ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2019

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1534

Metabolic and endocrine effects ofSGLT2 inhibition

PER LUNDKVIST

ISSN 1651-6206ISBN 978-91-513-0561-5urn:nbn:se:uu:diva-373522

Dissertation presented at Uppsala University to be publicly examined in Enghoff lecture-hall, Entrance 50, Uppsala academic hospital, Uppsala, Friday, 8 March 2019 at 13:15 for thedegree of Doctor of Philosophy (Faculty of Medicine). The examination will be conductedin English. Faculty examiner: Docent Michael Alvarsson (Department of Molecular Medicineand Surgery (MMK), K1, Karolinska institute, Stockholm, Sweden).

AbstractLundkvist, P. 2019. Metabolic and endocrine effects of SGLT2 inhibition. DigitalComprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1534.66 pp. Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-513-0561-5.

Obesity and type 2 diabetes (T2D) are two growing global health problems with similarcomorbidity profiles. SGLT2 inhibitors (SGLT2i) improve blood glucose control and canrelieve both T2D and obesity, as well as their associated health problems such as hypertension,kidney failure, and cardiovascular disease.

In paper I, 50 obese patients without diabetes were treated for 24 weeks with SGLT2idapagliflozin + GLP-1 receptor agonist (GLP-1RA) exenatide or placebo. They were examinedregarding body weight loss and body composition. The placebo-adjusted weight loss was 4.13kg, mostly attributable to adipose tissue loss.

In paper II, 43 completers of the study in paper I entered a 28-week extension phase in whichall participants received active treatment. We found major reductions in body weight, adiposetissue volume, blood pressure and prediabetes that were sustained at 52 weeks.

In paper III, 84 patients with T2D and non-alcoholic fatty liver disease underwent a 12-weektreatment with dapagliflozin, omega-3 (n-3) carboxylic acids (OM-3CA), the combination ofboth or placebo to assess effects on liver fat content. MRI showed significant reductions of liverfat versus baseline and, for the combination, versus placebo.

In paper IV: 15 metformin-treated patients with T2D were assessed for changes in plasmaglucagon levels following a single dose of dapagliflozin during experiments with stable versusfalling plasma glucose. Changes in glucagon levels could largely be explained by changes inglucose levels.

In conclusion, SGLT2 inhibition can lower body weight and cardiovascular risk factors inobese patients without diabetes when combined with GLP-1RA, and it can reduce liver fat inT2D patients, in particular when given together with OM-3CA. SGLT2i effects on glucagonsecretion can largely be explained by lower glucose levels rather than direct α-cell effects.

Keywords: SGLT2 inhibition, GLP-1 receptor agonism, DPP4 inhibition, NAFLD,prediabetes, type 2 diabetes, obesity, metabolic syndrome, glucagon.

Per Lundkvist, Department of Medical Sciences, Clinical diabetology and metabolism,Akademiska sjukhuset, Uppsala University, SE-75185 Uppsala, Sweden.

© Per Lundkvist 2019

ISSN 1651-6206ISBN 978-91-513-0561-5urn:nbn:se:uu:diva-373522 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-373522)

Dedicated to Hannah and Ingrid

"Life can only be understood backwards, but it must be lived forwards." /Søren Kierkegaard

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Lundkvist, P., Sjöström, CD., Amini, S., Pereira MJ., Johnsson E., Eriksson, JW. Dapagliflozin once-daily and exenatide once-weekly dual therapy: A 24-week randomized, placebo-controlled, phase II study examining effects on body weight and prediabetes in obese adults without diabetes. Diabetes Obes Metab. 2017; 19(1): 49–60.

II Lundkvist, P., Pereira MJ., Katsogiannos, P., Sjöström, CD., Johns-son E., Eriksson, JW. Dapagliflozin once daily plus exenatide once weekly in obese adults without diabetes: Sustained reductions in body weight, glycaemia and blood pressure over one year. Diabetes Obes Metab. 2017; 19(9): 1276–1288.

Eriksson, JW., Lundkvist, P., Jansson, PA., Johansson, L., Kvarn-ström, M., Moris, L., Miliotis, T., Forsberg, GB., Risérus, U., Lind, L., Oscarsson, J. Effects of dapagliflozin and n-3 carboxylic acids on non-alcoholic fatty liver disease in people with type 2 diabetes: a double-blind randomised placebo-controlled study.

IV Lundkvist, P., Pereira MJ., Kamble, PG., Katsogiannos, P., Lang-kilde AM., Esterline, R., Johnsson, E., Eriksson, JW. Glucagon lev-els during short-term SGLT2 inhibition are largely regulated by glu-cose changes in patients with type 2 diabetes. J Clin Endocrinol Metab. 2019; 104(1): 193–201.

Reprints were made with permission from the respective publishers.

Contents

Introduction ................................................................................................... 11

Obesity, type 2 diabetes and non-alcoholic fatty liver disease - definitions and prevalence ............................................................................................... 12

Obesity ...................................................................................................... 12Type 2 diabetes ......................................................................................... 12Non-alcoholic fatty liver disease .............................................................. 13

Pathobiology ................................................................................................. 14Obesity ...................................................................................................... 14Type 2 diabetes ......................................................................................... 15Non-alcoholic fatty liver disease .............................................................. 15

Impact on morbidity and mortality ................................................................ 17Obesity ...................................................................................................... 17Type 2 diabetes ......................................................................................... 17Non-alcoholic fatty liver disease .............................................................. 18

Prevention and therapy .................................................................................. 19Lifestyle .................................................................................................... 19Bariatric surgery ....................................................................................... 19Obesity medication ................................................................................... 20Antidiabetic medication ............................................................................ 21Treatments options for non-alcoholic fatty liver disease .......................... 22

SGLT2 inhibitors .......................................................................................... 23

Incretin-based therapies ................................................................................. 26

Omega-3 carboxylic acids ............................................................................. 28

Overview of investigated treatment effects ................................................... 29

Study aims and hypotheses ........................................................................... 30

Subjects and study design ............................................................................. 31Paper I ....................................................................................................... 31Paper II ..................................................................................................... 32Paper III .................................................................................................... 32

Paper IV .................................................................................................... 33

Methods ......................................................................................................... 35Body weight .............................................................................................. 35Magnetic Resonance Imaging .................................................................. 35Oral Glucose Tolerance Test .................................................................... 35Biochemical analyses ............................................................................... 36Experimental visit design ......................................................................... 36Statistical analyses .................................................................................... 36

Results ........................................................................................................... 38Paper I ....................................................................................................... 38Paper II ..................................................................................................... 38Paper III .................................................................................................... 39Paper IV .................................................................................................... 40

Discussion ..................................................................................................... 42Cardiometabolic risk factors ..................................................................... 42Cardiometabolic benefits .......................................................................... 47Incretins .................................................................................................... 47Safety ........................................................................................................ 48

Conclusions ................................................................................................... 50

Sammanfattning (summary in swedish) ........................................................ 51

Acknowledgements ....................................................................................... 53

References ..................................................................................................... 55

Figues and tables Figure 1: Pathobiology of type 2 diabetes ..................................................... 15Figure 2: SGLT2 inhibition, mechanism of action ....................................... 24Figure 3: Effects of investigated treatment modalities .................................. 29Figure 4: Design of studies I and II ............................................................... 32Figure 5: Design of study III ......................................................................... 33Figure 6: Design of study IV ......................................................................... 34Figure 7: Body weight (kg), 1-year change (studies I and II) ....................... 39Figure 8 A-B: Glucose and glucagon changes from baseline (study IV) ...... 41Figure 9: Proposed metabolic and endocrine effects of SGLT2 inhibition

and coadministered drugs ............................................................... 49

Table 1: Studies I and II participants clinical characteristics ........................ 31Table 2: Study III participants clinical characteristics .................................. 33Table 3: Study IV participants clinical characteristics .................................. 34Table 4: Treatment effects on body weight, abdominal adipose tissue, liver

fat and damage markers (study III) ................................................ 40

Abbreviations

α-MSH Alpha melanocyte stimulating hormone AASLD American association for the study of liver

diasease AE Adverse event aGLP-1 active GLP-1 AgRP Agouti-related peptide ALT Alanine transaminase AST Asparatate transaminase AUC Area under curve BMI Body mass index CART Cocaine amphetamine regulated transcript CI Confidence interval CRP C-reactive protein CV (D) Cardiovascular (disease) CVOT Cardiovascular outcome trial DAPA Dapagliflozin DHA Docosahexaenoic acid DPP4 (i) Dipeptidyl peptidase 4 (inhibitor) eGFR Estimated glomerular filtration rate ELISA Enzyme-linked immunosorbent assay EPA Eicosapentaenoic acid EXE Exenatide FGF21 Fibroblast growth factor 21 GBP Gastric bypass GCP Good clinical practise GLP-1 (RA) Glucagon-like peptide 1 receptor agonist GMR Geometric mean ratio HbA1c Glycosylated haemoglobin HDL High density lipoproteins IBT Incretin-based therapies IFG Impaired fasting glucose IGT Impaired glucose tolerance IL-6 Interleukin 6

LDL Low density lipoproteins LSM Least square means MA Monoamine MACE Major cardiovascular event MetS Metabolic syndrome MMRM Mixed model of repeated measurements MRI Magnetic resonance imaging NAFLD Non alcoholic fatty liver disease NASH Non alcoholic steatohepatitis NEFA Non estrified fatty acids NPY Neuropeptide Y n-3CA see OM-3CA OD Once daily OGTT Oral glucose tolerance test OH Hydroxyl group OM-3CA Omega-3 carboxylic acids PBO Placebo PDFF Proton density fat fraction PG Plasma glucose PNPLA3 Patatin-like phospholipase domain-

containing protein 3 POMC Proopiomelanocortin PYY Polypeptide Y QW Once a week (quantum vis) SAXA Saxagliptin SCD-1 Stearoyl-CoA desaturase 1 SD Standard deviation SEM Standard error of the mean SGLT Sodium glucose cotransporter SGLT2 (i) Sodium glucose cotransporter 2 (inhibitor) SubQ Subqutaneous T2D Type 2 diabetes mellitus Tmax Time to maximum concentration TNF-α Tumor necrosis factor-alpha TG Triglycerides Visc Visceral γ-GT Gamma-glutamyl transferase

11

Introduction

Adipose tissue or fat constitutes most of mans stored energy reserves and have no doubt been life-saving at times of famine or disease throughout history. However, following the agricultural revolution and subsequent industrialization, food supply has risen and labor-intensive work diminished. As this trend has seemingly con-tinued, today's easy calorie access and sedentary lifestyle have skewed energy balance, arguably transforming the once life-saving fat reserve into one of today's most significant threats to public health.

12

Obesity, type 2 diabetes and non-alcoholic fatty liver disease - definitions and prevalence

Obesity Overweight and obesity are defined as a body mass index (kg/m2; BMI) ≥ 25 and ≥ 30 respectively. According to world health organization, more than 1.9 billion of the global adult population were overweight, and 650 million were obese in 2016.1 Corresponding percentages in Sweden was 36% overweight and 15% obesity, a prevalence rising with age.2

BMI is used as a simple and yet powerful measure of over- or underweight3,4, but it can be misleading as it does not take body composition and fat distribution into account5. Body fat amount is in itself linked to the risk of metabolic complications of obesity5. Moreover, intraabdominal or visceral adipose tissue has proven more harmful than subcutaneous fat depots6,7,8,9. Body fat amount and distribution can be estimated with bioimpedance scales or more precisely with dedicated ultra-sound/magnetic resonance imaging (MRI) scans. The simple biometrics waist circumference and waist-to-hip ratio are also shown to correlate highly with radio-logical findings of visceral adiposity10.

Type 2 diabetes Type 2 diabetes mellitus (T2D) is a dysregulated metabolic state characterized by hyperglycemia due to insulin resistance together with a relative insulin deficiency11. High glucagon secretion12, increased renal glucose reabsorption13 and impaired incretin signaling14,15 have also been shown to contribute to the develop-ment of T2D.

Prediabetes often precedes T2D and is characterized by impaired fasting glucose (IFG; 5.6-6.9 mmol/l), impaired glucose tolerance at 120 min post glucose chal-lenge (IGT; 75 grams glucose, 7.8-11.0 mmol/l) or raised glycosylated hemoglo-bin (HbA1c; 39-47 mmol/mol)16. Impaired fasting glucose is linked with peripher-al insulin resistance (skeletal muscle and adipose tissue), while IGT is thought to reflect hepatic overproduction of glucose17. The reactive beta-cell hypertrophy and hyperinsulinemia transiently stave off T2D, but without lifestyle or medical inter-vention, a subsequent beta cell failure and hyperglycemia generally follows. Type 2 diabetes is diagnosed when one or more of the following criteria are met; fasting

13

plasma glucose (PG) ≥ 7 mmol/l, random or 2-hour OGTT PG ≥ 11.1 mmol/l, or HbA1c ≥ 48 mmol/mol16. The global prevalence of T2D in 2017 was estimated to be 451 million and may reach 693 million by year 204518. In Sweden, 3-5 %19 of the population have a T2D diagnosis and prevalence rises with age to around 9 % ≥ 65 years20.

Non-alcoholic fatty liver disease Around 75 % of people with T2D21,22 have pathological amounts of ectopic fat depositions in the liver. Liver fat is considered a disease when it exceeds 5.5 % and is then called non-alcoholic fatty liver disease (NAFLD). Around 25-30 % of NAFLD patients develop a more severe inflammatory liver condition called non-alcoholic steatohepatitis (NASH), accounting for 5-12 % in the entire popula-tion22,23.

14

Pathobiology

Obesity The etiology of overweight and obesity is complex and often a mix of genetic predisposition24 and the environment. Factors like high/low birthweight25,26, neu-roendocrine disorders27, pharmaceuticals28, activity levels29, psychosocial30, and socioeconomic status31 all contribute. Although reasons vary, a positive energy balance is the common denominator. Feeding behavior and energy metabolism play integral parts here and are both largely regulated via the arcuate nucleus lo-cated in the hypothalamus32. Furthermore, the amygdala and nucleus accumbens have been associated with satiety and food reward respectively33,34. Genetic predisposition to obesity is important, with twin and adoptee studies indi-cating a heritability of 55-85 percent24. However, the genetics are complex, heter-ogeneous and only explains small BMI differences overall35. Monogenic mutations can cause morbid obesity, affecting central appetite regulation in the hypothala-mus or amygdala36 (e.g., leptin signaling pathway37). A similar loss of function due to traumatic injury, tumor or radiotherapy may also induce overeating and obesity38. In health, many of the arcuate nucleus’ neuroendocrine neurons contribute to bal-ancing the energy homeostasis and are themselves governed by systemic circula-tion of metabolic peptide hormones. Anorexigenic peptides induce satiety and include glucagon-like peptide 1 (GLP-1)39, leptin40, insulin, and peptide YY (PYY)41. They promote satiety by stimulating α-melanocyte stimulating hormone (α-MSH) release via proopiomelanocortin/cocaine-amphetamine-regulated tran-script (POMC/CART) and through inhibition of neuropeptide Y/agouti-related peptide (NPY/AgRP) neurons. Ghrelin, on the other hand, is an orexigenic peptide that inhibits the POMC/CART and stimulates NPY/AgRP pathways respectively.42 Similarly, glucocorticoids stimulate appetite centrally but also induce peripheral insulin resistance and contribute towards visceral adiposity.43 While insulin has a net anabolic effect in practice and leptin supplementation only works in the genetically leptin-deficient, GLP-1 receptor agonists (GLP-1RA) has shown promise in obesity treatment clinically44,45.

