Obesity management among patients with type 2 diabetes and prediabetes: a focus on lifestyle modifications and evidence of antiobesity medications
ABSTRACT
Introduction: The prevalence of obesity has increased over the past three decades in the United States and worldwide. This article reviews landmark trials for lifestyle modifications as well as clinical evidence and implications for locaserin, phentermine with topiramate, buproprion with naltrexone, and liraglu- tide in patients with type 2 diabetes and obesity.Areas covered: A MEDLINE search, from 1970 to May 2017, was conducted using key search terms – lifestyle modifications, antiobesity medication, obesity and diabetes. Published clinical trials, in the English language, with the adult patient population of type 2 diabetes or prediabetes, were reviewed and critiqued.Expert commentary: Lifestyle modifications have shown to prevent the progression to type 2 diabetes mellitus. Pharmacologically, each medication has proven effect on both type 2 diabetes mellitus and obesity, specifically reducing weight from baseline by 4.7 kg, 10.2 kg, 5.0 kg, and 6.4 kg with lorcaserin, phentermine with topiramate, bupropion with naltrexone, and liraglutide, respectively. The most efficacious medication is phentermine with topiramate, but liraglutide has long-term evidence, up to 3 years, particularly in patient with prediabetes. A risk-benefit analysis should be completed to determine which specific medication should be initiated for a patient with type 2 diabetes
1.Introduction
According to the Obesity Medicine Association, obesity is defined as a ‘chronic, relapsing, multi-factorial, neurobehavioral disease’ [1]. Specifically, obesity is a disease state, particularly when an individual has excessive body fat; an endocrine or immune dysfunction resulting in metabolic syndrome; or has physical forces that cause stress to other tissues [1].In clinical practice, obesity is commonly defined by the body mass index (BMI), which accounts for weight over height (kilogram [kg] per meter squared [m2]). A BMI between 25 and 29.9 kg per m2 places the individual into the overweight classification, whereas a BMI of 30 kg per m2 or above cate- gorizes the individual in the obesity classification [1]. There are other measurements of weight to determine obesity, such as waist circumference, waist-to-hip ratio, and/or percent body fat. It is important to use a consistent measurement for weight in clinical practice. In addition, each weight measurement has different sensitivity and specificity in terms of correlation with metabolic syndrome [1]. Clinical trials for obesity include change in BMI and/or waist circumference from baseline to the end of the study period as secondary end points. The prevalence of obesity has doubled in 70 countries, since 1990, with an estimated 603 million adults affected by this dis- ease state. Based on worldwide statistics, a mortality rate of 4 million deaths – most deaths due to cardiovascular disease – is associated with a higher BMI [2]. In the United States, it is estimated that the overall prevalence of obesity for adults is 36.5% [3]. This estimation was based on data collected over a 3-year period (2011–2014).
This crude estimate determines that one in every three adults is obese in the United States [3]. However, two in every three adults is overweight [3]. In the United States, the prevalence of obesity has steadily increased over the past three decades [3,4]. As the incidence and preva- lence of obesity have increased, there also has been an increase in the number and percentage of adult Americans diagnosed with diabetes since the year 1958 to 2014 [4]. According to 2012 data, approximately 86 million Americans have prediabetes, which is an increase from 79 million in the year 2010 [5]. Worldwide, the prevalence of obesity has doubled since 1980 with 2014 data indicating that 1.9 billion and 600 million indivi- duals are overweight and obese, respectively [6].Obesity has been linked to metabolic issues; however, some individuals have a BMI above 30 kg/m2 without weight-related health consequences (e.g. hyperlipidemia, hypertension, diabetes) [7]. More research is needed to determine predictors among individuals with obesity who are considered metabolically healthy [7]. There is a connec- tion between obesity and type 2 diabetes, as excessive body weight accompanied by excess adiposity can result in insulin resistance and metabolic syndrome [8,9]. Weight loss among patients with type 2 diabetes and obesity can not only improve metabolic end points, including glucose levels, but also lower the risk of progression in patients with prediabetes [1,10–13]. The purpose of this article is to iden- tify and summarize landmark clinical trials for lifestyle mod- ifications and clinical evidence of newer anti-obesity medications, since 2012, in the long-term management of obesity among adult patients with type 2 diabetes mellitus and prediabetes.
2.Lifestyle modifications
For obesity management, the ultimate goal is to reduce obe- sity-related mortality and morbidity. It is also desired to pre- vent further weight gain while reducing body weight by 5– 10% from baseline. The general desired weight loss goal can be initially 5% from baseline [1,10–13]. In patients with obesity and diabetes, loss of more than 3% of total bodyweight can be considered satisfactory and thus clinically meaningful as opposed to the general definition of more than 5% of total bodyweight in patients without diabetes [14]. If this goal is achieved, then 10% from baseline can be the next step as large weight losses produce greater benefits [1]. The three main components for a successful weight loss program focus on lifestyle modifications and consist of healthy food choices, physical activity, and behavioral modifications. [1,10–13]In terms of nutritional approaches and strategies, there are several options that can be individualized for the patient. Specifically in patients with type 2 diabetes, low carbohydrate intake can improve glucose levels, hemoglobin A1c (HbA1c) concentrations, and triglycerides [15,16]. Low carbohydrate intake is defined as 50–150 g per day, as compared to a very low carbohydrate diet of 50 g or less per day.
