Phentermine, topiramate and their combination for the treatment of adiposopathy (‘sick fat’) and metabolic disease

Abstract

Positive caloric balance often causes pathologic adipocyte and adipose tissue anatomical and functional changes (termed adiposopathy or ‘sick fat’), which may lead to pathogenic adipocyte and adipose tissue responses and metabolic disease. Fat weight loss may improve adiposopathy, and thus improve metabolic disease in overweight patients. Unfortunately, the efficacy of non-surgical weight loss therapies is often limited due to redundant physiological systems that help ‘protect’ against starvation and/or negative caloric balance. One strategy to overcome these limitations is to combine weight loss drug therapies having complementary mechanisms of action, thereby affecting more than one physiologic process influencing body fat accumulation. Phentermine is a noradrenergic sympathomimetic amine approved for short-term treatment of obesity. Topiramate is a sulfamate-substituted monosaccharide derivative of the naturally occurring sugar monosaccharide D-fructose approved as a treatment for migraine headaches and seizure disorders. Although known to facilitate weight loss since its approval, topiramate monotherapy does not have a regulatory indication as an anti-obesity agent. Phentermine HCl/topiramate controlled-release (PHEN/TPM CR) is a combination agent containing immediate-release phentermine and controlled-release topiramate. Clinical trials involving thousands of patients demonstrate PHEN/TPM CR to be effective in improving the weight of patients, and also effective in improving adiposopathy-associated metabolic diseases. This review examines the pathophysiology of adiposopathy as a contributor to metabolic disease, the data supporting phentermine monotherapy, topiramate monotherapy and their combination as anti-obesity and anti-adiposopathy agents, and the preliminary evidence supporting PHEN/TPM CR as a generally well-tolerated and effective agent to improve metabolic disease.

Keywords::

• adiposity
• adiposopathy
• blood pressure
• cholesterol
• diabetes mellitus
• dyslipidemia
• hypertension
• lipids
• obesity
• phentermine
• PHEN/TPM CR
• Qnexa^®
• topiramate
• VI-0521

Figure 1. Simplified relationship between selected CNS factors^† on anorexigenic and orexigenic effectors within the hypothalamus.

CNS factors may increase anorexigenic POMC (which is cleaved to form melanocortins such as MSH) and CART expression, and decrease orexigenic expression of NPY and AgRP. Alterations in these appetitive effectors contribute to ‘second-order’ signaling through arcuate nucleus neuron projection to other hypothalamic regions. The net result is increased anorexigenic and increased catabolic effects, with decreased orexigenic and decreased anabolic effects. This figure is greatly simplified, and does not show all the interconnected signaling pathways between these CNS factors.

^†The reported effects of these CNS factors on POMC and NPY are not always consistent in the literature, with more research needed to confirm these effects.

^‡While CNS adiponectin may have anorexigenic effects [175], adiponectin may not cross the blood–brain barrier [176], so the physiological importance of adiponectin on the CNS is unclear [140].

^§Meant to reflect the fed versus non-fed state as reflected by relative physiologic increases or decreases in glucose levels. While the signaling may be glucose mediated, this is not necessarily meant to represent pathologic hyperglycemia.

AgRP: Agouti-related peptide; BDNF: Brain-derived neurotrophic factor; CART: Cocaine amphetamine-regulated transcript; CB1R: Cannabinoid 1 receptor; CRH: Corticotropin-releasing hormone; GLP-1: Glucagon-like peptide-1; MC3R: Melanocortin 3 receptor; MC4R: Melanocortin 4 receptor; MCH: Melanin-concentrating hormone; MSH: Melanocyte-stimulating hormone; NPY: Neuropeptide Y; POMC: Pro-opiomelanocortin; PYY 3–31: Peptide YY; TRH: Thyroid-releasing hormone; VIP: Vasoactive intestinal peptide.

Data taken from [23,173,174].

Figure 2. Relative importance of prophylactically treating migraine headaches versus treating adiposity, adiposopathy and metabolic disease.

^†Topiramate doses may be as low as 50 mg per day for prophylaxis of migraine headaches, and as high as 400 mg per day for treatment of seizure disorders.

^‡Both phentermine and topiramate are approved drugs. Neither phentermine nor topiramate have product information that warns of potential adverse experiences with their combined use.

^§The ‘low dose’ PHEN/TPM CR evaluated in clinical trials contained 23 mg of topiramate CR.

^¶It is yet to be proven that controlled-release topiramate affords additional efficacy or safety advantages over non-controlled-release formulations.

PHEN/TPM CR: Phentermine HCl and topiramate controlled-release.