15

Type 2 diabetes The development of type 2 diabetes is dependent on genetic46 and environmental factors47. Low birthweight48, ethnicity49, inactivity, disease and medications can all increase T2D risk. However, two of the most prominent T2D risk factors are obe-sity and large adipose tissue mass50. The body fat distribution also matters, with visceral fat correlating stronger with insulin resistance and cardiovascular (CV) disease than gluteal and femoral adiposity.51 Furthermore, ectopic fat – lipid ac-cumulation in non-adipose tissue – in skeletal muscle, heart, and liver seem to have a proinflammatory role and can decrease organ function and glucose dispos-al47. Interestingly, inflammation markers C-reactive protein (CRP), interleukin 6 (IL-6) and tumor necrosis factor-α (TNF-α) correlate to the prevalence of diabetes and prediabetes47,52.

Figure 1: Pathobiology of type 2 diabetes. Type 2 diabetes (T2D) is characterized by hy-perglycemia due to insulin resistance, subsequent beta cell failure, inherent hyperglucago-nemia, inappropriate hepatic glucose production as well as increased renal glucose reuptake and low levels of incretin hormone GLP-1, glucagon-like peptide 1.

Non-alcoholic fatty liver disease Non-alcoholic fatty liver disease is diagnosed when ectopic liver fat deposition exceeds 5.5 %. Around 25-30% of NAFLD patients also develop NASH, a more severe condition where liver lobes become inflamed and liver cells start to balloon.

16

This inflammation causes subsequent liver cell damage and increased hepatocyte injury markers in circulation. Why liver inflammation develops in some patients, but not others is incompletely known. There is however evidence that specific genes play a role in NASH sus-ceptibility, most notable of which is single nucleotide polymorphism in I148M in gene PNPLA353. Moreover, the inflammatory component seems linked to the ina-bility of liver cells to accommodate the increased flux of fatty acids causing toxic lipid intermediates like diacylglycerol and ceramides. Mechanisms of toxicity remain largely unknown but may include endoplasmic reticulum stress and oxida-tive stress due to mitochondrial overload and subsequent dysfunction.54

17

Impact on morbidity and mortality

Obesity Overweight and obesity negatively affect several CV and metabolic risk factors including hypertension, coronary heart disease, stroke, sleep apnea, osteoarthritis, hyperlipidemia, insulin-resistance, NAFLD and T2D. It further increases the can-cer risk of the breast, colon, prostate, endometrium, kidney, and gallbladder.1 Although not accounting for body composition, BMI in itself is a strong predictor of all-cause mortality, both above and below an optimal ratio of 22.5-25 kg/m2. This formes a J-shaped curve showing a substantial increase in excess mortality for each 5 kg/m2 increment of BMI. Median survival is decreased by 2-4 years at BMI 30-35 kg/m2, and 8-10 years at BMI 40-45 kg/m2. When broken down into categories, this excess mortality is driven mainly by vascular disease.4 The so-called metabolically healthy obese – pertaining to unaffected blood pressure, blood lipids and glucose metabolism – also display increased CV risk and all-cause mor-tality over ten years.55 Hence, large fat mass per se seemingly increases the risk of mortality and cardiovascular disease (CVD) beyond these surrogate metabolic markers.

Type 2 diabetes While lifestyle interventions can slow the progression of prediabetes into T2D56, further dysregulation is far more common. Type 2 diabetes has implications for many organ systems as it is a prominent risk factor for micro- and macrovascular disease, often affecting heart, liver, nervous system, arteries, kidney, and retina57. While CV-risk factor reduction has intensified in recent years, benefitting people with and without diabetes alike, the relative risk to die from CVD remains two- to threefold higher in the diabetes cohort58. More than 80 percent of T2D patients (in the U.S.) are also overweight or obese59, many of them with central or visceral adiposity. Such dysregulated metabolic states often occur together with hyperten-sion and deranged blood lipids clustering into the so-called Metabolic Syndrome (MetS)60.

18

Non-alcoholic fatty liver disease The link between NAFLD and metabolic disease with visceral adipose tissue, obe-sity and hepatic insulin resistance is strong, estimated to affect a majority of T2D-patients21,22. Patients with NAFLD have an increased risk of dying from liver dis-ease, malignancy and CVD and around one quarter of them develop NASH, rais-ing this risk even further23. Type 2 diabetes is associated with such disease pro-gression, as well as an increased risk of further developing liver fibrosis and hepa-tocellular carcinoma23,61. NASH has also been linked with higher risk of chronic kidney disease and CV events62.

19

Prevention and therapy

Except for lipase inhibitors and traditional hypoglycaemic agents respectively, treatment options for obesity and T2D (in Sweden) largely overlap. Lifestyle mod-ification and bariatric surgery target both obesity and T2D and antihypertensive and lipid-lowering drugs tackle the CV risk factors they often share60. Interesting-ly, the two new drug classes incretin-based therapies (IBT) and SGLT2 inhibitors (SGLT2i) have now been shown to similarly reduce body weight, CVD incidence, plasma glucose, and systolic blood pressure, but via different mechanisms of ac-tion63,64,65,66,67.

Lifestyle Lifestyle recommendations include a varied, healthy diet (e.g., mediterranean) and regular exercise most days to decrease body weight. These interventions can im-prove many risk factors of T2D including body weight and composition, glycemic control, lipid profile and peripheral insulin sensitivity. Lifestyle changes also low-er the risk for commonly associated comorbidities such as cancer and CVD. Long term study results cannot distinguish between various hypocaloric diet fads regard-ing weight loss.68 The combined diet and exercise interventions can improve cardiometabolic risk factors, increase insulin sensitivity and prevent progression into type 2 diabetes as compared to controls, especially over 60 years of age.69,70,71 However, exercise alone has little effect on body weight loss over time, probably as feeding increase in a compensatory manner.72 A negative energy balance is also accompanied by a decrease in energy expenditure while appetite for energy-dense foods increases, seemingly protecting against weight loss.73 This is perhaps a critical explanatory factor to why long term weight loss by lifestyle modification alone has proven so difficult to maintain.

Bariatric surgery Bariatric surgery has become a prevalent method to reduce body weight, with sur-gical procedures varying with body weight (at least BMI ≥ 35 kg/m2) and patient preference. The so-called Roux-en-Y gastric bypass (GBP) is, however, the most common variant in Sweden today. In GBP, a new and shorter alimentary canal is

20

formed by a small stomach pouch that is attached directly to the jejunum and thereby bypassing the remaining ventricle and duodenum. The pouch mechanical-ly only accommodates 15-30 ml at a time, forcing the patient to take small meals several times a day. The bypass itself reduces energy uptake from the intestine, lowering body weight but often also causing a deficient uptake of vitamin B12, iron, protein and folate. This malnutrition leads to the need for lifelong high pro-tein diet and vitamin supplementation.74 Following surgery, a weight loss of 0.5-1 kg per week is to be expected. Most of the weight reduction occurs during the first 6-12 months, with a plateau often reached after two years. Twelve years after surgery, average weight loss was 35 kg and T2D remission 51 % in those with the disease at baseline75. Complications of GBP include leakage or strictures of the GBP anastomosis, internal herniation, dumping, and gallstones.

Obesity medication The majority of obesity drugs to date suppress appetite through upregulation of monoamine turnover (MA; serotonin, noradrenaline, dopamine) in the CNS. These compounds induce satiety by either MA release, reuptake inhibition or in some cases via MA receptor agonism. Treatment efficacy after one year versus placebo range from approximately 2.5 – 9 kg weight loss but is frequently accompanied by high attrition rates and partial weight rebound in the second year. Side effects are often GI or nervous system related, and percent responders (~30-40 %) are gener-ally comparable between substances.76,77 Bupropion/naltrexone (Contrave, Mysimba) promote satiety via noradrenaline-dopamine reuptake inhibition while opioid receptor antagonist naltrexone increas-es its POMC/CART signal frequency. The result is a 1-year weight loss of ~5 kg versus placebo, even so, systolic blood pressure and pulse may increase. Common adverse events include nausea, vomiting, constipation, dry mouth, headache, diz-ziness, insomnia and anxiety.76,77 Sibutramine (Reductil) is a serotonin and noradrenaline reuptake inhibitor that was withdrawn from the market in 2010 following the CVOT study showing an in-creased risk of major adverse cardiovascular events (3-point MACE; including nonfatal stroke, nonfatal myocardial infarction, and CV death).76,77,78 Phentermine/topiramate (Qsymia) combines the noradrenaline release in central neurons by phentermine with anticonvulsant topiramate's weight reducing side effects. One years´ treatment yields a weight loss of 8.8 kg versus placebo as well as moderately improved glycemia, TGs, and systolic blood pressure. Common side effects include GI disturbance, cognitive and nervous system symptoms.76,77

21

Lorcaserin (Belviq) is a selective HT52c agonist inducing satiety through the POMC/CART pathway. It lowers body weight by 3.2 kg versus placebo after one year but is associated with a rebound in body weight the second year despite ongo-ing treatment. Commonly reported adverse effects are mainly attributed to the GI tract and nervous system. The safety trial CAMELLIA recently showed improved CV risk factors without significantly reducing MACE.76,77,79 Rimonabant (Acomplia, Zimulti) is an inverse agonist of cannabinoid receptor 1 (CB-1) inhibiting feeding signals in the hypothalamus. It lowered body weight 4.7 kg versus placebo with one-year treatment but was withdrawn in 2008 due to se-vere psychiatric side effects including anxiety, depression, and suicidal behavior.80 Orlistat (Xenical, Alli) is a pancreatic and gastric lipase inhibitor that reduces deg-radation and uptake of fat in the intestine. It lowers body weight moderately (-2.6 kg), can reduce prediabetes and plasma lipids. However, discontinuation of treat-ment is common due to unfavorable side effects such as soiling, fecal urgency, and incontinence as well as loss of efficacy over time.76,81

Antidiabetic medication Before the advent of IBTs and SGLT2is, antidiabetics consisted mainly of insulin, glucose-independent insulin secretagogues, and insulin sensitizers. As such, the price of glucose lowering has often been increased body weight as well as a higher risk of hypoglycemia. Insulin effectively decreases hyperglycemia while simultaneously increasing body weight and hypoglycemia frequency.82 Although newer insulin analogs (e.g., de-temir, glargine) have flatter profiles and (possibly explaining) fewer hypoglycae-mias as well as less weight gain, exogenous insulins have yet to show a significant decrease in CV disease and mortality83. Sulfonylurea (SU; Mindiab) is an insulin secretagogue, inducing insulin secretion by closing KATP channels in pancreatic β cells. The depolarization causes glucose-independent insulin release, in turn increasing body weight and the risk of hypo-glycemia.84 No dedicated outcome study has shown CV benefits of SU treatment. Metformin (Glucophage) is a biguanide that lowers hepatic glucose production and peripheral insulin resistance. It is the first-line medication for T2D, is body weight neutral, unsuitable with renal or hepatic impairment and has GI side ef-fects. Metformin was the first antidiabetic agent to reduce CV complications as seen with the pivotal UKPDS study85. However, the evidence is relatively weak and only valid for overweight patients and has since then been called into ques-tion86.

22

Thiazolidinediones (TZD; Actos, Avandia) alter gene transcriptions through PPARγ activation and thereby sensitize adipose tissue to insulin. TZDs increase adipogenesis and body weight but may also redistribute fat from visceral to subcu-taneous depots87. Common side effects include water retention and reduced bone mineral density while reported CV effects are contradictory88,89.

Treatments options for non-alcoholic fatty liver disease Currently, weight loss and exercise is the recommended treatment regime of NAFLD; there are no drugs approved with this indication23,54. However, according to the American Association for the study of liver disease (AASLD) guidelines 2018, there are a couple of compounds that may be considered in biopsy-proven NASH. One such drug is TZD pioglitazone (Actos) that can reduce liver fat and improve liver histology but is associated with certain cancers, bone fractures, weight gain, and fluid retention.89,90 Another one is vitamin E, which may be con-sidered in patients without diabetes or cirrhosis. There are however several con-cerns about the long-term safety of vitamin E, including the risk of hemorrhagic stroke, overall mortality, and prostate cancer.90 Obeticholic acid (OCA), a type of synthetic bile acid that activates the farnesoid X receptor, have also proven histo-logically visible NASH benefits versus placebo. OCA is currently not recom-mended for NASH and is associated with pruritus and increased LDL.90,91 Other agents in the treatment pipeline include omega-3 carboxylic acids (OM-3CA) that may be considered in NAFLD with hypertriglyceridemia. Newer antidi-abetic treatment modalities have also shown promise regarding NAFLD treatment. Although uncertain if and how GLP-1RA have any direct effects in the liver, sev-eral agents have shown positive effects on NASH. For instance, both exenatide and liraglutide have shown reduced liver fat, hepatocyte injury markers, and im-proved fibrosis or fibrosis score.92,93,94 SGLT2i has shown promising results on liver health in obese rodents as well as lowering of liver fat, hepatocyte injury markers and fibrosis score humans 90,95,96,97,98,99. However, the overall knowledge of SGLT2is effect on NAFLD and NASH remain limited.

23

SGLT2 inhibitors

The kidneys contribute to glucose homeostasis via gluconeogenesis (15-55 g/day) and glucose reabsorption through symports in the proximal convoluted tubule (180 g/day). Sodium glucose cotransporters 1 (SGLT1 high-affinity low capacity) and 2 (SGLT2 low-affinity high capacity) utilize the electrochemical sodium gradient to transport glucose against concentration gradients. SGLT2 is located primarily in segment S1 followed by SGLT1 in S3 of the proximal renal tubule. Ninety percent of filtered glucose is absorbed via SGLT2 with residuals escaping urinary excre-tion via SGLT1. Healthy kidneys reabsorb glucose linearly up to a concentration threshold around 11 mmol/l and after that succumb to glucosuria.13,100 However SGLT2 gene SLC5A2 is overexpressed in chronic hyperglycaemia101 (e.g., predia-betes, T2D), raising the reabsorptive threshold and furthering hyperglycemia as well as promoting renal hyperfiltration.102 Conversely, there are rare hereditary mutations of SLC5A2 (e.g., familial renal glucosuria) that cause varying loss of the reabsorptive function of SGLT2. The mutations result in glucosuria, polyuria and sometimes urogenital infections but seldom hypoglycaemia101. In the 19th century, similar hypoglycemic effects as with SLC5A2-mutations were achieved with administration of apple tree bark isolate phlorizin, an unspecific SGLT inhibitor. Its use as a therapeutic glucoside was however eventually aban-doned due to GI-related side effects, largely owing to the high density of SGLT1 in the small intestine.103 More recently, SGLT interest was rekindled, and first selective SGLT2 inhibitor dapagliflozin was produced in 2008104. The first in class highly selective SGLT2 inhibitor (SGLT2i), dapagliflozin (Forx-iga 10 mg) was introduced to the market in 2012 for the T2D indication and was soon followed by empagliflozin (Jardiance 25 mg) and canagliflozin (Invocana 100/300 mg). A fourth compound, ertugliflozin (Steglatro 5/15 mg) recently came to market in the US and EU. The four US/EU approved substances induce up to ~80 grams of glucosuria and a mild osmotic diuresis/natriuresis. Results show improved glycemia and insulin sensitivity105, reduced systolic/diastolic blood pres-sure 3-5/2 mmHg as well as slowed progression of kidney disease106. Furthermore, a consistent and maintained107 body weight loss around 2-3 kg106,108 may be mostly due to decreased fat mass109. Actual weight loss is however substantially smaller than the approximate -11 kg expected reduction with unchanged caloric intake110.

