Weight Management in Type 2 Diabetes: Current and Emerging Approaches to Treatment

Diabetes Care Volume 38, June 2015 1161

Weight Management in Type 2 Luc Van Gaal1 and Andre´ Scheen2 Diabetes: Current and Emerging Approaches to Treatment Diabetes Care 2015;38:1161–1172 | DOI: 10.2337/dc14-1630

Diabetes is a growing global health concern, as is obesity. Diabetes and obesity are intrinsically linked: obesity increases the risk of diabetes and also contributes to disease progression and cardiovascular disease. Although the beneﬁts of weight loss in the prevention of diabetes and as a critical component of managing the condition are well established, weight reduction remains challenging for individuals with type 2 diabetes due to a host of metabolic and psychological factors. For many patients, lifestyle intervention is not enough to achieve weight loss, and alternative options, such as pharmacotherapy, need to be considered. However, many traditional glucose-lowering medications may lead to weight gain. This article focuses on the potential of currently available pharmacological strategies and on emerging approaches in development to support the glycemic and weight-loss goals of individuals with type 2 diabetes. Two pharmacotherapy REVIEW types are considered: those developed primarily for blood glucose control that have a favorable effect on body weight and those developed primarily to induce weight loss that have a favorable effect on blood glucose control. Finally, the potential of combination therapies for the management of obese patients with type 2 diabetes is discussed.

Obesity and diabetes are intimately linked (1). Obesitydin particular abdominal obesitydis a major driver in the development of diabetes and cardiovascular disease (2), with the increasing prevalence of obesity mirrored by the rising prevalence of diabetes (3). In addition, obesity and overweight are associated with multiple comorbidities (4). Weight reduction, therefore, is a key therapeutic goal in both the prevention and management of type 2 diabetes (5). Weight reduction with intensive lifestyle intervention (ILI) has been shown to reduce the incidence of diabetes by 58% (6). For individuals with diabetes, studies

(Look AHEAD [Action for Health in Diabetes], N = 5,145) have shown that a loss of 1 – fi Department of Endocrinology, Diabetology 5 10% of body weight can improve tness, reduce HbA1c levels, improve cardio- and Metabolism, Antwerp University Hospital, vascular disease (CVD) risk factors, and decrease use of diabetes, hypertension, and Antwerp, Belgium lipid-lowering medications (7,8). Additional benefits of weight loss include reduction 2Department of Diabetes, Nutrition and Meta- of depression symptoms and remission or reduced severity of obstructive sleep apnea bolic Disorders and Clinical Pharmacology Unit, (9,10). Greater clinical improvements are observed with greater weight loss (8). University of Liege,` Liege,` Belgium Guidelines recommend lifestyle modifications as the foundation of weight loss. Corresponding author: Luc Van Gaal, luc.van [email protected]. Although ILI produces clinically beneficial weight loss for many patients, the reality is that ILI is difficult to achieve and maintain over the long-term for most patients. Even in Received 3 July 2014 and accepted 22 February 2015. an optimal clinical trial setting, such as Look AHEAD, one-third of all patients were © 2015 by the American Diabetes Association. unable to achieve at least 5% weight loss after 1 year (11). Most individuals with Readers may use this article as long as the work diabetes tend to lose weight over a period of 4–6months,andlose4–10% of their is properly cited, the use is educational and not baseline weight before experiencing a plateau in weight loss, generally followed by a for profit, and the work is not altered. 1162 Weight Management in Type 2 Diabetes Diabetes Care Volume 38, June 2015

weight regain (12). In Look AHEAD, half of the emotional and cognitive control re- (13,14). The resultant decrease in blood all patients who lost 5% of their body lated to food intake (22). These factors glucose levels corresponds with a weight after 1 year of ILI regained some mean that sustained weight loss can be decrease in glycosuria, a major con- or all of their initial weight loss by year 8 even more difficult to achieve for over- tributing factor to the weight gain (11). Long-term weight loss is still difficult weight and obese people with diabetes. observed in patients treated with con- to achieve for many patients, and alter- Although improvement of glucose conventional antihyperglycemic agents. native options, such as pharmacotherapy, trol remains the primary goal in pharma- Treatment with certain classes of ther- should be considered for patients who cological treatment of type 2 diabetes, apies and several baseline patient char- cannot lose weight with lifestyle modifi- avoidance of pharmacologically induced acteristics are predictive of weight gain cation alone. weight gain should also be considered (Table 2) (24). Many conventional glucose-lowering as a clinically important goal (23). Guide- Substantial increases in weight have agents commonly result in weight gain lines recommend lowering HbA1c to been observed in patients treated with (13,14). In addition to antihyperglyce- #6.5% (48 mmol/mol) or #7.0% (53 insulin (24). The mechanisms responsi- mic agents, some antipsychotic medica- mmol/mol), which often necessitates in- ble for insulin-induced weight gain are tions used to treat comorbid psychiatric sulin escalation or use of combination varied, complex, and partially unknown. disorders and antiepileptic drug deriva- therapies to achieve this goal (5). When Subcutaneous administration bypasses tives used to treat diabetic neuropathy considering combination therapies, clini- hepatic insulin sensors and leads to may lead to weight gain (15,16). Meta- cians should remain aware of the weight physiologically abnormal levels of insu- bolic, psychological, and behavioral fac- gain that is often associated with diabetes lin exposure at peripheral tissues, which tors also affect the ability of people with medications. may disrupt the homeostatic regulation diabetes to lose weight (17,18). In addi- The focus of this review article is to discuss of body weight (25). A comparative tion, homeostatic control of body weight the potential for pharmacotherapiesd study showed that subcutaneous deliv- is regulated by a complex neurohor- those currently available and those in ei- ery of insulin leads to more weight gain d monal system that involves a feedback ther late- or early-stage development to than intraperitoneal delivery (26). Pre- loop between the brain and peripheral support the glycemic and weight loss goals clinical research has shown that insulin tissues, and perturbations to this system of people with diabetes. Although we rec- has a role in the central nervous system, affect weight (19). Diet-induced weight ognize that bariatric surgery may offer a where it regulates satiety signals and loss increases the orexigenic hormones potential treatment solution for some pa- suppresses appetite, and it is suggested ghrelin and gastric inhibitory/glucose- tients (1), this topic will not be considered thatthesefunctionsmaybeimpaired dependent insulinotropic polypeptide in the present review. in type 2 diabetes (27). Insulin-induced (GIP) and decreases anorexigenic hor- hypoglycemia is a further factor; mild mones leptin, peptide YY (PYY), cholecys- WEIGHT GAIN WITH hypoglycemia in rats stimulates appe- tokinin (CCK), amylin, and glucagon-like CONVENTIONAL THERAPIES tite and leads to the increased consump- peptide 1 (GLP-1) (20,21). Improved Many pharmacological agents used tion of calories (28). Some patients glycemic control decreases glycosuria, in the treatment of diabetes directly participate in compensatory overeating which may impair weight loss. Functional contribute to weight gain through their because of their fear of hypoglycemia changes within the brain also affect glucose-lowering mechanisms (Table 1) (29). Preclinical data indicate that

Table 1—Antidiabetes therapies associated with weight gain Drug class Mechanism of action How mechanism of action leads to weight gain

Insulin* (25,27,28,30,32) c Regulates glucose metabolism c Causes hypoglycemia, a potent stimulus to feed c Lowers blood glucose by facilitating peripheral c Perceived risk of hypoglycemia may lead to glucose uptake, primarily by skeletal compensatory overeating muscle and fat c Glycemic control reverses the negative energy balance c Inhibits hepatic glucose production from glycosuria c Inhibits lipolysis c Insulin is an anabolic hormone and may lead to changes c Inhibits proteolysis in metabolism that encourage weight gain c Enhances protein synthesis Sulfonylurea (24,32) c Stimulates release of insulin from pancreatic c Glycemic control reverses the negative energy balance b-cells from glycosuria c May also have mild extrapancreatic effects c Causes hypoglycemia, a potent stimulus to feed such as increasing the sensitivity of peripheral tissues to insulin TZD (32–35) c Decreases insulin resistance in the periphery and c Increases adipocyte differentiation in the liver c Improves glycemic control c Redistributes fat (less visceral, more subcutaneous) c Increases plasma volume *Most insulins are associated with weight gain; however, insulin detemir has been shown to have a reduced weight-gain effect compared with other insulins (111). care.diabetesjournals.org Van Gaal and Scheen 1163

