Combating the dual burden: therapeutic targeting of common pathways in obesity and type 2 diabetes

Transcription

1 Combating the dual burden: therapeutic targeting of common pathways in obesity and type 2 diabetes André J Scheen, Luc F Van Gaal The increasing prevalence of obesity is contributing substantially to the ongoing epidemic of type 2 diabetes. Abdominal adiposity, a feature of ectopic fat syndrome, is associated with silent inflammation, abnormal hormone secretion, and various metabolic disturbances that contribute to insulin resistance and insulin secretory defects, resulting in type 2 diabetes, and induce a toxic pattern that leads to cardiovascular disease, liver pathologies, and cancer. Despite the importance of weight control strategies in the prevention and management of type 2 diabetes, long-term results from lifestyle or drug interventions are generally disappointing. Furthermore, most of the classic glucose-lowering drugs have a side-effect of weight gain, which renders the management of most overweight or obese people with type 2 diabetes even more challenging. Many anti-obesity pharmacological drugs targeting central control of appetite were withdrawn from the market because of safety concerns. The gastrointestinal lipase inhibitor orlistat was the only anti-obesity drug available until the recent US, but not European, launch of phentermine controlledrelease topiramate and lorcaserin. Improved knowledge about bodyweight regulation opens new prospects for the potential use of peptides derived from the gut or the adipose tissue. Combination therapy will probably be necessary to avoid compensatory mechanisms and potentiate initial weight loss while avoiding weight regain. New glucoselowering treatments, especially glucagon-like peptide-1 receptor agonists and sodium glucose cotransporter-2 inhibitors, offer advantages over traditional antidiabetic drugs by promoting weight loss while improving glucose control. In this Review, we explore the overlapping pathophysiology and also how various treatments can, alone or in combination, combat the dual burden of obesity and type 2 diabetes. Introduction The dramatic rise of the dual epidemics of type 2 diabetes and obesity is associated with increased mortality and morbidity and represents one of the most important public health challenges worldwide. 1 Type 2 diabetes is a complex disease in which genetic and epigenetic factors interact with a toxic environment that promotes the development of obesity. 2 Environmental risk factors include the consumption of high-calorie and high-fat foods, inadequate physical activity, and recently proposed alternative mechanisms (panel), to create a chronic energy imbalance. 3 Abdominal adiposity in particular is associated with a substantially increased risk for type 2 diabetes, and overall 80% of people with type 2 diabetes are overweight or obese. Striking parallel increases in the prevalence of obesity and type 2 diabetes support the importance of body fatness as a contributing factor to the occurrence of diabetes and its complications. 4 Furthermore, adipose tissue is an active endocrine and immune organ whose dysfunction (adiposopathy or so-called sick fat) is promoted by excessive caloric balance. Thus, targeting of adiposopathy and not only excess bodyweight should be viewed as a main objective in the management of obese patients with type 2 diabetes. 5 Both obesity and type 2 diabetes are associated with many medical complications, especially cardiovascular disease, 6 which substantially increase global health-care costs. 1 Aggressive treatment, especially prevention of weight gain and ideally facilitation of weight reduction, can minimise and reduce diabetes-associated complications. 7 However, weight loss and maintenance are challenging in obese people without diabetes, and even more so in obese people with type 2 diabetes. 8 The close link between type 2 diabetes and excess bodyweight emphasises the need to consider the weight effects of different treatment regimens, besides their effects on glucose homoeostasis. 9 A shift from a glucocentric to a weight-centric management of type 2 diabetes can be proposed, which emphasises the urgent need for new treatment strategies. 10,11 Some glucoselowering drugs cause weight loss and some (albeit fewer) anti-obesity drugs improve glucose tolerance; thus, in this Review, we explore the overlapping pathophysiology of obesity and type 2 diabetes and also how the various treatments can, alone or in combination, combat the dual burden of these diseases. We will focus on lifestyle and drug therapy; the notable success of bariatric or metabolic surgery in combating obesity and type 2 diabetes has been covered elsewhere. 12 Panel: Main contributors to the pathophysiology of obesity and type 2 diabetes Genetic predisposition and unhealthy lifestyle (sedentary lifestyle and overeating) Adipose tissue dysfunction leading to ectopic fat deposition Excess visceral fat (adiposopathy, associated with silent inflammation) Insulin resistance leading to β-cell burden and progressive β-cell dysfunction or defect Dysfunction in neurohumoral pathways (links between brain, endocrine pancreas, adipose tissue, and gastrointestinal tract) Potential role of environmental pollutants Potential interference of the intestinal microbiome Lancet Diabetes Endocrinol 2014; 2: Published Online February 19, S (14)70004-X University of Liège, Division of Diabetes, Nutrition and Metabolic Disorders and Clinical Pharmacology Unit, CHU Sart Tilman, Liège, Belgium (Prof A J Scheen MD); and University of Antwerp, Department of Endocrinology, Diabetology and Metabolism, Antwerp University Hospital, Antwerp, Belgium (Prof L F Van Gaal MD) Correspondence to: Prof André J Scheen, Division of Diabetes, Nutrition and Metabolic Disorders, Department of Medicine, CHU Sart Tilman (B35), B 4000 Liege, Belgium andre.scheen@chu.ulg.ac.be Vol 2 November