24

Figure 2: SGLT2 inhibition, mechanism of action. Sodium glucose cotransporter 2 (SGLT2) reabsorbs 90 % of filtered glucose. Inhibition of SGLT2 blocks glucose reab-sorption in renal proximal tubule. This inhibition results in urinary glucose excretion ~60-80 grams/day which in turn lowers plasma glucose, blood pressure and body weight. Na+, sodium.

SGLT2i-induced glucosuria lower HbA1c 6-13 mmol/mol compared with placebo, can raise LDL moderately and increases the frequency of genital mycotic infec-tions111,112,113,114. The osmotic diuresis increases blood viscosity and can exacerbate volume depletion, raising the risk of serious adverse events such as venous throm-boembolism as well as hypotension and acute kidney damage respectively. Other serious adverse events reported include euglycemic diabetic ketoacidosis (eu-DKA), lower limb amputations, bone fractures, and acute pancreatitis.115 In partic-ular euDKAs have been debated following findings that SGLT2i can increase glu-cagon105 and lipid oxidation116,117 and can have a direct effect on pancreatic α-cell glucagon release118. Two cardiovascular outcome studies (CVOT) with SGLT2i have shown reduced MACE to date. These studies are EMPA-REG (empagliflozin) and CANVAS (canagliflozin), where the former also lowered CV death.66,119 Dapagliflozin CVOT (DECLARE-TIMI 58) displayed significantly lowered hospitalization for heart failure but not reduced MACE or CV-death. Furthermore, the CVD-REAL observational study indicates that the SGLT2i class decreased risk for premature death120.

25

Mechanisms that are responsible for the apparently fast working CV and kidney-sparing properties of SGLT2i are incompletely known. A prevailing hypothesis focus on fluid volume changes, causing interstitial congestion relief through os-motic diuresis and a lower blood pressure121. Natriuresis is thought to contribute through increased tubuloglomerular feedback thereby alleviating hyperfiltration and glomerular pressure122. A shift towards cardiac substrate utilization towards energy efficient ketone bodies and an organ-sparing inhibition of Na+/H+ exchang-er 1 and 3 of the heart and kidney respectively are two other current theories117,123,124.

26

Incretin-based therapies

Incretins are a small group of metabolic hormones with hypoglycemic properties and include GLP-1 and glucose-dependent insulinotropic peptide (GIP). GLP-1 and GIP mainly originate from enteroendocrine L and K cells respectively and are secreted from the intestinal mucosa within minutes of feeding. These peptides act by delaying GI passage, stimulating glucose-dependent β-cell insulin release, downregulating gluconeogenesis by lowering glucagon as well as suppressing appetite.125 GLP-1 is lower in T2D and obesity, and dipeptidyl peptidase-4 (DPP4) upregulated in T2D according to some studies126,127,128. Native GLP-1 is a 30 amino acid neuropeptide cleaved from proglucagon, have a half-life of 1 - 3 minutes and is broken down primarily by DPP-4 before it is ex-creted via the kidneys.125 GLP-1 receptors (GLP-1R) have been found in the pancreas, GI-tract, kidneys, brain129, peripheral nervous system. GLP-1 also seem to have direct cardiovascular effects130,131,132. Incretin-based therapies to date include injectable GLP-1 receptor agonists (GLP-1RAs) and DPP4 inhibitors (DPP4i) taken orally, although an oral preparation of GLP-1RA semaglutide is currently in phase III133. GLP-1RA mimics endogenous GLP-1 while DPP4i prevents the enzymatic degradation of endogenous GLP-1 and GIP, thereby increasing their concentrations post-prandially134. Marketed GLP-1RAs are most often human GLP-1 molecules with altered amino acid sequence and added protein or lipid moieties for increased stability and half-life (90-98 % homology). Synthetic GLP-1 mimetics exenatide and lixisenatide are based on exendin 4, a peptide in the Gila monster saliva (50-53 % homology) and have similar GLP-1R-agonistic properties to human GLP-1.135,136,137 GLP-1RA and DPP4i similarly promote glycaemic control by stimulating insulin release, inhibiting glucagon secretion and lowering glucose levels in the post-prandial state138,139. Generally, GLP-1RAs are more potent than DPP4is but also have more side effects140, most notably increased GI discomfort and nausea141. GLP-1RAs further reduce systolic blood pressure by ~2-3 mmHg, increase heart rate by 2-3 bpm130, lower kidney disease65, TGs and LDL moderately and reduce body weight by 2-4 kg.132,142 Also, studies with increased GLP-1RA dosage and treatment time have shown further weight loss potential. Liraglutide 3.0 (Saxenda)

27

reduced body weight by 4.3% after three years of treatment after placebo-adjustment as well as improved glycemia and lipid profile63. Semaglutide 1 mg (Ozempic) weekly preparation show similar weight loss properties45. Although knowledge of central GLP-1 signaling is incomplete, weight loss effects have been linked with inhibition of NPY/AgRP 39. The combined effect of central appetite suppression and gastric emptying is likely to explain the weight loss with GLP-1RA treatment. In addition, GLP-1RA CVOTs have now demonstrated a reduced incidence of MACE when added to standard of care in patients with T2D. Four human GLP-1 RAs (albiglutide, dulaglutide, liraglutide, semaglutide) have displayed superiority versus placebo for 3-point MACE. Liraglutide also reduced all-cause mortality and CV death and semaglutide also decreased nonfatal strokes.64,65,143,144 Incretin mi-metics exenatide and lixisenatide have shown non-inferiority to placebo145,146.

28

Omega-3 carboxylic acids

Omega-3, or n-3, carboxylic acids, consist of fish oil derivatives eicosapentaenoic (EPA) and docosahexaenoic (DHA) fatty acids. OM-3CA can reduce triglycerides and liver fat without simultaneous body weight loss and is a treatment option in patients with both hypertriglyceridemia and NAFLD.147 Underlying mechanisms of action are incompletely understood, but increased basal metabolic rate and lipid oxidation seen with OM-3CA may matter.148 On the flipside, DHA has also been associated with raised levels of LDL, an acknowledged risk factor of coronary heart disease.149 There are no OM-3CA formulations for prescription in Sweden, meanwhile there are half a dozen of them available in the U.S. These include EPA/DHA mixtures Epanova, Omtryg, Lovaza (and two generics thereof) as well as EPA-only formu-lation Vascepa which recently showed ~25 % reduction MACE in a population with high CV-risk (REDUCE-IT).150 Although, no EPA/DHA-formulation has shown positive CV-outcomes to date, a study of cardiovascular benefit with Epa-nova (STRENGTH) is currently ongoing. OM-3CA side effects include GI-disturbances like nausea, burping, diarrhea, and abdominal pain.151

29

Overview of investigated treatment effects

Investigated treatments have different mechanisms of action, potentially increasing their effects when combined. Hence, giving SGLT2i together with incretin-based therapies (GLP-1RA or DPP4i) or OM-3CA are interesting therapy options for lowering of cardiometabolic risk factors and liver fat respectively.

Figure 3: Effects of investigated treatment modalities. Pharmacotherapy effects of interest for one or more of the treatments employed. Less established research questions in italics. SGLT2i = sodium glucose co-transporter 2 inhibitor; omega-3CA = omega-3 carboxylic acids; GLP-1RA = glucagon-like peptide-1 receptor agonist; DPP4i = dipeptidyl peptidase 4 inhibitor; systolic BP = systolic bloodpressure; MACE = major adverse cardiovascular events.

30

Study aims and hypotheses

The overall aim of this thesis is to improve understanding of how SGLT2 inhibi-tion affect glucose, lipid and energy metabolism as well as to elucidate underlying hormonal and metabolic mechanisms. To this aim, the following four tasks are performed: The first task is to evaluate the effects of SGLT2i with GLP-1RA on body weight and body composition in obese nondiabetic subjects compared with placebo. The hypothesis is that the combination treatment yields positive additive or synergistic effects on body weight and body composition. The second task is to evaluate the maintenance of body weight loss, body fat re-duction and decreased prediabetes, lipids and blood pressure with continuing SGLT2i and GLP-1RA treatment for 52 weeks. The hypothesis is that reductions in these outcome variables are maintained following one year of combination treatment. The third task is to evaluate how SGLT2i and OM-3CA, alone and in combination affect liver fat content in T2D patients with NAFLD compared with placebo. The hypothesis is that combination treatment can yield additive effects on liver fat deposition. The fourth task is to assess T2D patients for changes in glucagon levels, following a single dose of SGLT2i during experiments with stable versus falling plasma glucose. The hypothesis is that changes in glucagon levels could largely be ex-plained by changes in glucose levels.

31

Subjects and study design

Paper I A phase IIa, 24-week, double-blind, parallel group, placebo-controlled study of body weight changes in 50 obese subjects without diabetes (prediabetes, defined as either or both IFG and IGT, was 73.5 %) randomized to oral dapagliflozin 10 mg (DAPA) daily plus subcutaneous long-acting exenatide 2 mg once weekly (EXE) or placebo. Participants were instructed to follow a balanced diet and exer-cise moderately. Visits included screening, 1-2 weeks before randomization (week 0) with follow-up visits at weeks 4, 8, 12 and 24 as well as an optional 28-week open-label extension. Exploratory measures included body composition (by MRI), waist and hip circumference as well as hormones and substrates in fasting and during oral glucose tolerance tests (OGTT).

Table 1: Studies I and II participants clinical characteristics

SGLT2i + GLP-1RA (n=25) Placebo (n=24)

Age (yrs) 53 (13) 50 (12) Male sex, n (%) 10 (40) 9 (38) Body weight (kg) 106.4 (15.6) 102.7 (17.3) Body mass index (kg/m2) 35.8 (2.9) 35.0 (3.7) Waist circumference (cm) 117.6 (11.3) 114.4 (12.6) Total adipose tissue (l) 57.1 (9.7) 53.4 (7.0) Liver fat (%) 10.9 (10.6) 10.0 (8.5) HbA1c (mmol/mol) 37.2 (3.9) 38.1 (3.3) Impaired fasting glucose,a n (%) 16 (64) 17 (71) Impaired glucose tolerance,b n (%) 12 (48) 8 (33) Diastolic blood pressure (mmHg) 75.8 (10.6) 77.0 (12.6) Systolic blood pressure (mmHg) 133.7 (12.7) 133.8 (17.5)

Data are N or mean (SD). eGFR, estimated glomerular filtration rate; HbA1c, glycated hemoglobin; MDRD, modification of diet in renal disease formula; n, number of random-ized subjects. a fasting plasma glucose 5.6-6.9 mmol/l b 120-min plasma glucose 7.8-11 mmol/l.

32

Paper II A 28-week open-label extension study, in which all participants received active treatment, evaluating how changes from baseline in body weight, adipose tissue volume, glycemia, and systolic blood pressure was maintained in the continuing active treatment group. Outcome measures in the group switching from placebo to active treatment were evaluated from a confirmatory perspective. Participants were 38 of the study completers in paper I.

Figure 4: Design of studies I and II. Participants were randomized to 24 weeks of treat-ment with either SGLT2i dapagliflozin 10 mg once daily plus GLP-1RA exenatide 2 mg once weekly or placebo; QD, once daily; QW, once weekly. Open-label extension from week 24-52 in which all participants recieve active treatment.

Paper III A randomized, four group, placebo-controlled double-blind, double-dummy, mul-ticentre study of liver fat changes (measured as proton density fat fraction (PDFF) by MRI) in 84 subjects with T2D (HbA1c 58.9 mmol/mol) and NAFLD (PDFF 18 %) following 12-weeks of daily dosing of dapagliflozin 10 mg and OM-3CA Epa-nova 4 g alone and in combination compared with placebo. Exploratory outcome measures included changes in body weight, glycemic control, hepatocyte injury markers, oxidative stress biomarkers, and fatty acid metabolism.

33

Table 2: Study III participants clinical characteristics

Placebo OM-3 CA SGLT2i OM-3CA+ Total

SGLT2i (n = 21) (n = 20) (n = 21) (n = 22) (N = 84)

Age (years) 65.6 (6.1) 66.2 (5.9) 65.0 (6.5) 65.0 (5.4) 65.5 (5.9) Sex (male/female) 17/4 11/9 16/5 15/7 59/25 Weight (kg) 93.0 (12.2) 95.6 (13.7) 90.2 (8.7) 91.7 (12.9) 92.6 (12.0) BMI (kg/m2) 30.3 (3.1) 33.0 (4.1) 30.5 (2.8) 31.1 (3.6) 31.2 (3.5) Overweight /obesea 10/11 5/15 11/10 10/12 36/48

T2D dur (years) 6.5 (4.2) 6.3 (5.1) 6.7 (6.0) 8.5 (4.5) 7.0 (5.0) Data are reported as mean (SD), unless otherwise stated. Overweight, BMI 25–30 kg/m2; obese, BMI > 30 kg/m2. OM-3 CA, omega-3 carboxylic acids; SGLT2i, SGLT2 inhibi-tion; BMI, body mass index; T2D dur, type 2 diabetes duration.

Figure 5: Design of study III. A multi-center randomised placebo-controlled double-blind parallel-group study. Participants with type 2 diabetes and NAFLD were randomly as-signed 1:1:1:1 to treatments with SGLT2i dapagliflozin 10 mg and/or OM-3CA epanova 4 g or placebo. The primary endpoint was liver fat content assessed by MRI (proton density fat fraction) and biomarkers associated with T2D and NAFLD.

Paper IV A phase IV, randomized 3-treatment crossover, open-label study of glucagon changes following a single-dose of dapagliflozin 10 mg with 5-h isoglycemic clamp or saline infusion (experiment DG and D respectively). The third experi-ment was with saline infusion, exploring the effects of adding DPP4i saxagliptin 5 mg to SGLT2 inhibition (experiment DS). All experiments were immediately fol-lowed by 2-h OGTTs. The 15 participants had T2D and were treated with metfor-min. Exploratory outcomes included changes in lipid energy substrates and addi-tional hormones like insulin and aGLP-1.

34

Table 3: Study IV participants clinical characteristics

Baseline characteristics n=15

Gender, male/female (n) 12/3 Age (years) 67 (6) Type 2 diabetes duration (years) 7.4 (4.2) Body mass index (kg/m2) 27.1 (2.9) Body weight (kg) 87 (12) Lean mass (kg) 62 (11) Waist/hip ratio 1.0 (0.1) Waist (cm) 104 (7) Hip (cm) 105 (6) HbA1c (mmol/mol) 56.5 (5.8) Fasting plasma glucose (mmol/L) 9.6 (1.1) Metformin dose (mg) 1933 (594) eGFR MDRD (mL/min/1.73m2) 97 (18) Heart rate (bpm) 69 (9) Diastolic blood pressure (mmHg) 83 (9) Systolic blood pressure (mmHg) 142 (11)

Data are N or mean (SD). eGFR, estimated glomerular filtration rate; HbA1c, glycated hemoglobin; MDRD, modification of diet in renal disease formula; n, number of subjects.

Figure 6: Design of study IV. At the three visits participants received a single-dose of SGLT2i dapagliflozin 10mg accompanied by either of the following in randomized order (R): isoglycemic clamp (experiment DG); saline infusion (experiment D); or DPP4i sax-agliptin 5mg plus saline infusion (experiment DS). Directly after 5-h infusion periods, a 2-h oral glucose tolerance test (OGTT) was performed.

35

Methods

Body weight Body weight changes were measured with an electronic scale (TANITA) at base-line (I-IV) and 4, 8, 12 (III), 24, 38 and 52 weeks (I, II) with participants in light indoor clothing without shoes.

