One of the challenges of using anti-diabetic therapy in obese or overweight patients is the prospect of substantial iatrogenic weight gain with many widely used drug classes. Since lifestyle interventions including weight loss are usually very difficult to achieve and/or sustain, the practitioner frequently is left with few options when trying to avoid adverse effects of antidiabetic therapy that are counterproductive to efforts directed at weight loss.
One of the challenges of using anti-diabetic therapy in obese or overweight patients is the prospect of substantial iatrogenic weight gain with many widely used drug classes. Since lifestyle interventions including weight loss are usually very difficult to achieve and/or sustain, the practitioner frequently is left with few options when trying to avoid adverse effects of antidiabetic therapy that are counterproductive to efforts directed at weight loss.
New classes of anti-diabetic agents mimic or potentiate the activities of incretin hormones, including the injectable glucagon-like peptide-1 (GLP-1) mimetics and the oral dipeptidyl peptidase- 4 (DPP-4) inhibitors. These drugs have benefits for glucose-dependent insulin secretion, suppression of glucagon secretion, inhibition of gastric emptying, and appetite reduction.11–14 They are effective in lowering glycosylated hemoglobin (HbA1c) and are associated with either weight loss (GLP-1 mimetics) or weight neutrality (oral DPP-4 inhibitors). Thus, these drugs promise to be useful in obese/overweight patients with type 2 diabetes.
A new anti-obesity agent, the selective cannabinoid receptor antagonist rimonabant, is effective in reducing weight and improving a range of metabolic defects in patients with type 2 diabetes. Such new agents should improve our ability to treat overweight or obese patients. Effects of Traditional Oral Anti-diabetic Agents on Body Weight
Insulin Secretagogues
Sulfonylureas
As insulin secretion is relatively deficient in type 2 diabetes, use of insulin secretagogues is logical for patients in whom the β-cell defect is not too advanced. Treatment with sulfonylureas (SUs) reduces HbA1c by approximately 1–2%. SU treatment in the UK Prospective Diabetes Study (UKPDS) was associated with a significant reduction in microvascular complications and a trend toward reduction in myocardial infarction (MI), but no significant effect on diabetes-related or all-cause mortality.15Weight gain and hypoglycemia—the latter usually asymptomatic or mild—are common side effects with these agents.16–18 Longer-acting agents (e.g. chlorpropamide, glyburide (glibenclamide), and sustained-release glipizide) are more likely to cause hypoglycemia than shorter-acting agents.18 Weight gain averages about 1–4kg and generally stabilizes over six months. In the UKPDS, for example, weight gain was 1.7kg with glyburide and 2.6kg with chlorpropamide.Weight increases with SU treatment may be partly due to anabolic effects of increased plasma insulin levels or reduced loss of calories as glucose in the urine.17 Lesser degrees of weight gain are claimed for some SU formulations (e.g. extended-release glipizide).19 However, individual metabolic responses to these agents vary, and weight gain should be anticipated with any agents in this class. SUs are, therefore, widely regarded as being less attractive as first-line therapy for obese patients; however, they remain in widespread use, often in combination with other drugs, notably metformin. When used in obese patients, treatment with SUs should be accompanied by a weight control program of diet and exercise wherever possible.