Clinical overview

Adiposity promotes the most common metabolic diseases encountered in clinical medicine [1]. Many of these metabolic diseases are major atherosclerotic coronary heart disease (CHD) risk factors (e.g., diabetes mellitus, high blood pressure [BP] and dyslipidemia) [2,3]. The association between increased adiposity, metabolic disease and CHD risk is not absolute [4]. In some individuals, adiposity may elicit limited pathogenic adipose tissue endocrine and immune responses, which are made even less pathogenic if other body organs are able to mitigate these pathogenic adipose tissue responses. Such patients may develop fat-mass-related complications; however, adiposity-induced metabolic diseases may be avoided or delayed [5]. Conversely, even modest fat weight gain among genetically or environmentally susceptible patients may incite pathogenic adipose tissue endocrine and immune responses that lead to metabolic disease [6]. Thus, adiposity is most likely to ‘cause’ metabolic disease when fat weight gain results in pathogenic adipose tissue endocrine and immune adipose tissue responses in patients having other body organs (e.g., liver, muscle, brain etc.) with inadequate potential to prevent the pathologic metabolic consequences of dysfunctional adipose tissue [2].

Adipose tissue as a pathogenic organ

Clinicians often apply the suffix ‘pathos’ to describe diseased organs, such as cardiomyopathy, myopathy, encephalopathy, ophthalmopathy, retinopathy, enteropathy, nephropathy, neuropathy and dermopathy. ‘Adiposopathy’ is a term applied to reflect anatomic abnormalities of adipocytes and adipose tissue that arises from fat weight gain, and that results in pathogenic endocrine and immune adipose tissue responses adversely affecting metabolic health. In lay terms, ‘adiposopathy’ can be characterized as representing ‘sick fat’, which is an endocrine ‘disease’ [7].

Adipogenesis

The degree to which adiposity is manifest as adiposopathy largely depends on the effects of a positive caloric balance on adipogenesis. During positive caloric balance, adipocytes normally undergo increased recruitment, proliferation and differentiation; during a negative caloric balance, adipocytes shrink in size and/or may undergo apoptosis [8–13]. If, during positive caloric balance, increased adipogenesis provides adequate, added adipose tissue functionality, then patients may avoid adverse metabolic consequences associated with adiposity. Conversely, if adipogenesis is impaired, then inadequate recruitment, proliferation and/or differentiation may cause adipocytes and adipose tissue to become dysfunctional, pathogenic, and contributory to metabolic disease. Specifically, if during positive caloric balance the recruitment and proliferation of new fat cells is impaired, then existing fat cells may become excessively enlarged. For decades, data support excessive adipocyte hypertrophy as being associated with adipocyte and adipose tissue dysfunction, which contributes to metabolic disease [6,14–16].

Fat depots

The pathogenic potential of adipose tissue is not solely due to how calories are stored (adipocyte hypertrophy vs hyperplasia), but also due to where fat is stored [6]. Visceral or intraperitoneal (omental, mesenteric and umbilical), extraperitoneal (peripancreatic and perirenal) and intrapelvic (gonadal/epididymal and urogenital) adipose tissues reportedly have higher metabolic activity compared with subcutaneous, peripheral adipose tissue (truncal, gluteofemoral, mammary and inguinal) [6,17]. Due to differences in location and function, overweight patients with a disproportionate increase in visceral adiposity may have an increased risk of developing metabolic diseases [5,18–23]. Conversely, some overweight individuals who have relatively unimpaired adipogenesis predominantly in peripheral subcutaneous adipose tissue may avoid clinically meaningful adverse adipose tissue pathogenic responses, and thus remain metabolically healthy [6,18,24–27].

Although visceral adipose tissue accumulation is most often associated with metabolic disease [28], other fat depots also have pathogenic potential [6,19]. Adipose tissue has the potential for pathogenic responses from virtually all fat depots, such as significant immunologic abnormalities involving pericardial, perimuscular, perivascular, orbital and paraosseal adipose tissue [29]. During positive caloric balance, impaired adipogenesis in subcutaneous adipose tissue may lead to increased storage of fat in other depots, resulting in visceral adipose tissue accumulation and visceral adipocyte hypertrophy, which is particularly pathogenic. Furthermore, an increase in adipocyte size and adipose tissue accumulation in the subcutaneous abdominal region is most often associated with an increased, not decreased, risk of metabolic disease [6,22,30].

Free fatty acids

Circulating free fatty acid levels are determined by many factors, such as:

• • Whether the patient is in a fasting (postabsorptive) state;

• • The degree to which other body organs (e.g., muscle and liver) metabolize free fatty acids;

• • The degree to which adipose tissue stores free fatty acids as triglycerides.

If adipocytes and/or adipose tissue are impaired in their ability to adequately store free fatty acids, then the resultant increase in circulating free fatty acids may contribute to metabolic disease. Because it is the largest fat depot (making up ≥80% of body fat), subcutaneous adipose tissue produces the majority of the free fatty acids, possibly including the majority of free fatty acids in the portal system [19,31]. Nonetheless, an increase in visceral adipocyte hypertrophy and/or visceral adipose tissue often increases free fatty acid delivery to the liver through its portal drainage, which may promote hepatic ‘lipotoxicity’. The increased, lipotoxic free fatty acid delivery to the liver increases the risk of hepatic-mediated insulin resistance and dyslipidemia [32,33].