Magnetic Resonance Imaging In paper I-III, changes in body composition were measured by MRI using a 1.5T Achieva scanner (Philips Healthcare, Best, The Netherlands). All data was collect-ed and analyzed at a single center under blinded conditions. Liver volume and fat content were both quantified by dedicated whole liver single breath-hold scans. Liver volume was assessed by a trained professional following semi-automated segmentation. Liver fat content was quantified as the median PDFF after manual segmentation of axial image slices. In paper I and II, a semi-automated segmented whole body water-fat MRI scan was used to assess total adipose and lean tissue respectively as well as subcutane-ous and visceral abdominal adipose tissue depots. In paper III, abdominal fat content was assessed from 21-slice, 8 mm slice thick-ness of a single breath-hold axial scan at L4/L5. Subcutaneous and visceral adi-pose tissue were then separated and quantified by an automated technique.

Oral Glucose Tolerance Test Oral glucose tolerance testing was performed before and after study treatments in paper I, II (0, 15, 30, 60, 90, 120 and 180 min) and III (−15, 0, 30, 60 and 120 min) with blood sampling for glycemic, metabolic and exploratory labs to assess glucose tolerance, insulin sensitivity and energy substrate utilization.

36

Biochemical analyses Analyses of blood samples were performed in accordance with the clinical routine at the Department of clinical chemistry at Uppsala academic hospital or the Co-vance Laboratories (Madison, WI, USA). The in-house analysis included glucagon (ELISA, Mercodia AB, Uppsala, Sweden), aGLP-1 (electrochemiluminescence kit, MesoScale Discovery, Rockville, MD, USA) as well as substrate analyses glycerol (free glycerol reagent, Sigma Chemical Co, St. Louise, MO, USA) and nonesterified fatty acids (NEFA; fluorometric assay kit, Cayman Chemical, Ann Arbor, MI, USA).

Experimental visit design Paper IV was a 3-visit crossover study with the following experimental visit de-sign. Participants fasted overnight until baseline measurements at 8-9 am and im-mediately after that received a single-dose of dapagliflozin 10 mg accompanied by either of the following in randomized order: isoglycemic clamp controlled with variable glucose (10 %) infusion based on bedside measurements every 5-10 min. The amount of glucose infused was noted every 20 min during the 5-h glucose infusion for calculation of glucose infusion rate. (experiment DG); isotonic saline infusion (experiment D); or saxagliptin 5 mg plus isotonic saline infusion (experi-ment DS). Directly after 5-h infusion periods, a 2-h OGTT was performed. Concentrations of glucagon, insulin, aGLP-1, glucose, glycerol, and NEFA were measured at 0, 60, 120, 180, 300, 305 (and 315 for hormones and glucose), 330, 360, and 420 min. β-hydroxy (OH)-butyrate were measured at 0, 120, 300, and 420 min. Urine was collected, volume and glucose concentrations measured for infusion and OGTT periods respectively.

Statistical analyses All statistical analyses were performed with software SAS version 9.4 (SAS Insti-tute Inc., Cary, North Carolina) or SPSS version 24 (IBM Corp, Armonk, New York). 95% confidence intervals (CI) or standard error of the mean (SEM) were used and p-values < 0.05 were considered significant. Primary outcome measures were defined as change from baseline (II) and as com-pared to placebo (I, III, IV). Safety, full and per protocol (I, II only) analyses sets were analyzed (I-IV). Non-normally distributed data according to Shapiro-Wilks test was log transformed unless otherwise stated. Analyses were unadjusted for multiple comparisons (I, II, IV) or adjusted with Tukey’s method (III).

37

In paper I and II, least-squares mean (LSM) changes using a mixed model for re-peated measures with baseline value (continuous), and treatment (I only), week, treatment-by-week interaction and sex (categorical) as covariates. Similar model terms were used in analyses of covariance of MRI, fasting plasma glucose and ketones measurements. An unstructured matrix for the within-participant error variance-covariance was used. Proportional changes in prediabetes within and between groups were tested with paired McNemar (within the group) and Cochran–Mantel–Haenszel test (adjusted for treatment between groups) respec-tively. In paper III, log-transformed, 12-week LSM changes were analyzed in a mixed linear model with baseline value as a covariate, treatment as fixed effect and study center as a random effect. Dunnett’s test was used for pairwise comparisons versus placebo. Spearman’s rank correlation test was used to analyze pairwise correla-tions of the changes in outcome variables from baseline to week 12. The interac-tion between PNPLA3 genotype, baseline values, treatment (exploratory covari-ates), study site (random effect covariate) and the primary outcome variables were investigated using a mixed model. In paper IV, a paired Student's t-test was used to assess relative changes in AUC from baseline (calculated with the trapezoidal rule) in experiment DG and DS as compared with experiment D. A forward stepwise multiple regression analysis of pooled data from the two main visits (D and DG), included sex, subject, experi-ment type, change in plasma glucose and insulin to analyse their relative contribu-tion to change in glucagon.

38

Results

Paper I Following 24 weeks of combination treatment, participants decreased body weight by 4.13 kg, adipose tissue volume by 4.09 l and systolic blood pressure by 6.7 mmHg when adjusted for placebo. Percent of participants achieving weight loss ≥ 5 % and 10 % was 36 % and 12 % with active treatment compared with 4 % and 0 % with placebo respectively. The proportion of prediabetes improved significantly from 68% to 34.8% with active treatment.

Paper II In the continuing active treatment group, reductions at 24 weeks were sustained at 52 weeks, for body weight (−4.5 and −5.7 kg), total adipose tissue volume (−3.8 and −5.3 l), proportion with prediabetes (34.8 % and 35.3 %), and systolic blood pressure (−9.8 and −12.0 mmHg) respectively. Additionally, waist circumference, triglycerides and LDL were significantly lowered from baseline with active treat-ment. The placebo-group switching to active treatment attained similar results to those of the active treatment group in paper I.

39

Figure 7: Body weight (kg), 1-year change (studies I and II). Active treatment reduced body weight 5.69 kg from baseline, the placebo-group switching to active treatment show similar body weight decreases -4.15 kg compared with main study outcome with active treatment. DAPA+ExQW, dapagliflozin 10 mg + exenatide 2 mg; PBO, placebo; PBO→ DAPA+ExQW, placebo-group switching to active treatment dapagliflozin 10 mg + ex-enatide 2 mg at extension study start.

Paper III Following 12 weeks treatment, liver fat content was significantly reduced from baseline in all active treatment groups. Combination treatment (-21 %, adjusted p <0.05) but not monotherapy dapagliflozin (-13 %) or OM-3CA (-15 %) reduced liver fat versus placebo (-3 %). Presence of G-allele in the PNPLA3 was associated with larger live fat reductions with combination therapy, and smaller ones with monotherapies. SGLT2 inhibitors alone decreased liver cell injury markers including aminotrans-ferases, fibroblast growth factor 21 (FGF21) and γ-glutamyl transferase (γ-GT). Dapagliflozin-groups generally also lowered abdominal adipose tissue, waist cir-cumference, and body weight.

40

Table 4: Treatment effects on body weight, abdominal adipose tissue, liver fat and liver damage markers (study III)

Placebo OM-3CA SGLT2i

OM-3CA + SGLT2i

Body weight, kg B 92.9 (12.16) 95.6 (13.68) 90.3 (9.04) 91.6 (12.84)

C –0.27 (1.79) –0.16 (1.02) –2.44 (2.14)* –2.16 (1.30)*

Liver PDFF, % B 15.1 (6.5) 22.2 (11.0) 17.3 (9.1) 17.8 (9.2)

C –0.59 (1.86) –3.15 (2.88) –2.23 (3.30) –3.15 (3.49)*

SubQ adipose tissue, l B 3.91 (1.59) 4.93 (2.03) 3.84 (1.40) 4.10 (1.53)

C –0.06 (0.22) 0.05 (0.13) –0.29 (0.28)* –0.23 (0.22)*

Visc adipose tissue, l B 3.96 (1.06) 4.32 (1.20) 4.02 (1.03) 3.95 (1.02)

C 0.03 (0.32) 0.01 (0.20) –0.27 (0.25)* –0.17 (0.23)*

AST, µkat/l B 0.49 (0.22) 0.51 (0.17) 0.52 (0.19) 0.50 (0.17)

C –0.02 (0.12) 0.08 (0.15) –0.07 (0.09)* 0.02 (0.09)

ALT, µkat/l B 0.57 (0.21) 0.64 (0.25) 0.67 (0.25) 0.61 (0.29)

C –0.003 (0.15) 0.10 (0.28) –0.14 (0.14)* 0.001 (0.22)

Data are reported as: B, mean at baseline (SD); C, mean change at 12 weeks treatment (SD); *p<0.05 as significant difference in descriptive geometric mean ratio (95% CI) ver-sus placebo in a mixed model analysis. SubQ, subcutaneous. Visc, visceral. OM-3CA, omega-3 carboxylic acids, PDFF, proton-derived fat fraction.

Paper IV During 5-h infusions, levels of glucagon (-20 and -9 %), glucagon/insulin ratio (-25 and 37 %), glycerol, NEFA, and beta-OH-butyrate were lower when plasma glucose (3 versus -17 %) was stable versus allowed to fall. Insulin changed signif-icantly from baseline only as glucose fell (-31%). Change in glucose level was the main predictor of change in glucagon level according to multivariate analyses (r2

41

0.23). Urinary glucose excretion was numerically higher with stable versus falling glucose. With add-on DPP4i saxagliptin, glucagon and aGLP-1 were higher during infusions, and glucose rose less and glucagon fell more in subsequent OGTTs.

Figure 8 A-B: Glucose and glucagon changes from baseline (study IV). Data are means ± SEM. N=14 completers for dapagliflozin + isoglycemic clamp and saline, N=13 complet-ers for dapagliflozin + saxagliptin. ΔAUC, area under the curve for change in concentra-tion from time 0 to 300 mins (infusion) or from time 300 to 420 min (OGTT); DAPA, dapagliflozin; h, hour; mins, minute; NS, non-significant; OGTT, oral glucose tolerance test; SAXA, saxagliptin; SEM, standard error of mean.

42

Discussion

Obesity and type 2 diabetes often go with cardiovascular risk factors dyslipidemia, hypertension, and NAFLD. As disease prevalence grows, so does the need for efficacious and safe therapy options. SGLT2 inhibitors and GLP-1 receptor ago-nists are potentially game-changing in this respect, with proven pleiotropic posi-tive effects on body weight, blood pressure, MACE and promising liver fat results63–67, 90,92–95,97–99,106–109,154. OM-3CA is also shown to reduce liver fat and lower MACE147,150. These drug classes have fundamentally different mechanisms of action, and com-bining them may produce additive or even synergistic effects. Hence, SGLT2 in-hibitors with incretin-based therapies and OM-3CA are interesting therapy options for lowering CV risk and liver fat respectively.

Cardiometabolic risk factors

Body weight and body composition In studies I and II, obese non-diabetic participants given SGLT2i and GLP-1RA achieved ~ 4 kg placebo-adjusted weight loss at 24 weeks, a weight reduction that was maintained at 52 weeks. A comparable body weight reduction was observed in the placebo group switching to active treatment in study II. Participants receiv-ing SGLT2i with and without OM-3CA in study III lost ~ 2 kg body weight, mir-roring monotherapy results with SGLT2i in T2D-studies155. However, the extrapo-lated weight change from calorie loss with SGLT2i is around -11 kg, most of which remain unrealized.110 The calorie loss is seemingly compensated at a new and lower equilibrium, often attributed to an increased appetite.156 Weight loss in studies I and II were more substantial than previously reported with monotherapy SGLT2i.108 This is probably the effect of combining SGLT2i with GLP-1RA, and that these drugs have different and complementary mechanisms of action. Our weight loss results are in line with additive effects of each drug.108 In study I, a majority of weight loss occured in the first 12 weeks of treatment. This pattern was mirrored by the placebo group switching to active treatment in study II, and is in line with previous weight loss results with SGLT2i.157

43

Body weight effects of SGLT2i can probably be attributed to losses in both calo-ries and body water. Early effects are likely to mirror urinary fluid loss through natriuresis and osmotic diuresis, as reflected by the higher hematocrit previously reported.158 The concurrent glucosuria is shown to lower plasma glucose and insu-lin levels while sometimes raising glucagon.116 The resulting rise in gluca-gon/insulin ratio would stimulate glycogenolysis, potentially lowering mean hepat-ic glycogen stores, glycogen-bound water and hence reduce body weight marginal-ly. A rise in glucagon/insulin ratio with SGLT2i is also seen to stimulate lipid mobilization, β-oxidation, and ketone body formation.116 Increased utilization of fatty acid energy substrates with SGLT2i was reflected in our results in study IV. Comparing effects of SGLT2i with falling versus stabi-lized glucose, results showed increased glucagon/insulin ratio and rising glycerol, free fatty acids and ketone bodies. A partial shift towards lipid mobilization and oxidation is likely to be a primary contributor to long-term weight loss. Our find-ings in study IV reinforce previous results on SGLT2i and body composition107,109 and is confirmed by MRI changes in visceral and subcutaneous adipose tissue in studies I and II, as well as abdominal and liver fat lowering in study III. A large part of weight loss in studies I and II was due to decreased adipose tissue volume according to MRI. As lipids have more than twice the energy density of glucose, this can partly explain sustained energy requirements inspite of only moderate weight loss. Additionally, oxidation of ketone bodies is shown to be relatively energy efficient in terms of oxygen consumption, at least in the heart.117 In study IV, plasma concentrations of ketone body increased with SGLT2i and glucose lowering, most likely reflecting intracellular substrate availability. If aero-bic fuel efficiency of ketones can be extrapolated to other ketone-oxidizing tissues, energy expenditure could decrease overall and in part also contribute to an early tapering of weight loss. Body weight was maintained following the first 12 weeks of study I (-4.1 kg) and throughout study II (-4.5 to -5.7 kg). The placebo group switching to active treat-ment achieved a weight loss similar to that in study I, consolidating these findings. In comparison, a recent meta-analysis of approved obesity pharmacotherapies showed that placebo-adjusted one year weight loss was 2.6 - 8.8 kg (23-49 % ≥5 %).76 Varying designs and populations make cross-study comparisons inherently difficult. For instance, our study differed by lack of formal lifestyle intervention and a continuing placebo-group. That said, a continuing pattern from study I would yield 1-year weight loss of -5.3 kg (40 % ≥5 %) versus placebo. When the same drug combination was given to mostly obese T2D-patients for one year, re-sults were slightly lower than ours (DURATION-8: combined -3.3 kg, GLP-1RA -1.5, SGLT2i -2.3 kg).159

44

It seems possible that weight loss effects in studies I and II could be augmented by a formal lifestyle intervention. Raising the dose of the GLP-1RA component may also increase weigh loss, as seen with high-dose liraglutide 3.0 mg160. Individual weight loss varied in studies I and II, potentially due to differing GLP-1RA response, which in part may be genetically determined161. This suggests that patient genetics can help to predict GLP-1RA treatment response.

