Emerging Peptide Hormone Therapies for the Treatment of Obesity: Mechanisms of Action, Clinical Safety, and Efficacy

Introduction Prevalence and Risk of Obesity and Case Presentation George A. Bray, MD, MACP The Clinical Evidence for Behavioral Management of the Obese Patient Gary D. Foster, PhD Neurohormonal Pathophysiologyand Pharmacotherapy for theTreatment of Obesity Steven R. Smith, MD The Synergistic Management of Obesity and Diabetes Ken Fujioka, MD

Introduction

Obesity in the US is increasing at an alarming rate and according to recent data is now considered an epidemic, affecting more than one-third of adults in the United States. Most body systems are adversely affected by obesity, including the endocrine and cardiovascular systems, where diseases such as diabetes have the highest annual death rates.

In the face of this epidemic, how should the clinician approach the obese patient, particularly one with diabetes and/or metabolic syndrome? Is behavioral management a viable option? What is the neurohormonal basis for obesity, and what associated pharmacotherapies are available? How can obesity and diabetes be managed synergistically?

Vindico Medical Education sponsored a symposium at The Obesity Society 2008 Annual Scientific Meeting to have a practical discussion of these issues. A key word in the title of this monograph is “emerging.” This topic is a very dynamic area for research with significant advancements allowing projections of what may be possible over the next few years as we progress from benchtop to bedside. Accordingly, this discussion will include not only state-of-the-art science but a look over the horizon as well. In addition, the issues are considered in relation to an actual patient case. The proceedings from the experts’ discussions provided the material for this CME monograph, and I thank the speakers for sharing their expertise on these topics and for participating in the preparation of this CME monograph, which summarizes key points and provides the reader with important information about the rapidly developing field of peptide hormone therapies for the treatment of obesity.

George A. Bray, MD, MACP
Course Director

George A. Bray, MD, MACP Boyd Professor Chief, Division of Clinical Obesity and Metabolism Professor, Pennington Biomedical Research Center Baton Rouge, LA Steven R. Smith, MD Professor and Assistant to the Executive Director for Clinical Research Pennington Biomedical Research Center Baton Rouge, LA Gary D. Foster, PhD Professor of Medicine and Public Health Director, Center for Obesity Research & Education Temple University School of MedicinePhiladelphia, PA Ken Fujioka, MD Director Nutrition and Metabolic Research Center Center for Weight Management Scripps Clinic Department of Diabetes and Endocrinology San Diego, CA

Prevalence and Risk of Obesity and Case Presentation

George A. Bray, MD, MACP

Two major health surveillance systems capture obesity data in the US, both of which are supported by the Public Health Service.^1,2 The National Health and Nutrition Examination Survey (NHANES) provides direct measurements, whereas Behavioral Risk Factor Surveillance System (BRFSS) data are based on telephone surveys. Based on direct measurements, the prevalence of obesity (body mass index [BMI] ?30) in the US doubled between 1980 and 2000 and, in 2006, included 34% of adults (Figure 1). Although BRFSS data parallel NHANES data, the prevalence of obesity in BRFSS is approximately 10% lower than NHANES, possibly reflecting the effect of survey respondents overestimating their height or underestimating their weight. Accordingly, when considering the status of obesity, it is important to consider the source of the data.

Figure 1. Prevalence of Obesity: NHANES vs BRFSS Prevalence of obesity in the US has more than doubled since 1980. Source: CDC. MMWR. 2008;57:765-768; Ogden CL, et al. JAMA. 2008;299:2401-2405.

Obesity is an excess of body fat, which results from an imbalance between energy intake and expenditure. When a person is in a positive balance, the energy is stored in adipocytes, resulting in subcutaneous or visceral fat accumulation. The products of the hypertrophied visceral adipocytes are implicated in the disease entities that are associated with obesity (Figure 2).^3-5 The entire body is affected by obesity, and 4 of the top 5 causes of death in the US, including cardiovascular disease and diabetes, are related to obesity.⁶ Cardiovascular disease and diabetes are of major importance, with patients at risk for diabetes mellitus being particularly susceptible to increases in BMI.^7,8 The multi-hormonal control of body weight, vis-à-vis the role of fat, gut, and pancreatic islet-derived signals, is discussed in detail later in this monograph.

Figure 2. Adverse Cardiometabolic Effects of Products of Adipocytes Many products of hypertrophied adipocytes contribute to an adverse cardiometabilic profile. Source: Lyon CJ, et al. Endocrinology. 2003;144:2195-2200; Trayhurn P, et al. Br J Nutr. 2004;92:347-355; Eckel RH, et al. Lancet. 2005;365:1415-1428.

What Is Metabolic Syndrome?