In addition to its contribution to the portal system, subcutaneous adipose tissue contributes to most of the systemic circulating free fatty acids [32,33], which are potentially lipotoxic to the muscle and pancreas, thus decreasing insulin sensitivity in muscle and possibly decreasing insulin secretion from the pancreas [34–36]. In summary, while visceral adipose tissue accumulation is most associated with metabolic disease, both visceral and subcutaneous adipose tissue have pathogenic potential [19].

Adipose tissue as an active endocrine & immune organ

Anatomically, excessive adipocyte hypertrophy (due to excessive fat content) adversely affects intracellular organelles, as evidenced by increased markers of intracellular endoplasmic reticulum stress and mitochondrial dysfunction [37,38]. This supports the concept of ‘sick fat’ [2,7], wherein adiposity lead to pathologic adipocyte and adipose tissue anatomic changes, which leads to pathogenic adipose tissue endocrine and immune responses, which then contribute to metabolic disease.

Adipose tissue is an active endocrine organ whose disruption can contribute to metabolic disease [6,7,37,39–42]. Adipose tissue is also an active immune organ whose disruption can contribute to metabolic disease [6,29,37,43–48]. Biopsy of the periumbilical subcutaneous (abdominal) region supports adiposity as increasing adipose tissue macrophage content and/or increasing markers of macrophage activation, representing potential pathogenic adipose tissue immune responses [49]. Pathogenic alterations in adipose tissue endocrine and immune factors associated with adiposopathy contribute to metabolic disease, while improvements in adipose tissue endocrine and immune factors may improve metabolic disease [6,23,25,50].

Interaction or ‘cross-talk’ with other body tissues

Adiposopathy alone does not cause metabolic disease. It is the interaction or ‘cross-talk’ with other body organs that substantially determines whether pathogenic adipocyte and adipose tissue endocrine and immune responses ultimately ‘cause’ metabolic disease [2]. Many non-adipose tissue body organs modulate the pathogenic endocrine and immune responses from adipose tissue during positive caloric balance [2]. One of the best described pathogenic responses of dysfunctional adipose tissue (‘sick fat’) is the increased net release, and thus increased circulating levels, of free fatty acids. Normally, after meals, circulating free fatty acids are substantially reduced (suppressed), due to increased insulin secretion that stimulates free fatty acid storage in adipose tissue and other body tissues. As insulin levels decrease in the hours after initial release, free fatty acid levels gradually rise [51]. If, during positive caloric balance, fat weight gain results in excessive adipocyte hypertrophy and adipose tissue dysfunction, then this may result in insulin resistance and impaired storage of free fatty acids within adipocytes. The disruptive imbalance between lipolysis and lipogenesis in adipose tissue increases both fasting and postprandial circulating free fatty acids, relative to those without insulin resistance [52]. If the liver is limited (‘inflexible’) in its ability to metabolize the increased delivery of free fatty acids, then intrahepatic lipid deposition (often reflected by ‘fatty liver’) may promote hepatic insulin resistance and dyslipidemia [53]. Conversely, if the liver is ‘hyperflexible’, in its ability to manage the pathogenic responses of adipose tissue through genetic predisposition, or through the use of therapeutic agents (e.g., peroxisome proliferator-activated receptor [PPAR]-γ agonists), then the adverse metabolic disease consequences of increased free fatty acids may be averted [2,52,54].

Similarly, some patients have an inherent or acquired impairment in the ability to metabolize intramuscular fat [53]. A lack of ‘flexibility’ on the part of skeletal muscle may result in an inability to metabolize the adiposopathic increased free fatty delivery from the circulation. If so, then this may result in ectopic free fatty acid storage in muscle and the accumulation of ‘toxic’ intramyocellular lipids such as diacylglycerol, fatty acyl CoA and ceramides [2]. This ‘lipotoxicity’ promotes insulin resistance in muscle [55–57].

Therapeutically, hypocaloric nutritional intervention would not be expected to affect the inherent ‘flexibility’ for muscle to metabolize free fatty acids. However, such intervention would be expected to improve adipose tissue function, improve free fatty acid storage in adipose tissue, reduce circulating free fatty acids, reduce lipid delivery to skeletal muscle, and thus reduce lipid deposition in skeletal muscle. As a result, reduced fatty acid delivery and reduced lipid deposition in skeletal muscle may improve its insulin sensitivity [57]. Similarly, some patients may have familial or genetic limitations in their ability to secrete insulin. The disease of Type 2 diabetes mellitus is often revealed when adiposity results in pathogenic adipose tissue endocrine and immune responses, including an increased release of free fatty acids. This exacerbates genetic limitations of the pancreas, potentially resulting in an even further impairment in insulin secretion [58,59]. In addition, adipose tissue is also known to have substantial interactions and cross-talk with other body organs, such as the immune system, heart and vasculature, brain, endocrine glands, intestine, kidneys, and other body organs, all with potential physiologic and pathologic metabolic consequences [2].

Adiposopathy

The goal of treating adiposity is to improve patient health. If the patient has fat mass-related morbidities, then fat weight loss may improve such conditions [60]. If the patient has adiposopathy-induced metabolic disease (e.g., Type 2 diabetes mellitus, high BP and/or dyslipidemia), then fat weight loss interventions may reduce adipocyte hypertrophy, reduce visceral adiposity, improve adipocyte and adipose tissue pathogenic endocrine and immune responses, and thereby improve metabolic disease (Table 1)[50,54].