2 Complex interplay between obesity and type 2 diabetes Effect of weight changes on type 2 diabetes The excess risk for diabetes with even modest weight gain is substantial and absolute weight gain during adulthood is a significant independent risk factor for type 2 diabetes. 7 Intentional weight loss is associated with reduced insulin resistance and a subsequent reduction in glucolipotoxicity, which improves overall glucose homoeostasis. 4,13 In both the Finnish and US diabetes prevention programmes, weight loss gradually reduced the risk of type 2 diabetes, and even modest weight loss substantially reduced the incidence of diabetes an effect that was found to persist long term. 14,15 In the recent Look AHEAD trial in overweight or obese patients with type 2 diabetes, weight loss was greater in the intensive lifestyle intervention group than in the control group throughout the study, and was accompanied by greater reductions in HbA 1c, which enabled less insulin to be used in the intervention group than in the control group. 16 Effect of glucose control on bodyweight regulation Achievement and maintenance of acceptable bodyweight is more difficult in overweight or obese patients with type 2 diabetes than in those without the disease. 17 Among other factors, hyper insulinaemia could contribute to this phenomenon through the anti-lipolytic and anabolic effects of insulin. 8 For a given weight loss during a very-low-calorie diet, the decrease in adiposity per unit of weight loss is attenuated in patients with type 2 diabetes, possibly because of the anti-lipolytic effect of hyperinsulinaemia. 18 Furthermore, improved glucose control results in decreased glucosuria and thereby limits energy wasting. 8 Thus, loss of weight and fat mass is challenging in patients with type 2 diabetes. 7,19 Although reduction of hyperglycaemia remains the main goal in the pharmacological treatment of type 2 diabetes, avoidance of weight gain can be a clinically important secondary goal. 20 With the exception of metformin, traditional glucose-lowering drugs for type 2 diabetes (sulfonylureas, glitazones, and insulin) can further increase weight in overweight or obese patients, 21 which can undermine the benefits of improved glycaemic control. 7 Although a link between drug-induced weight gain and increased cardiovascular risk has not yet been established, pharmacological approaches that do not cause weight gain or promote weight loss for example those that target the gut (incretin-based therapies) 22 or the kidneys (sodium glucose cotransporter-2 [SGLT2] inhibitors) 23 might be a valuable alternative in selected patients, despite their higher cost. 21 Comorbidities shared and magnified by both obesity and diabetes Adipose tissue releases many bioactive mediators that affect bodyweight homoeostasis, insulin resistance, lipids, blood pressure, coagulation, fibrinolysis, and inflam mation, which together lead to oxidative stress, endothelial dysfunction, and atherosclerosis. 6 Therefore, obesity is not only a major risk factor for the development of type 2 diabetes but also predisposes individuals to comorbidities including hypertension, dyslipidaemia, and sleep apnoea. Together, obesity and comorbid dis orders increase the risk for cardiovascular disease, the major cause of morbidity and mortality in type 2 diabetes. Conversely, moderate weight loss (>5% initial bodyweight), which has been associated with improvements in glycaemic parameters, also reduces the occurrence of other comorbidities, including dyslipidaemia and hypertension, culminating in a reduced risk of cardiovascular disease. 24 For all new medications developed as anti-obesity drugs (or as antidiabetic agents), their effect on cardiovascular risk factors (and ideally cardiovascular outcomes) is important to analyse, in addition to their specific weightloss effects or glucose-lowering benefits. 5 Non-alcoholic fatty liver disease is also often associated with adiposity (especially abdominal adiposity), dyslipidaemia, and insulin resistance. Type 2 diabetes is an additional risk factor for progressive liver disease and liver-related death in patients with non-alcoholic fatty liver disease. Non-alcoholic steatohepatitis is present in roughly 10% of patients with type 2 diabetes and is associated with an increased risk for cirrhosis, hepatocellular carcinoma, and liver-related death. 25 A growing body of evidence supports a connection between type 2 diabetes, obesity, and cancer. 26 Several meta-analyses of epidemiological data show that people with type 2 diabetes are at increased risk of developing many different types of cancer and have a raised cancer mortality risk. Several common risk factors, including obesity, could underlie the association between diabetes and cancer. Abdominal adiposity, as part of the ectopic fat syndrome, has been shown to play a part in creating insulin resistance and a systemic pro-inflammatory environment, which could result in the development of both diabetes and cancer. 26 Finally, obstructive sleep apnoea is a highly prevalent disorder often associated with central obesity, which could worsen insulin resistance and lead to impaired glucose tolerance and type 2 diabetes. 27 Improvement of sleep apnoea with weight reduction can also improve insulin sensitivity and glucose profile. As discussed, obesity is widely known to worsen cardiovascular complications associated with diabetes 5,6 but it could also worsen microangiopathy. Obesity is an independent risk factor for kidney changes, since it is associated with an increased risk of albuminuria and glomerulosclerosis, and worsens the course of chronic kidney disease. 28 People with type 2 diabetes with a higher BMI and larger neck circumference are more likely to have diabetic retinopathy and more severe stages of this disorder than are people with a low BMI. 29 Finally, cardiovascular autonomic neuropathy is common in people with diabetes, and central obesity also seems to Vol 2 November 2014