Metabolic syndrome consists of a group of risk factors associated with heart disease, stroke, and diabetes, and is common in overweight and obese people.⁹ According to the American Heart Association and the National Heart, Lung, and Blood Institute, metabolic syndrome is characterized by the presence of at least 3 of the following factors (for persons with diabetes, at least 2)¹⁰:

• Increased waist circumference (men ?40 inches; women ?35 inches)
• Elevated fasting glucose (?100 mg/dL)
• Reduced HDL (men <40 mg/dL; women <50 mg/dL)
• Elevated triglycerides (?150 mg/dL)
• Elevated blood pressure (?130/85 mm Hg)

The predominant type of obesity associated with metabolic syndrome is central obesity, that is, in people whose fat accumulates around the abdomen as compared to around the hips. Significantly, the deposition of visceral fat as opposed to subcutaneous fat most likely contributes to the condition.¹¹

Weight loss, as would be expected, can reverse all components of metabolic syndrome except HDL cholesterol. If weight loss is stabilized, HDL cholesterol should increase. Before discussing in detail the behavioral and pharmacologic therapies that target the energy balance formula, we review a clinical case, and consider management options as we proceed through the discussions.

References

A. Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey. Available at: http://www.cdc.gov/nchs/nhanes.htm. Accessed October 18, 2008.
B. Centers for Disease Control and Prevention. Behavioral Risk Factor Surveillance System. Available at: http://www.cdc.gov/brfss/index.htm. Accessed October 18, 2008.
C. Lyon CJ, et al. Endocrinology. 2003;144:2195-2200.
D. Trayhurn P, et al. Br J Nutr. 2004;92:347-355.
E. Eckel RH, et al. Lancet. 2005;365:1415-1428.
F. Mokdad AH, et al. JAMA. 2004;291:1238-1245.
G. Chan JM, et al. Diabetes Care. 1994;17:961-969.
H. Colditz G, et al. Ann Intern Med. 1995;122:481-486.
I. Metabolic syndrome. American Heart Association. Available at: http://www.americanheart.org/presenter.jhtml?identifier=4756. Accessed October 18, 2008.
J. Grundy SM. J Clin Endocrinol Metab. 2004;89:2595-2600.
K. You T, et al. J Clin Endocrinol Metab. 2004; 89:5517–5522.

Case Description

The patient in the case is a supervisor with a basically sedentary job. He eats out several nights a week, and exercises infrequently other than the little work-related walking. He was diagnosed with diabetes when he was 42. He smoked until his coronary artery bypass graft (CABG) surgery 7 years ago, after which he gained 25 lb. He eats more when he feels sad. Although his BMI is slightly below obese, he meets the criteria for metabolic syndrome.

We will revisit the patient after each discussion focusing on the behavioral management of the obese patient, the neurohormonal pathophysiology and pharmacotherapy for the treatment of obesity, and the synergistic management of obesity and diabetes.

Case Presentation

KEY: CABG – coronary artery bypass graft, CAD – coronary artery disease

Laboratory Values

KEY: ALT – alanine transaminase, GFR – glomerular filtration rate, HbA1C – glycosylated hemoglobin, HDL-C – high density lipoprotein cholesterol, LDL-C – low density lipoprotein cholesterol

The Clinical Evidence for Behavioral Management of the Obese Patient

Gary D. Foster, PhD

The social stigma surrounding obesity—often endorsed by the patients themselves—is that obesity is primarily the result of moral ineptitude. Additionally, approximately one-third of physicians, psychologists, and dietitians believe obesity relates to lack of willpower. Accordingly, before beginning any treatment for the obese patient, it is important to identify and modify any negative attitudes or behaviors that may have an untoward impact on the patient-provider relationship. Behavior modification is as important for the healthcare provider as it is for the patient.

The American Diabetes Association (ADA) Medical Nutrition Therapy (MNT) position statement, updated in 2008, acknowledges the importance of behavior modifications in their nutrition recommendations and interventions for people with and at risk for type 2 diabetes.¹ The recommendations reinforce the importance of controlling body weight:

• Weight loss is recommended for all individuals who have or are at risk for diabetes.
• Low-carbohydrate or low-fat calorie diets may be effective in the short-term (up to 1 year).
• Bariatric surgery may be considered for diabetic patients with a BMI ?35 kg/m² and can result in marked improvements in glycemia.
• Weight loss medications may be considered in the treatment of overweight individuals with type 2 diabetes, and can help achieve a 5% to 10% weight loss when combined with lifestyle modifications.
• Physical activity and behavior modifications are important in weight loss programs, and are most helpful in maintaining weight loss.