However, reducing body fat is not the only way to improve adipose tissue function. Adding adipocyte or adipose tissue functionality is another approach. PPAR-γ agonists may improve metabolic diseases normally associated with an increase in body fat (Type 2 diabetes mellitus and dyslipidemia) by increasing adipogenesis (recruitment, proliferation and differentiation of new adipocytes) [25,52]. PPAR-γ agonists illustrate how adding (functional) adipose tissue may be an effective treatment for metabolic diseases otherwise associated with adiposity and adiposopathy.

Conversely, not all weight loss interventions improve metabolic disease. Antiretroviral therapy for HIV is sometimes associated with ‘HIV lipodystrophy’, wherein weight loss is manifest by a disproportionate loss of subcutaneous adipose tissue relative to visceral adipose tissue. These HIV drug therapies may promote insulin resistance and dyslipidemia [61], and illustrate how even weight loss can be associated with pathogenic adipocyte and adipose tissue dysfunction resulting in metabolic disease.

Currently, bariatric surgery appears to be the most effective intervention to reduce adiposity and adiposopathy through improvement in adipose tissue function and perhaps other mechanisms [50]. Perhaps the most cost-effective treatment for adiposity and adiposopathy in the individual, willing patient is appropriate nutrition and increased physical activity (Table 1)[2]. However, a challenge for the clinician and patient is that public health weight loss nutritional and physical activity initiatives, as well as commercial nutritional weight loss plans, have met with limited success in both individuals and populations [62]. Other weight loss therapeutic options include anti-obesity drug therapies (Table 1)[54]. Targeting weight loss is medically justified because even just a 5% weight loss in overweight patients may improve pathogenic adipose tissue responses [63]. Thus, while many overweight patients may want (and sometimes expect) therapeutic interventions that result in weight loss of up to 100 lbs or more [7], demonstrable and significant metabolic benefits can be achieved with ‘only’ a 5–10% reduction in body weight [64,65].

Weight loss challenges

Unfortunately, despite available therapeutic options, genetic, physiologic, environmental and behavioral factors impede effective management of the unabated obesity epidemic [201].

One potential physiologic impediment may be related to adipogenesis. Physiologically, during fat weight loss, a reduction in adipogenic markers does not necessarily reflect adipocyte dysfunction [23,50]. Instead, it is physiologically appropriate that adipogenic processes diminish when the need for additional adipocytes and adipose tissue is diminished, as occurs during active weight loss. However, the potential for adipogenesis may be increased, as evidenced by enhanced preadipocyte adipogenic and anti-apoptotic protein expression [66,67]. An increase in adipogenic potential may be a contributing physiologic cause for the increased potential for all-too-common fat regain among those who lose body fat.

Another important factor contributing to the lack of persistence with weight loss is that the human body is not very efficient in self-regulating against excessive energy storage. Conversely, the human body is often entirely too efficient in self-regulating against too little available energy through multiple redundant mechanisms intended to prevent starvation. It appears that conserving energy during a negative caloric balance represents a more urgent priority for survival than expending energy during positive caloric balance. While this may be advantageous in environments of food scarcity and when high physical activity is necessary for survival, it is potentially disadvantageous when food and transportation conveniences are readily available.

The major components of total daily energy expenditure include resting energy expenditure, non-resting energy expenditure (e.g., physical activity and exercise) and the thermal effect of food [68]. Data suggest that having a low resting metabolic rate (i.e., having low resting energy expenditure) is not a major ‘cause’ of obesity in most patients, despite the common belief otherwise [69]. Instead, resting energy expenditure is largely dependent upon the BMI. While individual variabilities exist, formulae that include BMI (and age), such as the Harris–Benedict equation, yield reasonable mean resting energy expenditure predictions for normal and obese individuals [70]. Thus, in those without secondary causes of adiposity, patients with higher BMI have higher resting energy expenditure, while patients with lower BMI have lower resting energy expenditure. During dynamic weight loss (negative caloric balance), the resting metabolic rate often decreases below what is appropriate for BMI, and may [71] or may not [72] return to what is appropriate for BMI shortly upon stabilization of body weight and a return to energy balance. Non-resting energy expenditure may also decrease during weight loss due to improved skeletal muscle work efficiency [73]. Whatever degree of conservation of energy occurs during dynamic weight loss, diminished total energy expenditure limits fat weight reduction during times of negative caloric balance, which is a limitation perhaps mitigated by engaging in routine physical exercise [74].