Liver fat Non-alcoholic fatty liver disease is strongly associated with visceral adiposity, T2D, and hepatic insulin resistance and is in itself a CV risk factor.62 SGLT2 inhi-bition can contribute towards reduced liver fat by urinary calorie loss and a nega-tive energy balance. Although mechanisms remain unclear, both OM-3CA and GLP-1RA can also reduce liver fat, and GLP-1RA can also lower liver cell dam-age markers.92,93 Change in liver fat were exploratory endpoints in studies I and II. However, no significant changes were detected, as have previously been reported in preclinical studies with SGLT2i and GLP-1RA monotherapies92,94–97. These findings may be partly due to lack of power and an unselected material regarding hepatic steatosis. However, since these drugs have partly contrary effects on the glucagon/insulin ratio – an established regulator hepatic glucose and lipid metabolism – GLP-1RA may impair the shift towards increased lipid oxidation seen with monotherapy SGLT2i. Lower hepatic insulin resistance at baseline in this non-diabetic popula-tion may also contribute to smaller liver fat reductions than in T2D patients. In study III, we investigated liver fat changes with mono- and dual therapy SGLT2i and OM-3CA versus placebo. Participants had T2D and NAFLD at base-line and reduced liver fat percent with both monotherapies versus the starting val-ue and in combination versus placebo. SGLT2i also reduced liver cell damage markers, not seen with dual treatment, indicating complex drug interaction effects on liver cells. Monotherapy SGLT2i decreases plasma glucose concentrations, leading to a rise in glucagon/insulin ratio, as seen in study IV. This hormonal response, in turn, is well documented to mobilize stored energy through glycogenolysis, gluconeogen-esis, lipolysis and lipid oxidation162. OM-3CA can similarly shift the fatty acid balance from synthesis towards oxidation but without concurrent weight loss116,117,147. These effects together can explain the numerically larger liver fat reduction seen with combination therapy SGLT2i and OM-3CA versus monother-apy arms in study III. The largely improved hepatic biomarker profile with SGLT2i in study III is in line with decreased lipotoxicity and oxidative stress, as supported by reduced γ-GT

45

levels. The same effect was not seen when both SGLT2i and OM-3CA were given. While OM-3CA can upregulate transaminase transcription and thereby dampen SGLT2is transaminase reduction, this does not explain other ameliorated positive results seen with monotherapy SGLT2i. FGF21 was also reduced with monothera-py SGLT2i. High FGF21 levels have been associated with NASH and mitochon-drial dysfunction. As such, lowered FGF21 indicates NASH resolution and allevi-ated metabolic stress.163,164 Liver fat treatment response in study III differed with PNPLA3 I148M genotype, an association seen previously53. In our study, G allele-carriers responsed more than others to combined treatment but less to monotherapy OM-3CA. Hence, PNPLA3 genetic analyses are warranted in future NAFLD studies.

Glucose SGLT2 inhibitors lower the threshold for urinary glucose excretion from -11 to -4 mmol/L, thereby decreasing plasma glucose. As the mechanism is dependent on the glucose and sodium concentration gradients, monotherapy SGLT2i is associat-ed with little-added risk of hypoglycemia but have been reported in some studies to increase glucagon118,153. A majority of participants in studies I and II had prediabetes at inclusion, and those receiving active treatment had improved glycemic measures after 24 and 52 weeks. Participants in study III had T2D, and subgroups receiving SGLT2i showed significant HbA1c reductions after twelve weeks. These results were ex-pected and in line with the established glucosuric effects of SGLT2i. While OM-3CA had no apparent effect on glycemia, it notably decreased stearoyl-CoA de-saturase 1 (SCD-1) and δ-6 desaturase and increased δ-5 desaturase indices, an enzyme activity pattern associated with lowered propensity to T2D develop-ment165. In study IV, single dose SGLT2i was given to participants with T2D to examine effects on glucagon, other hormones, and energy substrates. In these experiments, plasma glucose was the controlled factor. Following SGLT2i dosing, glucagon levels decreased. This drop was more pronounced with stabilized versus falling plasma glucose. Concomitantly, glucagon/insulin ratio decreased in isoglycemia and increased as glucose fell. During the isoglycemic experiment in study IV, 45-50 grams of intravenous glu-cose was required to keep glucose at baseline levels. The concurrent urinary glu-cose excretion (UGE) was significantly lower, averaging 17 grams. The apparent difference between administered and excreted glucose may be due to sustained glucose availability and oxidation. This discrepancy is reflected by the simultane-ous decrease in glucagon/insulin ratio, likely to move the balance away from net

46

endogenous glucose production towards energy storing glycogenesis and lipogen-esis. In comparison, UGE during saline infusion was 13 grams, less than with the glu-cose clamp. This difference probably reflects ongoing glucose oxidation and ex-cretion, lowering glucose concentration gradient over time (plasma glucose de-creased 2.8 mmol/l). As moderate hyperglycemia was maintained even with saline infusion, this may explain the lack of glucagon stimulation166. Findings on gluca-gon baseline values and trajectories with SGLT2i have differed between stud-ies154,153, and in comparison our baseline values were rather high153,154. Indeed, glucagon level and outcome is shown to vary both with the participants metabolic control167 and with the choice of glucagon assay168, both of which may influence these results. Although some uncertainty remains, recently reported results support our findings on glucagon changes with SGLT2i169. Changes in insulin and c-peptide largely mirrored glucose trajectories as expected. In study IV, plasma glucose was the controlled variable by experimental design, and glucagon was seen to fall significantly more with stable versus falling glucose. In multivariate analyses, glucose level was the most important factor in explaining glucagon changes. Taken together, this suggests that glucose levels rather than SGLT2 inhibition per se mattered most to glucagon secretion.

Lipids A blood lipid profile with high triglycerides and LDL is associated with an in-creased risk of cardiovascular disease. In T2D patients, monotherapy SGLT2i generally lower triglycerides but can raise LDL170,171 whereas GLP-1RAs lower them both132. In studies I and II, triglycerides and LDL decreased significantly from baseline to 52 weeks. This lowering could reflect a dominating GLP-1RA effect on lipids, and the glucagon-lowering effect of GLP-1RA may also dampen SGLT2is lipid mobi-lization. A lower insulin-resistance in this nondiabetic cohort and a stabilized body weight after one year may also contribute to the improved lipid profile. In study III, the balance of plasma lipids changed in participants receiving OM-3CA whereas SGLT2i showed no such apparent effects on fatty acid composition. However, butyrylcarnitine levels increased in the SGLT2i treatment group possi-bly reflecting increased fatty acid metabolism. In study IV, single dose SGLT2i was accompanied by stable versus falling glucose by study design. Fatty acid energy substrates were generally lower with stable compared with falling plasma glucose. This difference can be accounted for by higher carbohydrate availability and thereby more glucose oxidation in isoglyce-

47

mia. Conversely, relatively higher levels of fatty acid substrates with falling glu-cose and SGLT2i are signs of a shift towards lipid mobilization. These changes are probably mediated by the concurrent differences in insulin lev-els and glucagon/insulin ratio. Our findings are in line with previously published results172,116 and support that increased lipid mobilization and oxidation is second-ary to SGLT2i’s glucose lowering effect.

Blood pressure Blood pressure was an exploratory outcome variable in studies I and II, in which SGLT2i and GLP-1RA were combined for one year of treatment. Previous results show reductions of 3-5 and 2-5 mmHg for SGLT2i and GLP-1RA respectively106,142. In studies I and II, systolic but not diastolic blood pressure was significantly lowered at 24 weeks and maintained at 52 weeks. Changes were more substantial than seen previously with monotherapies. The seemingly increased efficacy in blood pressure lowering can be explained by their differing and seem-ingly complementary mechanisms of action. SGLT2 inhibitors reduce blood pres-sure through induced natriuresis and osmotic diuresis. Although incompletely known, mechanisms behind GLP1-RAs blood pressure lowering effects may in-volve natriuresis173 but also a complementary vasodilation174.

Cardiometabolic benefits In studies I and II, long-term cardiometabolic risk factors were generally im-proved, lowering body weight, lipid profile, HbA1c and systolic blood pressure. However, swiftly attained positive results in CV-outcome trials suggest that pro-tective mechanisms involve other pathways. For SGLT2i this may include fluid congestion relief 121 and increased cardiac ketone body utilization117, while GLP-1RA seemingly has direct effects on heart and vasculature function.132 In study III, reduced liver fat and inflammation may also contribute towards CV-protection long-term. However, opinion and research outcomes regarding cardio-vascular benefits with OM-3CAs currently diverge. While EPA-only formulation Vascepa recently announced significant MACE reductions in the US150, the Euro-pean medical agency have reported that they no longer consider any OM-3CAs effective in secondary CVD-prevention.175

Incretins Feeding stimulates GLP-1 secretion, which generally lowers glucagon and raises insulin levels. DPP4 inhibitors hinder GLP-1 breakdown. In study IV, active GLP-1 levels decreased irrespective of plasma glucose changes while rising with the addition of DPP4i. Unexpectedly, glucagon concomitantly rose with active GLP-1,

48

differing from results in previous longer-term studies. Varying choices of gluca-gon assay may impact these differing outcomes. It is also possible that physiologi-cal adaptations to chronic dosing per se have an impact on glucagon trajectories, partly explaining differences. The changes in hormones and energy substrates during 2-h OGTT largely reflect different starting values and should be interpreted with caution. Notably, active GLP-1 rose in these conditions, while glucose rose numerically less with DPP4i added. At the same time, insulin and lipid energy substrates did not change signifi-cantly. These results are generally in line with previous results seen with DDP4i although the lack of insulin increase appears surprising.

Safety Pharmacological weight loss therapies have been associated with safety issues76. In contrast, SGLT2i and GLP-1RA can both reduce MACE-incidence and have otherwise good safety and tolerability profiles64,66,67,120. No new or unexpected safety issues were identified in any of these four studies. Notably, no hypoglyce-mic episodes were noted with active treatment in studies I or II, despite giving two antihyperglycemic agents to a non-diabetic study population. However as ex-pected, more nausea and injection site issues were reported with GLP-1RA treat-ment. AE frequency and related discontinuations were generally as previously reported.

49

Figure 9: Proposed metabolic and endocrine effects of SGLT2 inhibition and coadminis-tered drugs. SGLT2 inhibition (SGLT2i) induces urinary glucose excretion and lowers plasma glucose, which in turn raises the glucagon/insulin ratio. The resulting energy deficit stimulates appetite to replenish energy substrate stores. This is partly counteracted by de-creased appetite and slowed gastric emptying through GLP-1 receptor agonism (GLP-1RA). Raised glucagon/insulin ratio also stimulates hepatic glucose production and causes a partial shift towards lipid mobilization, in turn lowering lipids stored in adipose tissue and the liver. Both omega-3 carboxylic acids (OM-3CA) and GLP-1RA seem to have liver fat lowering qualities, possibly via increased energy expenditure and indirect effects re-spectively, but mechanisms remain unknown.

50

Conclusions

In papers I and II, SGLT2 inhibitor dapagliflozin and GLP-1 receptor agonist ex-enatide reduced body weight, prediabetes and systolic blood pressure in an obese non-diabetic population without new safety concerns. Lipid profile was also im-proved from baseline after one year of treatment. This was a first-ever placebo-controlled study of this drug combination in a prediabetic population and suggests a potential role in T2D and CVD prevention. A dose-finding study for GLP-1 re-ceptor agonism together with SGLT2 inhibition could yield further weight loss. Cardiovascular outcome studies of this drug combination would be of interest for both obesity- and T2D-cohorts. In paper III, dapagliflozin and omega-3 (n-3) carboxylic acids lowered liver fat content in overweight T2D patients with NAFLD, in particular when given togeth-er. Dapagliflozin also lowered hepatocyte injury biomarkers, a promising finding for NASH prevention and treatment. Further studies with liver biopsies in a NASH population are needed to validate these findings. In paper IV, with single-dose dapagliflozin, changes in glucose can explain chang-es in glucagon levels. The impact of SGLT2 inhibition on glucagon and fatty acid mobilization seems to be mainly in response to glucose lowering via urinary ex-cretion rather than through direct drug effects on pancreatic α-cells. This finding indicates largely intact counter-regulatory mechanisms in the maintenance of glu-cose homeostasis with an SGLT2 inhibitor present. Future crossover studies in-cluding a placebo arm and at varying glucose levels could further elucidate SGLT2 inhibitors glucagon effects.

51

Sammanfattning (summary in swedish)

Fetma och typ 2 diabetes (T2D) är två mycket stora och växande globala hälso-problem. Trots att sjukdomarnas bakomliggande orsaker är heterogena och mång-faceterade har båda tillstånden liknande samsjukligheter. Det inbegriper höga blodfetter, högt blodtryck, fettlever, nedsatt njurfunktion och framförallt ökad risk för kardiovaskulär sjukdom och dödlighet. SGLT2-hämmare förbättrar blodsock-erkontrollen genom att framkalla glukosutsöndring med urinen och har visat sig kunna lindra både T2D- och fetmahälsoproblem såsom just övervikt/fetma, högt blodtryck, njursvikt och hjärt-kärlsjukdom. Artikel I-II: I dessa två studier undersöktes den kombinerade effekten av 24 veck-ors behandling med SGLT2-hämmaren dapagliflozin och GLP-1 receptor agonis-ten exenatid på i första hand kroppsvikt, men även kroppssammansättning, blod-glukos och systoliskt blodtryck hos 50 obesa vuxna utan diabetes. Behandlingen tolererades väl och skillnaden i kroppsvikt var -4,1 kg efter 24 veckor, mängden fettväv minskade nästan lika mycket enligt magnetkameraundersökning. Även prediabetiska (måttligt förhöjda) blodsockervärden och systoliskt blodtryck mins-kade signifikant. 38 av dessa deltagare gick in i den efterföljande 28 veckor långa studieförlängningen, alla deltagare fick nu aktiv behandling. Förlängningsstudiens resultat visade bibehållet minskad kroppsvikt (-5.7 kg), volym fettväv, blodsocker- och blodtrycksnivå efter 52 veckors behandling. Även blodfettsprofilen förbättra-des från studiestart med ett års behandlling. Ovanstående resultat pekar mot att läkemedelskombinationen SGLT2-hämmare och GLP-1-receptor agonist kan förebygga T2D och hjärtkärlssjukdom. Detta är extra intressant eftersom man har påvisat minskad hjärtkärlsjuklighet hos T2D-patienter med båda läkemedelstyperna var för sig. Artikel III: I denna 12-veckorsstudie utvärderades effekten av dapagliflozin, fria omega-3 (n-3) karboxylsyror var för sig och tillsammans på leverns fettinnehåll jämfört med placebo. 84 personer med T2D, övervikt/fetma och fettlever deltog. Procent leverfett minskade med kombinationsbehandlingen jämfört med placebo och gentemot startvärdet med vardera preparat enskilt. Vidare minskade leverska-demarkörer med dapagliflozin ensamt. Resultaten är sammantaget lovande och motiverar mer forskning avseende preparatens roll i framtida fettleverbehandling.

52

Artikel IV: Här studerades effekten av endos dapagliflozin på glukagon (blod-sockerhöjande hormon), andra hormoner samt energislag, vid stabilt respektive fallande blodsocker. 15 patienter med T2D deltog. Glukagon sjönk från startnivån både vid stabilt och fallande blodsocker, men sjönk mindre när glukos sjönk. Vid sjunkande blodsocker ökade mängden fettsyror i blodet. Tillägg av en DPP4-hämmaren saxagliptin höjde oväntat glukagon-nivån i fasta. Effekterna av SGLT2-hämning på glukagon och fettsyramobilisering framstår som en fysiologisk reaktion på glukosminskning via urinutsöndring snarare än genom direkta läkemedelseffekter på bukspottkörtelns α-celler. Resultatet tycks visa i stort sett intakta motreglerande mekanismer vid upprätthållandet av glukosbalan-sen i närvaro av en SGLT2-hämmare. Sammanfattningsvis kan SGLT2-hämmare sänka kroppsvikt och kardiovaskulära riskfaktorer hos obesa patienter utan diabetes när de kombineras med en GLP-1 receptor agonist, samt minska leverfett hos T2D-patienter, framförallt när de ges tillsammans med fria omega-3-karboxylsyror. Effekten av SGLT2-hämmare på glukagonfrisättning kan till stor del förklaras av glukosförändringar.