Principles and Strategies for Behavior Management

The behavioral treatment of obesity consists of a set of principles and practices to modify eating and activity.² From a behavioral point of view, the approach to weight management does not necessarily depend on the etiology of obesity. Whether the patients are genetically predisposed, are behaviorally predisposed, or are responding to a gene-environment interaction, all patients need a set of skills to manage their weight.

The Behavior Chain

The approach to the behavioral treatment of obesity must consider that there is a chain of events—the behavior to be modified is preceded by antecedents and followed by consequences—and must recognize the non-behavioral causes of obesity. The short-term rewards of eating high-fat, calorie-dense foods may include feelings of pleasure, satiety, and comfort. However, the long-term consequences, including elevated lipids, insulin resistance, and weight gain are often disregarded. Getting patients to focus on the antecedents—the triggers and cues that precede the behavior—is more effective than countering the short-term consequences.

It is best to avoid asking, “Why?” when talking to patients about how much they eat; rather ask, “How did it happen?” Clinicians should feel free to ask as many questions as they need to get a picture of how the eating happened. This series of events can be summarized in a behavior chain (Figure 1).³ Such chains are useful for patients and practitioners to think about how things happened rather than why.

From the pursuit of how, rather than why, the clinician can acquire a broader understanding of how to prevent unhealthy behaviors in the future. The clinical challenge is to determine where the links in this chain can be broken.

Behavioral treatment, therefore, is saying, “How did this behavior happen?” Beyond that, to instill some therapeutic optimism, the clinician should progress to, “What would you do differently if the same situation happens in the future?”

Figure 1. The Behavior Chain Understanding a behavior chain helps patients and practitioners learn how behaviors happen. Source: Brownell KD. The LEARN Program for Weight Management. American Health Publishing Co. 2004.

Breaking the Chain: Food Intake Diaries

If patients were to do only one thing to manage their weight, it should be write down what they eat. The data supporting this are clear, for both short- and long-term efficacy. People who record what they eat achieve greater weight loss than people who do not record their meals. For example, in a randomized trial, patients who were in the highest third of adherence to completing food intake records lost more than twice as much weight (18.1 kg vs 7.7 kg, respectively) as those in the lower third of adherence to recording their food intake.⁴

Self-monitoring and recording of food intake should not record every detail, rather it should be simple and include types of foods, portion sizes, calories (goal is to reduce by 500 kcal/d), times, places, activities, thoughts, and moods.²

Breaking the Chain: Portion-Controlled Meals

Regardless of whether a person has diabetes or metabolic syndrome, the approach to losing weight is the same, and using portion-controlled meals is a significant, easy, and data-supported component of a weight-loss plan: its importance cannot be over-emphasized. If the amount of choice is decreased, structure is increased and consequently, weight loss is increased. Structure gives a person a sense of control, and control gives a sense of self-efficacy, which drives behavior chains. Portion-controlled meals include fixed-portions and calorie amounts, resulting in reduction of choices and contact with problem foods. They are convenient to use, satisfy appetite (monotony and sensory specific satiety), and facilitate dietary adherence.

The Achilles’ heel of behavioral treatment is adherence. People know what to do, but they are less clear about how to get it done. Portion-controlled meals can help with that.

In a classic study, patients randomized to an isoenergetic diet in which 2 of the 3 meals and both snacks were replaced with commercially-prepared liquid meals (shakes, soups, hot chocolate) and snack replacement bars lost more weight (7.8%) than patients randomized to a conventional food diet (1.5%),^5,6 and they maintained their weight loss when they continued to replace 1 meal and 1 snack every day for 4 years.⁵

Look AHEAD and Intensive Lifestyle Intervention

In the Look AHEAD (Action For Health in Diabetes) trial, 5,145 people were randomized to either diabetes support and education (DSE) or intensive lifestyle intervention (ILI), an aggressive behavioral treatment program.⁷ After 1 year, weight loss has been remarkable. The ILI group lost an average of 8.6% of body weight, compared with 0.7% in the DSE group, (Figure 2, P<0.001).

Figure 2. Look AHEAD Trial 1-year Weight Loss Progressive behavioral management imporoves weight loss. Source: Look AHEAD Research Group. Diabetes Care. 2007;30:1374-1383.

Combination Therapy: Pharmacologic and Behavioral

Behavioral therapy modifies the external environment, including stimulus control, meal planning, and self-monitoring. Pharmacotherapy modifies the internal environment by decreasing hunger, increasing satiety, or affecting nutrient absorption. When behavioral and pharmacologic therapies are combined, a synergistic effect may result compared with using drugs alone. For example, in a 4-arm randomized trial, patients who received both therapy and support, combining behavioral and pharmacotherapeutic approaches to weight loss, lost significantly more weight than those in the other 3 groups (Figure 3, P<0.001).⁴

In summary, behavioral therapy has a documented and important place in weight loss programs. By using the concept of a behavior chain, healthcare practitioners can focus on how a behavior happened rather than why. Two important strategies for breaking the chain include documenting food intake and using portion-controlled meals. Evidence shows that behavior therapy has a synergistic effect when combined with pharmacotherapy.