A patient with type 2 diabetes or prediabetes and obesity could focus on increasing intake of low glycemic index foods or reducing daily carbohydrate intake [15,16]. Aerobic activity is recom- mended for all patients with obesity. The recommended amount of aerobic activity is 150 min per week of moderate- intensity activity or 75 min per week of vigorous-intensity activity [17]. Aerobic activity has been shown to reduce weight while improving metabolic end points and preventing further weight gain [17]. To prevent further weight regain and pro- mote more weight loss, aerobic activity can be increased in terms of frequency or time to more than 300 min per week of moderate-intensity or 150 min per week of vigorous-intensity [17]. Anaerobic activity is considered resistance or strength training and often more useful in reducing percent body fat and building muscle [18]. Leisure time and occupational activ- ities can be accounted for as a portion of a patient’s physical activity regimen [19,20]. In patients with diabetes, prolonged sitting should be discouraged and interrupted every 30 min with short bouts of physical activity [21]. Overall, individuals with type 2 diabetes or prediabetes should be encouraged to engage in physical activity based on type, time, intensity, enjoyment level, and frequency. Behavioral modifications should be individualized for a patient with type 2 diabetes or prediabetes and obesity. It is important to have adequate assessment of the patient’s current and daily behaviors before providing individualized recommendations to overcome beha- viors or barriers [1].
3.Landmark trials for lifestyle modifications
The initial publication of the Diabetes Prevention Program (DPP) studied the effects of lifestyle modifications or treatment with metformin in preventing or delaying type 2 diabetes [22]. The study included both males and females of different ethnic groups who were at high risk for diabetes. This risk included a BMI of 24 kg/m2 or higher (22 kg/m2 or higher for Asians), fasting plasma glucose (FPG) of 95–125 mg/dL and 2 h after a 75-gram oral glucose load of 140–199 mg/dL. Eligible patients were placed into three groups: standard lifestyle recommen- dations plus metformin titrated to a goal of 850 mg twice daily; standard lifestyle recommendations plus placebo; or intensive lifestyle modifications. Participants were followed for an average of 2.8 years [22].Results were statistically significant for the primary out- come of diabetes prevention. The crude incidence rates were 11.0, 7.8, and 4.8 cases/100 person-years in the placebo, met- formin, and intensive lifestyle groups, respectively. There was a 58% reduction in the incidence of diabetes in the intensive lifestyle vs. placebo group, a 31% reduction in the metformin vs. placebo group, and a 39% reduction when intensive life- style was compared to metformin. Dietary changes and exer- cise were implemented across all study groups, with the intensive lifestyle recommendation having the greatest impact on the patient population. As a specific example, daily energy intake decreased with a mean (±SE) of 249 ± 27 kcal in the placebo, 296 ± 23 kcal in the metformin, and 450 ± 26 kcal in the intensive lifestyle modification group. Overall, the results were most likely due to weight loss among the groups – 0.9 kg with placebo, 2.1 kg with metformin, and 5.6 kg with intensive lifestyle interventions (p < 0.001) [22].
The DPP study showed that intensive lifestyle modifications can lead to prevention of diabetes, greater decrease in BMI, and overall increase in health outcomes. The study is applic- able to patients with diabetes as there would be a substantial reduction in weight and an overall increase in health for obese patients as type 2 diabetes can be prevented or delayed in high-risk patients with intensive lifestyle intervention and met- formin [22].
The 10-year follow-up of diabetes and incidence and weight loss in the Diabetes Prevention Program Outcome Study (DPPOS) sought to see the long-term effects of the DPP [23]. All partici- pants in the DPP program were eligible for follow-up. From randomization to the follow-up study, cumulative incidences were lower in the lifestyle and metformin groups compared to placebo. However, during the DPPOS period, the incidences did not differ significantly between the groups. During the 10-year period, the intensive lifestyle group originally lost weight, then partly regained weight although still weighing about 2 kg less than weight at randomization. The metformin group maintained their weight loss, of about 2.5 kg, throughout the 10-year study period . Patients aged 45 years and older had more sustained weight loss from randomization throughout the DPPOS than those less than 45 years of age. Patients between 60 and 85 years of age at randomization lost weight throughout the DPPOS study in both the metformin and placebo groups. This data suggested that there is marginal benefit in the long term with lifestyle intervention alone since patients initially lost weight, but failed to maintain it. In obese patients, long-term weight loss may be more efficacious when lifestyle intervention is added to other types of therapies [23].The Finnish Diabetes Prevention Study (DPS) sought to see the difference between the effects of short- and long-term changes in diet and behavior on the prevention of diabetes [24]. Five hundred twenty-two middle-aged and overweight subjects with impaired glucose tolerance were randomized to either a usual care or intensive lifestyle intervention group. The usual care group received usual instructions for diet and exercise while the intensive group received personalized, face-to-face sessions with a nutritionist periodically.