An example of a neuroendocrine mechanism that helps account for this conservation involves leptin. Leptin is an adipocyte-released cytokine whose blood levels increase with adipocyte size and number. In obese patients with leptin deficiency, leptin administration reduces food intake and can result in dramatic weight loss [75,76]. However, most evidence suggests leptin approximates its maximum physiological effects at the high end of normal physiologic levels [5]. Thus, in those without leptin deficiency, the supraphysiologic levels of leptin found with obesity in humans do not effectively counteract persistent or worsening obesity. However, during weight loss, leptin levels decrease and low leptin levels can significantly favor positive caloric balance through increased appetite, among other possible mechanisms (Figure 1)[77]. Increased physical activity, such as through increased physical exercise, may enhance leptin sensitivity [78], reduce the so-called ‘leptin resistance’ [79], and better allow for improved longer term weight loss maintenance [25]. Thus, in addition to improving cardiovascular health [80], preserving lean body mass [81,82] and potentially improving metabolic parameters during weight loss [83], increased physical activity, such as through routine physical exercise, may also favorably affect leptin activity. This represents yet another potential justification for recommending increased physical activity, along with appropriate nutrition, towards the goal of maximizing chances of successful and persistent weight loss.

Physiological barriers such as these help to account for the frequent failure of nonsurgical weight loss interventions and a high rate of fat regain among overweight patients who have lost weight. Although not always effective [84], a pharmaceutical approach to overcome these multiple barriers is through combining weight loss drug therapies having complementary mechanisms of action, thereby potentially affecting more than one physiologic process [23]. Several combination anti-obesity strategies are in development [3]. An example of one such complementary approach is the combined use of phentermine and topiramate. The remainder of this review will focus on how both these agents, alone and in combination, reduce body weight, improve adipocyte and adipose tissue function (as described previously), and improve metabolic disease.

Phentermine

Indication & history

Phentermine is an approved anti-obesity agent indicated as an adjunct to appropriate nutrition and physical exercise for short-term (up to 12 weeks) treatment of obesity. Significantly due to its generic status, phentermine is the most commonly prescribed prescription appetite suppressant [3]. Phentermine resin was first approved by the US FDA in 1959 [85]. Phentermine resin formulations allow for a slower gastrointestinal release after ingestion. Administration of phentermine resin often begins with a 15 mg dose, titrated to 30 mg per day if needed.

In the 1970s, phentermine hydrochloride (HCl) was developed, with (depending upon the manufacturer) doses ranging from 8 to 37.5 mg, which is generally equivalent to 6.4–30 mg of phentermine resin. The phentermine HCl salt easily dissociates in the GI tract, resulting in immediate-release of phentermine drug; phentermine HCl is absorbed from the GI tract approximately three-times faster than phentermine resin [86]. Theoretically, immediate-release phentermine HCl has a more intense appetite-suppressant effect compared with phentermine resin, but for a shorter duration of time.

Mechanism of action

Anti-obesity therapies promote weight loss generally through decreased gastrointestinal absorption of energy (e.g., orlistat [3] and some bariatric surgeries [50]), increased energy expenditure and/or decreased energy (caloric) intake. Thyroid hormone is a classic example of an orally administered agent that increases energy expenditure through its CNS and systemic effects [87–90]. As perhaps the first anti-obesity agent, thyroid hormone was used as early as 1893 to induce weight loss [87]. However, excessive use of thyroid hormone has toxic effects upon the CNS and multiple other body organs. Dextrothyroxine was one of the five treatment groups evaluated in the Coronary Drug Project. The thyroid treatment arm in this study was stopped due to excessive mortality [91]. Toxicity due to excessive thyroid hormone is why the product information has a black warning against the use of thyroid hormone for weight loss [87].

Similarly, amphetamines were used as early as 1938 as weight loss treatments [92]. The mechanism of action of amphetamines includes an increase in CNS dopamine and norepinephrine (both catecholamines), and serotonin (an indolamine) activity. Clinical trials in humans suggest sympathomimetic agents increase energy expenditure [93,94]. In animals, amphetamines are reported to increase energy expenditure [95], such as through increasing thermogenesis [96]. Other animal data suggest that while amphetamines may increase locomotor activity, they may suppress the overall resting metabolic rate [97]. Although one might suppose that amphetamines (like other stimulants) increase the metabolic rate, the reported human data are remarkably scarce regarding the effect of amphetamines on energy expenditure. One study suggested that dextroamphetamine or methylphenidate treatment of children with attention-deficit hyperactivity disorder tended to increase the resting metabolic rate but decrease physical activity, resulting in an overall decrease in energy expenditure. Thus, at least at doses typically prescribed for weight loss in obese patients, the long-term effects of amphetamines on total energy expenditure are unclear. What does seem to be clear is that administration of these monoamine neurotransmitters reduces appetite, as would be expected through their decreases in neuropeptide Y [98], increases in pro-opiomelanocortin [99] and increases the anorexigenic peptide located in the hypothalamus, termed ‘cocaine and amphetamine-regulated transcript’ (CART) [100], among other potential effects (Figure 1). As with thyroid hormone, amphetamines have potential toxic CNS and other adverse systemic effects, including an increase in BP, tachycardia and euphoria. Amphetamines also have a high potential for abuse and addiction [87].