Figure 3. Percent Reduction in Initial Weight on Pharmacotherapy and Behavior Modification Pharmacotherapy with lifestyle intervention leads to greater weight loss. Source: Wadden TA, et al. N Engl J Med. 2005;353:2111-2120.

References

1. Bantle JP, et al. Diabetes Care. 2008;31:S61-S78.
2. Foster GD, et al. AJCN. 2005;82:230S-235S.
3. Brownell KD. The LEARN Program for Weight Management. American Health Publishing Co. 2004.
4. Wadden TA, et al. N Engl J Med. 2005; 353:2111-2120.
5. Ditschuneit HH, et al. Am J Clin Nutr. 1999;69:198-204.
6. Flechtner-Mors M, et al. Obes Res. 2000;8:399-402.
7. Look AHEAD Research Group, Diabetes Care. 2007; 30:1374-1383.

Discussion

How would you approach the case study patient in terms of behavior modification, considering his known behaviors?

Gary D. Foster, PhD: I would manage the patient in 2 ways. First, I would assess the patient’s opinion of his most problematic eating situations. If you determine what the patient believes is the most important, they’re going to be more motivated to change it. Second, I would ask, “How does this happen?” not “Why do you overeat?” Jumping to a solution and prescribing behavioral modifications without listening to the patient’s point of view will not be as effective as first listening to the patient and making recommendations based on his or her attitudes, beliefs, and desires.

Would you use portion control for this gentleman?

Foster: Portion control would be reasonable. Again, it starts with a conversation with the patient. If he or she is willing to start meal replacement, I would go there pretty quickly. If he or she is not willing, I would probably make a more judicious recommendation based on where I believe adherence would be best, and the best return on investment in terms of the exchange in calories.

Let’s assume the patient is now participating in Look AHEAD, working toward the 8.5% weight loss from portion control and exercise. What is the algorithm for managing medications in the context of a portion-controlled diet?

Steve R. Smith, MD: I would stop the sulfonylurea, and follow up weekly or every other week thereafter. Self-monitoring of blood glucose is important to ensure the patient’s glycemia remains under control.

Ken Fujioka, MD: For insulin, if he says he can do portion control, I put the pressure on him, because if he does not adhere to meal replacements, his blood sugar will elevate markedly. When starting a patient on meal replacement I have cut insulin in half, and sometimes I have had to cut it more.

Please discuss the use of the word “obesity.” Some people think that it may be pejorative.

Foster: It does not make sense to me to change the word when used in the context of the medical condition. However, when talking to patients, comments or questions such as, “What do you think about your weight?” or “I’ve noticed that your weight’s changed since I last saw you.” are probably the most nonjudgmental, unloaded questions.

Neurohormonal Pathophysiology and Pharmacotherapy for the Treatment of Obesity

Steven R. Smith, MD

The control of eating behavior is not restricted to cognitive, behavioral, and environmental factors. A very complex network of signals senses the energy in the periphery, adipose tissue, liver, and skeletal muscle and translates those energy signals into peptide signals that are transmitted to the brain to help control hunger, appetite, satiety, and food-seeking and other behaviors.

The Major Signals and Their Origins

Taste is an important signal in overeating. However, other major signals come from fat cells that secrete endocrine hormones and adipokines, such as leptin, which serves as an energy sensor and travels throughout the body, including to the brain (Figure 1).¹ The liver also has several circuits that connect with the brain, and signals from the pancreas (eg, amylin) are elicited during a meal. Signals also originate from several portions of the gut, most importantly the L-cells in the small intestine and colon, which secrete peptides (eg, glucagon-like peptide 1 [GLP-1], oxytomodulin, and peptide YY_3-36) that travel to the brain to control metabolism and food intake.

Figure 1. Multi-Hormonal Control of Body Weight: Role Of Fat-, Gut-, and Islet-derived Signals Major signals in weight control. Source: Modified from Badman MK, Flier JS. Science. 2005;307:1909-1914.

Peripheral Peptides and the Brain

How does the brain sense these peripheral signals to integrate the internal environment with the external environment? Receptors for these peptides have been demonstrated to be in the hypothalamus and the hindbrain. The arcuate nucleus in the hypothalamus is an important control center that helps integrate many of these signals, and these circuits provide important insight into how peripheral signals like GLP-1 can influence areas within the brain, specifically the hypothalamus.