The intervention group received specific guidance on topics such as diabetes risk factors, saturated fat, fiber, physical activity along with practical goals for each patient. Dietary advice was based on a 3-day food record that patients had assembled at baseline. Use of a very-low-calorie diet (VLCD) for 2–5 weeks or as a substi- tute was allowed for patients to increase their weight loss. Exercise intervention was individualized per patient’s fitness capacity by a nutritionist during the counseling sessions. The intervention group was recommended to increase aerobic capacity and cardiorespiratory fitness. These activities were supervised, progressive, and individually tailored for each patient. Results of the study show a mean reduction of 4.5 kg in the intervention group vs. 1.0 kg in the control group at year 1. Significant greater improvements in FPG, 2-hour plasma glucose, and HbA1c were also seen in the intervention group vs. control in year 1. The study shows the benefits of intensive lifestyle modifications tailored to patient’s goals and treatment. The beneficial changes include patients’ diet, physical activity, blood glucose, lipid concentrations, and a highly significant reduction in diabetes incidence. For patients with prediabetes and obesity, similar benefits may be seen with weight reduction as it may help glucose tolerance and achieve better health outcomes [24].The China Da Qing Diabetes Prevention (CDQP) Study sought to see the long-term effects of lifestyle intervention on the prevention of diabetes, macro- and microvascular complica- tions, and mortality [25]. The study began in 1986 with 577 adults from various clinics in China, randomized to a lifestyle intervention group which included either diet, exercise or diet plus exercise. Weight loss occurred in both groups and was 0.9 kg in the control group vs. 1.9 kg in the intervention group.
The cumulative incidence for first cardiovascular event was 41% in the lifestyle intervention group, compared to 44% in the control group. The cumulative cardiovascular mortality (12% versus 17% in the control group) as well as the cumulative incidence for all-cause mortality (25% versus 29% in the control group) was lower in the lifestyle intervention group. All three outcomes were not statistically significant, but the study had limited statistical power [25].The CDQDP study showed a reduction in diabetes inci- dence when active interventions were provided to participants [25]. Compared to control, the intervention group had a 51% lower incidence (hazards ratio [HR] 0.49; 95% confidence inter- val [CI] 0.33–0.73) in the active intervention group between 1986 and 1992 and a 43% lower incidence (HR 0.57; 95% CI 0.41–0.81) over the 20-year period. Overall, the study showed that diabetes can be prevented with lifestyle interventions for up to 20 years. Although the study did not detect a statistical difference in first cardiovascular event, cumulative cardiovas- cular mortality, or all-cause mortality, further studies are war- ranted to determine the benefit of lifestyle intervention for cardiovascular outcomes. The findings suggest that early life- style interventions might play a change in behavior, allowing patients to have better long-term outcomes. The effects on insulin secretion or sensitivity with lifestyle interventions may also be maintained over a long period of time, as it would delay the onset or diagnosis of diabetes [25]. After 23 years of follow-up, the CDQDP study showed a reduced incidence of diabetes (72%, p = 0.001), cardiovascular mortality (11.9%, p = 0.033), and mortality (28.1%, p = 0.049), compared to placebo [26]. This long-term evidence indicates that lifestyle interventions can reduce the risk of mortality and metabolic disease [26].
On the basis that intensive lifestyle intervention may produce weight loss, the Look AHEAD (Action for Health in Diabetes) sought to see if weight loss produced any benefits for cardi- ovascular health [27]. Patients were randomized to an inter- vention group composed of decreased calorie intake and increased physical activity or a control group with diabetes support and education. The planned follow-up for the study was 13.5 years, but was stopped short because of futility analysis. The authors suggested that the study either lacked power or the weight loss in the intervention group was not at a threshold to significantly affect cardiovascular outcomes. However, the data does suggest some positive outcomes for patients with type 2 diabetes who were obese. The intensive lifestyle group had significantly greater decreases in weight (p < 0.001) and waist circumference (p < 0.001) while fitness (p < 0.001) improved over 10 years, compared to the control group. In the first year, the intervention group produced a mean 8.6% weight loss vs. 0.7% in the control group. Throughout the trial, the intervention group produced a mean 6.0% weight loss vs. 3.5% in the control. While the Look AHEAD trial did not show a reduction in mortality, it did show that a 5% or 10% weight loss at year 1 can lead to maintenance of weight loss at year 4, indicating long-term success. The study also concluded that intensive lifestyle inter- ventions produced meaningful reductions in hemoglobin A1c, especially in the first year of the study. Of note, the study showed some partial remission of diabetes for the participants receiving lifestyle interventions, indicating positive short-term and long-term benefits for patients with type 2 diabetes and obesity [27]. More recently, a post hoc analysis of the Look AHEAD trial associated greater weight loss (≥10%) with reduc- tion in the incidence of cardiovascular events [28]. This finding validates the importance of sustained weight loss over a long period of time with lifestyle modifications.