Rapid-acting Secretagogues
The non-sulfonylurea secretagogues repaglinide and nateglinide were developed to have a rapid onset and short metabolic half-life, resulting in preferential targeting of post-prandial hyperglycemia. These agents reduce HbA1c to a degree broadly similar to that observed with SUs (about 1–2%) with repaglinide having a somewhat greater effect in this regard.16–18 The drugs must be taken before each main meal. As with other insulin secretagogues, hypoglycemia and weight gain are the most common side effects, although both appear to be reduced in frequency compared with SUs. In a study involving 576 patients with type 2 diabetes, less weight gain was observed with repaglinide than with glyburide (2.5 kg versus 3.6 kg, respectively) among pharmacotherapy-naïve patients—although previously treated patients did not exhibit less weight gain with repaglinide.20 In general, no weight loss can be expected in patients switched from an SU to repaglinide. Nateglinide is a D-phenylalanine derivative with a faster onset and shorter duration of action than repaglinide, a meglitinide derivative. In a large-scale placebo-controlled trial, weight gains were 0.3 kg with nateglinide 60 mg tid and 0.9 kg with 120 mg tid.21 In a direct comparison with repaglinide, weight gain was 0.7 kg with nateglinide 360 mg/day and 1.8 kg with repaglinide 6 mg/day.22Insulin Sensitizers
Thiazolidinediones
Thiazolidinediones (TZDs) act to increase insulin sensitivity in adipose tissue, muscle, and the liver.23 Rosiglitazone and pioglitazone reduce HbA1c by about 0.5–1.5%.16–18 However, these agents are associated with weight gain. A small proportion of patients develop leg edema, which is attributed to fluid retention; heart failure is a concern in vulnerable patients (and is discussed below). Weight gain is generally similar to that seen with SUs (i.e. about 1–4 kg, with stabilization over six to 12 months). Of note, weight gain may be increased with concomitant treatment with other medications, including insulin.24 The weight gain with TZDs is partly attributable to adipogenesis, but recent data suggest that fluid retention may in fact be the main cause.25 There is some evidence that fat is redistributed away from visceral deposits, which are more closely associated with insulin resistance, to subcutaneous deposits.16–18 TZDs have complex effects on atherogenic lipid profiles with some notable differences between the two agents.26 In the secondary prevention PROspective PioglitAzone Clinical Trial In MacroVascular Events (PROactive) trial in high-risk patients with type 2 diabetes, pioglitazone was associated with a statistically significant 16% reduction in the secondary end-point of all-cause mortality, non-fatal MI, or stroke. However, an increased frequency of heart failure and edema without heart failure, as well as a 4 kg weight increase compared with placebo, detracted from the positive impact on vascular events.27,28 In the EU, TZDs are contraindicated in patients with a history of heart failure or current evidence of heart failure; their use in combination with insulin is contraindicated because of a perceived increased risk of heart failure. Recent US guidelines urge a cautious approach to TZD use in patients with evidence of heart failure. TZDs are also contraindicated in patients with active liver disease, although their effects on the liver are still under investigation; reduced levels of hepatic transaminases have been reported in several studies.
Metformin
Metformin, the only widely available biguanide, acts primarily by reducing hepatic glucose production and thereby reducing fasting hyperglycemia in the presence of a sufficient amount of insulin. The drug improves post-receptor insulin signaling to increase cellular insulin sensitivity.17,29 Metformin does not stimulate insulin secretion and may decrease hyperinsulinemia, an action regarded as potentially beneficial in obese patients. The agent reduces HbA1c by approximately 1–2%.16,17,29 Metformin also results in small beneficial changes in the atherogenic lipid profile and may exert positive effects on other cardiovascular risk factors.17In the UKPDS, metformin therapy in obese patients (mean body mass index (BMI) approximately 32 kg/m2) randomized to the drug after an inadequate response to diet was associated with statistically significant reductions of 32% in any diabetes-related end-point, 39% in MI, and 42% in diabetes-related mortality.30 In the Diabetes Prevention Program, metformin therapy reduced risk of diabetes by 31% in obese patients with impaired glucose tolerance.7 Metformin is not associated with weight gain and can sometimes promote modest weight loss. In the UKPDS, obese patients randomized to metformin gained 1–2 kg compared with gains of 5–7 kg in patients receiving SU or insulin treatment.30 In the Diabetes Prevention Program, metformin was associated with weight loss of 2.1 kg.7
Co-administration of metformin with insulin has the potential to reduce insulin-associated weight gain.31The mechanisms of weight neutrality or weight loss associated with metformin in this context are not completely understood. Suggestions include a reduction in hyperinsulinemia, a reduced risk of hypoglycemia and hence lower calorie consumption, or decreased food intake as a consequence of the wellrecognized gastrointestinal (GI) side effects of the drug; these often include a degree of anorexia.