The key neurons of the arcuate nucleus are pro-opiomelanocortin (POMC) and agouti-related peptide/neuropeptide Y (AgRP/NPY), which send projections into other areas within the hypothalamus, particularly the lateral hypothalamus and the periventricular nucleus. One exciting area of research is how these signals not only modulate food intake, but also control peripheral metabolism, specifically in the liver.

Lower centers of the brain, or the hindbrain, can also sense nutrients and other signals from the periphery.² The hindbrain is not behind the blood-brain barrier. Therefore, it has access to many peripheral peptides, and a specific anatomical area in the hindbrain, the nucleus tractus solitarii (NTS) in the area postrema, senses many fuels—insulin, glucose, and other peptides—and communicates with the arcuate nucleus and other parts of the forebrain to coordinate behaviors.

Pancreatic Amylin and Pramlintide, its Synthetic Analog

Amylin is a 37-amino acid neuroendocrine peptide that is co-secreted with insulin by the pancreas. Importantly, amylin inhibits glucagon secretion, a hormone that is elevated and dysregulated in obesity and type 2 diabetes.

Pramlintide is a modified peptide with a more favorable pharmacology, and is approved in the US for treatment of patients with diabetes when administered with insulin. Pramlintide binds to neurons in the area postrema and the central nucleus of the amygdala, and has been shown in several studies to reduce body weight. In a recent randomized, placebo-controlled study of obese patients, pramlintide (180 µg, TID) before meals reduced body weight by 2.1% (P<0.001) after 44 days, compared with +0.1% for placebo.³

Food intake, measured over a 24-hour period on days 3 and 43 of the study, was substantially reduced in patients who received pramlintide on both days. These results provide direct pharmacologic evidence that targeting these circuits can help patients reduce weight in the clinical trial setting.

A subsequent dose-finding study in 411 obese patients included structured lifestyle intervention (LEARN) and either placebo or pramlintide (120, 240, or 360 µg) given before breakfast and dinner, or the same doses three times daily at breakfast, lunch, and dinner.⁴ In this 16-week study, a dose-dependent reduction in body weight occurred in the patients treated twice daily, with the patients receiving pramlintide 360 µg BID achieving an average 6.1 kg weight loss compared with 4.8 kg, 3.8 kg, and 2.6 kg in the 240 µg, 120 µg, and placebo groups, respectively (P<0.01). Weight loss in the 3 TID pramlintide dose groups ranged from 4.5 kg to 6.0 kg, but was not dose-dependent. In a 1-year study, 60% of patients achieved at least a 5% weight loss at 12 months, and 30% to 40% of patients achieved at least a 10% weight loss,⁵ and many of the patients who continued on the highest dose for 2 years were able to maintain their weight loss.⁶These results support the potential for this therapeutic approach to produce beneficial weight loss results and weight loss maintenance.

Intestinal L-cells and GLP-1

GLP-1 is released post-prandially from L-cells of the gut and is a potent incretin, or stimulator of insulin secretion. GLP-1 activates neurons in the NTS that project to key regions in the hypothalamus. The widely distributed dipeptidyl peptidase IV (DPP-IV) cleaves and inactivates GLP-1. Exenatide is a GLP-1 analog that is resistant to DPP-IV inactivation and is approved for the treatment of type 2 diabetes by subcutaneous injection twice daily.

Data from an early study using graded intravenous doses of exenatide showed a dose-dependent decrease in both food and calorie intake.⁷ More recently, data from the pivotal studies comparing exenatide, insulin aspart 70/30, and insulin glargine in patients with type 2 diabetes showed an antihyperglycemic effect of all treatments (Figure 2).^8-10However, weight gain was consistently observed in the insulin groups, compared with an approximate 5-lb reduction in body weight observed in patients taking exenatide.

Liraglutide, a longer-acting once weekly GLP-1 analog is in development and has shown adequate glucose control and mean weight loss of 3.8 kg over 15 weeks.¹¹ In addition, a 30-week study comparing twice daily with once weekly exenatide in nearly 300 patients showed similar weight losses of approximately 3.7 kg in each group.¹² In a separate study, liraglutide (1.9 mg/day) produced significantly greater weight loss after 14 weeks compared with placebo (3 kg vs 1.8 kg).¹³ In other recently reported studies, weight loss with liraglutide alone or in combination with oral anti-diabetic agents (metformin, glimepiride, or metformin with glimepiride) was greater for patients with higher baseline BMI.^14-18

Figure 2. Changes in Glycemia and Weight in 3 Head-to-Head Studies: Exenatide Vs. Insulin Exenatide results in both improved glycemia and a weight reduction. Source: Heine RJ, et al. Ann Intern Med. 2005;143:559-569; Barnett AH, et al. Clin Ther. 2007;29:2333-2348; Nauck MA, et al. Diabetologia. 2007;50:259-267.