4.Anti-obesity medications
When considering pharmacological therapy for obesity manage- ment, it is important to remember that medications are only indicated as adjunct therapy with non-pharmacological interven- tions. There are advantages and disadvantages for each estab- lished or newer medication for short-term or long-term weight loss management. Table 1 summarizes the pharmacology of the newest agents since June 2012; Table 2 provides clinical gui- dance for these four anti-obesity medications [29–33]. Overall, patients that have a BMI above 30 kg/m2 or a BMI above 27 kg/ m2 with a weight-related comorbid condition can be considered as candidates for pharmacological therapy [1,10–13]. Before a medication can be approved for obesity or weight loss manage- ment, the US FDA has mandated that the active medication show a greater than 5% placebo-subtracted difference [23]. Another outcome could be more than 35% of patients lose 5% of weight from baseline or double the portion of patients losing 5% of weight from baseline, in comparison to the placebo group [34]. These outcomes need to be statistically significant for considera- tion of approval. The FDA also provides guidance on duration of obesity trials, particularly in the evaluation of adverse effects or special patient populations (e.g. pediatric).
In the United States, there are several sets of guidelines for the management of obesity. Some guidelines vary in terms of aggressive treatment as well as the target for weight loss [1,10– 13,35]. For example, the American College of Cardiology, American Heart Association, and The Obesity Society focus on BMI as the target and have a low benefit-to-risk analysis and cost-effectiveness in terms of using pharmacologic treatment (i.e. less aggressive) [10]. In comparison, the American Association of Clinical Endocrinologists focus on weight-related complications and utilize a high benefit-to-risk analysis and cost-effectiveness in terms of pharmacologic treatment (i.e. more aggressive) [13].Focusing on the FDA-approved dosing, lorcaserin 10 mg by mouth twice daily does have evidence for obesity manage- ment in three Phase III clinical trials [36–38]. Focusing on patients with diabetes, lorcaserin was investigated in the BLOOM-DM study in adult patients (18–65 years) if they had a diagnosis of type 2 diabetes treated with metformin with or without a sulfonylurea [38]. The patients had to have HbA1c between 7% and 10% and BMI of 27 to 45 kg/m2. Lastly, the patients needed to be able to complete lifestyle modifications. Similar to the other two lorcaserin trials, education and coun- seling on lifestyle modifications were given to every patient, which included a reduction of caloric intake by 600 calories per day as well as encouragement of physical activity for 30 min per day. Patients were randomized to one of three groups which included lorcaserin 10 mg by mouth twice daily (n = 256), lorcaserin 10 mg by mouth once daily (n = 95), or placebo (n = 252) for 1 year. It is important to note that the authors used 84% power and needed 250 patients in each group, when calculating a 40% dropout rate [38].
Of the lorcaserin twice-daily group, 66% of the patients completed the study; 8.6% of the patients dropped out due to an adverse event [38]. Within the placebo group, 62.1% completed the study. The average patient was female, Caucasian (61%), middle-aged (mean: 53 years), with an aver- age BMI of 36 kg/m2. Eighty-two percent of the patients had a HbA1c value less than 9% (mean = 8.1%) at baseline with a fasting blood glucose (FBG) level of 159 mg/dL. Ninety-two percent of the participants were receiving metformin while 50% were receiving a sulfonylurea. Overall, lorcaserin 10 mg twice daily was significantly better in terms of weight loss from baseline and achievement of 5% or 10% weight loss from baseline, compared to placebo (p < 0.001 for three primary outcomes). Table 3 summarizes the specific data points for primary outcomes. In regard to the secondary out- comes, lorcaserin 10 mg twice daily produced a 27.4 mg/dL reduction, compared to 11.9 mg/dL reduction with placebo for FBG levels (p < 0.001). In addition, lorcaserin produced approximately a 1% reduction in HbA1c concentrations, com- pared to 0.4% reduction in the placebo group (p < 0.001). Patients receiving lorcaserin were more likely to experience hypoglycemia (29.3% vs. 21%, respectively), than placebo. Among the patients experiencing hypoglycemia, a majority of the patients were taking a sulfonylurea [38].From the BLOOM-DM study, there was a lack of a dose response between the once-daily and twice-daily lorcaserin groups, which was a different finding from the BLOSSOM study [37,38]. There was no difference in baseline characteristics, but the lack of a dose response could be due to an older population and a larger portion of men in the BLOOM-DM study than in the BLOSSOM study [37,38]. The BLOOM-DM study specifically focused on patients with type 2 diabetes while the other lorca- serin studies focused on overweight or obese adults with hyper- tension, hyperlipidemia, cardiovascular disease, impaired glucose tolerance, or sleep apnea. There was no report of chan- ging antidiabetic agents during the BLOOM-DM study, but one would assume that sulfonylurea dose was reduced due to hypo- glycemia. Overall, the BLOOM-DM provided evidence for lorca- serin that it can lead to weight loss among patients with type 2 diabetes, leading to improved glucose homeostasis. Therefore, patients with type 2 diabetes may become less insulin-resistant, based on the incidence of hypoglycemic episodes [38].