In the USA, the authority to prescribe certain controlled substances is defined by Drug Enforcement Agency (DEA) scheduling. Schedule I controlled substances include drugs with a high potential for abuse and with no accepted medical use, such as heroin, peyote and others. Most physicians do not have a medical need to apply for the special permission required to prescribe Schedule I drugs. DEA Schedule II drugs include those with a high potential for abuse, but which have medical use. Common examples include cocaine (for topical use), methadone, morphine and amphetamines. Therapeutic forms of amphetamines include amphetamine, dextroamphetamine and methamphetamine. Depending upon the preparation, amphetamines have an indicated use to treat attention-deficit disorder and narcolepsy. Some amphetamines continue to have a regulatory indication to treat obesity, although they are rarely used for this purpose. Instead, the product information for amphetamines usually has a black box warning regarding the high potential for abuse, and risk of sudden death and serious cardiovascular events.

Phentermine is commonly described as a noradrenergic and possibly dopaminergic [101] sympathomimetic amine, although some question whether phentermine causes clinically meaningful release of dopamine from central neurons, at least relative to amphetamine [102]. Phentermine is chemically related to amphetamine, but does not have its addictive potential. Phentermine is listed as a Schedule IV drug (along with other drugs such as barbiturates, benzodiazepines and librium) because it has a low potential for abuse, and because it has a medical use for short-term treatment of obesity. Figure 1 describes first- and second-order hypothalamic signaling, which may be influenced by increased CNS noradrenergic activity [103], as might be promoted by phentermine. While phentermine increases sympathetic activity (which reduces appetite), little evidence suggests that its weight loss effects are substantially due to increasing resting energy expenditure [104].

Adverse experiences

Phentermine is contraindicated where sympathomimetic drug administration may pose a significant risk, such as in patients with unstable cardiovascular disease, moderate-to-severe high BP, hyperthyroidism and unstable cardiac dysrhythmias. Phentermine is not recommended in patients treated with monoamine oxidase inhibitors or in patients with a history of drug abuse (including excessive alcohol consumption). Phentermine should be used with caution in patients concurrently treated with other agents that affect the CNS agents, such as agents that increase adrenergic responses (i.e., decongestants). Phentermine’s side effects include palpitations, tachycardia, increase in BP, tremor, diaphoresis, overstimulation, anxiety, irritability, restlessness, dizziness, insomnia, euphoria, dysphoria, headache, dryness of the mouth and gastrointestinal complaints (i.e., diarrhea and constipation) [3].

Phentermine, adiposopathy & metabolic disease

Although phentermine is among the oldest approved anti-obesity agents, it has among the least data with respect to its effects upon adiposopathy. Limited to no published data exist on phentermine monotherapy effects upon adipogenesis, free fatty acids and important adipose tissue endocrine and immune factors (Table 1). Shorter term data suggest that the weight loss associated with phentermine may have some favorable lipid effects [105], with variable effects regarding BP [105,106]. However, published data on the long-term (≥1 year) effects of phentermine monotherapy on metabolic disease are lacking.

Topiramate

Indication & history

Topiramate is a sulfamate-substituted monosaccharide derivative of the naturally occurring sugar monosaccharide D-fructose. It was originally developed as a treatment for Type 2 diabetes mellitus [60,107]. Despite promising early animal data, the direct antihyperglycemic effects of topiramate (independent of weight loss) were not confirmed in humans [108]. However, some authors continue to believe topiramate has insulin secretagogue and insulin sensitization potential [109].

Topiramate has undergone evaluation and development at least since 1979. Topiramate was approved in the USA as a marketed pharmaceutical in 1996 [110]. Topiramate has a US regulatory indication for prophylaxis of migraine headaches and treatment of seizure disorders. Most applicable to this review is that while topiramate does not have a regulatory indication for weight loss [111], some clinicians have used topiramate ‘off label’ for the purpose of weight reduction. This is because early on, even at the time of approval, the treatment emergent ‘adverse’ experiences described in its development reportedly included: “mild, dose-related weight loss was associated with topiramate therapy” [112]. Since approval, clinical trial data have supported topiramate as having weight loss effects [113–119].

An illustrative study was a placebo-controlled, 6-month clinical trial of topiramate (not controlled-release) in obese subjects starting with 16 mg once a day, which was increased to 16 mg twice a day, and then increased weekly by 16 mg twice a day until reaching the target dose of 64, 96, 192 or 384 mg per day [116]. Mean weight loss from baseline for placebo was -2.6%. At the respective doses of topiramate, weight loss was -5.0, -4.8 and -6.3%. This suggested that most of the weight loss with topiramate occurred at the lower topiramate doses. A total of 21% of topiramate subjects withdrew because of adverse experiences (compared with 11% of placebo-administered patients). According to the authors: “The most frequent adverse events were related to the central or peripheral nervous system, including paresthesia, somnolence, and difficulty with memory, concentration, and attention. Most events were dose-related, occurred early in treatment, and usually resolved spontaneously” [116].