Adverse Effects of GLP-1 Analogs

Pancreatitis has been reported in patients treated with exenatide. However, the incidence of pancreatitis in patients on exenatide is not different from pancreatitis in the general population, and no definite causal relationship has been proven. Long-term safety studies are needed to clarify the relationship between GLP-1 analogs and pancreatitis.

Future Directions

Targeting just one pathway may not be sufficient for opimal weight management. Thus, combinations of these peptide hormones may be justified. For example, pre-clinical studies of amylin plus leptin have shown a significant additive response in both reduced food intake and body weight loss.²⁰ Amylin, through its hindbrain signal, affects short-term satiety; whereas leptin’s hypothalamic signal allows endocrine control of long-term adiposity. In humans, an additive weight loss effect of metreleptin and pramlintide was achieved that was significantly greater than either drug as monotherapy.²¹

In summary, interesting and complex peripheral neuropeptide hormone systems feed into multiple circuits in the brain, which work synergistically to regulate body weight and food intake. Pramlintide has been shown to reduce body weight and food intake, with sustained weight loss when combined with lifestyle intervention. Exenatide and liraglutide dose-dependently decrease food intake, and combination peptide therapies demonstrate a significant and robust weight loss effect in obese patients. These data and ongoing research support the prospect that even greater weight loss may be achieved pharmacologically in patients who need improved glycemic control and body weight.

References

1. Badman MK, Flier JS. Science. 2005;307:1909-1914.
2. Grill HJ. Obesity. 2006;14:216S-221S.
3. Smith SR, et al. Am J Physiol Endocrinol Metab. 2007; 293:E620–E627.
4. Aronne LJ, et al. Obesity. 2006;14:A17. Abstract #53-OR.
5. Smith SR, Aronne LJ, et al. Diabetes Care. 2008;31:1816-1823.
6. Kesty N. Obesity Society 2008. Presentation #323-P.
7. Gutzwiller JP, et al. Gut. 1999;44:81–86.
8. Heine RJ, et al. Ann Intern Med. 2005;143:559-569.
9. Barnett AH, et al. Clin Thera. 2007;29:2333-2348.
10. Nauck MA, et al. Diabetologia. 2007;50:259-267.
11. Kim D, et al. Diabetes Care. 2007 Jun;30(6):1487-93.
12. Drucker DJ, et al. American Diabetes Association 2008. Abstract #107-OR.
13. Vilsbøll T, et al. Diabetes. 2006;55(Suppl. 1):A27.
14. Garber et al. Diabetes. 2008;57(Suppl. 1):LB3.
15. Nauck et al. Diabetes. 2008;57(Suppl. 1):A150.
16. Marre et al. Diabetes. 2008; 57(Suppl. 1):A4.
17. Zinman et al. Poster presentation at EASD 2008.
18. Russell-Jones et al. Diabetes. 2008;57(Suppl. 1):A159.
19. Roth JD, et al. Proc Natl Acad Sci U S A. 2008;105:7257–7262.
20. Weyer C, et al. American Diabetes Association 2008. Abstract #1738-P.

Discussion

The patient in the case study has diabetes, and his current medications are glyburide and metformin. Would you consider adding exenatide?

Steven R. Smith, MD: Yes. A drug like exenatide, which helps reduce body weight without a risk of hypoglycemia, is a rational approach for this patient. I would withdraw or reduce the sulfonylurea, which is a real risk for hypoglycemia and weight gain in this particular patient.

Could you address the pancreatitis issue in more detail?

Smith: The issue at hand is whether the cases of pancreatitis reported in patients on exenatide exceed what we would expect from people with diabetes or those in the same age category. Safety data from from the pivotal randomized clinical trials show that there is no evidence for increased risk of pancreatitis. However, once a drug is approved for use, we do not know how many patients are treated, how long, and what their characteristics are compared with the general population. Well-designed, large, randomized clinical trials are needed to determine whether pancreatitis is a risk with these drugs.

What about peptide YY?

Smith: Peptide YY works differently than the GLP-1 analogs and pramlintide. However, the clinical data to date suggests that peptide YY is going to be a difficult therapeutic agent because of the nausea and vomiting that has been associated with every clinical trial that interacts with the PPY receptor system. I am slightly more skeptical about peptide YY than some of the other peptide therapies that have nausea as a side effect.

The Synergistic Management of Obesity and Diabetes

Ken Fujioka, MD

How should diabetic patients be managed? Until this year, the idea was, “Just get the blood sugar down. Do not worry about weight; get the glucose under control, and everything will fall in place.” Microvascular diseases, such as myocardial infarction and cardiovascular disease, are the leading cause of death among diabetics, and traditional wisdom was that better blood sugar control will reduce macrovascular disease such as heart disease.