Phentermine with topiramate has evidence in weight reduction from three clinical trials [39,44,45]. Focusing on obese patients with comorbidities, the 1-year CONQUER study was a rando- mized, placebo-controlled trial to determine the efficacy of con- trolled-release phentermine plus topiramate. [39] These patients were 18–70 years of age, with BMI of 27–45 kg/m2, two or more comorbidities, and waist circumference indicating obesity. Patients were randomized to either placebo (n = 979), low-dose phentermine 7.5/topiramate 46 mg (n = 488), or high-dose phentermine 15/topiramate 92 mg (n = 981). All patients were educated and counseled on lifestyle modifications following the Lifestyle, Exercise, Attitudes, Relationships, and Nutrition (LEARN) manual and were recommended a daily caloric reduction of 500 calories per day [39].Discontinuation rates were 36% in the 15/92 mg group, 31% in the phentermine 7.5/46 mg group, and 43% in the placebo group [39]. Discontinuation rates due to adverse events were lower with the phentermine and topiramate groups (3.0% with 15/92 mg), compared to placebo (12.8%). The average patient was female, Caucasian (86%), middle-aged (mean 51 years), with an average BMI of 36.4 kg/m2. Sixty-two percent had diabetes mellitus or impaired glucose tolerance. Overall, primary out- comes were all statistically significant in the 15/92 mg group, when compared to placebo (see Table 3). In a subgroup analysis of patients with type 2 diabetes, the average weight change was −8.8% with 15/92 mg of phentermine/topiramate, whereas patients in the placebo group lost approximately 1.9% from baseline. Secondary outcomes of waist circumference, systolic blood pressure (SBP), FBG, and HbA1c were statistically signifi- cant in the 15/92 mg of phentermine and topiramate groups vs. placebo (p < 0.0001). From the CONQUER study, 68% of patients had impaired glucose tolerance or diabetes. The maximum dose of phentermine and topiramate resulted in fewer patients progressing to diabetes (relative risk [RR] 0.47; 95% CI 0.25– 0.88) vs. placebo (RR 0.78; 95% CI 0.4–1.5). In patients with type 2 diabetes or prediabetes, there was a statistically significant improvement with phentermine 15 mg/topiramate 92 mg, com- pared to placebo (p < 0.001) [39].
The SEQUEL trial was a year-long extension of the CONQUER trial [45]. Patients were eligible if they completed the CONQUER trial; enrollment was in the original assigned group of placebo, the moderate–high dose of phentermine and topiramate, or the maximum dose of phentermine and topiramate. Of those who completed the CONQUER trial, 78.1% continued for an addi- tional 52-week period. Patients’ mean weight loss was 10.5% in the maximum dose group vs. 1.8% in placebo group (p < 0.0001). More patients achieved ≥5% weight loss with phentermine and topiramate (79% vs. 30%, p < 0.0001 when compared to placebo). The annualized incidence rate for patients without diabetes at baseline were 3.7% and 0.9% in the placebo and maximum-dose phentermine and topiramate groups, respectively. The data indicates a 54% risk reduction for the moderate-dose and 76% reduction with the maximum dose of phentermine and topiramate. Although not FDA-approved for the prevention of diabetes, the end result of the study does suggest benefit in this area [45].The OB-202 study was a double-blind, placebo-controlled trial to assess the safety and efficacy of phentermine and topiramate in patients with obesity and type 2 diabetes [46]. Patients’ ages ranged from 18 to 70 years with BMI of 27 to 45 kg/m2 and HbA1c of 7% to 12%. Those who had SBP >150 mmHg, diastolic blood pressure >95 mmHg, history of glaucoma, or participated in a formal weight-loss program were excluded. Patients were actively treated and managed for their comorbidities, including options to add, remove, or adjust medications for type 2 diabetes, hypertension, and/or dyslipidemia. All subjects received lifestyle intervention such as daily exercise along with reduced intake of 500 calories per day. The average patient had diabetes for 9 years, with 60% taking metformin, compared to 32.3% taking a sulfo- nylurea, and a mean baseline HbA1c of 8.7%. Patients were also excluded from trials if on insulin or glucagon-like peptide-1 receptor agonist. Those who tolerated the 28-week trial were eligible to enroll in the DM-230 28-week continuation trial. The DM-230 patients were continued in the original assigned group from the OB-202 study. Seventy-nine percent of patients com- pleted the OB-202 trial while 12% discontinued due to noncom- pliance. One hundred thirty patients enrolled in the DM-230 extension trial with 92.3% completing the study. The most com- mon reason for discontinuation was loss to follow-up (3.8%) [46].