Another study evaluated the long-term effects of topiramate (again, not controlled-release) in obese subjects participating in a nonpharmacologic weight-loss program [114]. Topiramate produced significant weight loss of -7.0, -9.1 and -9.7% at respective doses of 96, 192 and 246 mg per day, compared with a loss of -1.7% in the placebo group. As before, most of the weight loss was achieved at the lower topiramate doses. Importantly, topiramate significantly improved BP, glucose and insulin levels. Regarding safety and tolerability, the authors stated: “The most common adverse events more frequently observed in topiramate-treated subjects occurred mostly during the titration phase and were related to the central or peripheral nervous system and included paresthesia, difficulty with concentration/attention, depression, difficulty with memory, language problems, nervousness, and psychomotor slowing.” The study was stopped early because the study sponsor elected to: “develop a new controlled-release formulation with the potential to enhance tolerability and simplify dosing” [114]. However, this controlled-release monotherapy development program was later abandoned, presumably because of a lack of consistent benefits and sufficient efficacy, balanced against the adverse experiences found with higher topiramate monotherapy doses thought to be required to obtain an approvable weight-loss indication [60].

Mechanism of action

Topiramate is a ‘neurostabilizer’ [107]. Signaling from CNS neuron to neuron is propagated by action potentials, which result from ion (i.e., sodium and potassium) flux across voltage-gated ion channels in neuron cell walls. Seizures occur when excessive firing disrupts the normal neural signaling flow from postsynaptic depolarization, to action potential generation, to neurotransmitter release [120]. Although not yet definitively proven, topiramate may reduce seizure activity by modifying excitatory voltage-activated sodium and calcium channels, antagonizing α-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid kainate (AMPA/KA) receptors, and enhancing γ-aminobutyric acid (GABA) receptor-mediated inhibitory currents [121].

Similarly, although yet to be proven, topiramate has several potential mechanisms accounting for weight loss. As it is a CNS-acting drug, topiramate may have CNS effects that alter caloric balance. Topiramate’s antagonism of AMPA/KA receptors may reduce compulsive or addictive food cravings [122], which is supported by topiramate’s effectiveness in improving binge eating disorder [123] and effectiveness in reducing other addictive behaviors [124]. Topiramate’s activation of GABA receptors may decrease night-time and deprivation-induced feeding [125]. Decrease feeding during caloric deprivation would favor persistence of weight loss. While animal studies suggest topiramate may increase levels of neuropeptide Y (NPY) in the hypothalamus (an effect that would seem to promote positive caloric balance [126,127]), topiramate may also increase levels of hypothalamic corticotropic-releasing hormone, which may be catabolic (Figure 1)[128]. Topiramate may also increase hypothalamic galanin, with unclear body weight implications [128,129]. While some studies suggest leptin levels decrease with topiramate-induced weight reduction [119], other reports suggest leptin levels may either not significantly decrease, or only modestly decrease with topiramate-induced weight loss, which is another effect that may favor persistence of weight loss [130]. Anorexia is commonly reported with topiramate; however, topiramate may not reduce objective measures of appetite during weight loss [131]. A lack of increased hunger and appetite during weight loss may reflect a cumulative effect of the anorectic effects of topiramate, balanced against the orexigenic effect of topiramate-promoted fat loss or related energy deficit through other mechanisms [131]. Again, this would favor persistence of weight loss. Regarding energy expenditure, topiramate may decrease energy storage and usage efficiency, and thus increase energy expenditure [132]. Topiramate may increase thermogenesis [133].

Applicable to adipocyte function, topiramate (as well as related sulfamates) may inhibit adipocyte mitochondrial carbonic anhydrase isozyme V [134], inhibiting carbonic anhydrase-mediated lipogenesis [135,136]. Topiramate is also associated with decreased lipoprotein lipase activity in white adipose tissue, which would limit free fatty acid substrate for lipogenesis [137]. Conversely, topiramate may increase lipoprotein lipase activity in brown adipose tissue, which may represent increased thermogenesis and increase lipoprotein lipase activity in skeletal muscle, further supporting the potential for substrate oxidation [137]. Another effect on the level of the adipocyte is that topiramate may increase adiponectin levels [138], which favorably affects many peripheral physiologic processes relevant to metabolic disease [139]. While the clinical significance is unclear, increased circulatory adiponectin may have CNS effects regarding energy balance [140].

Finally, clinical trials support topiramate as altering taste, which could conceivably affect caloric intake [141], and topiramate is a carbonic anhydrase inhibitor, which may have anorexigenic effects (see later). Overall, topiramate has multiple potential mechanisms that could promote both weight loss, and persistence of weight loss.