Overall, these characteristics of metformin make it a preferred therapy for obese patients with type 2 diabetes. Use of the drug is limited, however, by risk for lactic acidosis—a rare condition but one associated with a high case-fatality rate. For this reason, metformin is limited to patients with renal function sufficient to avoid drug accumulation and is contraindicated in patients with cardiac or respiratory insufficiency (e.g. congestive heart failure (CHF)), other conditions associated with hypoxia or reduced perfusion, hepatic dysfunction, alcoholism, or history of metabolic acidosis.16–18 Use of metformin is also limited by the high frequency of GI adverse effects; a lesser proportion of patients cannot tolerate metformin at higher doses, whereas a minority are unable to tolerate the drug at any dose.
α-Glucosidase Inhibitors
The α-glucosidase inhibitors (AGIs) reduce the rate of polysaccharide digestion in the proximal small intestine, thereby reducing postprandial hyperglycemia, by inhibiting intestinal α-glucosidase enzymes.16–18 These agents (acarbose, miglitol, voglibose) are taken three times daily with meals that contain digestible carbohydrates.They reduce HbA1c by 0.5–0.8%. Acarbose treatment was associated with a 25% reduction in risk of progression to diabetes in patients with impaired glucose tolerance in the Study to Prevent Non-Insulin-Dependent Diabetes Mellitus (STOP-NIDDM) trial.32 AGIs pose little risk for hypoglycemia as monotherapy, but hypoglycemia that occurs when they are used in combination with other drugs, such as SUs, must be treated with oral glucose rather than complex carbohydrates. AGIs may be associated with a small reduction in body weight. Carbohydrate malabsorption and associated marked weight loss do not occur with these agents, since carbohydrate absorption is moved more distally along the intestinal tract. However, greater amounts of undigested oligosaccharides are passed to the large bowel, resulting in troublesome GI side effects, including abdominal discomfort, flatulence, and diarrhea. GI side effects have resulted in discontinuation of AGIs in 25–45% of patients in clinical trials. Insulin
Insulin, the most effective glucose-lowering drug, is sometimes the treatment of choice in patients with type 2 diabetes and is often required when glycemic goals cannot be obtained using oral agents.33
However, the adverse effects of insulin often lead to sub-optimal results in clinical practice. Most important from the perspective of the patient is the substantial weight gain, ranging from 2 kg to nearly 10 kg with various formulations, that so often accompanies insulin treatment.34–37 This weight gain poses particular problems for overweight or obese patients who usually have pre-existing insulin resistance, weight-related health issues, and longerterm health risks. Gains in body weight appear to increase in proportion to daily insulin dose and plasma insulin concentration.24 The magnitude of weight gain may, to some extent, also reflect the timing of insulin injections, co-administration of oral anti-diabetic agents, and risk of hypoglycemia. In fact, the risk of severe hypoglycemia is relatively low in obese insulin-resistant patients treated with insulin in the early course of the disorder.33
New Agents
The incretin hormone GLP-1, secreted by the enteroendocrine L cells of the GI tract in response to eating, is rapidly degraded by the proteolytic enzyme DPP-4. GLP-1 mimetics (or GLP-1 receptor agonists) are parenteral formulations that resist degradation by DPP-4 and competitively bind to Gprotein- coupled cellular GLP-1 receptors on islet α- and β-cells,13,14 thereby mimicking the effects of endogenous GLP-1. The DPP-4 inhibitors are orally active agents (incretin enhancers) that selectively bind to DPP-4 and block the action of the enzyme in degrading GLP-1; this results in persistent higher levels of active GLP-1 after eating. As already mentioned, the biologic actions of GLP-1 include improvements in glucose-dependent β-cell function, suppression of glucagon secretion, reduced gastric emptying, and increased satiety. In addition, GLP-1 has been shown to stimulate β-cell expansion in preclinical studies, a finding that has also been observed with chronic DPP-4 inhibition using the DPP-4 inhibitor vildagliptin in animal models.11,12,38GLP-1 Mimetics—Exenatide, Liraglutide
In 2005 exenatide, the first licensed incretin-based treatment, was approved as a twice daily subcutaneous injection before meals in patients who are receiving metformin and/or SU and who have not achieved glycemic control. The addition of exenatide to metformin and/or SU reduced HbA1c by approximately 1.0% and produced significant weight loss ranging from 0.9 kg (when added to metformin/SU) to 2.5 kg (when added to metformin) at the approved 5 μg and 10 μg doses in six-month studies.39–41 In a relatively small group of overweight metformin-treated patients followed for 82 weeks, there was maintained reduction in HbA1c and continued weight loss (5.3 kg).42 In a comparison with insulin glargine, both treatments reduced HbA1c by 1.1%, and exenatide produced a 2.3 kg weight loss compared with a weight gain of 1.8 kg with insulin glargine.43 Severe hypoglycemia has been infrequent with use of exenatide, but hypoglycemia occurred in 14–36% of patients receiving SU or metformin/SU; accordingly, a dose reduction in SU is recommended when the agents are combined.44 In addition to its need for twice-daily injection, exenatide use may be limited by a high frequency of GI adverse events, including nausea and, less commonly, vomiting (see Table 1). A new longacting formulation to be administered once weekly is in development that may offer substantially better prospects in terms of patient acceptability.