The primary outcome of the OB-202/DM-230 trial was change in HbA1c after 56 weeks [46]. The phentermine 15 mg and topir- amate 92 mg groups had a greater decrease in HbA1c of 1.6%, compared to 1.2% reduction in the placebo group (p < 0.05). Secondary end points included percent of patients achieving HbA1c goal of ≤7.0% (53% treatment vs. 40% placebo; p < 0.05) and percent achieving HbA1c goal of ≤6.5% (32% treatment vs. 16% placebo; p < 0.05). FBG change was approximately −42 mg/dL with treatment (−27 mg/dL with placebo; p < 0.05). The mean weight change from baseline was −9.6% in treatment vs. −2.6% in placebo (p < 0.0001). Sixty-five percent of patients in the treatment group achieved 5% weight loss vs. 24% in placebo (p < 0.0001). Following this trend, 37% in the treatment group achieved 10% weight loss vs. 9% in the placebo (p < 0.0001). Fifty-eight hypogly- cemic events were recorded during both trial periods (5 in placebo vs. 12 with phentermine and topiramate). These events were most likely related to concomitant antidiabetic agents in which 90% of events included a sulfonylurea. Overall, the trials provided great evidence for adjunctive pharmacotherapy to intensive lifestyle modification for long-term weight management and improve- ment in metabolic end points [46]. Overall, phentermine and topiramate, as combination ther- apy for weight loss, provided patients with type 2 diabetes and prediabetes with an option for improvement in metabolic end points, such as weight, FBG, and HbA1c. In addition, patients may be able to reduce the dose or discontinue main- tenance medications if weight-loss goals are achieved. Bupropion 360 mg plus naltrexone 32 mg sustained-released has four placebo-controlled trials, demonstrating efficacy for weight loss among patients who are overweight or obese [40,47–49]. The COR-DM study is specific for patients with type 2 diabetes and obesity [40]. To be included in the trial, these patients were 18–70 years of age, with a BMI of 27–45 kg/m2, HbA1c of 7–10%, and FBG less than 270 mg/dL. Patients were educated and counseled on lifestyle modification, based on the American Diabetes Association booklets, and were recommended physical activity of at least 30 min a day and a daily reduction of 500 calories. Patients enrolled were not taking any oral antidiabetic medications or had to be stable on them for ≥3 months prior to randomization. Most medications for hypertension and dyslipi- demia were allowed as long as they were stabled for at least 4 weeks prior to randomization. These patients were randomized in a 2:1 fashion to either the maximum dose of bupropion and naltrexone (360/32 mg per day) or placebo. Assuming a 33% attrition rate, 350 patients were needed, and the study was set at a 99% power and two-sided significance level of 0.05 to detect a difference in mean weight loss between groups. However, 335 patients were included in the final analysis for the bupropion and naltrexone group, compared to 170 patients in the placebo group. The average patient was a 54-year-old Caucasian (~80%), with a BMI of 36.4 kg/m2 and mean HbA1c of 8.0%. More than 50% of patients were females and taking metformin (77%), sulfonylurea (48%), or thiazolidinedione (31%) for diabetes management [40].The primary outcomes from the study were all statistically significant (see Table 3) [40]. Secondary outcomes indicated favored treatment with bupropion and naltrexone as patients in this group had a lower waist circumference, decreased SBP, FBG, and HbA1c. However, only waist circumference (p = 0.006) and reduction in HbA1c (p < 0.001) were statistically significant. During the study, fewer patients in the treatment group required an increase in dose or addition of oral antidiabetic drugs vs. placebo (22.3% vs. 35.2%, respectively, p < 0.01). Other outcomes in the study included favorable change in HDL-cholesterol (+3.0 mg/dL in treatment vs. −0.3 mg/dL in placebo, p < 0.001) and triglycerides (−11.2 mg/dL in treatment vs. −0.8 mg/dL in placebo, p = 0.007) [40]. The drug’s efficacy for obesity management may be clouded in controversy due to cardiovascular evidence [50]. Based on the LIGHT trial at 25% completion, it showed a 41% decrease in major adverse cardiovascular events such as cardiovascular death, non- fatal stroke, or myocardial infarction (HR 0.59, 95% CI 0.39–0.90), and the sponsor company filed a claim for cardiovascular benefit [50]. However, a second analysis at 50% completion shows no benefit between treatment and placebo (HR 0.88, 95% CI 0.57– 1.34) [50]. The trial was terminated before completion, and a final analysis at 64% completion revealed no benefit between treat- ment and placebo groups as well (HR 0.95, 95% CI 0.65–1.38) [50]. The controversy with premature public disclosure of trial results pre-empted the FDA to mandate for more cardiovascular evi- dence. For patients with type 2 diabetes and obesity, bupropion and naltrexone may be a viable option for long-term weight management, but there may be underutilized due to speculated cardiovascular risk among a high-risk patient population.Among all of the anti-obesity