53

Acknowledgements

This thesis is based on the hard work of many coworkers to whom I am very grate-ful. Many thanks to Uppsala University and the department of medical sciences for accepting me as a PhD student and teaching me the fundamentals of scientific research. Jan Eriksson, main supervisor – many thanks for the provided opportunities to learn and grow with these interesting, rewarding and at times, challenging tasks. Your broad knowledge, ongoing scientific curiosity and capacity for multitasking are out of the ordinary. Lars Lind, your role as co-supervisor has been appreciated, providing useful feed-back when needed. It was a pleasure to work with you in the effect II study. Maria Joao Pereria – your role as co-supervisor has been much appreciated! You have always been positive and found time to help out with whatever the projects required, regardless of it being late evenings, vacation or the weekend. I am very grateful to have had you as a co-supervisor. Joey Lao Börjesson, a co-supervisor at the start of my doctoral studies that later moved on to form your own research team. I enjoyed our collaboration during both the dapalost study and the development and preparation of the dapaglucagon study. All previous and current members of the clinical diabetes research team – many thanks for all your hard work and for making our time together enjoyable. Petros Katsogiannos – thank you for your work as study physician and for provid-ing useful input on most of my work. You have always been a very helpful and friendly senior colleague and I hope to work more with you in the future. Niclas Abrahamsson, thank you for your support as a senior clinican and for providing a friendly and fun environment to work in.

54

The team of research nurses, Carola, Tanya, Caroline W, Anna, Sofia, Caroline M, Monika, Lovisa. Many thanks for all of your hard work, flexibility and for all the laughs we have shared over the years. All the members of the clinical diabetes research laboratory: Thank you all for fruitful collaborations throughout these four years. Your work has been instrumen-tal to this thesis, I am very grateful. A special thanks to Prasad Kamble for the many hours spent in the lab on these projects. I admire your work ethic and drive to succeed. Many thanks to David Sjöström, Eva Johnsson, Anna-Marie Langkilde, Russell Esterline and Jan Oscarsson at AstraZeneca R&D, Mölndal, Sweden – it has been a pleasure working with you all! Thank you to the department of clinical chemistry at Uppsala academic hospital and to Covance laboratories for the excellent lab work provided. Thanks to all staff involved at the Uppsala academic hospital radiology department and Antaros for performing state of the art imaging as well as providing invaluable expertise in it’s interpretation. Springer publishing – thank you for all your supporting work on and around the work on these publications. A special thanks to senior medical writer Julian Mar-tins, your keen eye for detail, prowess in statistics and experience has really been a valuable resource during the dapalost project. Hannah and Ingrid – my beloved wife and daughter. Thank you for always being there for me and for putting up with me, I love you! Mom and dad, thank you for literally everything! Thank you for all your support, encouragement and unconditional love. Funding: This thesis work has been supported by AstraZeneca AB, the Swedish Diabetes Foundation, the Ernfors Foundation, EXODIAB, ALF grants of the Upp-sala academic hospital – I thank you all!

55

References

1. WHO Fact sheet obesity and overweight. who.int. http://www.who.int/en/news-room/fact-sheets/detail. Published 2018. Accessed December 2, 2018.

2. Folkhälsomyndigheten. https://www.folkhalsomyndigheten.se/folkhalsorapportering- statistik/statistikdatabaser-och-visualisering/nationella-folkhalsoenkaten/levnadsvanor/ overvikt-och-fetma/. Published 2016. Accessed December 2, 2018.

3. van Dis I, Kromhout D, Geleijnse JM, Boer JM a, Verschuren WMM. Body mass index and waist circumference predict both 10-year nonfatal and fatal cardiovascular disease risk: study conducted in 20,000 Dutch men and women aged 20-65 years. Eur J Cardiovasc Prev Rehabil. 2009;16(6):729-734. doi:10.1097/HJR.0b013e328331dfc0.

4. MacMahon S, Baigent C, Duffy S, et al. Body-mass index and cause-specific mortality in 900 000 adults: Collaborative analyses of 57 prospective studies. Lancet. 2009;373(9669):1083-1096. doi:10.1016/S0140-6736(09)60318-4.

5. Smith SR, Lovejoy JC, Greenway F, et al. Contributions of total body fat, abdominal subcutaneous adipose tissue compartments, and visceral adipose tissue to the metabolic complications of obesity. Metabolism. 2001;50(4):425-435. doi:10.1053/meta.2001.21693.

6. Després JP, Moorjani S, Lupien PJ, Tremblay A, Nadeau A, Bouchard C. Regional distribution of body fat, plasma lipoproteins, and cardiovascular disease. Arterioscler Thromb Vasc Biol. 1990. doi:10.1161/01.ATV.10.4.497.

7. Britton KA, Massaro JM, Murabito JM, Kreger BE, Hoffmann U, Fox CS. Body fat distribution, incident cardiovascular disease, cancer, and all-cause mortality. J Am Coll Cardiol. 2013. doi:10.1016/j.jacc.2013.06.027.

8. Klingberg S, Mehlig K, Lanfer A, Björkelund C, Heitmann BL, Lissner L. Increase in waist circumference over 6 years predicts subsequent cardiovascular disease and total mortality in nordic women. Obesity. 2015;23(10):2123-2130. doi:10.1002/oby.21203.

9. Koster A, Leitzmann MF, Schatzkin A, et al. Waist Circumference and Mortality. Am J Epidemiol. 2008;167(12):1465-1475. doi:10.1093/aje/kwn079.

10. Borruel S, Moltó JF, Alpañés M, et al. Surrogate markers of visceral adiposity in young adults: Waist circumference and body mass index are more accurate than waist hip ratio, model of adipose distribution and visceral adiposity index. PLoS One. 2014;9(12). doi:10.1371/journal.pone.0114112.

11. Defronzo RA. Banting Lecture. From the triumvirate to the ominous octet: a new paradigm for the treatment of type 2 diabetes mellitus. Diabetes. 2009;58(4):773-795. doi:10.2337/db09-9028.

12. Baron AD, Schaeffer L, Shragg P, Kolterman OG. Role of hyperglucagonemia in maintenance of increased rates of hepatic glucose output in type II diabetics. Diabetes. 1987. doi:10.2337/diab.36.3.274.

56

13. Alsahli M, Gerich JE. Renal glucose metabolism in normal physiological conditions and in diabetes. Diabetes Res Clin Pract. 2017;133:1-9. doi:10.1016/j.diabres.2017.07.033.

14. Nauck M, Stöckmann F, Ebert R, Creutzfeldt W. Reduced incretin effect in type 2 (non-insulin-dependent) diabetes. Diabetologia. 1986. doi:10.1007/BF02427280.

15. Knop FK, Vilsbøll T, Højberg P V., et al. Reduced incretin effect in type 2 diabetes: Cause or consequence of the diabetic state? Diabetes. 2007. doi:10.2337/db07-0100.

16. Classification and diagnosis of diabetes: Standards of medical care in Diabetesd2018. Diabetes Care. 2018. doi:10.2337/dc18-S002.

17. Abdul-Ghani MA, Tripathy D, DeFronzo RA. Contributions of beta-cell dysfunction and insulin resistance to the pathogenesis of impaired glucose tolerance and impaired fasting glucose. Diabetes Care. 2006;29(5):1130-1139. doi:10.2337/diacare.2951130.

18. Cho NH, Shaw JE, Karuranga S, et al. IDF Diabetes Atlas: Global estimates of diabetes prevalence for 2017 and projections for 2045. Diabetes Res Clin Pract. 2018. doi:10.1016/j.diabres.2018.02.023.

19. Berne C, Fritz T. Diabetes mellitus. In: Läkemedelsboken. Läkemedelsverket; 2018:1-28. doi:10.1016/B978-0-08-097037-0.00064-6.

20. Lind L, Siegbahn A, Ingelsson E, Sundström J, Ärnlöv J. A detailed cardiovascular characterization of obesity without the metabolic syndrome. Arterioscler Thromb Vasc Biol. 2011. doi:10.1161/ATVBAHA.110.221572.

21. El-Serag HB, Tran T, Everhart JE. Diabetes Increases the Risk of Chronic Liver Disease and Hepatocellular Carcinoma. Gastroenterology. 2004;126(2):460-468. doi:10.1053/j.gastro.2003.10.065.

22. Williams CD, Stengel J, Asike MI, et al. Prevalence of nonalcoholic fatty liver disease and nonalcoholic steatohepatitis among a largely middle-aged population utilizing ultrasound and liver biopsy: A prospective study. Gastroenterology. 2011;140(1):124-131. doi:10.1053/j.gastro.2010.09.038.

23. Rinella ME. Nonalcoholic fatty liver disease: a systematic review. JAMA. 2015;313(22):2263-2273. doi:10.1001/jama.2015.5370.

24. Wardle J, Carnell S, Haworth CMA, Plomin R. Evidence for a strong genetic influence on childhood adiposity despite the force of the obesogenic environment. Am J Clin Nutr. 2008;87(2):398-404. doi:10.1007/s12160-009-9116-5.

25. Rooney K, Ozanne SE. Maternal over-nutrition and offspring obesity predisposition: targets for preventative interventions. Int J Obes. 2011;35(7):883-890. doi:10.1038/ijo.2011.96.

26. Ravelli AC, Osmond C. Obesity at the age of 50 years in men and women exposed to famine prenatally. Am J Clin Nutr. 1999;70:811-816. doi:10.1056/ nejm197608122950701.

27. Incollingo Rodriguez AC, Epel ES, White ML, Standen EC, Seckl JR, Tomiyama AJ. Hypothalamic-pituitary-adrenal axis dysregulation and cortisol activity in obesity: A systematic review. Psychoneuroendocrinology. 2015. doi:10.1016/ j.psyneuen.2015.08.014.

28. Leslie WS, Hankey CR, Lean MEJ. Weight gain as an adverse effect of some commonly prescribed drugs: A systematic review. QJM. 2007. doi:10.1093/ qjmed/hcm044.

57

29. Hill JO, Wyatt HR, Peters JC. Energy balance and obesity. Circulation. 2012;126(1):126-132. doi:10.1161/CIRCULATIONAHA.111.087213.

30. Vámosi M, Heitmann BL, Kyvik KO. The relation between an adverse psychological and social environment in childhood and the development of adult obesity: A systematic literature review. Obes Rev. 2010. doi:10.1111/j.1467-789X.2009.00645.x.

31. Rossen LM. Neighbourhood economic deprivation explains racial/ethnic disparities in overweight and obesity among children and adolescents in the USA. J Epidemiol Community Health. 2014;68(2):123-129. doi:10.1136/jech-2012-202245.

32. Williams KW, Elmquist JK. From neuroanatomy to behavior: Central integration of peripheral signals regulating feeding behavior. Nat Neurosci. 2012. doi:10.1038/nn.3217.

33. Douglass AM, Kucukdereli H, Ponserre M, et al. Central amygdala circuits modulate food consumption through a positive-valence mechanism. Nat Neurosci. 2017. doi:10.1038/nn.4623.

34. Boekhoudt L, Roelofs TJM, de Jong JW, et al. Does activation of midbrain dopamine neurons promote or reduce feeding? Int J Obes. 2017;41(7):1131-1140. doi:10.1038/ijo.2017.74.

35. Speliotes E, Willer C, Berndt S. Association analyses of 249,796 individuals reveal 18 new loci associated with body mass index. Nat Genet. 2010;42(11):937-948. doi:10.1038/ng.686.Association.

36. Anderberg RH, Anefors C, Bergquist F, Nissbrandt H, Skibicka KP. Dopamine signaling in the amygdala, increased by food ingestion and GLP-1, regulates feeding behavior. Physiol Behav. 2014;136:135-144. doi:10.1016/j.physbeh.2014.02.026.

37. Farooqi IS, O’Rahilly S. Mutations in ligands and receptors of the leptin–melanocortin pathway that lead to obesity. Nat Clin Pract Endocrinol Metab. 2008;4(10):569-577. doi:10.1038/ncpendmet0966.

38. Hochberg I, Hochberg Z. Expanding the definition of hypothalamic obesity. Obes Rev. 2010;11(10):709-721. doi:10.1111/j.1467-789X.2010.00727.x.

39. Secher A, Jelsing J, Baquero AF, et al. The arcuate nucleus mediates GLP-1 receptor agonist liraglutide-dependent weight loss. J Clin Invest. 2014;124(10):4473-4488. doi:10.1172/JCI75276.

40. Bertile F, Raclot T. The melanocortin system during fasting. Peptides. 2006;27(2):291-300. doi:10.1016/j.peptides.2005.03.063.

41. Zanchi D, Depoorter A, Egloff L, et al. The impact of gut hormones on the neural circuit of appetite and satiety: A systematic review. Neurosci Biobehav Rev. 2017. doi:10.1016/j.neubiorev.2017.06.013.

42. Bouret SG. Formation of Projection Pathways from the Arcuate Nucleus of the Hypothalamus to Hypothalamic Regions Implicated in the Neural Control of Feeding Behavior in Mice. J Neurosci. 2004;24(11):2797-2805. doi:10.1523/JNEUROSCI.5369-03.2004.

43. Berthon BS, MacDonald-Wicks LK, Wood LG. A systematic review of the effect of oral glucocorticoids on energy intake, appetite, and body weight in humans. Nutr Res. 2014. doi:10.1016/j.nutres.2013.12.006.

44. Nuffer WA, Trujillo JM. Liraglutide: A New Option for the Treatment of Obesity. Pharmacotherapy. 2015. doi:10.1002/phar.1639.

58

45. Madsbad, Sten; Holst JJ. Glycaemic control and weight loss with semaglutide in type 2 diabetes. Lancet Diabetes Endocrinol. 2017;5(5):315-317. doi:10.1016/S2213-8587(17)30094-3.

46. Klein BE, Klein R, Moss SE, Cruickshanks KJ. Parental history of diabetes in a population-based study. Diabetes Care. 1996;19(8):827-830. doi:10.2337/diacare.19.8.827.

47. Connolly V, Unwin N, Sherriff P, Bilous R, Kelly W. Diabetes prevalence and socioeconomic status: A population based study showing increased prevalence of type 2 diabetes mellitus in deprived areas. J Epidemiol Community Health. 2000. doi:10.1136/jech.54.3.173.

48. Whincup PH, Kaye SJ, Owen CG, et al. Birth weight and risk of type 2 diabetes: a systematic review. JAMA J Am Med Assoc. 2008;300(24):2886-2897. doi:10.1001/jama.2008.886.

49. Carulli L, Rondinella S, Lombardini S, Canedi I, Loria P, Carulli N. Review article: diabetes, genetics and ethnicity. Aliment Pharmacol Ther. 2005;22 Suppl 2:16-19. doi:10.1111/j.1365-2036.2005.02588.x.

50. Chan JM, Rimm EB, Colditz GA, Stampfer MJ, Willett WC. Obesity, fat distribution, and weight gain as risk factors for clinical diabetes in men. Diabetes Care. 1994;17(9):961-969. doi:10.2337/diacare.17.9.961.

51. Biggs ML, Mukamal KJ, Luchsinger J a, et al. Association between adiposity in midlife and older age and risk of diabetes in older adults. JAMA. 2010;303(24):2504-2512. doi:10.1097/OGX.0b013e3182022112.

52. Kahn SE, Cooper ME, Del Prato S. Pathophysiology and treatment of type 2 diabetes: Perspectives on the past, present, and future. Lancet. 2014;383(9922):1068-1083. doi:10.1016/S0140-6736(13)62154-6.