3 have an important role in the early pathogenesis of this complication. 30 Thus, targeting of both obesity and diabetes reduces the incidence and severity of most of these comorbid disorders, increasing patient quality of life and lessening the strain on health-care systems. Pathophysiology and common therapeutic pathways The common pathophysiology of obesity and type 2 diabetes is quite complex (figure 1), and whether there is a so-called common soil, or whether obesity merely leads to type 2 diabetes, continues to be discussed. Genetic predisposition, epigenetic factors (programming in early life), and environmental exposure have a major role in the development of both obesity and type 2 diabetes. 31,32 Although epigenetic factors have attracted much interest in recent years, few data suggest that the available interventions (lifestyle and drugs) can cause epigenetic modifications. 32 Overall, type 2 diabetes develops in obese people because of inadequate islet β-cell and adipose tissue responses to chronic fuel excess, which results in nutrient spillover and metabolic stress that damages several organs. 3,13,33 Furthermore, environmental pollutants 34 and specific bacterial intestinal colonisation 35 have been proposed as additional risk factors for obesity, β-cell dysfunction, and the development of type 2 diabetes. Thus, reversal of over-nutrition (contributing to weight loss), lessening of adipose tissue disturbances, and protection or healing of β cells should be treatment priorities. 5,31 Increasing evidence suggests that dysregulation of complex neurophysiological pathways (including hormonal from adipocytes, pancreatic islets, and the gastrointestinal tract) contributes to the pathogenesis of obesity and type 2 diabetes (figure 1). For example, the neurohormonal control of bodyweight involves a complex interplay between long-term adiposity signals (eg, leptin), and short-term satiation signals (eg, amylin) (figure 1). 36 The adipocyte-secreted leptin is known as a key appetite-regulating hormone that affects food intake, energy expenditure, and behaviour. Furthermore, leptin is not only a key player in controlling bodyweight but also affects the insulin glucose axis. Dysfunction of the crosstalk between adipose tissue and the insulin axis plays a role in the development of hyperinsulinaemia and type 2 diabetes. 37 Amylin, a hormone co-secreted with insulin, exerts effects that complement those of insulin to regulate blood glucose concentrations, help to avoid weight gain, and even promote weight loss. 36 Incretins gut-derived hormones released in response to nutrient ingestion might be perturbed in obesity and type 2 diabetes. 38 For example, glucagon-like peptide-1 (GLP-1) can reduce food intake through both central (hypothalamus) and peripheral (stomach) actions, but also substantially enhances insulin secretion an effect that is essential to limit postprandial hyperglycaemia. 38 These observations open up new prospects for better integrative management of obesity and type 2 diabetes. 22 Adipokines (eg, leptin and adiponectin) Pro-inflammatory cytokines NEFA Excess visceral (ectopic) fat Adiposopathy Genetics (polygenic diseases) Epigenetics (fetal/neonatal programming) Environment (unhealthy diet, sedentary lifestyle, and pollutants) Insulin Amylin β-cell burden, dysfunction, or apoptosis Clinical challenges The main clinical objectives in tackling of the dual burden of obesity and type 2 diabetes are: in normoglycaemic overweight or obese patients, the primary prevention of type 2 diabetes, especially in patients at high risk (eg, those with impaired glucose tolerance); in overweight or obese patients with type 2 diabetes, reduction in bodyweight to facilitate glucose control (and lessen other risk factors) or at least avoid further weight gain; and the selection of effective treatments for both obesity and type 2 diabetes in the same patient while reducing associated cardiovascular risk factors and improving quality of life and life expectancy. 19 Similar to lifestyle changes that positively affect both bodyweight and glucose control (figure 2), other therapeutic interventions, especially pharma cological approaches, should ideally target both bodyweight and glucose regulation in an integrative approach. 8,19 However, long-term head-to-head comparisons of the effects of various treatment options on overall patient s prognosis are crucial but are still absent for most approaches. In particular, there is an urgent need for more data about young obese people a population in which the prevalence of obesity and type 2 diabetes is increasing. Blood glucose control becomes increasingly chal lenging in the obese patient with type 2 diabetes after failure of metformin monotherapy (a glucose-lowering drug that does not induce weight gain or hypo glycaemia). 11,19 Attempts to reach fasting glucose, post prandial glucose, or HbA 1c targets with other therapies can lead to a higher risk of hypoglycaemia, weight gain, or both, which can worsen overall prognosis. 8 Fear of these outcomes might favour physicians therapeutic inertia and reduce patient compliance, contributing to the fact that targets are largely unmet worldwide. Incretin-based therapies already offer Ghrelin GLP-1 GIP Cholecystokinin Oxyntomodulin Microbiota changes Gut barrier dysfunction Figure 1: Complex pathophysiology of obesity and type 2 diabetes, involving genetics, epigenetics, and environment burdens interfering with various target organs interconnected by complex neurohormonal pathways NEFA=non-esterified fatty acids. GLP-1=glucagon-like peptide-1. GIP=glucose-dependent insulinotropic polypeptide. Vol 2 November

4 A B Energy intake Carbohydrates Lipids Proteins (Alcohol) Glucose inflow Dietary intake Carbohydrates Hepatic glucose production Glycogenolysis/gluconeogenesis Lifestyle Bodyweight Lifestyle Energy expenditure Resting metabolic rate Diet-induced thermogenesis Physical activity Glucose outflow Tissue uptake Non-insulin dependent Insulin dependent Urinary loss (glucosuria) strategies with follow-up for at least 6 months have been assessed in people with type 2 diabetes, with varying results. 40 Dietary interventions are most effective in promoting initial weight loss. Furthermore, energy restriction improves glycaemic control within days of initiation, independent of weight loss. However, the net improvement of glucose profile is usually small 1 year after weight loss. Very-low-calorie diets lead to better initial weight loss and glycaemic control but yield no better long-term results than does more moderate energy restriction treatment. loss seems to be mediated through changes in energy intake rather than diet composition. 41 Similarly, diet composition has little effect on glycaemic control independent of total calorie intake. 40 However, recent data suggest that prebiotic and probiotic approaches can target the gut microbiome and thereby improve host metabolism. 