Table 2—Predictors of weight gain* through the activation of peroxisome – Baseline patient characteristic proliferator activated receptor-g (PPAR-g) predictive of weight gain Result (33). Activation of PPAR-g promotes pre- adipocyte differentiation into smaller, Age c Patients #65 years are more likely to gain weight mature adipocytes. Although these c Patients .65 years are more likely to lose weight smaller adipocytes are more insulin sen- Ethnicity c Caucasians are more likely to gain weight sitive, PPAR-g activation triggers an in- Smoking status c Current smokers are more likely to gain weight crease in adiponectin secretion from Baseline HbA1c c Patients with HbA1c .7.2% (55 mmol/mol) are more likely to gain weight these cells, which may also contribute to the insulin-sensitizing effects of TZDs. c Patients with HbA1c #7.2% (55 mmol/mol) are more likely to lose weight Increases in appetite and water retention *Based on results from van Dieren et al. (24). may also be factors that contribute to TZD-associated weight gain (34). Al- though TZDs may lead to an increase in fat mass, there is a shift of fat distribu- insulin inhibits lipolysis and promotes hours (32), which increases the risk of hy- tion from visceral to subcutaneous adi- lipogenesis (30). Generally, insulin has poglycemia and can then cause patients to pose depots, which may contribute to limited effects on resting metabolic rate, participate in compensatory overeating. theimprovedhepaticandperipheral and it is unlikely that insulin-induced Reduction of glycosuria is another poten- tissue sensitivity to insulin observed glucose improvement affects weight tial mechanism for weight gain with sulfo- with TZD treatment (35). through changes in basal energy expen- nylureas. Less signiﬁcant weight-gain Among traditional glucose-lowering diture (31). effectshavebeendemonstratedwith agents, metformin is the only agent Sulfonylureas are insulin secretagogs meglitinidesdanother class of insulin that can be considered weight neutral that lead to minimal weight gain, com- secretagogsdpresumably due to their (Table 3) and may even give rise to min- pared with insulin, through many of the shorter duration of action and associated imal weight loss (24). The favorable ef- same mechanisms that occur with insulin lower risk of hypoglycemia. fect on weight observed with metformin use (24). The insulin secretion after sulfo- Thiazolidinediones (TZDs) enhance use may be due to its ability to reduce nylurea administration lasts for several glucose uptake by peripheral tissues energy intake (31).

Table 3—Antidiabetes therapies that are weight neutral or have weight-loss potential How mechanism of action leads to weight Drug class Mechanism of action loss/weight neutrality

a-Glucosidase inhibitors (32) c Reversibly inhibits membrane-bound intestinal c Weight loss due to inhibition of carbohydrate a-glucoside hydrolase enzymes digestion and delayed gastric emptying via GLP-1 c Delays glucose absorption c Increases GLP-1 secretion Amylin mimetics (32,68) c Induces satiety c Weight loss due to increased satiety and decreased c Slows gastric emptying caloric intake c Decreases hepatic glucose output by suppressing postprandial secretion of glucagon Biguanides/metformin (31,32) c Decreases hepatic glucose production c May have an anorectic effect c Decreases glucose production c Decreases intestinal absorption of glucose c Improves insulin sensitivity by increasing peripheral glucose uptake and utilization GLP-1R agonists (98) c Binds and activates the human GLP-1R c Weight loss due to inhibition of gastric emptying c Enhances glucose-dependent insulin secretion c Decreased calorie ingestion through central by the pancreatic b-cell nervous system c Increases intracellular cAMP leading to insulin c Reduced acid secretion release in the presence of elevated glucose concentrations c Increases satiety DPP-4 inhibitors (98) c Increases and prolongs active incretin levels c Slight reduction in caloric intake compensating c Increases insulin release and decreases for reduction in glycosuria glucagon levels in the circulation in a glucose-dependent manner SGLT2 inhibitors (55) c Binds to SGLT2 receptors and prevents c Calorie loss due to increased renal glucose reabsorption of ﬁltered glucose excretion c Lowers renal threshold for glucose c Increases renal glucose excretion 1164 Weight Management in Type 2 Diabetes Diabetes Care Volume 38, June 2015

ANTIDIABETES THERAPIES WITH as nausea and vomiting, although rare, hypoglycemia, nausea, diarrhea, and na- WEIGHT-LOSS POTENTIAL occurred as the most commonly re- sopharyngitis; however, hypoglycemia The ability to target poor glycemic control ported adverse events (AEs), and most was much lower in patients not on con- and overweight/obesity simultaneously AEs were of mild-to-moderate severity comitant sulfonylurea therapy. represents the ideal approach to manage- and transient (46). The weight loss ob- Although GLP-1R agonists are gener- ment of type 2 diabetes (23). Several served with exenatide once-weekly was ally safe and tolerable, postmarketing therapies show promise in this area (Ta- independent of these gastrointestinal reports of acute pancreatitis with GLP- ble 3), the most promising of which are AEs. Because exenatide stimulates insu- 1R agonist use have led to pancreatitis considered below. lin secretion in a glucose-dependent being listed under the warnings and pre- manner, there was a limited occurrence cautions in exenatide (twice-daily and GLP-1 Receptor Agonists of major and minor hypoglycemia across once-weekly) and liraglutide U.S. pre- GLP-1 is an endogenous peptide hor- clinical trials (46). Liraglutide (1.8 mg scribing information (53), and the U.S. mone produced in the gut in response q.d.) has similarly been shown to reduce Food and Drug Administration (FDA) to nutrient absorption. The insulino- HbA1c (–1.18%) and weight (–3.24 kg) and the European Medicines Agency tropic action of GLP-1 is dependent on from baseline at 26 weeks (47,48). Lira- (EMA), despite a reassuring position glucose (36), which means its activity glutide (1.8 mg q.d.) is generally well statement, continue to investigate this should not be associated with hypogly- tolerated, and the most frequently re- safety signal (54). The current evidence cemia. GLP-1 exerts its effects through ported AEs were gastrointestinal (48). regarding a potential link between pan- binding to the GLP-1 receptor (GLP-1R), In a 26-week study of liraglutide (1.8 creatitis and GLP-1R agonist therapies which is expressed on pancreatic b-cells mg q.d.) compared with exenatide (10 has been conflicting, and large-scale, on- (37). GLP-1/GLP-1R signaling increases mg b.i.d.), gastrointestinal AEs resolved going, prospective studies will hopefully b-cell sensitivity to glucose and sup- more quickly, and fewer cases of minor address questions surrounding a possi- presses glucagon secretion from pancre- hypoglycemia were seen in the liraglu- ble association. atic a-cells (38,39). In addition, GLP-1 tide treatment arm (25.5%) than in the Although exenatide and liraglutide exerts extra pancreatic effects, such as exenatide arm (33.6%) (48). Episodes of are the established GLP-1R agonists in reducing hepatic glucose production minor hypoglycemia were thought to be the U.S. and Europe, lixisenatide and albi- and inhibiting gastric emptying (39). mainly due to the concomitant medica- glutide have been approved in Europe, GLP-1 action at the hypothalamus pro- tions used (sulfonylureas). and albiglutide and dulaglutide in the motes satiety (40). Because native GLP-1 Liraglutide (1.2 mg and 1.8 mg) has U.S. Other GLP-1R agonists, such as is rapidly inactivated in vivo, GLP-1R been investigated in combination with in- semaglutide and additional exenatide agonists were developed that mimic sulin, metformin, sulfonylurea, metfor- extended-release formulations (i.e., once- the actions of GLP-1 in vivo but are re- min plus rosiglitazone, or metformin monthly and once-yearly formulations) sistant to enzymatic degradation and in- plus glimepiride (49). Significant reduc- are under clinical development (53). activation by dipeptidyl peptidase-4 tions in HbA1c overbaselinewereob- Clinical trial data show that these ther- (DPP-4) (41). GLP-1R agonists exert di- served within 8 weeks of treatment with apies provide reductions in HbA1c levels verse actions on multiple-target tissues liraglutide combination therapy (plus and body weight in people with type 2 and lead to a reduction in blood glucose metformin, glimepiride, or metformin diabetes. and in body weight (42,43). plus rosiglitazone; P , 0.0001 for all com- Exenatide and liraglutide were the binations) and were maintained until Sodium-Glucose Cotransporter 2 first two GLP-1R agonists available for week 26 (50). The addition of liraglutide Inhibitors the treatment of type 2 diabetes. In a to metformin or metformin plus rosiglita- Another new class of antidiabetes ther- meta-analysis of randomized controlled zone led to weight reductions, but liraglu- apies that show potential for weight loss trials (RCTs) with exenatide twice-daily tide plus sulfonylurea treatment was (although they are not indicated for and once-weekly (trial durations of 12– weight neutral. As with monotherapy, weight loss per se) are the sodium- 52 weeks), the overall reduction in most AEs with liraglutide combination glucose cotransporter 2 (SGLT2) in- HbA1c from baseline was –1.1% (43). therapy were gastrointestinal in nature hibitors (55). In individuals with type 2 However, clinicians should consider the (51). Major hypoglycemia was reported diabetes, SGLT2 expression is increased results from studies of intention-to- only when liraglutide was used in combi- in the renal proximal tubular cells, lead- treat populations wherever possible, nation with a sulfonylurea. Similarly, ex- ing to increased renal glucose reab- because the weight-loss responses to enatide once-weekly has been studied in sorption, which ultimately aggravates treatment may differ compared with combination with other antihyperglyce- hyperglycemia. SGLT2 inhibitors reduce completer populations (44). The once- mic agents with study durations of 24– blood glucose mainly through increased weekly extended-release formulation 30 weeks (52). In combination with met- glycosuria, although indirect mechanisms of exenatide has consistently demon- formin, metformin plus sulfonylurea, have also been reported (23,55). strated weight-loss properties across sulfonylurea with or without TZD, or A meta-analysis of 10 RCTs showed multiple clinical trials (45,46), showing metformin plus TZD, treatment with ex- that dapagliflozin (1–50 mg per day) was ameanweightlossof–2.67 kg versus enatide once-weekly led to significant associated with a reduction in baseline comparator drugs (i.e., exenatide twice- improvements from baseline in HbA1c HbA1c of –0.53% in patients with type 2 daily, insulin, liraglutide, pioglitazone) levels and body weight with all combina- diabetes (56). In all studies, dapagliflozin (43). Gastrointestinal side effects, such tions. The most common AEs were monotherapy significantly lowered HbA1c care.diabetesjournals.org Van Gaal and Scheen 1165