medications, liraglutide has the most evidence for weight loss among patients with obesity and other comorbid conditions such as diabetes and prediabetes [41,42,51–54]. In the SCALE Diabetes study, liraglutide was eval- uated in adult patients with not only a stable weight, but also a BMI equal to or above 27 kg/m2 [41]. Patients had a diagnosis of type 2 diabetes with a HbA1c of 7–10% with one to three oral antidiabetic agents. If a patient was taking a sulfonylurea, the dose was decreased by 50% to reduce the risk of hypoglycemia. All patients received education and counseling on balanced meals of 30% fat, 20% protein, and 50% carbohydrates. In addi- tion, all patients were encouraged to exercise more than 150 min per week and reduce caloric intake by 500 calories per day. Patients were randomized in a 1:1:2 ratio as 212 patients received placebo, 211 patients received liraglutide 1.8 mg per day, and 423 patients received liraglutide 3 mg per day for a total of 56 weeks. The average patient in the SCALE Diabetes study was middle-aged (55 years), female (48.5%), Caucasian (83%), with a mean baseline BMI of 37 kg/m2. In the placebo group, patients had a slightly shorter history of diabetes (mean: 6.7 years) com- pared to the liraglutide 3 mg group (mean: 7.5 years). Fifty percent of the patients were taking metformin, comparatively to 21% taking metformin and sulfonylurea at baseline [41]. For the primary outcomes, please refer to Table 3 for specific results for liraglutide 3 mg and placebo [41]. For secondary out- comes, liraglutide did statistically improve waist circumference, SBP, FBG, and HbA1C when compared to placebo. It is important to note that pulse was not reported at baseline; however, liraglu- tide did increase heart rate by 2 bpm while placebo decreased heart rate by −1.4 bpm. Based on long-term evidence, arrhyth- mias may occur with liraglutide (5 cases per 100 patient years) vs. placebo (2 cases per 100 patient years) [41]. Liraglutide 3 mg was studied in the SCALE PreDiabetes study, which randomized adult patients with hypertension or hyperlipi- demia and BMI equal to or above 27 kg/m2 to a treatment or placebo group [42]. Similar to the SCALE Diabetes study, patients were educated on balanced meals, physical activity, and reduc- tion of caloric intake. Randomization took place in a 2:1 ratio in which 1244 patients received placebo compared to 2487 patient who received liraglutide 3 mg for 56 weeks. In this particular study, a majority of patients were female (78%), younger than the SCALE Diabetes study (mean age 45 years), and Caucasian (85%) [41,42]. The average BMI at baseline was 38.3 kg/m2. Based on the patient population, the mean baseline HbA1C was 5.6% as 61% of patients had prediabetes. After 56 weeks of therapy, weight loss showed improvement in metabolic end points, such as the reduction or delay in the onset of type 2 diabetes mellitus. The progression to diabetes after 56 weeks resulted in 7.2% of patients with liraglutide vs. 20.7% of patients receiving placebo (p < 0.001). The SCALE PreDiabetes study was continued for 160 weeks, in which only 2% of patients receiving liraglutide 3 mg had a diagnosis of type 2 diabetes at the end of the study, vs. 6% in the placebo group [53]. The mean time to the diagnosis of diabetes with liraglutide 3 mg in patients with prediabetes was 99 weeks vs. 87 weeks with placebo. These outcomes are most likely attributed to sustainable weight loss from baseline [53]. 5.Conclusion Obesity is a lifelong chronic disease. The motivation to start the process may not be found within certain patients. The goal in weight loss is to decrease the risk of obesity-related complica- tions and cardiovascular disease. Achievable goals and expecta- tions should be set and individualized for each patient. Tailoring to each patient is paramount because patients have different preference for foods and lifestyle activities. The hard part of obesity management is weight loss, but the hardest part is maintenance of weight loss, as seen patients regaining some weight in DPPOS. Lifestyle modifications must be part of every patient’s weight-loss plan, but chronic drug therapy may be warranted for sustained weight loss in some patients. Before selecting an adjunct therapy for patients, the efficacy and safety profile, along with the price of the anti-obesity medication, must be considered as it may deter patients from compliance. While not discussed in detail within this manuscript, bariatric surgery is an effective option for those who do not respond to lifestyle modifications and pharmacotherapy. 