42 Pool of plasma glucose Figure 2: Main mechanisms controlling (A) bodyweight and (B) blood glucose and the crucial role of lifestyle (before consideration of pharmacological interventions) Bodyweight changes affect glucose control, whereas glucose control can affect bodyweight changes in patients with type 2 diabetes. In addition to lifestyle, genetics can also play a part in controlling appetite and energy expenditure (resting metabolic rate, thermogenesis, and energy metabolism). some benefits in terms of both weight changes and risk of hypoglycaemia. 22 Nevertheless, additional new strategies, with low risk of hypoglycaemia and weight gain (and, if possible, with some weight loss), should be developed to better control glucose; ideally, such strategies should also improve other cardiovascular risk factors. 10,22 Worsening trends in obesity and type 2 diabetes create a serious conundrum: how can blood glucose, blood pressure, and lipids be controlled when so many antidiabetic drugs cause weight gain and thereby exacerbate other cardiovascular risk factors associated with type 2 diabetes? Thus, a clear need exists for clinicians to understand the risks and benefits of different treatment options to minimise cardiovascular risk in obese patients with type 2 diabetes. 11 In the following sections, we discuss the evidence underlying treatments for diabetes and obesity that can help to inform the clinician as to the most appropriate strategy (lifestyle, monotherapy, or combination therapy) for each individual patient. Lifestyle intervention in the management of obesity and type 2 diabetes Diet alone Obesity is usually caused by excess calorie intake in relation to energy expenditure, 3 so its treatment should mainly focus on healthy diet and increased physical activity, especially in the presence of type 2 diabetes (figure 2). 39 However, implementation and maintenance of the lifestyle changes associated with weight loss can be challenging for many patients. 24 Various weight loss Physical activity alone Exercise training consisting of aerobic exercise, resistance training, or a combination of both significantly reduced HbA 1c in patients with type 2 diabetes, 43 but a significantly greater change in BMI was not noted when exercise groups were compared with control groups. Importantly, BMI might not be a highly relevant measure in exercise studies: exercise can increase muscle mass while decreasing fat mass (which is good for patient health), but such changes are not necessarily represented by a change in BMI. In another study, a 12-week aerobic exercise programme, without a hypocaloric diet, reduced visceral adipose tissue an effect that might improve overall metabolic profile. 44 Diet and physical activity combined Lifestyle combining a healthy hypocaloric diet and physical activity is the mainstay of management to improve the metabolic profile of overweight or obese individuals with prediabetes or type 2 diabetes. The clinical effectiveness of such an approach in the prevention or delay of type 2 diabetes in people at high risk has been established from randomised controlled trials of structured interventions. 14,15 However, translation into routine practice generally has a smaller effect on the risk reduction for type 2 diabetes. 45 In a meta-analysis of randomised controlled trials, HbA 1c changes in patients with type 2 diabetes corresponded to changes in bodyweight, and both changes were rather small overall. 46 However, although these benefits are small from a population perspective, some people can achieve substantial improvements in several metabolic parameters. It must also be mentioned that lifestyle might not be sufficient to reduce cardiovascular risk in patients with type 2 diabetes. The recent Look AHEAD trial assessed whether an intensive lifestyle intervention for weight loss would reduce cardiovascular morbidity and mortality in overweight or obese patients with type 2 diabetes. 16 After a median follow-up of 9 6 years, and Vol 2 November 2014

5 despite substantial improvements in bodyweight and various cardiovascular risk factors and a partial remission of type 2 diabetes, 47 no difference was recorded in the primary outcome: a composite of death from cardiovascular causes, non-fatal myocardial infarction, non-fatal stroke, or admission to hospital for angina. 16 The fact that a difference between groups was not recorded might be due to an insufficient weight reduction in the intervention group versus the control group, or progressively better medical management of cardiovascular risk factors in routine medical care. 16 So far, no controlled study is yet available to show that lifestyle intervention can reduce the incidence and severity of diabetes-related complications. Anti-obesity drugs with positive effects on diabetes Although the obesity epidemic is continually expanding, at very high costs for health-care systems, very few options are available for the pharmacotherapy of obesity because most anti-obesity drugs developed so far have poor efficacy and safety profiles. 48 However, several such drugs have shown potential in the prevention and management of type 2 diabetes, 49 although the long-term health benefits remain unclear. Most drugs used as anti-obesity agents have been withdrawn because of safety issues. 50 Nevertheless, these drugs supported the proof-of-concept that pharmacological approaches to promote weight loss can help to improve blood glucose control and attenuate other cardiovascular risk factors in patients with type 2 diabetes, and that some of these metabolic effects occur independent of weight loss. This was first the case for fenfluramine and its stereoisomer dexfenfluramine, although these drugs, which mainly act by releasing serotonin, were only assessed in small, short-term clinical trials in patients with type 2 diabetes, 49 and are not reviewed here. In 2012, the US Food and Drug Administration approved two new drugs, lorcaserin and phentermine topiramate extended release, which might help to combat the dual burden of obesity and type 2 diabetes. 24,48,51 However, these two drugs did not receive approval from the European Medicine Agency, so the overall political environment for anti-obesity drug treatment seems to differ between the USA and Europe. 52 Although head-to-head comparison trials are not available, lorcaserin is probably less effective but better tolerated than phentermine topiramate extended release in obese patients with or without type 2 diabetes. 