compared with placebo (P , 0.01). Da- As with monotherapy, SGLT2 inhibi- (–2.57 kg, P , 0.00001), although there pagliflozin therapy was associated tors combined with other antihypergly- was some heterogeneity in the weight- with a –1.63-kg reduction in body cemic agents (i.e., metformin, insulin, loss data across studies. Pramlintide was weight and had a favorable weight profile sulfonylurea, TZD) have been found to associated with a higher incidence of compared with placebo and metformin. reduce HbA1c (dapagliflozin [1–50 mg]: mild-to-moderate, mainly transient, nau- Dapagliflozin-induced body weight has –0.73%; canagliflozin [50–300 mg]: sea than control. Some studies have re- been shown to occur through reductions –0.97%) and body weight (overall: ported the incidence of hypoglycemia in fat mass, visceral fat, and subcutaneous –0.59 kg) (55,66). When used in combi- (mild to moderate) to be higher with fat (57,58). Although dapagliflozin mono- nation therapy, dapagliflozin and cana- pramlintide versus placebo, whereas therapy did not lead to hypoglycemia, the gliflozin were associated with an others have reported the converse (69). incidence of hypoglycemic events in- increased risk of genital tract infections, creasedwhendapagliflozin was com- whereas dapagliflozin was also associ- ANTIOBESITY bined with sulfonylureas and insulin ated with a modest increased risk of PHARMACOTHERAPIES (56). Dapagliflozin was also associated UTI. However, the number of hypogly- Although several antiobesity agents have with an increased risk of mild urinary cemic episodes experienced by patients been withdrawn from the market because and genital tract infections compared treated with SGLT2 inhibitors did not of safety concerns, five are now avail- with placebo. differ from placebo. able in the U.S.dorlistat, lorcaserin, Multiple studies have demonstrated Empagliflozin was approved in 2014 phentermine plus topiramate, naltrexone that canagliflozin (100 and 300 mg q.d.) by the FDA and EMA to improve glyce- plus bupropion (NB), and liraglutide (3.0 fi treatment resulted in signi cant HbA1c mic control in adults with type 2 diabe- mg) (Table 4) for chronic weight manage- fi reductions compared with placebo or ac- tes, and bene cial effects on HbA1c and mentdand one (orlistat) is currently tive comparator (55). After 26 weeks, can- weight have been observed with empa- available in Europe, with liraglutide (3.0 agliflozin (100 and 300 mg q.d.) reduced gliflozin as monotherapy or combination mg) and NB having recently received fa- – HbA1c levels from baseline by 0.77% and therapy (55). A meta-analysis of 10 RCTs vorable opinions from the Committee for – fl 1.03%, respectively (59). Canagli ozin found that mean changes in HbA1c were Medicinal Products for Human Use. These has demonstrated dose-related reduc- –0.62% for empagliflozin (10 mg) and pharmacotherapies have been demon- tions in baseline body weight: –2.2% –0.66% for empagliflozin (25 mg) com- strated to help people with type 2 diabe- and –3.3% after 26 weeks and –3.3% pared with placebo (67). The incidence tes achieve their weight-loss goals and – fl and 4.4% after 52 weeks for 100 and of hypoglycemia with empagli ozin provide them with an HbA1c-reducing 300 mg, respectively, in individuals with treatment was similar to placebo. Mean benefit(70–73). diabetes (59,60). Canagliflozin (50–300 weight change from baseline was –1.85 mg) has also been reported to have bene- and –1.84 kg, with 10- and 25-mg empa- Orlistat ficial weight effects in overweight or obese gliflozin doses, respectively, compared Orlistat is indicated for obesity manage- individuals without diabetes (61). Overall, with placebo. Although an increase in ment, including weight loss and weight the reported incidence of hypoglycemia the incidence of UTIs was not ob- maintenance, when used in conjunction after 26 weeks with canagliflozin (100 served, the risk of genital tract infec- with a reduced-calorie diet, and reduction and 300 mg) was low (;3%) and similar tion was increased with empagliflozin of the risk of weight regain after prior to the incidence reported with placebo versus placebo. weight loss (74). Orlistat functions as an (59), except in patients on background sul- Additional SGLT2 inhibitors, such as antiobesity agent by inhibiting gastroin- fonylurea (62). The incidence of mild gen- ipragliflozin (approved in Japan) and to- testinal lipases, thereby reducing absorp- ital mycotic infections and urinary tract fogliflozin, are in clinical development tion of dietary fat (75). In a 4-year study of infections (UTIs) was higher with canagli- (55). Initial study data have also shown obese patients without diabetes, orlistat flozin treatment than with placebo (59). reductions in HbA1c levels and in body (120 mg t.i.d.) plus lifestyle changes pro- In addition to their effect of reducing weight in people with diabetes. duced moderate weight loss (–5.8 kg HbA1c and body weight, dapagliflozin and from baseline vs. –3.0 kg from baseline canagliflozin showed beneficial effects on Pramlintide with lifestyle changes alone) and resulted blood pressure in individuals with type 2 Pramlintide acetate is a synthetic analog in a greater reduction in the incidence of diabetes (55). These effects may be due to of human amylin that has been shown to type 2 diabetes over lifestyle changes increased glucose and sodium excretion in reduce HbA1c and body weight in patients alone (6.2% vs. 9.0%, a –37.3% reduction; the urine with SGLT2 inhibitors (23). Al- with diabetes (68). It is indicated for the P = 0.0032) (76). though increased glucose excretion con- management of type 2 diabetes in the Orlistat also provided weight-loss ben- tributes to the weight loss observed with U.S. but is not available in Europe. A meta- efits for patients with diabetes. After 52 SGLT2 inhibitor treatment, weight reduc- analysis showed that in patients with type weeks of orlistat treatment (120 mg t.i.d.) tion is often limited to ,4 kg after even 2 diabetes, pramlintide is associated combined with a reduced-calorie diet 52 weeks of treatment (55). This attenua- with a small but significant reduction in and a weight-management program, tion of weight loss may occur because HbA1c (–0.33%, P = 0.0004) that is consis- obese patients with type 2 diabetes SGLT2 inhibitor–induced glycosuria is ac- tent over time (–0.3 to –0.42%; weeks achieved –5.0% reduction in weight companied by compensatory hyperphagia, 12–52) (69). Pramlintide was associated from baseline versus –1.8% with placebo as demonstrated in animal studies (63) and with a significant reduction in weight (P , 0.0001) and –1.1% HbA1c reduction suggested by human studies (64,65). from baseline compared with control versus –0.2% with placebo (P , 0.0001) 1166 Weight Management in Type 2 Diabetes Diabetes Care Volume 38, June 2015

Table 4—Antiobesity therapies currently approved for chronic weight management Drug Mechanism of action How mechanism of action leads to weight loss

Orlistat (75–77) c Inhibits gastrointestinal and pancreatic c Prevents absorption of dietary fat lipases Lorcaserin (71) c Selectively stimulates 5HT2C c Promotes feelings of satiety and regulates appetite Phentermine plus topiramate (80,84) c Phentermine acts on hypothalamus to c Promotes feelings of satiety and regulates stimulate norepinephrine release from appetite adrenal glands c The precise mechanisms by which c Topiramate acts on multiple cellular phentermine plus topiramate produce targets as an antiepileptic agent weight loss is unknown Naltrexone sustained release plus bupropion c Increases levels of dopamine and POMC c Suppresses appetite sustained release (73,86) neuronal activity c Increases secretion of melanocortins, which c Blocks opioid receptors on POMC mediate anorectic effects and regulate neurons, preventing feedback inhibition energy balance of these neurons and further increasing POMC activity Liraglutide (3.0 mg) (98) c Binds and activates the human c Weight loss due to inhibition of gastric GLP-1R emptying c Enhances glucose-dependent insulin c Decreases calorie ingestion through central secretion by the pancreatic b-cell nervous system c Increases intracellular cAMP leading to c Reduces acid secretion insulin release in the presence of elevated glucose concentrations c Increases satiety