The ACCORD (Action to Control CardiOvascular Risk in Diabetes) and ADVANCE (Action in Diabetes) trials were large trials that investigated the impact of glucose control on heart disease. They enrolled more than 20,000 patients.¹ The objective was to keep glucose at near-normal levels, and patients were followed for 3.5 to 5 years. Despite intensive glucose control with a thiazolidinedione (TZD), sulfonylurea, metformin, and insulin, there was no reduction in heart disease at the end of the studies. The ACCORD trial, in fact, was stopped 17 months early because of an increase in deaths. There was also a 3.5 kg mean weight gain in the intensive glucose controlled group, with one-third of patients gaining more than 10 kg.

Accordingly, now when clinicians treat their patients with diabetes, they should be attentive to preventing or reversing weight gain. Unfortunately, many diabetes drugs are associated with weight gain (Table 1).²

Table 1. Anti-Hyperglycemic Agents for Treating Type 2 Diabetes Source: Nathan DM, et al. Diabetes Care. 2006;29:1963-1972.

Patients—particularly diabetic patients—are well-aware when they are gaining weight. When they begin therapy with insulin, a sulfonylurea, or a TZD, their weight begins increasing very quickly, and continues to increase. It is frustrating and distressing, both to patients and their physicians, and getting patients to lose weight is an important part of their diabetic management. If they are losing weight, they will have better glycemic control and may require fewer diabetic medications, with a theoretically lower cost.

Although drugs for short-term weight loss (diethylpoprion, phentermine, phendimetrazine, and benzphetamine) have been available for almost 50 years, only 2 drugs are approved for the long-term management of obesity: sibutramine, a norepinephrine and serotonin uptake inhibitor, and orlistat, a lipase inhibitor. These drugs should be considered for diabetic patients, because weight loss and maintenance of weight loss are essential for obese diabetics. Side effects of all weight-loss drugs must be considered, particularly when managing patients with heart disease.

Sibutramine

Approved by the FDA in 1997, sibutramine increases satiety, although it may have a small metabolic effect, primarily by attenuating the decline in metabolic rate that occurs during weight loss.^{3, 4}

In a 6-month trial of 175 obese diabetic patients randomized to sibutramine or placebo, weight loss was 0.5% for the placebo group, while the sibutramine patients achieved an approximate 4% decrease in weight from baseline: for those who completed the trial, weight loss increased slightly to 4.5%. Total body weight loss of 5% and 10% was achieved by 33% and 8% of study completers on sibutramine, respectively, and not in any placebo patients (Figure 1).⁵ Moreover, those who achieved at least 5% or 10% total body weight loss from baseline had a greater reduction in A1C (0.53% and 1.65%, respectively) and greater mean change from baseline in fasting plasma glucose relative to placebo. Clinicians must remember, however, to be diligent in monitoring blood sugar, to assure patients don’t become hypoglycemic; especially when the patient is taking a concomitant sulfonylurea.

Combination Therapy with an Amylin Mimetic

Combining pharmacotherapy with different but complimentary mechanisms of action may improve weight loss. A recent 4-arm, randomized, placebo-controlled multicenter trial evaluated the weight-loss effect and safety and tolerability of co-administration of pramlintide with sibutramine or phentermine in overweight or obese subjects.⁶ As previously discussed, sibutramine is approved for the long-term management of obesity. Phentermine, approved in 1959 for the short-term management of obesity, acts through the monoaminergic system (norepinephrine, dopamine) to decrease food intake and body weight, without a substantial metabolic effect.^3,7-10 In addition, all participants participated in the LEARN lifestyle intervention that included counselling on diet, physical activity, behavioral modifications, and a recommended diet that provided an approximate 500 kcal/day deficit. After a 1-week placebo run-in period, patients were randomized to treatment for 24 weeks.

Figure 1. Sibutramine in Diabetes – Weight Loss Weight loss with sibutramine. Source: Fujioka K, Lee MW. Nutr Clin Pract. 2007;22:50-54.

Starting at week 2, weight loss in both combination groups was significantly greater compared with placebo or pramlintide alone (Figure 2). By week 24, both combination groups lost an average of approximately 11.3 kg (P<0.001), compared with patients in the placebo group ( 2.2 kg) and the pramlintide group (3.7 kg).

Figure 2. Increased Weight Loss with Combination Treatment Combination therapies can improve weight loss. Source: Aronne LJ, et al. Diabetes. 2008; 57(Suppl 1):A30.

In this study, 24-hour ambulatory blood pressure and heart rate was also monitored. Diastolic blood pressure and heart rate were significantly elevated with both combination treatments compared with placebo and pramlintide alone. Additionally, systolic blood pressure was significantly higher in subjects treated with pramlintide plus phentermine compared with subjects treated with pramlintide alone. The elevations in blood pressure and heart rate associated with the combination treatments are consistent with previous results with sibutramine and phentermine used individually.