48 Cardiovascular outcome data will be invaluable to establish the place of both drugs in the treatment of overweight or obese patients with type 2 diabetes. 48,53 Sibutramine Sibutramine inhibits the reuptake of both norepinephrine and serotonin. It induces dose-dependent weight loss in overweight or obese patients, with improvements in some cardiovascular risk factors, especially lipid profile (increased HDL cholesterol) 54 and glucose profile. 55 In obese patients with type 2 diabetes, 56 the reductions in bodyweight, waist circumference, fasting blood glucose, and HbA 1c were significantly greater after sibutramine (10 20 mg/day) than after placebo (appendix p 1). However, the overall effect size on HbA 1c was modest ( 0 28%, 95% CI 0 13 to 0 42; p=0 0002), with some heterogeneity between studies. 57 Because sibutramine can increase heart rate and blood pressure, concerns have been raised about its cardiovascular safety. 58 The SCOUT trial in overweight or obese people with preexisting cardiovascular disease, type 2 diabetes, or both (84% of participants had type 2 diabetes) showed a significantly higher risk of a primary cardiovascular outcome event in the sibutramine group than in the placebo group. 59 However, greater weight loss was associated with reduced overall cardiovascular risk in both groups. 60 Sibutramine was withdrawn from the market because of an uncertain benefit risk balance, even in patients with type 2 diabetes. Rimonabant Rimonabant, a selective cannabinoid type 1 receptor blocker, reduces bodyweight and improves cardiovascular and metabolic risk factors in non-diabetic overweight or obese patients. 61 Rimonabant 20 mg per day produced a clinically meaningful reduction in bodyweight and improved HbA 1c and various cardiovascular and metabolic risk factors in patients with type 2 diabetes inadequately controlled on various glucose-lowering treatments (appendix p 1). In a pooled efficacy analysis, rimonabant produced weight loss and significant improvements in several cardiometabolic risk factors, such as waist circumference, HbA 1c, HDL cholesterol, triglycerides, and blood pressure, and some of these effects were partly independent of weight loss. 62 However, rimonabant increased the risk of psychiatric adverse events. 63 The cardiovascular outcome study CRESCENDO was stopped prematurely because of a higher risk of suicide in the rimonabant 20 mg group than in the placebo group, 64 leading to the withdrawal of the drug from the market. Orlistat Since the withdrawal of (dex)fenfluramine, rimonabant, and sibutramine, orlistat which acts through a peripheral rather than central mechanism remains the only approved anti-obesity drug for long-term treatment in most countries. Orlistat, a gastrointestinal lipase inhibitor, has been assessed in combination with a mildly reduced-calorie and reduced-fat diet in overweight or obese patients with 57 or without 65 type 2 diabetes. Orlistat (120 mg three times per day) has proven its efficacy and tolerability in 6 12-month randomised placebo-controlled trials in overweight or obese patients with various background treatments (appendix p 2). Orlistat is a useful adjunctive treatment for weight loss and reduction See Online for appendix Vol 2 November

6 in waist circumference and to improve glycaemic control, β-cell function and insulin resistance indices, serum lipid concentrations, inflammatory markers, and, in some trials, blood pressure. More patients given orlistat improved from diabetic status to normal or impaired glucose tolerance compared with those given placebo. In a 4-year randomised controlled trial, orlistat plus lifestyle changes produced slightly greater weight loss and resulted in a greater reduction in the incidence of type 2 diabetes (hazard ratio 0 627, 95% CI ; p=0 0032) in obese patients than lifestyle changes alone. 66 The preventive effect was essentially explained by the difference in participants with impaired glucose tolerance. The tolerance profile for orlistat is acceptable, although mild-to-moderate, and mostly transient, gastrointestinal events were reported. 66 A retrospective analysis of pooled data suggested that orlistat improves glycaemic control more than would be predicted by weight loss alone, 67 in agreement with an indirect comparison between orlistat and sibutramine. 57 Postulated mechanisms underlying this intrinsic effect of orlistat on glucose control include improved insulin sensitivity, incomplete digestion of dietary fat, reduction of postprandial plasma non-esterified fatty acids, decreased visceral adipose tissue, and partial stimulation of GLP-1 secretion. 67 Phentermine topiramate combination Topiramate, a sulfamate-substituted monosaccharide whose complex mechanism of action remains poorly understood, is approved as a treatment for migraine headaches, bipolar disease, and seizure disorders. Although known to facilitate weight loss an effect confirmed in randomised controlled trials in obese people 68 topiramate monotherapy does not have a regulatory indication as an anti-obesity drug. Nevertheless, randomised controlled trials of topiramate (96 mg or 192 mg per day) in overweight or obese patients with type 2 diabetes (appendix pp 3 4) showed that topiramate significantly reduced bodyweight and HbA 1c in a dose-dependent manner, with an HbA 1c reduction more than predicted from observed weight loss. However, topiramate 96 mg twice daily did not significantly increase insulin-stimulated glucose uptake, despite reductions in bodyweight, body fat, and HbA 1c. 69 Use of the new controlled release formulation of topiramate extended release confirmed significant reductions in bodyweight and HbA 1c in patients with type 2 diabetes (appendix pp 3 4). However, the CNS and psychiatric adverse event profile of topiramate makes it unsuitable as a monotherapy for obesity and type 2 diabetes. Phentermine, a well-known sympathomimetic amine that acts as stimulant and appetite suppressant, was combined with topiramate extended release as a fixeddose combination to further decrease appetite and increase satiety. 70 Several randomised controlled trials showed it to be effective for weight loss and to cause improvements in adiposopathy-associated metabolic diseases, including dysglycaemia. 