(70). Retrospective analysis of seven as hypertension, dyslipidemia, or type 2 phentermine plus topiramate and NB, studies of orlistat (120 mg t.i.d.) con- diabetes (78). In overweight or obese despite a smaller amount of weight firmed that orlistat-treated patients patients without diabetes, lorcaserin loss (71,73,80). This suggests that the had significantly greater decreases in treatment provided a 25.81% reduction antihyperglycemic effect of lorcaserin body weight than the placebo group from baseline body weight after 1 year may be due to more than weight loss (–3.77 vs. –1.42 kg, P , 0.0001) and versus 22.16% with placebo (P , 0.001) alone. Although hypoglycemia was larger mean decreases in HbA1c than pla- (79). Lorcaserin is a selective small-molecule slightly more frequent in the lorcaserin cebo (–0.74% vs. –0.31%, P , 0.0001) agonist of the 5-hydroxytryptamine 2C treatment groups than in the placebo (77). For patients with minimal weight serotonin receptor (5HT2C), which regu- group, no severe hypoglycemia was loss (,1% of baseline body weight), orli- lates mechanisms related to satiety, in- reported (71). No evidence of in- stat still provided a significantly greater gestive behavior, glucose tolerance, and creased depression, suicidal thoughts, decrease in HbA1c than placebo (–0.29% hepatic insulin sensitivity (71). Unlike or echocardiogram-detected valvular re- vs. –0.14%, P = 0.008). Potential mecha- previously available antiobesity agents, gurgitations were found in the lorcaserin nisms to explain the better glycemic con- lorcaserin has a low affinity for the treatment arms. Overall, the most com- trol independent of weight loss may be 5HT2B receptor subtype, whose activa- mon AEs with lorcaserin were headache, the improvement of insulin sensitivity, tion has been linked to the development back pain, nasopharyngitis, and nausea. slower/incomplete digestion of dietary of valvular heart disease (78). fat, reduction of postprandial plasma In a phase 3 study of subjects with Phentermine Plus Topiramate nonesterified fatty acids, decreased vis- type 2 diabetes treated with metformin Phentermine is a norepinephrine- and ceral adipose tissue, and stimulation of or a sulfonylurea, lorcaserin treatment dopamine-releasing agent (with a lower GLP-1 secretion. Orlistat was generally resulted in mean weight changes of effect with dopamine vs. norepineph- well tolerated, and gastrointestinal ef- 24.5% (twice daily) and 25.0% (once rine) approved for the short-term treat- fects were the most commonly reported daily) compared with 21.5% with pla- ment of obesity (80). Topiramate has AEs, but all events were considered mild cebo at week 52 (71). Participants also several pharmacological mechanisms or moderate (75,76). showed a significant improvement in of action and has been assessed as a sin- glycemic control: HbA1c decreased gle agent for weight reduction in obese Lorcaserin 20.9% and 21.0% from baseline with patients with and without type 2 diabe- Lorcaserin is indicated as an adjunct to a lorcaserin twice daily and once daily, re- tes and hypertension (81–83). reduced-calorie diet and increased spectively, versus 20.4% with placebo A phase 3 study examined the efficacy physical activity for chronic weight (P , 0.001 for each lorcaserin dose). It of the combination of phentermine 7.5 management in adults with an initial is interesting to note that the reductions mg/topiramate 46.0 mg (PHEN 7.5/TPM 2 2 BMI of $30 kg/m (obese) or $27 kg/m in HbA1c observed with lorcaserin are 46.0) or phentermine 15.0 mg/topiramate (overweight) in the presence of at least one equal to or higher than those observed 92.0 mg (PHEN 15.0/TPM 92.0) on weight-related comorbid condition such with other antiobesity agents, such as weight loss after 56 weeks (80). Patients care.diabetesjournals.org Van Gaal and Scheen 1167

with type 2 diabetes (a subgroup of 388 vomiting but was not associated with in- newer therapies are also overviewed subjects in this study [N = 2,487]) creased depression, suicidal thoughts, or below. The potential of these therapies achieved weight reductions of 26.8% hypoglycemia. NB may represent a novel in patients with type 2 diabetes, as well with PHEN 7.5/TPM 46.0 and 28.8% pharmacological approach for the treat- as their cardiovascular safety, will need with PHEN 15.0/TPM 92.0 versus ment of obesity, but further studies are to be established. 21.9% with placebo (80). For patients required to assess its effects on cardio- with diabetes, significantly greater re- vascular outcomes, because systolic FUTURE PROSPECTS IN CLINICAL DEVELOPMENT ductions in HbA1c (–0.4 mmol/L) were blood pressure and pulse rate have seen with both doses of PHEN/TPM been found to be higher with NB than As our knowledge of the physiology of than with placebo (–0.1 mmol/L). In the with placebo (78). appetite and energy homeostasis im- 108-week extension study, both doses of proves, so too will our ability to under- PHEN/TPM were associated with signifi- Liraglutide stand how therapies might be combined cant and sustained weight loss (–9.0%, Liraglutide (3.0 mg q.d.) has been to provide effective weight manage- P , 0.0001; placebo, –2.0%) (84). PHEN shown to provide weight-reduction ment in patients with type 2 diabetes, 7.5/TPM 46.0 and PHEN 15.0/TPM 92.0 benefits for obese patients; after 20 while also minimizing AEs. Therapies led to reductions in HbA1c of –0.4% and weeks, the placebo-subtracted reduc- that are currently under clinical investi- –0.2%, respectively, in contrast to the tion in weight from baseline with liraglu- gation are included in Table 5. placebo group (0%) (84). Furthermore, tide (3.0 mg) treatment was –4.4 kg (P = in patients without diabetes, the two 0.003) (87). A further study showed that GLP-1R Agonists doses of PHEN/TPM led to 54% and after 1 year, subjects who received lira- Similar to liraglutide (3.0 mg), exena- 76% reductions, respectively, in the pro- glutide (3.0 mg) lost –5.8 kg more than tide (10 mg b.i.d.) has been shown to fi gression of subjects to type 2 diabetes, the placebo group, and after 2 years, provide weight-reduction bene ts in compared with placebo (84). pooled participants who completed the obese people with or without prediabe- Phentermine plus topiramate was study on liraglutide (2.4/3.0 mg) tes (92). After 24 weeks, the placebo- well tolerated; constipation, paresthe- maintained a weight loss of –7.8 kg subtracted difference in percentage – , sia, and dry mouth were the most com- (88). The weight reductions observed weight reduction was 3.3% (P monly reported treatment-emergent with liraglutide (3.0 mg q.d.) primarily 0.001); exenatide-treated subjects lost – – AEs (84). However, the FDA has re- result from reductions in fat mass and 5.1 kg from baseline versus 1.6 kg quired a Risk Evaluation and Mitigation body fat percentage (including visceral with placebo (92). GLP-1R agonists for Strategy for phentermine plus topira- fat) rather than in lean tissue mass oral delivery are also currently under mate to educate prescribers and pa- (88,89). Similar to liraglutide (1.8 mg) investigation in preclinical and clinical tients on the increased risk of orofacial treatment in patients with diabetes, studies (93). clefts in infants exposed to phenter- the most common AEs with liraglutide mine plus topiramate during the first (3.0 mg) treatment in obese patients GLP-1R Agonist Combination trimester of pregnancy (85). were gastrointestinal and consistent Therapies with the known physiological effects Because GLP-1R agonists and basal insu- Naltrexone Sustained Release Plus of GLP-1R agonists (88). Liraglutide lins offer complementary pharmaco- Bupropion Sustained Release (3.0 mg) was approved by the FDA in logic effects on prandial and fasting Naltrexone is an opioid receptor antag- December 2014 for chronic weight glycemia (94), there is growing clinical onist, whereas bupropion is a norepineph- management in addition to a reduced- interest in combinations of these two rine and dopamine reuptake inhibitor calorie diet and physical activity and is agents. The combination of exenatide (86). The combination increases pro- now undergoing EMA regulatory review (10 mg b.i.d.) with insulin glargine (ap- opiomelanocortin (POMC) neuronal firing, for the treatment of obesity. proved in the U.S. and Europe) led to which may have anorectic effects. The greater reductions in HbA1c levels, com- combination provides greater weight Safety Concerns pared with insulin glargine alone loss than monotherapy with either agent Recent safety concerns about an in- (–1.74% vs. –1.04%). Treatment with ex- or placebo (86), which is an effect con- creased risk of major cardiac AEs have enatide and insulin glargine led to a firmed in overweight/obese patients led to market withdrawal of existing an- weight decrease of –1.8 kg, whereas with type 2 diabetes, hypertension, or tiobesity medications or a lack of new insulin glargine alone led to a weight hyperlipidemia (23). In patients with treatments being approved (90). Assess- increase of 1.0 kg. The number of hypo- type 2 diabetes, NB therapy significantly ment of cardiovascular safety has now glycemic events between groups did decreased HbA1c from baseline (–0.6% emerged as a major consideration for all not differ significantly. vs. –0.1% with placebo, P , 0.001) new antiobesity and glucose-lowering Liraglutide with insulin degludec (73). NB-treated patients experienced agents under current review by the (IDegLira)dnow approved in Europed significantly greater weight reduction FDA (91). Given the significant need for is another combination currently being from baseline than patients in the pla- effective and safe weight-loss medica- investigated for the treatment of type cebo group (–5.0% vs. –1.8%, P , 0.001). tion, it is perhaps not surprising that 2 diabetes. Initial clinical data show Compared with placebo, treatment with many more antiobesity therapies are in that IDegLira led to greater reductions NB was associated with a higher development, as detailed in the the re- in HbA1c (–1.9%) versus insulin deglu- incidence of nausea, constipation, and cent article by Rodgers et al. (74); these dec (–1.4%) or liraglutide (–1.3%) alone 1168 Weight Management in Type 2 Diabetes Diabetes Care Volume 38, June 2015