Adapted from Byetta PI.44
Liraglutide is an acylated GLP-1 mimetic in development that has a half-life long enough to permit once-daily subcutaneous dosing. In a recent 14-week study in obese patients, liraglutide once daily significantly reduced HbA1c at doses of 0.65, 1.25, and 1.9 mg (1.7% reduction at highest dose); liraglutide therapy was associated with dose-dependent reductions in body weight of 3.0 kg at the highest dose compared with a loss of 1.2 kg with placebo.45 GI adverse effects were common but appeared to decrease over time.DPP-4 Inhibitors – Sitagliptin, Vildagliptin
Of the various oral DPP-4 inhibitors in development sitagliptin and vildagliptin are at the most advanced stage, the f
rmer having been approved for use in patients with type 2 diabetes in October 2006 in the US. Both agents have been shown to be effective in reducing HbA1c as monotherapy and when added to metformin, SU, or TZD therapy in patients with type 2 diabetes.12,46–52 HbA1c reductions generally have ranged from 0.5% to 1.1%, with greater reductions seen in patients with higher baseline HbA1c.12 A pooled analysis of 1,301 drug-naïve patients receiving vildagliptin 100 mg/day indicates reductions of 1.1% overall, 1.3% in patients with baseline HbA1c >8.0%, and 1.7% in those with baseline levels >9.0%.12
Modeling studies have indicated improved β-cell function during treatment with these agents.11–13,53,54 A 52-week meal-test study showed improved β-cell function (insulin secretion measured as suprabasal Cpeptide area under the curve divided by glucose area under the curve during standardized meal) at 12 weeks in patients who had vildagliptin added to metformin compared with metformin; this improvement was maintained over 52 weeks.53 In another study, improved secretory tone was observed, whereby treatment increased the insulin secretory rate at any given glucose level compared with placebo.54
The DDP-4 inhibitors appear to have side effect profiles similar to placebo12 (see Table 2) and are not associated with any marked risk for hypoglycemia as monotherapy. For example, in trials hypoglycemia in vildagliptin-treated patients has been mild and infrequent, with rates similar to those observed with rosiglitazone (one event in each group over six months)51 and metformin (<1% of patients over one year)48; in an add-on study with insulin, vildagliptin treatment was associated with a decrease in the frequency and severity of hypoglycemia compared with ongoing insulin treatment as monotherapy.55
*Pooled data from monotherapy trials with 50mg bid and 100mg qd. Adapted with permission from Nathwani A. Presented at: American Diabetes Association 66th Scientific Sessions,Washington, DC, June 13, 2006.60
The DPP-4 inhibitors generally appear to be weight neutral,12,46,49,51 although some significant reductions in body weight have been observed in comparative studies. For example, vildagliptin produced a 0.3kg reduction in weight compared with a 1.6kg increase with rosiglitazone;51 among obese patients, vildagliptin was associated with a 1.1kg decrease and rosiglitazone with a 1.7kg increase. Similarly, a 52-week study in metformin-treated patients with inadequate glycemic control showed that sitagliptin was associated with a 1.5kg decrease compared with a 1.1kg increase with glipizide.12 Other potential benefits of these medications are suggested by a significant reduction in GI side effects with the addition of vildagliptin to metformin, compared with ongoing metformin treatment as monotherapy.49
Overall, the findings with oral DPP-4 inhibitors are promising. Although they appear to lack the consistently significant reduction in body weight associated with GLP-1 mimetics, the weight neutrality of these agents—together with the absence of a need for injection, the apparent absence of significant GI adverse effects, and the low risk for hypoglycemia— suggests utility particularly, although not exclusively, in obese patients. If experience in clinical practice substantiates the benefits suggested in these data, clinicians will have an exciting new option for obese patients with type 2 diabetes.Anti-obesity Drugs—Rimonabant
