks, and change in OSA severity by weight change category

9. Figure E2. Change in polysomnographic and quality of life endpoints by weight change category.

10. Figure E3. Nocturnal heart rate at baseline, week 12 and week 32. Nocturnal heart rate was derived from the electrocardiogram component of the polysomnography assessment. The timing of the polysomnographic recording varied between participants; for most participants, the recording was performed between 21:00 and 07:00 hours.

11. List of investigators in the SCALE Sleep Apnea study group

The following principal investigators participated in the conduct of this trial: Louis Aronne, Adam Blackman, Joe Blumenau, Richard Bogan, Mardik Donikyan, Jane Dyonzak, Helene Emsellem, Milton Erman, Neil Feldman, Gary Foster, David Fried, June Fry, Mark Goetting, Alexander Golbin, Timothy Grant, Steven Hull, Andrew Jamieson, Stephen Kreitzer, Mitchell Lee, Daniel Lorch, Mortimer Mamelak, Abe Marcadis, Curtis Mello, Emmanuel Mignot, Polly Moore*, Adam Moscovitch, Daniel Norman, Vikas Pandith, James Perlstrom, Russell Rosenberg, Markus Schmidt, Andrew Schreiber, Paula Schweitzer, David Seiden, Manuel Suarez, Stephen Thein, Thomas Wadden, Robert Wagner, Charles Wells, Sean Wharton, David Winslow, Esther Yoon *investigator was replaced by Esther Yoon

Reference 1 Defining and reporting hypoglycemia in diabetes: a report from the American Diabetes Association Workgroup on Hypoglycemia. Diabetes Care 2005; 28: 1245–1249.