Table 5—Future prospects combination treatment with pramlintide/ fi Mechanism of action and/or potential for metreleptin led to a signi cant earlier, sus- Drug class/combination weight loss tained, and greater weight loss than treatment with pramlintide or metreleptin Long-acting basal insulin/GLP-1 c Acts on receptors for GLP-1 alone (100). However, a subsequent trial analog (94,95) c GLP-1 action suppresses appetite, compensating for a potential insulin-induced weight increase was recently halted due to safety con- Pramlintide/metreleptin (100) c Leptin has a pivotal role in energy metabolism by cerns (101). inhibiting food intake and increasing energy Polyethylene glycolated (PEG)-leptin, expenditure along with PEG-GLP-1/glucagon, may be c Amylin analogs slow gastric emptying another potential combination therapy PEG–leptin/GLP-1/glucagon (102) c Leptin has a pivotal role in energy metabolism by option (102). This combination is an in- inhibiting food intake and increasing energy triguing potential antihyperglycemic op- expenditure tion, because preclinical data indicate c GLP-1/glucagon coagonism restores leptin that PEG-leptin and PEG–GLP-1/glucagon responsiveness c Improves glucose and lipid metabolism coagonism can restore leptin responsiveness, which is often reduced when leptin Unimolecular dual-incretin c Acts on receptors for both GLP-1 and GIP agonist (103) c Lowers postprandial glucose through pancreatic is used alone. Responsiveness to leptin is b-cell insulin secretion associated with decreased food intake, c GLP-1 action suppresses appetite improved glucose tolerance and insulin c Increases satiety sensitivity, and with decreased trigly- c Decreases food intake cerides and lower plasma cholesterol c Decreases fat mass concentrations. These may be the con- Unimolecular triple-incretin c Acts on receptors for GLP-1, GIP, and glucagon tributing factors that lead to the weight agonist (104) c Lowers postprandial glucose through pancreatic b-cell insulin secretion loss observed with leptin/GLP-1/glucagon c GLP-1 action suppresses appetite coagonism. These results suggest that c Decreases food intake the pharmacology of leptin in combina- c Decreases fat mass tion with other agents, such as GLP-1R c Increases energy expenditure agonists and amylin analogs, warrants additional study as a potential antihyperglycemic therapy that is associated with (95). IDegLira also provided a modest FUTURE PROSPECTS IN weight loss. weight loss of –0.5 kg from baseline PRECLINICAL DEVELOPMENT Another potential therapy is the com- to week 26, a –2.2-kg reduction, com- Given the role of leptin and amylin in bination of amylin analogs and GLP-1R pared with insulin degludec. IDegLira controlling food intake and energy ex- agonists. Because both agents can slow also resulted in significantly fewer penditure and the role of incretins gastric emptying, it is possible that these hypoglycemic episodes than insulin (GLP-1)inglucoseandweightcontrol two agents combined may have synergis- degludec. (97,98), that many of the therapies in tic effects, but the gastrointestinal toler- A combination of insulin glargine preclinical development involve these ance should be evaluated. with lixisenatide has also been investi- different hormones is no surprise. Ther- gated (96). The addition of lixisenatide apies that are currently being studied are Unimolecular Dual- or Triple-Incretin to insulin glargine produced greater re- included in Table 5. Receptor Agonists ductions in HbA1c (–0.32%; P , 0.0001) Another incretin pathway compound in and postprandial hyperglycemia (differ- Peptide Hormone Combination early-stage development is a peptide ence vs. placebo, –3.2 mmol/L; P , Therapies that acts as an agonist at both the GLP-1 0.0001) compared with insulin glargine Because results with recombinant hu- and GIP receptors (103). A preclinical study alone. The addition of lixisenatide also man leptin or metreleptin (human leptin indicates that this dual agonist has the po- had a favorable effect on body weight analog) have been disappointing in re- tential to enhance the antihyperglycemic (difference vs. placebo –0.89 kg; ducing HbA1c levels and weight for and antiobesity effects observed with P = 0.0012). Nausea, vomiting, and obese patients with type 2 diabetes monoagonism because it affects adiposity- symptomatic hypoglycemia were more (97), approaches are now focused on induced insulin resistance and pancre- commonly reported with lixisenatide leptin-related synthetic peptides, such atic insulin deficiency. A recent study in than with insulin glargine alone. as leptin receptor antagonists or leptin- rodents found that a new monomeric pep- Another potential future option is a related synthetic peptide analogs or tide triagonist, simultaneously acting at SGLT2 inhibitor/GLP-1R agonist combi- mimetics, and leptin combination three key metabolically related peptide nation (55). The effect of GLP-1R ago- therapies (99). Initial preclinical and hormone receptors (GLP-1, GIP, glucagon), nists on satiety may weaken the clinical data suggest that leptin and provided additional glucose control and compensatory “overeating” mecha- amylindtwo hormones involved in the weight-reducing benefits over dual coa- nism observed with SGLT2 inhibition control of satietydhave additive effects gonism (104). Extensive clinical investiga- and enhance weight loss by SGLT2 (99). A proof-of-concept RCT in over- tion into the efficacy and safety of inhibitors (63). weight/obese subjects showed that coagonist therapy for the treatment of care.diabetesjournals.org Van Gaal and Scheen 1169