Rimonabant is a selective cannabinoid-1 receptor blocker that has been approved for use in Europe as an adjunct to diet and exercise in obese patients or overweight patients with cardiometabolic risk factors such as type 2 diabetes or dyslipidemia. While it is not primarily a glucose-lowering agent, the effect of rimonabant on body weight, glycemia, and a range of cardiometabolic risk factors may make it useful in obese patients with diabetes. The Rimonabant-in-Obesity (RIO) series of trials in overweight or obese individuals has shown that rimonabant consistently reduces weight and waist circumference and improves lipid profile and other aspects of cardiovascular risk.56–59
The RIO-Diabetes trial compared rimonabant 5 mg or 20 mg with placebo for one year in more than 1,000 overweight or obese patients with type 2 diabetes who were receiving metformin (~65%) or SU (~35%).59 The 20 mg dose produced a weight reduction of 5.3 kg and reduced HbA1c by 0.6%. At this dose, a significantly greater proportion of patients with metabolic syndrome at baseline no longer met the criteria for metabolic syndrome at one year (although there was no significant difference between rimonabant and placebo with regard to proportion of patients developing metabolic syndrome during the study). Discontinuation rates were high in both rimonabant and placebo groups in all of the RIO trials, a finding that has been observed in some other anti-obesity drug trials. Although discontinuation rates were similar in all patient groups in RIO-Diabetes (see Table 3), discontinuation due to adverse events was more frequent in patients receiving rimonabant 20 mg (15%) compared with placebo (5%); the most frequent reasons for discontinuation in the 20-mg group were depressed mood disorders (11 patients, 3%), nausea (five patients, 1.5%), and dizziness (three patients, 0.9%). There is concern over the frequency of depression and depressed mood disorders observed in rimonabant-treated patients; monitoring for such effects is prudent, and additional study and experience are necessary to quantify any risks associated with these effects.
Reprinted from Scheen AJ, et al., “Efficacy and tolerability of rimonabant in overweight or obese patients with type 2 diabetes: a randomized controlled study”, Lancet (2006);368: pp. 1660–167259, with permission from Elsevier.
Conclusion
Obesity is a major—and modifiable—contributor to an alarming increase in type 2 diabetes worldwide. Overweight and obese patients should be encouraged to start and maintain a diet and exercise program to lose weight; this should help in moving towards glycemic goals, and improving other metabolic risk factors. Unfortunately, although lifestyle interventions can be effective in achieving metabolic control, patients find it very difficult to maintain lifestyle changes. The recent consensus statement of the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD) recommends initial treatment with lifestyle intervention and metformin—where safe and tolerated—for all patients with type 2 diabetes, whether obese or not. The statement also emphasizes the need for rapid addition of complementary medications or changes in regimens to promptly achieve and maintain glycemia at levels as close to normal as possible.18 When considering treatment of overweight and obese patients with type 2 diabetes, practitioners must remain aware of the potential for weight gain that is associated with many anti-diabetic medications. Avoiding iatrogenic weight gain is a first step towards improved metabolic control and, it is to be hoped, better long-term outcomes. For obese patients with multiple cardiometabolic risk factors rimonabant may prove to be a useful intervention. Other new agents such as the incretin-enhancing oral DPP-4 inhibitors and the incretin mimetics also represent major advances in therapy.These agents seem likely to have considerable utility in overweight and obese patients with type 2 diabetes.