53. Romeo S, Kozlitina J, Xing C, et al. Genetic variation in PNPLA3 confers susceptibility to nonalcoholic fatty liver disease. Nat Genet. 2008. doi:10.1038/ng.257.

54. Sunny NE, Bril F, Cusi K. Mitochondrial Adaptation in Nonalcoholic Fatty Liver Disease: Novel Mechanisms and Treatment Strategies. Trends Endocrinol Metab. 2016;xx(0):1-11. doi:10.1016/j.tem.2016.11.006.

55. Kramer CK, Zinman B, Retnakaran R. Are metabolically healthy overweight and obesity benign conditions?: A systematic review and meta-analysis. Ann Intern Med. 2013;159(11):758-769. doi:10.7326/0003-4819-159-11-201312030-00008.

56. Diabetes Prevention Program Research Group. Reduction in the Incidence of Type 2 Diabetes with Lifestyle Intervention or Metformin. N Engl J Med. 2002;346(6):393-403. doi:10.1056/NEJMoa012512.

57. WHO. Diabetes Fact Sheet. NMH Fact Sheet Febr 2010. 2010. doi:10.1017/CBO9781107415324.004.

58. DIA. Diabetes in America. National Institute of Diabetes and Digestive and Kidney Diseases. doi:10.1046/j.1365-2141.1998.00677.x.

59. CDC. National Diabetes Statistics Report, Estimates of Diabetes and Its Burden in the United States. 2017:p.8. https://www.cdc.gov/diabetes/pdfs/data/ statistics/national-diabetes-statistics-report.pdf.

60. Kaur J. A comprehensive review on metabolic syndrome. Cardiol Res Pract. 2014;2014. doi:10.1155/2014/943162.

59

61. Deshpande AD, Harris-Hayes M, Schootman M. Epidemiology of Diabetes and Diabetes-Related Complications. Phys Ther. 2008;88(11):1254-1264. doi:10.2522/ptj.20080020.

62. Adams LA, Anstee QM, Tilg H, Targher G. Non-alcoholic fatty liver disease and its relationship with cardiovascular disease and other extrahepatic diseases. Gut. 2017;66(6):1138-1153. doi:10.1136/gutjnl-2017-313884.

63. le Roux CW, Astrup A, Fujioka K, et al. 3 years of liraglutide versus placebo for type 2 diabetes risk reduction and weight management in individuals with prediabetes: a randomised, double-blind trial. Lancet. 2017;389(10077):1399-1409. doi:10.1016/S0140-6736(17)30069-7.

64. Marso SP, Daniels GH, Brown-Frandsen K, et al. Liraglutide and Cardiovascular Outcomes in Type 2 Diabetes. N Engl J Med. 2016;375(4):311-322. doi:10.1056/NEJMoa1603827.

65. Marso SP, Bain SC, Consoli A, et al. Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes. N Engl J Med. 2016;375(19):1834-1844. doi:10.1056/NEJMoa1607141.

66. Neal B, Perkovic V, Mahaffey KW, et al. Canagliflozin and Cardiovascular and Renal Events in Type 2 Diabetes. N Engl J Med. 2017;377(7):644-657. doi:10.1056/NEJMoa1611925.

67. Zinman B, Wanner C, Lachin JM, et al. EMPA-REG. N Engl J Med. 2015. doi:10.1056/NEJMoa1504720.

68. Fetma. In: Läkemedelsboken. Läkemedelsverket; 2016. https:// lakemedelsboken.se/kapitel/nutrition/fetma.html?search=kost&id=Kost#Kost.

69. Gillett M, Royle P, Snaith a, et al. Non-pharmacological interventions to reduce the risk of diabetes in people with impaired glucose regulation: a systematic review and economic evaluation. Health Technol Assess. 2012;16(33):1-236, iii-iv. doi:10.3310/hta16330.

70. Balk EM, Earley A, Raman G, Avendano EA, Pittas AG, Remington PL. Combined diet and physical activity promotion programs to prevent type 2 diabetes among persons at increased risk: A systematic review for the community preventive services task force. Ann Intern Med. 2015;163(6):437-451. doi:10.7326/M15-0452.

71. Selph S, Dana T, Blazina I, Bougatsos C, Patel H, Chou R. Screening for Type 2 Diabetes Mellitus: A Systematic Review for the U.S. Preventive Services Task Force. Ann Intern Med. 2015;162(11):765. doi:10.7326/M14-2221.

72. Swift DL, Johannsen NM, Lavie CJ, Earnest CP, Church TS. The role of exercise and physical activity in weight loss and maintenance. Prog Cardiovasc Dis. 2014. doi:10.1016/j.pcad.2013.09.012.

73. C.L. M, H.L. P, R.M. F, J. P. Attenuating the biologic drive for weight regain following weight loss: Must what goes down always go back up? Nutrients. 2017. doi:10.3390/nu9050468.

74. Abrahamsson N. On the Impact of Bariatric Surgery on Glucose Homeostasis. 2016. 75. Adams TD, Davidson LE, Litwin SE, et al. Weight and Metabolic Outcomes 12

Years after Gastric Bypass. N Engl J Med. 2017;377(12):1143-1155. doi:10.1056/NEJMoa1700459.

76. Khera R, Murad MH, Chandar AK, et al. Association of Pharmacological Treatments for Obesity With Weight Loss and Adverse Events. JAMA. 2016;315(22):2424. doi:10.1001/jama.2016.7602.

60

77. Nuffer W, Trujillo JM, Megyeri J. A Comparison of New Pharmacological Agents for the Treatment of Obesity. Ann Pharmacother. 2016;50(5):376-388. doi:10.1177/1060028016634351.

78. James WPT, Caterson ID, Coutinho W, et al. Effect of Sibutramine on Cardiovascular Outcomes in Overweight and Obese Subjects. N Engl J Med. 2010. doi:10.1056/NEJMoa1003114.

79. Bohula EA, Wiviott SD, McGuire DK, et al. Cardiovascular Safety of Lorcaserin in Overweight or Obese Patients. N Engl J Med. 2018. doi:10.1016/S0304-3940(99)00087-7.

80. Christensen R, Kristensen PK, Bartels EM, Bliddal H, Astrup A. Efficacy and safety of the weight-loss drug rimonabant: a meta-analysis of randomised trials. Lancet. 2007;370(9600):1706-1713. doi:10.1016/S0140-6736(07)61721-8.

81. Sahebkar A, Simental-Mendía LE, Reiner Ž, et al. Effect of orlistat on plasma lipids and body weight: A systematic review and meta-analysis of 33 randomized controlled trials. Pharmacol Res. 2017;122:53-65. doi:10.1016/j.phrs.2017.05.022.

82. Turner R. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet. 1998;352(9131):854-865. doi:10.1016/S0140-6736(98)07037-8.

83. Philis-Tsimikas A, Charpentier G, Clauson P, Ravn GM, Roberts VL, Thorsteinsson B. Comparison of once-daily insulin detemir with NPH insulin added to a regimen of oral antidiabetic drugs in poorly Controlled Type 2 Diabetes. Clin Ther. 2006;28(10):1569-1581. doi:10.1016/j.clinthera.2006.10.020.

84. Proks P, Reimann F, Green N, Gribble F, Ashcroft F. Sulfonylurea stimulation of insulin secretion. In: Diabetes. Vol 51. ; 2002. doi:10.2337/diabetes.51.2007.S368.

85. Holman RR, Paul SK, Bethel MA, Matthews DR, Neil HAW. 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med. 2008;359(15):1577-1589. doi:10.1056/NEJMoa0806470.

86. Boussageon R, Supper I, Bejan-Angoulvant T, et al. Reappraisal of metformin efficacy in the treatment of type 2 diabetes: A meta-analysis of randomised controlled trials. PLoS Med. 2012;9(4). doi:10.1371/journal.pmed.1001204.

87. Nakamura T, Funahashi T, Yamashita S, et al. Thiazolidinedione derivative improves fat distribution and multiple risk factors in subjects with visceral fat accumulation--double-blind placebo-controlled trial. Diabetes Res Clin Pract. 2001;54(3):181-190. doi:10.1016/S0168-8227(01)00319-9.

88. Liu J, Wang L-N. Peroxisome proliferator-activated receptor gamma agonists for preventing recurrent stroke and other vascular events in patients with stroke or transient ischaemic attack. In: Cochrane Database of Systematic Reviews. ; 2015. doi:10.1002/14651858.CD010693.pub3.

89. R. B, S.A. H, K. B, et al. A placebo-controlled trial of pioglitazone in subjects with nonalcoholic steatohepatitis. N Engl J Med. 2006. doi:10.1056/NEJMoa060326.

90. Chalasani N, Younossi Z, Lavine JE, et al. The diagnosis and management of nonalcoholic fatty liver disease: Practice guidance from the American Association for the Study of Liver Diseases. Hepatology. 2018;67(1):328-357. doi:10.1002/hep.29367.

91. Sumida Y, Yoneda M. Current and future pharmacological therapies for NAFLD/NASH. J Gastroenterol. 2018. doi:10.1007/s00535-017-1415-1.

61

92. Armstrong MJ, Gaunt P, Aithal GP, et al. Liraglutide safety and efficacy in patients with non-alcoholic steatohepatitis (LEAN): A multicentre, double-blind, randomised, placebo-controlled phase 2 study. Lancet. 2016. doi:10.1016/S0140-6736(15)00803-X.

93. Armstrong MJ, Houlihan DD, Rowe IA, et al. Safety and efficacy of liraglutide in patients with type 2 diabetes and elevated liver enzymes: Individual patient data meta-analysis of the LEAD program. Aliment Pharmacol Ther. 2013. doi:10.1111/apt.12149.

94. Shao N, Kuang HY, Hao M, Gao XY, Lin WJ, Zou W. Benefits of exenatide on obesity and non-alcoholic fatty liver disease with elevated liver enzymes in patients with type 2 diabetes. Diabetes Metab Res Rev. 2014. doi:10.1002/dmrr.2561.

95. Kuchay MS, Krishan S, Mishra SK, et al. Effect of empagliflozin on liver fat in patients with type 2 diabetes and nonalcoholic fatty liver disease: A randomized controlled trial (E-LIFT Trial). Diabetes Care. 2018. doi:10.2337/dc18-0165.

96. Ito D, Shimizu S, Inoue K, et al. Comparison of Ipragliflozin and Pioglitazone Effects on Nonalcoholic Fatty Liver Disease in Patients With Type 2 Diabetes: A Randomized, 24-Week, Open-Label, Active-Controlled Trial. Diabetes Care. 2017;40(10):1364 LP-1372. http://care.diabetesjournals.org/content/40/10/ 1364.abstract.

97. Komiya C, Tsuchiya K, Shiba K, et al. Ipragliflozin improves hepatic steatosis in obese mice and liver dysfunction in type 2 diabetic patients irrespective of body weight reduction. PLoS One. 2016. doi:10.1371/journal.pone.0151511.

98. Shibuya T, Fushimi N, Kawai M, et al. Luseogliflozin improves liver fat deposition compared to metformin in type 2 diabetes patients with non-alcoholic fatty liver disease: A prospective randomized controlled pilot study. Diabetes, Obes Metab. 2018. doi:10.1111/dom.13061.

99. Takase T, Nakamura A, Miyoshi H, Yamamoto C, Atsumi T. Amelioration of fatty liver index in patients with type 2 diabetes on ipragliflozin: an association with glucose-lowering effects. Endocr J. 2017. doi:10.1507/endocrj.EJ16-0295.

100. DeFronzo RA, Davidson JA, del Prato S. The role of the kidneys in glucose homeostasis: A new path towards normalizing glycaemia. Diabetes, Obes Metab. 2012;14(1):5-14. doi:10.1111/j.1463-1326.2011.01511.x.

101. Mogensen CE. Maximum tubular reabsorption capacity for glucose and renal hemodynamcis during rapid hypertonic glucose infusion in normal and diabetic subjects. Scand J Clin Lab Invest. 1971;28(1):101-109. doi:10.3109/ 00365517109090668.

102. Prié D. Familial renal glycosuria and modifications of glucose renal excretion. Diabetes Metab. 2014;40(6):S12-S16. doi:10.1016/S1262-3636(14)72690-4.

103. Ehrenkranz JRL, Lewis NG, Kahn CR, Roth J. Phlorizin: A review. Diabetes Metab Res Rev. 2005;21(1):31-38. doi:10.1002/dmrr.532.

104. Meng W, Ellsworth BA, Nirschl AA, et al. Discovery of dapagliflozin: A potent, selective renal sodium-dependent glucose cotransporter 2 (SGLT2) inhibitor for the treatment of type 2 diabetes. J Med Chem. 2008;51(5):1145-1149. doi:10.1021/jm701272q.

105. Merovci A, Solis-Herrera C, Daniele G, et al. Dapagliflozin improves muscle insulin sensitivity but enhances endogenous glucose production. J Clin Invest. 2014;124(2):509-514. doi:10.1172/JCI70704.

62

106. Minze MG, Will K, Terrell BT, Black RL, Irons BK. Benefits of SGLT2 Inhibitors beyond glycemic control - A focus on metabolic, cardiovascular, and renal outcomes. Curr Diabetes Rev. 2017;13:1-9. doi:10.2174/ 1573399813666170816142351.

107. Bolinder J, Ljunggren Ö, Johansson L, et al. Dapagliflozin maintains glycaemic control while reducing weight and body fat mass over 2 years in patients with type 2 diabetes mellitus inadequately controlled on metformin. Diabetes Obes Metab. 2014;16(2):159-169. doi:10.1111/dom.12189.

108. Cai X, Ji L, Chen Y, et al. Comparisons of weight changes between sodium-glucose cotransporter 2 inhibitors treatment and glucagon-like peptide-1 analogs treatment in type 2 diabetes patients: A meta-analysis. J Diabetes Investig. 2017;8(4):510-517. doi:10.1111/jdi.12625.

109. Bolinder J, Ljunggren Ö, Kullberg J, et al. Effects of dapagliflozin on body weight, total fat mass, and regional adipose tissue distribution in patients with type 2 diabetes mellitus with inadequate glycemic control on metformin. J Clin Endocrinol Metab. 2012;97(3):1020-1031. doi:10.1210/jc.2011-2260.

110. Ferrannini G, Hach T, Crowe S, Sanghvi A, Hall KD, Ferrannini E. Energy Balance After Sodium–Glucose Cotransporter 2 Inhibition. Diabetes Care. 2015;38(9):1730-1735. doi:10.2337/dc15-0355.

111. Mikhail N. Place of sodium-glucose co-transporter type 2 inhibitors for treatment of type 2 diabetes. World J Diabetes. 2014. doi:10.4239/wjd.v5.i6.854.

112. European Medicines Agency. Invokana EPAR Product Information.; 2013. doi:10.1016/0305-750X(79)90089-5.

113. European Medicines Agency. Jardiance EPAR Product Information.; 2014. doi:10.1016/0305-750X(79)90089-5.

114. European Medicines Agency. Forxiga EPAR Product Information.; 2012. doi:10.1016/0305-750X(79)90089-5.

115. Ueda P, Svanström H, Melbye M, et al. Sodium glucose cotransporter 2 inhibitors and risk of serious adverse events: nationwide register based cohort study. BMJ. 2018. doi:10.1136/bmj.k4365.

116. Daniele G, Xiong J, Solis-Herrera C, et al. Dapagliflozin enhances fat oxidation and ketone production in patients with type 2 diabetes. Diabetes Care. 2016. doi:10.2337/dc15-2688.

117. Ferrannini E, Mark M, Mayoux E. CV protection in the EMPA-REG OUTCOME trial: A thrifty substrate hypothesis. Diabetes Care. 2016;39(7):1108-1114. doi:10.2337/dc16-0330.