In summary, patients who are placed on diabetic medications that also assist in weight loss are more apt to be adherent to therapy, and the degree of weight loss correlates with improvements in A1C. Pharmacotherapy can help reduce weight in diabetics and should be considered as part of the armamentarium for healthcare providers who treat obese diabetic patients. Clinicians should be aggressive when managing weight in the diabetic patient and attentive to medication side effects. Consideration should be given to combination therapies for improved weight loss.

References

1. Dluhy RG, McMahon GT. N Engl J Med. 2008;358:2630-2633.
2. Nathan DM, et al. Diabetes Care. 2006;29:1963-1972.
3. Fujioka K, Lee MW. Nutr Clin Pract 2007;22:50-54.
4. Finer N. Int J Obes Relat Metab Disord. 2002;26(Suppl. 4):S29-33.
5. Fujioka K, et al. Diabetes Obes Metab. 2000;2:175-87.
6. Aronne LJ, et al. Diabetes. 2008; 57(Suppl 1):A30.
7. Chapman I, et al. Diabetologia. 2005;48:838-848.
8. Ioannides-Demos LL, et al. Drugs. 2005;65:1391-1418.
9. Kruger DF, Gloster MA. Drugs. 2004;64:1419-1432.
10. Neff LM, Aronne LJ. Curr Artheroscler Rep. 2007;9:454-462.

Discussion

In the case, the patient’s medications are metformin, glyburide, fenofibrate, a statin for lipids, and hydrochloro-thiazide and an ACE inhibitor for his blood pressure, but there is no weight loss medication. (1) Would you want to put him on a weight loss medication? (2) Would you want to change his glyburide?

Ken Fujioka, MD: I would not use phentermine or sibutramine, because they pose too much risk in this cardiac patient. Glyburide is the first medication I would remove, and I would add an agent that promotes weight loss and improves glucose control, like pramlintide or exenatide.

The Look AHEAD trial used orlistat for some of the patients who had trouble maintaining weight loss. Would you consider orlistat an option?

Fujioka: I do not often use orlistat because it is not an effective weight loss agent. In this patient, orlistat could be considered later, but I would first work with the patient—refer him to a dietitian to help improve his eating behavior.

How would you manage a patient on exenatide who loses a significant amount of weight, but has an increase in A1C of 1% to 1.5%?

Fujioka: If A1C is increasing, that is generally a compliance issue. Either the patient is overeating or, more likely, he or she is not taking the injections. This is where a physician must be proactive in managing the patient, for example by showing them the dosage pen, explaining how it works, and discussing expectations.

Noncompliance with other medications should also be considered. The patient may be compliant with their exenatide regimen but noncompliant with other diabetes drugs. Because the patient loses weight, he or she may stop taking other diabetes drugs and A1C subsequently increases.

How do we diagnose obesity? Who should we screen for obesity? What are the risks?

George A. Bray, MD, MACP: I believe BMI is still the metric of choice. BMI combined with a visual inspection of the patient should be sufficient to provide a good indication of obesity and relative risk of morbidity and mortality; waist circumference is a good criterion to supplement BMI. Studies suggest that, for cardiovascular risk and for diabetes, waist circumference or waist-to-height ratio is a better measure than BMI for the prediction of future risk. At the moment, however, from a treating practitioner’s point of view, I am unaware of another metric that works as well and is as easy to measure as BMI.

Steve R. Smith, MD: The clear risks of obesity are metabolic risks, type 2 diabetes, and cardiovascular disease, and endocrine disorders such as polycystic ovarian disease must also be considered. Also, we should not forget about the increased risk of cancer attributable to obesity, and the effect of orthopedic issues. The definition of the health risks associated with diabetes should be expanded.

Fujioka: I agree. Waist circumference is an efficient measure of risk. However, it is difficult to get doctors to do it. Regarding screening for obesity, it should be done by whomever the patient sees most often, whether it is his or her nurse practitioner, primary care physician, internist, or family physician. Many women see only their OB/GYN; therefore, the OB/GYN should be screening and making the diagnosis for those women. Only one-third of patients who clearly are overweight or obese are told they are “obese” or “overweight.”

Should schoolchildren be required to have their BMI determined and reported to their parents?

Bray: I think it’s a great idea to screen, and it is even better to give parents feedback. We give feedback about eyesight and hearing, and we should do the same for overweight and obesity. I believe that not screening children borderlines on irresponsible, given the increase in prevalence of childhood obesity. It is easy to measure and can be communicated in nonjudgmental language.