patients with obesity and type 2 diabetes is diabetes and consider weight-neutral 2. Van Gaal LF, Mertens IL, De Block CE. Mech- now required. or weight-reducing medications that anisms linking obesity with cardiovascular dis- – can complement the patient’sdesire ease. Nature 2006;444:875 880 3. Obesity and overweight. Fact sheet No. 311 POTENTIAL THERAPEUTIC for a healthier lifestyle. [Internet]. Geneva, Switzerland: World Health TARGETS For patients struggling to achieve or Organization International. Available from Owing to the complex pathophysiology maintain their weight-management http://www.who.int/mediacentre/factsheets/ of diabetes, additional therapeutic tar- objectives, concomitant antiobesity fs311/en/index.html. Accessed 2 Jan 2014 4. Guh DP, Zhang W, Bansback N, Amarsi Z, gets are under investigation as potential medications can be considered, with ’ Birmingham CL, Anis AH. The incidence of co- agents for glycemic control, many in the aim of reducing patients body morbidities related to obesity and overweight: combination with GLP-1R agonists (105). weight and glycemic targets. Recent ap- a systematic review and meta-analysis. BMC These possible agentsdsuch as GLP-1R ag- provals of therapies that provide both Public Health 2009;9:88 onist/PYY, fibroblast growth factor 21 with glycemic control and weight reduction, 5. Inzucchi SE, Bergenstal RM, Buse JB, et al. and the healthy pipeline of antiobesity Management of hyperglycemia in type 2 diabe- or without GLP-1R agonist, and GLP-1R/ tes, 2015: a patient-centered approach: update glucagon coagonistsdmay offer the po- medications, bode well for a wider to a position statement of the American Diabe- tential to normalize glucose levels but are choice in the future, with some agents tes Association and the European Association still in early development (105,106). PYY is targeting the central nervous system to for the Study of Diabetes. Diabetes Care 2015; – an incretin hormone that also has a role in reduce food intake and others targeting 38:140 149 the hormonal pathways involved in 6. Knowler WC, Barrett-Connor E, Fowler SE, satiety (106). The associated hypothesis et al.; Diabetes Prevention Program Research is that PYY may further enhance the glu- weight regulation and glucose homeo- Group. Reduction in the incidence of type 2 di- cose-lowering and weight-reducing ef- stasis. The emergence of a range of phar- abetes with lifestyle intervention or metformin. fects of GLP-1R agonists. Fibroblast macotherapies with varying modes of N Engl J Med 2002;346:393–403 growth factor 21 has broad metabolic action, coupled with ongoing improve- 7. Wing RR; Look AHEAD Research Group. Long- term effects of a lifestyle intervention on weight ments in our knowledge of the physiology effects, including enhancing insulin sen- and cardiovascular risk factors in individuals sitivity, decreasing triglyceride concen- of appetite and energy homeostasis, with type 2 diabetes mellitus: four-year results trations, and inducing weight loss, and provides the prospect of a rational com- of the Look AHEAD trial. Arch Intern Med 2010; this activity acts additively with GLP-1 bination therapy that is both effective 170:1566–1575 (107,108). By combining two peptides and tolerable. 8. Wing RR, Lang W, Wadden TA, et al.; Look AHEAD Research Group. Benefits of modest with different effects, GLP-1R/glucagon weight loss in improving cardiovascular risk coagonism may normalize adiposity and factors in overweight and obese individuals glucose tolerance through fat loss, de- Acknowledgments. The authors are grateful with type 2 diabetes. Diabetes Care 2011;34: creased food intake, and increased en- to Dr. Jennifer Chang of AXON Communica- 1481–1486 tions for writing assistance in the development ergy expenditure, while minimizing 9. Rubin RR, Wadden TA, Bahnson JL, et al.; of the manuscript. Look AHEAD Research Group. Impact of inten- hypoglycemic risk (109). Funding. L.V.G. received grant support from sive lifestyle intervention on depression and Another agent under clinical in- National Research Funds, Belgium, and also health-related quality of life in type 2 diabetes: vestigation as an antiobesity agent is received grant support for hepatic research the Look AHEAD Trial. Diabetes Care 2014;37: beloranib, a fumagillin-class methionine from the European Union consortium (Hepadip 1544–1553 and Resolve consortia). aminopeptidase-2 inhibitor that has re- 10. Foster GD, Borradaile KE, Sanders MH, Duality of Interest. Writing assistance for the et al.; Sleep AHEAD Research Group of Look cently completed phase 2 trials (110). manuscript was funded by Novo Nordisk. Novo AHEAD Research Group. A randomized study Because beloranib treatment is associ- Nordisk was also provided with the opportunity on the effect of weight loss on obstructive sleep ated with rapid weight loss and improve- to perform a medical accuracy review. L.V.G. is, apnea among obese patients with type 2 diabe- or has been, a member of the advisory boards ments in lipids (110), beloranib could tes: the Sleep AHEAD study. Arch Intern Med and speakers bureaus of AstraZeneca/Bristol- – likely also have a beneficial effect in the 2009;169:1619 1626 Myers Squibb, Boehringer Ingelheim, Eli Lilly, 11. Look AHEAD Research Group. Eight-year treatment of overweight/obese patients Janssen, Merck Sharp & Dohme, Novartis, Novo weight losses with an intensive lifestyle inter- fi – with type 2 diabetes. Nordisk, and Sano (in the period 2010 vention: the Look AHEAD study. Obesity (Silver Further research with all of these tar- 2013). A.S. has received lecture or adviser Spring) 2014;22:5–13 fees from AstraZeneca/Bristol-Myers Squibb, gets is required to determine their suit- 12. Pi-Sunyer FX. Weight loss in type 2 diabetic Boehringer Ingelheim, Eli Lilly, Janssen, Merck patients. Diabetes Care 2005;28:1526–1527 ability as antihyperglycemic agents. fi Sharp & Dohme, Novartis, Novo Nordisk, Sano , 13. UK Prospective Diabetes Study (UKPDS) – and Takeda (in the period 2010 2013). A.S. also Group. Intensive blood-glucose control with sul- CONCLUSIONS received an unrestricted research grant from phonylureas or insulin compared with conven- Although lifestyle interventions aimed Novo Nordisk and Novartis. No other potential tional treatment and risk of complications in fl at prompting weight loss are important con icts of interest relevant to this article were patients with type 2 diabetes (UKPDS 33). Lan- reported. in the management of type 2 diabetes cet 1998;352:837–853 Author Contributions. L.V.G. and A.S. conceived fi fi 14. Fonseca V, McDuf e R, Calles J, et al.; and the bene ts of weight reduction are and designed the manuscript, analyzed and inter- ACCORD Study Group. Determinants of weight irrefutable, most patients remain over- preted the data, drafted and revised the paper, gain in the action to control cardiovascular risk fi weight or obese. A shift in the approach and approved the nal version for publication. in diabetes trial. Diabetes Care 2013;36:2162– to weight management in people with 2168 type 2 diabetes is clearly needed. Health References 15. Cabrera J, Emir B, Dills D, Murphy TK, 1. Van Gaal LF, De Block CE. Bariatric surgery to Whalen E, Clair A. Characterizing and under- care practitioners should consider the treat type 2 diabetes: what is the recent evi- standing body weight patterns in patients weight effects of pharmacotherapy in dence? Curr Opin Endocrinol Diabetes Obes treated with pregabalin. Curr Med Res Opin the management of patients with 2012;19:352–358 2012;28:1027–1037 1170 Weight Management in Type 2 Diabetes Diabetes Care Volume 38, June 2015