71 Nevertheless, the higher dose of phentermine plus topiramate was associated with a slight increase in depression- and anxiety-related adverse events and in heart rate. 72 The efficacy and safety of two doses of phentermine plus topiramate extended release (7 5 plus 46 mg and 15 plus 92 mg) as an adjunctive therapy to lifestyle for weight loss and metabolic risk reduction were assessed in people who were overweight or obese with two or more risk factors (hypertension, dyslipidaemia, diabetes or prediabetes, or abdominal obesity). 72 In a diabetes subanalysis of this 1-year CONQUER trial, metformin-treated patients with type 2 diabetes reported greater weight loss and reductions in HbA 1c with both doses of phentermine plus topiramate than with placebo (appendix pp 3 4). 72 These positive results were confirmed in a 2-year extension period (SEQUEL trial). 73 In patients with type 2 diabetes at baseline, HbA 1c did not change in the placebo group (despite net increases in glucose-lowering drugs), whereas treatment with low and high strengths of the fixed-dose combination led to HbA 1c reductions (appendix pp 3 4). 73 In a post-hoc analysis in overweight participants with prediabetes, metabolic syndrome, or both, phentermine plus topiramate extended release produced significant weight loss and substantially reduced progression to type 2 diabetes after 108 weeks (compared with placebo, phentermine 7 5 mg plus topiramate 46 mg reduced type 2 diabetes incidence by 70 5%, and phentermine 15 mg plus topiramate 92 mg reduced incidence by 78 7% [both p<0 05 vs placebo]). loss was accompanied by improvements in several cardiometabolic disease risk factors. 74 Lorcaserin Oral lorcaserin, a selective serotonin 5-HT 2C receptor agonist that causes reduced food intake and increased satiety, is indicated in the USA as an adjunct to diet and exercise in the chronic weight management of obese (BMI 30 kg/m²) or overweight (BMI 27 kg/m²) adults with at least one weight-related comorbidity (eg, dyslipidaemia, hypertension, or type 2 diabetes). 75 In a systematic review and meta-analysis of 1-year randomised controlled trials in obese adults, weight loss of 3 23 kg was reported with lorcaserin (10 mg once or twice daily) compared with placebo. 76 Compared with placebo, lorcaserin reduced waist circumference, blood pressure, cholesterol, and triglycerides, without affecting heart rate. Adverse events were headache, nausea, and dizziness. 75,76 The 1-year BLOOM-DM study assessed lorcaserin in patients with type 2 diabetes who were given metformin, a sulfonylurea, or both. 77 loss was amplified (more patients lost 5% bodyweight) and HbA 1c and fasting glucose reductions were greater with the two doses of lorcaserin (10 mg once or twice daily) than with placebo (appendix p 4). 77 Slightly more patients taking lorcaserin than placebo reduced their overall use of oral antidiabetic Vol 2 November 2014

7 drugs, whereas fewer patients taking lorcaserin increased the total daily dose of glucose-lowering drugs. Insulin resistance index was reduced more substantially with lorcaserin than with placebo, but no significant betweengroup differences were recorded in terms of lipid profile, blood pressure, or inflammatory markers. Symptomatic hypoglycaemia occurred slightly more frequently with lorcaserin than with placebo. 77 Antidiabetic drugs with positive effects on obesity Pramlintide Pramlintide is an analogue of the pancreatic hormone amylin, which is deficient in patients with type 2 diabetes. 36 Through mechanisms similar to those of amylin, pramlintide ( μg three times daily) improves overall glycaemic control and reduces bodyweight in patients with type 2 diabetes (figure 3, appendix p 5). In addition to reducing postprandial glucose concentrations, pramlintide treatment is also associated with improvements in markers of oxidative stress and cardiovascular risk. Pramlintide has been shown to increase satiety and therefore decrease caloric intake through a central mechanism. 36 It is generally well tolerated, with the most frequent treat ment-emergent adverse event being transient nausea. Preliminary trials assessing the use of pramlintide for weight loss in obese patients without diabetes have shown weight loss of up to 8 kg after 1 year. However, these trials were limited by inconsistent study design, dosing, and patient populations. 78 New research is focusing on weight loss effects with a combined peptide approach in obese people (see later). 36,79 GLP-1 receptor agonists and DPP-4 inhibitors Incretin-based therapies exploit the insulinotropic actions of glucose-dependent insulinotropic polypeptide and GLP-1 for the treatment of type 2 diabetes (figure 3, appendix p 5). These therapies include GLP-1 receptor agonists (GLP-1 RAs) and dipeptidyl peptidase-4 (DPP-4) inhibitors (also called gliptins), the enzyme that inactivates these two incretin hormones in the body. 38 neutrality of DPP-4 inhibitors while improving glucose control 80 could be an advantage in the management of overweight or obese patients with type 2 diabetes compared with other glucose-lowering drugs that are known to promote weight gain. 21 Furthermore, these new compounds do not expose patients to hypoglycaemia (except when added to insulin or sulfonylureas) 80 and might also exert favourable effects on various cardiovascular risk factors. 81 Similar positive effects are observed with GLP-1 RAs, 82 which have the additional value of promoting weight loss in obese patients with type 2 diabetes. 22 Furthermore, the effects of GLP-1 RAs on bodyweight are also apparent in nondiabetic obese people, which means they could potentially be useful to treat obesity. 83 According to a systematic review of randomised controlled trials in overweight or obese patients with or without type 2 diabetes, 84 treatment with GLP-1 RAs (as monotherapy or combination therapy: exenatide twice daily, exenatide once weekly, or liraglutide once daily) results in greater weight loss than in control groups (placebo, oral antidiabetic drugs, or insulin). For example, weight loss in the GLP-1 RA groups for patients with diabetes was 2 8 kg (95% CI 3 4 to 2 3) and for patients without diabetes was 3 2 kg (95% CI 4 3 to 2 1). In patients with type 2 diabetes, GLP-1 RAs also have beneficial effects on blood pressure, plasma cholesterol, and glycaemic control (appendix p 5). 