118. Bonner C, Kerr-Conte J, Gmyr V, et al. Inhibition of the glucose transporter SGLT2 with dapagliflozin in pancreatic alpha cells triggers glucagon secretion. Nat Med. 2015;21(5):512-517. doi:10.1038/nm.3828.

119. Zinman B, Wanner C, Lachin JM, et al. Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes. N Engl J Med. 2015;373(22):2117-2128. doi:10.1056/NEJMoa1504720.

120. Kosiborod M, Cavender MA, Fu AZ, et al. Lower Risk of Heart Failure and Death in Patients Initiated on SGLT-2 Inhibitors Versus Other Glucose-Lowering Drugs: The CVD-REAL Study. Circulation. 2017:249-259. doi:10.1161/ CIRCULATIONAHA.117.029190.

63

121. Hallow KM, Helmlinger G, Greasley PJ, McMurray JJV, Boulton DW. Why do SGLT2 inhibitors reduce heart failure hospitalization? A differential volume regulation hypothesis. Diabetes, Obes Metab. 2018. doi:10.1111/dom.13126.

122. Heerspink HJL, Perkins BA, Fitchett DH, Husain M, Cherney DZI. Sodium Glucose Cotransporter 2 Inhibitors in the Treatment of Diabetes Mellitus: Cardiovascular and Kidney Effects, Potential Mechanisms, and Clinical Applications. Circulation. 2016. doi:10.1161/CIRCULATIONAHA.116.021887.

123. Wanner C, Marx N. SGLT2 inhibitors: the future for treatment of type 2 diabetes mellitus and other chronic diseases. Diabetologia. 2018.

124. Vallon V, Thomson SC. Targeting renal glucose reabsorption to treat hyperglycaemia: the pleiotropic effects of SGLT2 inhibition. Diabetologia. 2017. doi:10.1007/s00125-016-4157-3.

125. Holst JJ. The physiology of glucagon-like peptide 1. Physiol Rev. 2007;87(4):1409-1439. doi:10.1152/physrev.00034.2006.

126. Röhrborn D, Wronkowitz N, Eckel J. DPP4 in diabetes. Front Immunol. 2015;6(JUL). doi:10.3389/fimmu.2015.00386.

127. Toft-Nielsen M-B, Damholt MB, Masbad S, et al. Determinants of the Impaired Secretion of Glucagon- Like Peptide-1 in Type 2 Diabetic Patients. J Clin Endocrinol Metab. 2001;86(July):3717–3723. doi:10.1210/jcem.86.8.7750.

128. Holst JJ, Schwartz TW, Lovgreen NA, Pedersen O, Beck-Nielsen H. Diurnal profile of pancreatic polypeptide, pancreatic glucagon, gut glucagon and insulin in human morbid obesity. Int J Obes. 1983;7(6):529-538. http://europepmc.org/ abstract/med/6360922.

129. Mayo KE, Miller LJ, Bataille D, et al. International Union of Pharmacology. XXXV. The glucagon receptor family. Pharmacol Rev. 2003;55(1):167-194. doi:10.1124/pr.55.1.6.

130. Drucker DJ, Nauck MA. The incretin system: glucagon-like peptide-1 receptor agonists and dipeptidyl peptidase-4 inhibitors in type 2 diabetes. Lancet. 2006. doi:10.1016/S0140-6736(06)69705-5.

131. Muskiet MHA, Tonneijck L, Smits MM, et al. GLP-1 and the kidney: From physiology to pharmacology and outcomes in diabetes. Nat Rev Nephrol. 2017. doi:10.1038/nrneph.2017.123.

132. Nauck MA, Meier JJ, Cavender MA, El Aziz MA, Drucker DJ. Cardiovascular actions and clinical outcomes with glucagon-like peptide-1 receptor agonists and dipeptidyl peptidase-4 inhibitors. Circulation. 2017. doi:10.1161/ CIRCULATIONAHA.117.028136.

133. Novo Nordisk. Oral semaglutide shows superior improvement in HbA1c vs empagliflozin in the PIONEER 2 trial. Co Announc. 2018;(47):45-47. https://globenewswire.com/news-release/2018/05/29/1513377/0/en/Oral-semaglutide-shows-superior-improvement-in-HbA1c-vs-empagliflozin-in-the-PIONEER-2-trial.html.

134. Holst JJ. The physiology of glucagon-like peptide 1. Physiol Rev. 2007;87(4):1409-1439. doi:10.1152/physrev.00034.2006.

135. Eng J, Kleinman WA, Singh L, Singh G, Raufman JP. Isolation and characterization of exendin-4, an exendin-3 analogue, from Heloderma suspectum venom. J Biol Chem. 1992. doi:10.1007/s10529-011-0745-y.

64

136. Nielsen LL, Young AA, Parkes DG. Pharmacology of exenatide (synthetic exendin-4): A potential therapeutic for improved glycemic control of type 2 diabetes. Regul Pept. 2004. doi:10.1016/j.regpep.2003.10.028.

137. Werner U, Haschke G, Herling AW, Kramer W. Pharmacological profile of lixisenatide: A new GLP-1 receptor agonist for the treatment of type 2 diabetes. Regul Pept. 2010. doi:10.1016/j.regpep.2010.05.008.

138. Unger JR, Parkin CG. Glucagon-like peptide-1 (GLP-1) receptor agonists: Differentiating the new medications. Diabetes Ther. 2011;2(1):29-39. doi:10.1007/s13300-010-0013-5.

139. Vella A. Mechanism of Action of DPP-4 Inhibitors—New Insights. J Clin Endocrinol Metab. 2012;97(8):2626-2628. doi:10.1210/jc.2012-2396.

140. Wang T, Gou Z, Wang F, Nling M, Zhai S Di. Comparison of GLP-1 analogues versus sitagliptin in the management of type 2 diabetes: Systematic review and meta-analysis of head-to-head studies. PLoS One. 2014. doi:10.1371/journal.pone.0103798.

141. Htike ZZ, Zaccardi F, Papamargaritis D, Webb DR, Khunti K, Davies MJ. Efficacy and safety of glucagon-like peptide-1 receptor agonists in type 2 diabetes: A systematic review and mixed-treatment comparison analysis. Diabetes, Obes Metab. 2017. doi:10.1111/dom.12849.

142. Sun F, Wu S, Chai S, Yang Z, Yu K, Zhan S. Impact of GLP-1RA on heart rate , blood pressure and hypertension among type 2 diabetes: A systematic review and network meta-analysis. Value Heal. 2014;17(7):A719-A720.

143. Hernandez AF, Green JB, Janmohamed S, et al. Albiglutide and cardiovascular outcomes in patients with type 2 diabetes and cardiovascular disease (Harmony Outcomes): a double-blind, randomised placebo-controlled trial. Lancet. 2018. doi:10.1016/S0140-6736(18)32261-X.

144. Eli Lilly and company. Trulicity® (dulaglutide) demonstrates superiority in reduction of cardiovascular events for broad range of people with type 2 diabetes. Co Announc. 2018. http://www.lillydiabetes.com/.

145. Holman RR, Bethel MA, Mentz RJ, et al. Effects of Once-Weekly Exenatide on Cardiovascular Outcomes in Type 2 Diabetes. N Engl J Med. 2017:NEJMoa1612917. doi:10.1056/NEJMoa1612917.

146. Pfeffer MA, Claggett B, Diaz R, et al. Lixisenatide in Patients with Type 2 Diabetes and Acute Coronary Syndrome. N Engl J Med. 2015. doi:10.1056/NEJMoa1509225.

147. Parker HM, Johnson NA, Burdon CA, Cohn JS, O’Connor HT, George J. Omega-3 supplementation and non-alcoholic fatty liver disease: A systematic review and meta-analysis. J Hepatol. 2012;56(4):944-951. doi:10.1016/j.jhep.2011.08.018.

148. Couet C, Delarue J, Ritz P, Antoine JM, Lamisse F. Effect of dietary fish oil on body fat mass and basal fat oxidation in healthy adults. Int J Obes. 1997;21(November 1996):637-643. doi:10.1038/sj.ijo.0800451.

149. Wilson PWF, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation. 1998. doi:10.1161/01.CIR.97.18.1837.

150. Amarin Corporation plc. REDUCE-ITTM Cardiovascular Outcomes Study of Vascepa® (icosapent ethyl) Capsules Met Primary Endpoint. Co Announc. 2018:3-6. https://investor.amarincorp.com/news-releases/news-release-details/reduce-ittm-cardiovascular-outcomes-study-vascepar-icosapent.

65

151. Ito MK. A Comparative Overview of Prescription Omega-3 Fatty Acid Products. P T. 2015.

152. Yoneda YSYSM. Novel antidiabetic medications for non-alcoholic fatty liver disease with type 2 diabetes mellitus. Hepatol Res. 2017;47(4):266-280.

153. Merovci A, Solis-Herrera C, Daniele G, et al. Dapagliflozin improves muscle insulin sensitivity but enhances endogenous glucose production. J Clin Invest. 2014;124(2):509-514. doi:10.1172/JCI70704.

154. Ferrannini E, Muscelli E, Frascerra S, et al. Metabolic response to sodium-glucose cotransporter 2 inhibition in type 2 diabetic patients. J Clin Invest. 2014;124(2):499-508. doi:10.1172/JCI72227.uptake.

155. Zaccardi F, Webb DR, Htike ZZ, Youssef D, Khunti K, Davies MJ. Efficacy and safety of sodium-glucose co-transporter-2 inhibitors in type 2 diabetes mellitus: systematic review and network meta-analysis. Diabetes, Obes Metab. 2016. doi:10.1111/dom.12670.

156. Horie I, Abiru N, Hongo R, et al. Increased sugar intake as a form of compensatory hyperphagia in patients with type 2 diabetes under dapagliflozin treatment. Diabetes Res Clin Pract. 2018. doi:10.1016/j.diabres.2017.11.016.

157. Frias JP, Guja C, Hardy E, et al. Exenatide once weekly plus dapagliflozin once daily versus exenatide or dapagliflozin alone in patients with type 2 diabetes inadequately controlled with metformin monotherapy (DURATION-8): a 28 week, multicentre, double-blind, phase 3, randomised control. Lancet Diabetes Endocrinol. 2016;4(12):1004-1016. doi:10.1016/S2213-8587(16)30267-4.

158. List JF, Woo V, Morales E, Tang W, Fiedorek FT. Sodium-glucose cotransport inhibition with dapagliflozin in type 2 diabetes. Diabetes Care. 2009. doi:10.2337/dc08-1863.

159. Guja C, Frias JP, Ahmed AA, et al. DURATION-8 randomised controlled trial 1-year results: efficacy and safety of once-weekly exenatide plus once-daily dapagliflozin versus exenatide or dapagliflozin alone. Diabetol Conf 53rd Annu Meet Eur Assoc study diabetes, EASD 2017 Port. 2017. doi:10.1007/s00125-017-4350-z.

160. 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(1):11-22. doi:10.1056/NEJMoa1411892.

161. Pereira MJ, Lundkvist P, Kamble PG, et al. A Randomized Controlled Trial of Dapagliflozin Plus Once-Weekly Exenatide Versus Placebo in Individuals with Obesity and Without Diabetes: Metabolic Effects and Markers Associated with Bodyweight Loss. Diabetes Ther. 2018. doi:10.1007/s13300-018-0449-6.

162. Jones BJ, Tan T, Bloom S. Minireview: Glucagon in stress and energy homeostasis. Endocrinology. 2012. doi:10.1210/en.2011-1979.

163. Jiang S, Yan C, Fang QC, et al. Fibroblast growth factor 21 is regulated by the IRE1α-XBP1 branch of the unfolded protein response and counteracts endoplasmic reticulum stress-induced hepatic steatosis. J Biol Chem. 2014. doi:10.1074/jbc.M114.565960.

164. Koliaki C, Szendroedi J, Kaul K, et al. Adaptation of Hepatic Mitochondrial Function in Humans with Non-Alcoholic Fatty Liver Is Lost in Steatohepatitis. Cell Metab. 2015. doi:10.1016/j.cmet.2015.04.004.

66

165. VESSBY B, GUSTAFSSON I-B, TENGBLAD S, BOBERG M, ANDERSSON A. Desaturation and Elongation of Fatty Acids and Insulin Action. Ann N Y Acad Sci. 2006. doi:10.1111/j.1749-6632.2002.tb04275.x.

166. Gylfe E. Glucose control of glucagon secretion-’There’s a brand-new gimmick every year’. Ups J Med Sci. 2016;121(2):120-132. doi:10.3109/ 03009734.2016.1154905.

167. Lee M, Kim M, Park JS, et al. Higher glucagon-to-insulin ratio is associated with elevated glycated hemoglobin levels in type 2 diabetes patients. Korean J Intern Med. 2017;[Epub ahea:1-10. https://doi.org/10.3904/kjim.2016.233.

168. Kramer CK, Zinman B, Choi H, Connelly PW, Retnakaran R. Impact of the Glucagon Assay When Assessing the Effect of Chronic Liraglutide Therapy on Glucagon Secretion. J Clin Endocrinol Metab. 2017;102(8):2729-2733. http://dx.doi.org/10.1210/jc.2017-00928.

169. Sach-Friedl S, Augustin T, Magnes C, Ekardt E, Eberl A, Narath S, Brunner M, Korsatko S, Evehlikova E, Treiber G PT. Effect of dapagliflozin, saxagliptin, and the combination of the both on glucagon, endogenous glucose production (EGP) and glycerol in patients with type 2 diabetes. In: European Association for the Study of Diabetes; 14 September 2017. Lisbon, Portugal; 2017:890.

170. Yanai H, Hakoshima M, Adachi H, et al. Effects of Six Kinds of Sodium-Glucose Cotransporter 2 Inhibitors on Metabolic Parameters, and Summarized Effect and Its Correlations With Baseline Data. J Clin Med Res. 2017. doi:10.14740/ jocmr3046w.

171. Basu D, Huggins LA, Scerbo D, et al. Mechanism of Increased LDL (Low-Density Lipoprotein) and Decreased Triglycerides With SGLT2 (Sodium-Glucose Cotransporter 2) Inhibition. Arterioscler Thromb Vasc Biol. 2018. doi:10.1161/ATVBAHA.118.311339.

172. Ferrannini E, Baldi S, Frascerra S, et al. Shift to Fatty Substrate Utilization in Response to Sodium-Glucose Cotransporter 2 Inhibition in Subjects Without Diabetes and Patients With Type 2 Diabetes. Diabetes. 2016;65(5):1190-1195. doi:10.2337/db15-1356.

173. Lovshin JA, Barnie A, DeAlmeida A, Logan A, Zinman B, Drucker DJ. Liraglutide promotes natriuresis but does not increase circulating levels of atrial natriuretic peptide in hypertensive subjects with type 2 diabetes. Diabetes Care. 2015. doi:10.2337/dc14-1958.

174. Irace C, De Luca S, Shehaj E, et al. Exenatide improves endothelial function assessed by flow mediated dilation technique in subjects with type 2 diabetes: results from an observational research. Diab Vasc Dis Res. 2013;10(1):72-77. doi:10.1177/1479164112449562.

175. European Medicines Agency. Omega-3 Fatty Acid Medicines No Longer Considered Effective in Preventing Heart Disease.; 2018. https:// www.ema.europa.eu/documents/referral/omega-3-fatty-acid-medicines-omega-3-fatty-acid-medicines-no-longer-considered-effective-preventing_en.pdf.

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1534

Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofMedicine. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Medicine”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-373522

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2019