16. Scheen AJ, De Hert MA. Abnormal glucose 31. Makimattila¨ S, NikkilaK,Yki-J¨ arvinen¨ H. 46. De Block CEM, Van Gaal LF. Efficacy and metabolism in patients treated with antipsy- Causes of weight gain during insulin therapy safety of exenatide once weekly: an overview chotics. Diabetes Metab 2007;33:169–175 with and without metformin in patients with of the DURATION trials. Expert Rev Endocrinol 17. Toft UN, Kristoffersen LH, Aadahl M, type II diabetes mellitus. Diabetologia 1999; Metab 2012;7:611–623 vonHuthSmithL,PisingerC,JørgensenT. 42:406–412 47. Niswender K, Pi-Sunyer X, Buse J, et al. Diet and exercise intervention in a general 32. Hermansen K, Mortensen LS. Bodyweight Weight change with liraglutide and comparator populationdmediators of participation and changes associated with antihyperglycaemic therapies: an analysis of seven phase 3 trials adherence: the Inter99 study. Eur J Public Health agents in type 2 diabetes mellitus. Drug Saf from the liraglutide diabetes development pro- 2007;17:455–463 2007;30:1127–1142 gramme. Diabetes Obes Metab 2013;15:42–54 18. The DCCT Research Group. Weight gain as- 33. Yki-Jarvinen¨ H. Thiazolidinediones. N Engl J 48. Buse JB, Rosenstock J, Sesti G, et al.; LEAD-6 sociated with intensive therapy in the Diabetes Med 2004;351:1106–1118 Study Group. Liraglutide once a day versus Control and Complications Trial. Diabetes Care 34. Wilding J. Thiazolidinediones, insulin resis- exenatide twice a day for type 2 diabetes: 1988;11:567–573 tance and obesity: finding a balance. Int J Clin a 26-week randomised, parallel-group, multina- 19. Maclean PS, Bergouignan A, Cornier MA, Pract 2006;60:1272–1280 tional, open-label trial (LEAD-6). Lancet 2009; Jackman MR. Biology’s response to dieting: 35. Miyazaki Y, Mahankali A, Matsuda M, et al. 374:39–47 the impetus for weight regain. Am J Physiol Effect of pioglitazone on abdominal fat distribu- 49. Fadini GP, Simioni N, Frison V, et al. Inde- Regul Integr Comp Physiol 2011;301:R581– tion and insulin sensitivity in type 2 diabetic pendent glucose and weight-reducing effects of R600 patients. J Clin Endocrinol Metab 2002;87: liraglutide in a real-world population of type 2 20. Sumithran P, Prendergast LA, Delbridge E, 2784–2791 diabetic outpatients. Acta Diabetol 2013;50: et al. Long-term persistence of hormonal adap- 36. Goke¨ R, Wagner B, Fehmann HC, Goke¨ B. 943–949 tations to weight loss. N Engl J Med 2011;365: Glucose-dependency of the insulin stimulatory 50. Gallwitz B, Vaag A, Falahati A, Madsbad S. – 1597 1604 effect of glucagon-like peptide-1 (7-36) amide Adding liraglutide to oral antidiabetic drug ther- 21. de Luis DA, Gonzalez Sagrado M, Conde R, on the rat pancreas. Res Exp Med (Berl) 1993; apy: onset of treatment effects over time. Int J – Aller R, Izaola O. Decreased basal levels of 193:97–103 Clin Pract 2010;64:267 276 glucagon-like peptide-1 after weight loss in 37. Tornehave D, Kristensen P, Rømer J, 51. Blonde L, Russell-Jones D. The safety and fi obese subjects. Ann Nutr Metab 2007;51: Knudsen LB, Heller RS. Expression of the GLP-1 ef cacy of liraglutide with or without oral anti- – 134 138 receptor in mouse, rat, and human pancreas. diabetic drug therapy in type 2 diabetes: an 22. Rosenbaum M, Sy M, Pavlovich K, Leibel RL, J Histochem Cytochem 2008;56:841–851 overview of the LEAD 1-5 studies. Diabetes – Hirsch J. Leptin reverses weight loss-induced 38. Zander M, Madsbad S, Madsen JL, Holst JJ. Obes Metab 2009;11(Suppl. 3):26 34 changes in regional neural activity responses Effect of 6-week course of glucagon-like pep- 52. Pencek R, Blickensderfer A, Li Y, Brunell SC, to visual food stimuli. J Clin Invest 2008;118: tide 1 on glycaemic control, insulin sensitivity, and ChenS.Exenatideonceweeklyforthetreat- 2583–2591 ment of type 2 diabetes: effectiveness and tol- beta-cell function in type 2 diabetes: a parallel- 23. Scheen AJ, Van Gaal LF. Combating the dual erability in patient subpopulations. Int J Clin group study. Lancet 2002;359:824–830 burden: therapeutic targeting of common path- Pract 2012;66:1021–1032 39. Willms B, Werner J, Holst JJ, Orskov C, ways in obesity and type 2 diabetes. Lancet Di- 53. Samson SL, Garber A. GLP-1R agonist ther- Creutzfeldt W, Nauck MA. Gastric emptying, abetes Endocrinol 2014;2:911–922 apy for diabetes: benefits and potential risks. glucose responses, and insulin secretion 24. van Dieren S, Czernichow S, Chalmers J, Curr Opin Endocrinol Diabetes Obes 2013;20: after a liquid test meal: effects of exogenous et al. Weight changes and their predictors 87–97 glucagon-like peptide-1 (GLP-1)-(7-36) amide amongst 11 140 patients with type 2 diabetes 54. Egan AG, Blind E, Dunder K, et al. Pancre- in type 2 (noninsulin-dependent) diabetic in the ADVANCE trial. Diabetes Obes Metab atic safety of incretin-based drugsdFDA and patients. J Clin Endocrinol Metab 1996;81: 2012;14:464–469 EMA assessment. N Engl J Med 2014;370:794– 327–332 25. Russell-Jones D, Khan R. Insulin-associated 797 40. Toft-Nielsen MB, Madsbad S, Holst JJ. Con- weight gain in diabetesdcauses, effects and 55. Scheen AJ. Pharmacodynamics, efficacy coping strategies. Diabetes Obes Metab 2007; tinuous subcutaneous infusion of glucagon-like and safety of sodium-glucose co-transporter 9:799–812 peptide 1 lowers plasma glucose and reduces type 2 (SGLT2) inhibitors for the treatment 26. Liebl A, Hoogma R, Renard E, et al.; Euro- appetite in type 2 diabetic patients. Diabetes of type 2 diabetes mellitus. Drugs 2015;75: – pean DiaPort Study Group. A reduction in severe Care 1999;22:1137 1143 33–59 hypoglycaemia in type 1 diabetes in a random- 41. Kolterman OG, Kim DD, Shen L, et al. Pharma- 56. Zhang M, Zhang L, Wu B, Song H, An Z, Li S. ized crossover study of continuous intraperito- cokinetics, pharmacodynamics, and safety of exe- Dapagliflozin treatment for type 2 diabetes: neal compared with subcutaneous insulin natide in patients with type 2 diabetes mellitus. a systematic review and meta-analysis of ran- – infusion. Diabetes Obes Metab 2009;11:1001– Am J Health Syst Pharm 2005;62:173 181 domized controlled trials. Diabetes Metab Res 1008 42. Tatarkiewicz K, Sablan EJ, Polizzi CJ, Rev 2014;30:204–221 27. Woods SC, Lotter EC, McKay LD, Porte D Jr. Villescaz C, Parkes DG. Long-term metabolic 57. Bolinder J, Ljunggren O,¨ Kullberg J, et al. Chronic intracerebroventricular infusion of in- benefits of exenatide in mice are mediated Effects of dapagliflozin on body weight, total sulin reduces food intake and body weight of solely via the known glucagon-like peptide 1 re- fat mass, and regional adipose tissue distribu- baboons. Nature 1979;282:503–505 ceptor. Am J Physiol Regul Integr Comp Physiol tion in patients with type 2 diabetes mellitus 28. Cai XJ, Evans ML, Lister CA, et al. Hypoglyce- 2014;306:R490–R498 with inadequate glycemic control on metfor- mia activates orexin neurons and selectively in- 43. Esposito K, Mosca C, Brancario C, Chiodini min. J Clin Endocrinol Metab 2012;97:1020– creases hypothalamic orexin-B levels: responses P, Ceriello A, Giugliano D. GLP-1 receptor ago- 1031 inhibited by feeding and possibly mediated by the nists and HBA1c target of ,7% in type 2 diabe- 58. Bolinder J, Ljunggren O,¨ Johansson L, et al. nucleus of the solitary tract. Diabetes 2001;50: tes: meta-analysis of randomized controlled Dapagliflozin maintains glycaemic control while 105–112 trials. Curr Med Res Opin 2011;27:1519–1528 reducing weight and body fat mass over 2 years 29. Yki-Jarvinen¨ H, Ryysy L, Kauppila M, et al. 44. Dushay J, Gao C, Gopalakrishnan GS, et al. in patients with type 2 diabetes mellitus inade- Effect of obesity on the response to insulin Short-term exenatide treatment leads to sig- quately controlled on metformin. Diabetes therapy in noninsulin-dependent diabetes nificant weight loss in a subset of obese Obes Metab 2014;16:159–169 mellitus. J Clin Endocrinol Metab 1997;82: women without diabetes. Diabetes Care 59. Stenlof¨ K, Cefalu WT, Kim KA, et al. Efficacy 4037–4043 2012;35:4–11 and safety of canagliflozin monotherapy in sub- 30. Eriksson H, Ridderstrale˚ M, Degerman E, 45. Diamant M, Van Gaal L, Guerci B, et al. Ex- jects with type 2 diabetes mellitus inadequately et al. Evidence for the key role of the adipocyte enatide once weekly versus insulin glargine for controlled with diet and exercise. Diabetes cGMP-inhibited cAMP phosphodiesterase in the type 2 diabetes (DURATION-3): 3-year results of Obes Metab 2013;15:372–382 antilipolytic action of insulin. Biochim Biophys an open-label randomised trial. Lancet Diabetes 60. Stenlof¨ K, Cefalu WT, Kim KA, et al. Long- Acta 1995;1266:101–107 Endocrinol 2014;2:464–473 term efficacy and safety of canagliflozin care.diabetesjournals.org Van Gaal and Scheen 1171