84 GLP-1 RAs are associated with nausea, diarrhoea, and vomiting, and a slight increase in pulse rate, but not with hypoglycaemia. 84 In a post-hoc analysis of 26-week data from seven randomised controlled trials assessing liraglutide ( mg once daily) in type 2 diabetes, patients given liraglutide experienced additional HbA 1c reduction beyond that expected by observed weight loss. 85 In patients with type 2 diabetes not well controlled on oral glucose-lowering drugs (mainly metformin, sulfonylureas, or both) and compared with insulin glargine, 86 the addition of GLP-1 RAs showed almost Sulfonylureas, glinides GLP-1 RAs gain SGLT2 inhibitors DPP-4 inhibitors loss loss neutral Stomach and intestine Pancreatic β cells Kidney Lipase inhibitors loss Intestine Glucose control Brain loss Leptin, amylin Anorectic drugs α-glucosidase inhibitors neutral Adipose tissue Liver Muscle gain loss* Glitazones, insulin gain Metformin Insulin Figure 3: Several pharmacological approaches used to control hyperglycaemia in type 2 diabetes, with a focus on the drugs effects on bodyweight GLP-1 RA=glucagon-like peptide-1 receptor agonist. DPP-4=dipeptidyl peptidase-4 inhibitor. SGLT2=sodium glucose cotransporter 2. *Indicates modest effect. Vol 2 November

8 similar (or better) blood glucose control but with a lower incidence of hypoglycaemia and less weight gain (in fact, a significant weight loss was recorded). The combination of a GLP-1 RA (mainly targeting postprandial hyperglycaemia) and basal insulin (mainly targeting fasting hyperglycaemia) could also be highly effective for optimum glucose control while limiting weight gain, an adverse effect typically associated with insulin therapy in patients with type 2 diabetes. 87 Liraglutide 88,89 also seems to have promising effects on bodyweight in overweight or obese adults without type 2 diabetes. Monotherapy with liraglutide at a dose of mg once per day induced significant and sustained weight loss, improved some obesity-related risk factors (especially blood pressure), and reduced the incidence of prediabetes and metabolic syndrome over 1 2 years. 88,89 Notably, results were more favourable with liraglutide than with orlistat (120 mg three times per day orally), which was used as an active comparator. Several studies investigated whether weight loss with GLP-1 RAs corresponds to fat loss and which mechanisms might be implicated (appendix p 6). Overall, studies showed significant reductions in waist circumference, total fat, and visceral fat, proportional to weight loss; this effect resulted from decreased appetite and reduced energy intake (with some changes in food preference), with almost no or only a slight increase in energy expenditure. Finally, GLP-1 RAs were shown to improve various cardiovascular risk factors in overweight or obese patients with type 2 diabetes, especially blood pressure, postprandial lipid profile, endothelial dysfunction, inflammatory markers, and oxidative stress. 81 All these pleiotropic effects might lead to improvement in cardiovascular prognosis, although this hypothesis remained to be confirmed in ongoing large prospective randomised controlled trials. 81 SGLT2 inhibitors SGLT2 inhibitors (including dapagliflozin, cana gliflozin, and empagliflozin) increase glucose excretion, independent of insulin secretion or action, by inhibiting the renal reabsorption of glucose. 90 Dapagliflozin (5 or 10 mg) has been approved in Europe for the treatment of type 2 diabetes as an adjunct to diet and exercise, in combination with other glucose-lowering medicinal products, including insulin, and as a monotherapy for metformin-intolerant patients, 91 whereas canagliflozin was approved by the US Food and Drug Administration in April, 2013, and by the European Medicines Agency in November, Other SGLT2 inhibitors (notably empagliflozin) are in the late phase of clinical development. In several trials with patients with type 2 diabetes, SGLT2 inhibitors reduced HbA 1c (by 0 5% to 1 5%) when used as either monotherapy or combined with any other glucose-lowering agent, including insulin, with a weight loss of up to 3 kg (figure 3) and a low frequency of hypoglycaemia (appendix p 5). Cardiometabolic benefits included a reduction in systolic blood pressure and triglycerides (appendix p 5). Because of side-effects (urinary tract and genital infections), a potential risk of dehydration (elderly patients on diuretics), and reduced glucose-lowering activity in patients with renal impairment, appropriate patient selection and close monitoring will be important. Further clinical studies should show the effect of SGLT2 inhibitors on diabetic complications and cardiovascular events, and their long-term tolerability and safety profile. So far, SGLT2 inhibitors are the only oral glucoselowering drugs that also promote significant weight loss. 23 Dapagliflozin induced weight loss, mainly by reducing fat mass, visceral and subcutaneous adipose tissue in patients with type 2 diabetes inadequately controlled on metformin. 93 However, experimental data from rodents suggested that the persistent urinary glucose excretion induced by dapagliflozin was accompanied by compensatory hyperphagia. 94 These results from animals agree with findings in human beings showing a lower weight loss than that predicted by the recurrent waste of calories through sustained increased glucosuria. Therefore, weight loss induced by SGLT2 inhibitors might also need dietary intervention or other adjunctive therapies limiting food intake. The future for pharmacology of type 2 diabetes and obesity Novel pharmacological treatments are under investigation as potential treatments for obesity and type 2 diabetes. 48,50 There are two main avenues of investi gation: the first targets the CNS to reduce food intake (with drugs including naltrexone plus bupropion, tesofensine, and zonisamide), but again with the risk of frequent adverse events; 95 the second is more innovative and targets complex interrelated hormonal pathways (brain, gut, and adipose tissue) involved in weight regulation and glucose homoeostasis (figure 1). 96,97 Naltrexone sustained release plus bupropion sustained release The efficacy of existing centrally acting obesity pharmacotherapies is limited by compensatory mechanisms that mitigate weight loss and by potential sideeffects, necessitating combined therapies. Opioid receptor antagonism (naltrexone) plus pro-opiomelanocortin activation (bupropion) causes more weight loss than does monotherapy with either agent or placebo, 98 an effect confirmed in overweight or obese patients with comorbid type 2 diabetes, hypertension, or hyperlipidaemia. 99,100 Overall, placebo-subtracted mean weight loss averaged 4 7% (range %) with naltrexone bupropion after 1 year, with more patients achieving at least 5 10% weight loss and improvement of waist circumference, triglycerides, high-density lipoprotein, Vol 2 November 2014