6.Expert commentary Based on the evidence, phentermine with topiramate is most efficacious as an anti-obesity medication for chronic weight loss in patients with type 2 diabetes. However, liraglutide 3 mg has a significant amount of evidence in this population, including those with prediabetes. It is important to consider the numbers needed to treat in comparison to the numbers needed to harm when prescribing any of these anti-obesity agents for weight loss in patients with type 2 diabetes and prediabetes (see Table 4) [43]. While an agent may be warranted at times, it is important to note that these drugs have only been compared to placebo in patients with diabetes. Liraglutide 3 mg was compared to orlistat as an active comparator group; however, this study focused on patients that were overweight or had obesity and was not limited kg: kilograms; N: no; NNH: numbers needed to harm (most common adverse event); NNT: numbers needed to treat (based on achievement of 5% weight loss from baseline); PO: by mouth; SC: subcutaneous; Y: yes. Adapted from: Belviq (lorcaserin) [prescribing information]. Woodcliff Lake, NJ: Eisai Inc; December 2014.; Belviq XR (lorcaserin) [prescribing information]. Woodcliff Lake, NJ: Eisai Inc; July 2016.; QSYMIA (phentermine/topiramate) [prescribing information]. Mountain View, CA: Vivus, Inc; October 2014.; Contrave (naltrexone and bupropion) [prescribing information]. La Jolla, CA: Orexigen Therapeutics Inc.; September 2016.; Saxenda (liraglutide) [prescribing information]. Plainsboro, NJ: Novo Nordisk; September 2016.; MacDaniels JS, Schewartz TL. Effectiveness, tolerability, and practical application of the newer generation anti-obesity medications. Drugs in Context 2016;5:212291 to a specific comorbid condition [51]. Use of these agents for obesity management can span from 1 to 2 years, in which patients lost a significant amount of weight from baseline to year 1. However, in the second year, some patients still receiving the active drug gained a little bit of weight but not to baseline weight. Liraglutide has evidence for the longest period of time of 3 years [53]. It is important to consider cost before prescribing any of these anti-obesity agents. High copayments lead to higher cost to the patient, which alternately can decrease the absolute ben- efit of these medications. If the co-pay price is too high, then it may be more meaningful for the patient to invest in nonpharma- cologic interventions (i.e. gym membership). Therefore, the true benefit of anti-obesity medication may be overlooked and not fill the gap of obesity management due to a high cost–benefit analysis. Future studies should include a variety of patients, such as those with a short or long history of diabetes. In addition, the benefit of anti-obesity medications added to antidiabetic regimens known to cause weight gain would provide informa- tion on the counteractive effect of these agents. For future trials, it would be important to consider a secondary end point of post- prandial blood glucose as the anti-obesity agents work on decreasing satiety, ultimately leading to decreased food consumption.For patients with type 2 diabetes or at risk of diabetes who continue to struggle or have an unsuccessful trial of an anti-obesity agent, it is important to consider bariatric surgery as the next high- potent, high-cost option for weight loss [10–13]. Worldwide, bar- iatric and metabolic surgeries are increasingly being performed to effectively aid weight loss in patients with severe obesity, as well as improvement in metabolic outcomes [55]. The positive effects of bariatric surgery, especially with improvement in type 2 diabetes, have expanded the eligibility criteria for metabolic surgery to patients with diabetes and BMI of 30–35 kg/m2 [55,56]. Conceptually aligned with the positive weight-independent out- comes on type 2 diabetes and its remission, the term ‘metabolic surgery’ or ‘diabetes surgery’ was coined to describe bariatric surgery in patients with less severe obesity than the traditionally eligible population [55]. Gastrointestinal interventions (i.e. sleeve gastrectomy) are currently recognized as metabolic procedures with pathophysiological effects beyond weight loss [55]. 7.Five-year view In the next 5 years, there can be more evidence regarding current and investigational medications for obesity management. The current agents, as described in this publication, focus on satiety. There are no medications that further increase energy expenditure in order to promote weight loss. While there may never be approved medications focusing on energy expenditure, it is impor- tant to consider the investigational agents that would have a different mechanism of action toward chronic weight loss. Some of these investigational agents would target ghrelin (gastrointest- inal hormone that promotes hunger); amylin (pancreatic hormone that promotes gastric emptying and satiety); and peptide YY 2 peptide (gastrointestinal hormone that supports gastric emptying and suppresses appetite). The potential combination of investiga- tional agents with currently approved agents could lead to more successful weight loss, greater than 5% or 10% from baseline. Combinations of certain medications should be considered to promote dramatic weight loss. One final agent, under research, is semaglutide once daily; semaglutide is a glucagon-like peptide-1 receptor agonist with evidence for diabetes management. However, when used at higher doses, it has been shown to be an efficacious agent for obesity management. Data and evidence for investigational ages should be summarized in the next 3 years in phase 2 and Liraglutide 3 trials.