monotherapy in patients with type 2 diabetes 74. Rodgers RJ, Tschop¨ MH, Wilding JP. Anti- combination therapy for the treatment of obe- inadequately controlled with diet and exercise: obesity drugs: past, present and future. Dis sity: ‘a new kid on the block’? Ann Med 2011;43: findings from the 52-week CANTATA-M study. Model Mech 2012;5:621–626 249–258 Curr Med Res Opin 2014;30:163–175 75. Derosa G, Cicero AF, D’Angelo A, Fogari E, 87. Astrup A, Rossner¨ S, Van Gaal L, et al.; 61. Bays HE, Weinstein R, Law G, Canovatchel Maffioli P. Effects of 1-year orlistat treatment NN8022-1807 Study Group. Effects of liraglutide W. Canagliflozin: effects in overweight and compared to placebo on insulin resistance pain the treatment of obesity: a randomised, double- obese subjects without diabetes mellitus. Obe- rameters in patients with type 2 diabetes. J Clin blind, placebo-controlled study. Lancet 2009;374: sity (Silver Spring) 2014;22:1042–1049 Pharm Ther 2012;37:187–195 1606–1616 62. Usiskin K, Kline I, Fung A, Mayer C, 76. Torgerson JS, Hauptman J, Boldrin MN, 88. Astrup A, Carraro R, Finer N, et al.; NN8022- Meininger G. Safety and tolerability of canagli- Sjostr¨ om¨ L. XENical in the prevention of Diabe- 1807 Investigators. Safety, tolerability and sus- flozin in patients with type 2 diabetes mellitus: tes in Obese Subjects (XENDOS) study: a ran- tained weight loss over 2 years with the once- pooled analysis of phase 3 study results. Post- domized study of orlistat as an adjunct to daily human GLP-1 analog, liraglutide. Int J Obes – grad Med 2014;126:16–34 lifestyle changes for the prevention of type 2 (Lond) 2012;36:843 854 63. Devenny JJ, Godonis HE, Harvey SJ, Rooney diabetes in obese patients. Diabetes Care 89. Jendle J, Nauck MA, Matthews DR, et al.; – S, Cullen MJ, Pelleymounter MA. Weight loss 2004;27:155 161 LEAD-2 and LEAD-3 Study Groups. Weight loss 77. Jacob S, Rabbia M, Meier MK, Hauptman J. with liraglutide, a once-daily human glucagon- induced by chronic dapagliflozin treatment is Orlistat 120 mg improves glycaemic control in like peptide-1 analogue for type 2 diabetes attenuated by compensatory hyperphagia in type 2 diabetic patients with or without concur- treatment as monotherapy or added to metfor- diet-induced obese (DIO) rats. Obesity (Silver rent weight loss. Diabetes Obes Metab 2009;11: min, is primarily as a result of a reduction in fat Spring) 2012;20:1645–1652 361–371 tissue. Diabetes Obes Metab 2009;11:1163– 64. Ferrannini G, Hach T, Crowe S, Ferrannini E. 78. Rueda-Clausen CF, Padwal RS, Sharma AM. 1172 Energy balance following sodium-glucose co- New pharmacological approaches for obesity 90. Hiatt WR, Goldfine AB, Kaul S. Cardiovascu- transporter-2 (SGLT2) inhibition [EASD abstract management. Nat Rev Endocrinol 2013;9:467– lar risk assessment in the development of new 3]. Diabetologia 2014;57:S8 478 drugs for obesity. JAMA 2012;308:1099–1100 65. Perkins BA, Cherney DZ, Partridge H, et al. 79. Smith SR, Weissman NJ, Anderson CM, 91. U.S. Food and Drug Administration. Guid- Sodium-glucose cotransporter 2 inhibition and et al.; Behavioral Modification and Lorcaserin ance for industry: diabetes mellitusd evaluat- glycemic control in type 1 diabetes: results of an for Overweight and Obesity Management ing cardiovascular risk in new antidiabetic 8-week open-label proof-of-concept trial. Dia- (BLOOM) Study Group. Multicenter, placebo- therapies to treat type 2 diabetes [article online], betes Care 2014;37:1480–1483 controlled trial of lorcaserin for weight manage- 2008. Available from http://www.fda.gov/down 66. Berhan A, Barker A. Sodium glucose co- ment. N Engl J Med 2010;363:245–256 loads/Drugs/GuidanceComplianceRegulatoryInfor transport 2 inhibitors in the treatment of type 80. Gadde KM, Allison DB, Ryan DH, et al. Ef- mation/Guidances/ucm071627.pdf. Accessed 30 2 diabetes mellitus: a meta-analysis of random- fects of low-dose, controlled-release, phenter- April 2014 ized double-blind controlled trials. BMC Endocr mine plus topiramate combination on weight 92. Rosenstock J, Klaff LJ, Schwartz S, et al. Ef- Disord 2013;13:58 and associated comorbidities in overweight fects of exenatide and lifestyle modification on 67. Liakos A, Karagiannis T, Athanasiadou E, and obese adults (CONQUER): a randomised, body weight and glucose tolerance in obese et al. Efficacy and safety of empagliflozin for placebo-controlled, phase 3 trial. Lancet 2011; subjects with and without pre-diabetes. Diabe- type 2 diabetes: a systematic review and 377:1341–1352 tes Care 2010;33:1173–1175 meta-analysis. Diabetes Obes Metab 2014;16: 81. Wilding J, Van Gaal L, Rissanen A, 93. Araujo F, Shrestha N, Granja PL, Hirvonen J, 984–993 Vercruysse F, Fitchet M; OBES-002 Study Group. Santos HA, Sarmento B. Antihyperglycemic poten- 68. Younk LM, Mikeladze M, Davis SN. Pramlin- A randomized double-blind placebo-controlled tial of incretins orally delivered via nano and mi- tide and the treatment of diabetes: a review of study of the long-term efficacy and safety of crosystems and subsequent glucoregulatory the data since its introduction. Expert Opin topiramate in the treatment of obese subjects. effects. Curr Pharm Biotechnol 2014;15:609–619 Pharmacother 2011;12:1439–1451 Int J Obes Relat Metab Disord 2004;28:1399– 94. Buse JB, Bergenstal RM, Glass LC, et al. Use 69. Singh-Franco D, Perez A, Harrington C. The 1410 of twice-daily exenatide in basal insulin-treated effect of pramlintide acetate on glycemic con- 82. Rosenstock J, Hollander P, Gadde KM, Sun patients with type 2 diabetes: a randomized, trol and weight in patients with type 2 diabetes X, Strauss R, Leung A; OBD-202 Study Group. A controlled trial. Ann Intern Med 2011;154: – mellitus and in obese patients without diabetes: randomized, double-blind, placebo-controlled, 103 112 fi a systematic review and meta-analysis. Diabe- multicenter study to assess the ef cacy and 95.BuseJB,GoughSC,WooVC,etal. fi tes Obes Metab 2011;13:169–180 safety of topiramate controlled release in the IDegLira, a novel xed ratio combination of in- fi 70. Berne C; Orlistat Swedish Type 2 diabetes treatment of obese type 2 diabetic patients. sulin degludec and liraglutide, is ef cacious and – Study Group. A randomized study of orlistat in Diabetes Care 2007;30:1480 1486 safe in subjects with type 2 diabetes: a large, 83. Tonstad S, Tykarski A, Weissgarten J, et al. combination with a weight management pro- randomized phase 3 trial [abstract 65-OR]. Di- Efficacy and safety of topiramate in the treat- abetes 2013;62(Suppl. 1):A16 gramme in obese patients with type 2 diabetes ment of obese subjects with essential hyperten- 96. Riddle MC, Forst T, Aronson R, et al. Adding treated with metformin. Diabet Med 2005;22: sion. Am J Cardiol 2005;96:243–251 once-daily lixisenatide for type 2 diabetes inad- 612–618 84. Garvey WT, Ryan DH, Look M, et al. Two- equately controlled with newly initiated and 71. O’Neil PM, Smith SR, Weissman NJ, et al. year sustained weight loss and metabolic bene- continuously titrated basal insulin glargine: Randomized placebo-controlled clinical trial of fits with controlled-release phentermine/ a 24-week, randomized, placebo-controlled lorcaserin for weight loss in type 2 diabetes mel- topiramate in obese and overweight adults study (GetGoal-Duo 1). Diabetes Care 2013;36: litus: the BLOOM-DM study. Obesity (Silver (SEQUEL): a randomized, placebo-controlled, 2497–2503 – Spring) 2012;20:1426 1436 phase 3 extension study. Am J Clin Nutr 2012; 97. Cummings BP. Leptin therapy in type 2 di- 72. Garvey WT, Ryan DH, Bohannon NJ, et al. 95:297–308 abetes. Diabetes Obes Metab 2013;15:607–612 Weight-loss therapy in type 2 diabetes: effects 85. U.S. Food and Drug Administration. NDA 98. Lovshin JA, Drucker DJ. Incretin-based ther- of phentermine and topiramate extended re- 22580. QSYMIA (phentermine and topiramate apies for type 2 diabetes mellitus. Nat Rev En- lease. Diabetes Care 2014;37:3309–3316 extended-release) capsules: risk evaluation and mit- docrinol 2009;5:262–269 73. Hollander P, Gupta AK, Plodkowski R, et al.; igation strategy [article online], 2013. Available 99. Grasso P. Novel approaches to the treat- COR-Diabetes Study Group. Effects of naltrex- from http://www.fda.gov/downloads/drugs/drug ment of obesity and type 2 diabetes mellitus: one sustained-release/bupropion sustained- safety/postmarketdrugsafetyinformationfor bioactive leptin-related synthetic peptide ana- release combination therapy on body weight patientsandproviders/ucm312598.pdf. Accessed logs. Recent Pat Endocr Metab Immune Drug and glycemic parameters in overweight and 8 September 2014 Discov 2011;5:163–175 obese patients with type 2 diabetes. Diabetes 86. Katsiki N, Hatzitolios AI, Mikhailidis DP. Nal- 100. Moon HS, Chamberland JP, Mantzoros CS. Care 2013;36:4022–4029 trexone sustained-release (SR) + bupropion SR Amylin and leptin activate overlapping signalling 1172 Weight Management in Type 2 Diabetes Diabetes Care Volume 38, June 2015

pathways in an additive manner in mouse GT1-7 corrects obesity and diabetes in rodents. Nat 109. Day JW, Ottaway N, Patterson JT, et al. A hypothalamic, C2C12 muscle and AML12 liver cell Med 2015;21:27–36 new glucagon and GLP-1 co-agonist eliminates lines. Diabetologia 2012;55:215–225 105. Kahn SE, Cooper ME, Del Prato S. Patho- obesity in rodents. Nat Chem Biol 2009;5: 101. Tam CS, Lecoultre V, Ravussin E. Novel physiology and treatment of type 2 diabetes: 749–757 strategy for the use of leptin for obesity ther- perspectives on the past, present, and future. 110. Hughes TE, Kim DD, Marjason J, Proietto J, apy. Expert Opin Biol Ther 2011;11:1677–1685 Lancet 2014;383:1068–1083 Whitehead JP, Vath JE. Ascending dose-controlled 102. Clemmensen C, Chabenne J, Finan B, et al. 106. Troke RC, Tan TM, Bloom SR. The future trial of beloranib, a novel obesity treatment for GLP-1/glucagon coagonism restores leptin role of gut hormones in the treatment of obe- safety, tolerability, and weight loss in obese responsiveness in obese mice chronically main- sity. Ther Adv Chronic Dis 2014;5:4–14 women. Obesity (Silver Spring) 2013;21:1782– tained on an obesogenic diet. Diabetes 2014;63: 107. Kliewer SA, Mangelsdorf DJ. Fibroblast 1788 1422–1427 growth factor 21: from pharmacology to physi- 111. Meneghini L, Kesavadev J, Demissie M, 103. Finan B, Ma T, Ottaway N, et al. Unimolec- ology. Am J Clin Nutr 2010;91:254S–257S Nazeri A, Hollander P. Once-daily initiation of ular dual incretins maximize metabolic beneﬁts 108. Omar BA, Andersen B, Hald J, Raun K, basal insulin as add-on to metformin: a 26- in rodents, monkeys, and humans. Sci Transl Nishimura E, Ahren´ B. Fibroblast growth factor week, randomized, treat-to-target trial compar- Med 2013;5:209ra151 21 (FGF21) and glucagon-like peptide 1 contrib- ing insulin detemir with insulin glargine in 104. Finan B, Yang B, Ottaway N, et al. A ratio- ute to diabetes resistance in glucagon receptor- patients with type 2 diabetes. Diabetes Obes nally designed monomeric peptide triagonist deﬁcient mice. Diabetes 2014;63:101–110 Metab 2013;15:729–736