9 fasting insulin, and insulin resistance than those given placebo. Reductions in body fat mass and visceral adipose tissue were proportional to weight loss. 101 In patients with type 2 diabetes, naltrexone bupropion treatment significantly decreased HbA 1c by 0 5% more than did placebo (appendix p 4). Because systolic blood pressure and pulse rate were higher than with placebo, further studies are necessary to assess the effect of naltrexone bupropion on cardiovascular outcomes and it is not yet approved as an anti-obesity treament. 102 Leptin and leptin-related synthetic peptide analogues Leptin not only controls food intake and energy expenditure, but also modulates the insulin glucose axis. 37 Consequently, leptin replacement therapy raised hope for the treatment of obesity and type 2 diabetes. 103 Unfortunately, results with recombinant human leptin were rather disappointing, especially in overweight or obese patients with type 2 diabetes. 104 Metreleptin, an analogue of human leptin, improved insulin sensitivity, hypertriglyceridaemia, and hyperglycaemia in patients with lipodystrophy. 105 However, in obese patients with type 2 diabetes, metreleptin did not change bodyweight or circulating inflammatory markers and reduced HbA 1c only marginally. 106 New approaches in the development of anti-obesity and anti-diabetes pharmacophores are now focused on use of leptin-related synthetic peptides as leptin receptor antagonists or leptin-related synthetic peptide analogues or mimetics. 107 Peptide hormone combination treatments Experimental data suggest that leptin and amylin, two hormones involved in satiation control (see earlier), have additive effects. 108 In a proof-of-concept randomised controlled trial in overweight or obese patients, combination treatment with pramlintide and metreleptin led to significantly, early, and sustained greater weight loss than did treatment with pramlintide or metreleptin alone. 109 Although the pramlintide metreleptin combination was promising as a novel, integrated neurohormonal approach to obesity pharma cotherapy with or without type 2 diabetes, 109,110 the latest randomised controlled trial was stopped recently because of safety concerns. 111 Finally, instead of two or more agonists being used to treat obesity and type 2 diabetes, another promising option might be the use of single co-agonists. As examples, GLP-1 glucagon and GLP-1 amylin coagonists could be of potential interest. 112 Although substantial progress has been achieved in preclinical studies, the putative success and safety of co-agonist therapy for the treatment of patients with obesity and type 2 diabetes remains uncertain and needs extensive further clinical validation. 113 Another potential future option is an SGLT2 inhibitor GLP1-RA combination in an attempt to weaken the compensatory overeating mechanism that occurs with SGLT2 inhibition. 94 Search strategy and selection criteria We searched Medline, PubMed, the Cochrane library, and Google Scholar, for mainly original research articles published between January, 1973, and November, 2013, and focused on the dual treatment of obesity and type 2 diabetes. The main search terms used were obesity, type 2 diabetes, glucose-lowering, antidiabetic, anti-obesity, and weight loss. We identified full-text papers without imposing any language restrictions. Reference lists of original studies, narrative reviews, and previous systematic reviews and meta-analyses were also searched carefully. We limited our review mainly to randomised controlled trials and meta-analyses of randomised controlled trials (for specific criteria used to select studies for review, see restrictions in the legends of the tables). Conclusions The management of the obese patient with diabetes remains challenging, but, in any case, weight reduction should be regarded as a key objective. Lifestyle interventions for weight loss are recommended for most patients with type 2 diabetes to improve glycaemic control and reduce associated risk factors for com plications. Even modest weight loss can significantly improve glucose homoeostasis and reduce cardio metabolic risk factors, although achievement and especially maintenance of 5 10% bodyweight reduction remains difficult for many patients, and these surrogate endpoints are not always strengthened by hard outcome data. Concern is growing that the weight gain induced by some diabetes drugs diminishes their clinical benefits. New glucose-lowering agents (incretin-based therapies and SGLT2 inhibitors) are associated with some weight reduction and improvement of various cardiovascular risk factors. However, the long-term benefit risk profile of these new compounds remains to be established in large prospective studies. Finally, the place of anti-obesity drugs in the management of overweight or obese patients with type 2 diabetes has been limited by rather poor efficacy and safety concerns. A better understanding of weightregulating mechanisms has led to the identification of new targets for anti-obesity drugs. In particular, peptide hormones control bodyweight and glucose homoeostasis by engaging peripheral and central metabolic signalling pathways responsible for the maintenance of bodyweight and euglycaemia. Notably, therapeutic attempts to normalise bodyweight and glycaemia with single drugs alone have generally been disappointing, which paves the route for the use of combined therapies in the future. Personalised treatment in the management of patients with type 2 diabetes will become increasingly important with (and perhaps a consequence of) the increasing number of drugs available. The presence of obesity can certainly affect the physician s choice of the best pharmacological approach for a patient with type 2 diabetes. Vol 2 November