Adverse experiences

Many of the adverse experiences reported with topiramate are dose-related [112]. As with some other sulfa-based drugs (e.g., acetazolamide, hydrochlorothiazide and cotrimoxazole) topiramate can cause acute myopia, sudden onset of blurring of vision, redness of the sclera, photophobia and discomfort of the eyes resulting from secondary angle-closure glaucoma [142,143]. Angle-closure glaucoma occurs when the aqueous humor produced in the anterior chamber (the space between the iris/lens and cornea) by ciliary processes is unable to be drained through the trabecular meshwork due to closure of the angle between the iris and the cornea. The increased intraocular pressure may adversely affect various portions of the eye, such as the cornea, sclera, iris, lens and optic nerve. This potential adverse experience usually occurs within hours to weeks after topiramate initiation. Treatment involves stopping the topiramate [144]. The increased risk of secondary angle-closure glaucoma with topiramate is in contrast to primary angle-closure glaucoma, wherein treatment may include procedures such as iridoplasty or medications such as acetazolamide [145]. Acetazolamide may improve glaucoma by inhibiting carbonic anhydrase isozymes in the ciliary processes of the eye, reducing bicarbonate and aqueous humor secretion, and thus reducing the elevated intraocular pressure [146]. This is somewhat paradoxical, given that topiramate, like acetazolamide, is an inhibitor of carbonic anhydrase, and suggests that the secondary angle-closure glaucoma with topiramate is not related to its inhibition of carbonic anhydrase. Finally, the commonality of carbonic anydrase inhibitors helps explain why both acetazolamide and topiramate may be effective in treating conditions such as idiopathic intracranial hypertension [147].

However, while topiramate-associated glaucoma may not be related to mild inhibition of carbonic anhydrase, other adverse experiences may be related to this effect. Carbonic anhydrase is a ubiquitous family of metalloenzymes that catalyze carbon dioxide and water to bicarbonate and protons. In human metabolism, nutrients (e.g., carbohydrates, protein and fats) are metabolized by body tissues to generate sources of energy (e.g., adenosine triphosphate), water and carbon dioxide. Carbonic anhydrase enzymes help transport carbon dioxide by converting tissue carbon dioxide to bicarbonate, which is then carried by red blood cells to the lungs, where pulmonary carbonic anhydrase converts the bicarbonate back to carbon dioxide, which then undergoes respiratory exchange for oxygen. The types of topiramate adverse experiences most likely due to carbonic anhydrase inhibition include metabolic acidosis [148], increased risk of kidney stones [149] and oligohidrosis (that generally reverses after stopping topiramate), which may contribute to hyperthermia (mostly described in children) [150]. Other adverse experiences that may be related to carbonic anhydrase inhibition include an increase in paresthesias, which is among the most common adverse experiences associated with topiramate, as with other carbonic anhydrase inhibitors [151]. Additional adverse experiences sometimes associated with topiramate as well as other carbonic anhydrase inhibitors include anorexia, fatigue, somnolence and depression [152].

Special mention should be made of the potential CNS adverse experience of topiramate, which, as noted before, include difficulty with concentration, difficulty with memory and depression [117]. These potential adverse experiences are important with any drug affecting the CNS, and particularly important with regards to anti-obesity drug development. Rimonabant is a cannabinoid receptor 1 (CB1) antagonist that underwent extensive, if not comprehensive, clinical trial evaluation. Rimonabant was not only effective in generating weight loss, but also improved adiposopathy and many metabolic diseases, such as dyslipidemia and Type 2 diabetes mellitus [153]. However, rimonabant never obtained US regulatory approval because the FDA was more concerned about the potential for CNS adverse experiences than it was impressed by the numerous and significant improvements in metabolic disease – suggesting that the current criteria for anti-obesity drug development may need to be re-evaluated [154]. Paramount in the termination of rimonabant and several other CB1 antagonist development programs was a modest increase in depression and anxiety [155].

Because topiramate is a CNS-acting drug, and is being considered as a component of an anti-obesity agent, an understanding of its CNS effects becomes paramount. Publication of potential suicidality associated with topiramate is basically limited to case reports [156]. Because the frequency of suicide with individual anticonvulsants was not sufficient for proper statistical analyses, in 2008 the FDA mandated a warning labeling for many anticonvulsant medications regarding the increased risk of suicidal thoughts and behaviors, which included labeling applicable to topiramate. A subsequent cohort study of a large database including 297,620 new episodes of treatment with an anticonvulsant agent identified 26 completed suicides, 801 attempted suicides and 41 violent deaths. This analysis was not designed to determine if the anticonvulsants caused, or contributed, to suicides, or if these findings were simply incidental to the use of the anticonvulsants. Topiramate was chosen as the reference drug in this study. This exploratory analysis suggested that gabapentin, lamotrigine, oxcarbazepine and tiagabine were associated with an increased risk of suicidal acts or violent deaths compared with topiramate [157]. This might be interpreted as suggesting topiramate had the least potential to cause suicides as compared with the other approved and commonly prescribed anticonvulsant therapies. Finally, it is noteworthy that this study evaluated topiramate (not controlled-release) at doses typically used for anticonvulsant therapy.

Topiramate, adiposopathy & metabolic disease

Although topiramate does not have an indication for weight loss, it is interesting that topiramate appears to have more published data than phentermine (which does have an indication for weight loss) with respect to parameters associated with adiposopathy. While the effects upon adipogenesis are unknown, the medical literature does report data suggesting that topiramate has effects upon adipose tissue factors that would generally be considered favorable in preventing, or improving, metabolic disease (Table 1).

Phentermine & topiramate controlled-release combination

Rationale & history

Common themes in the development of pharmacologic therapies for metabolic diseases (e.g., obesity, diabetes mellitus, hypertension and dyslipidemia) include:

• • Drug treatment is intended to be an adjunct to appropriate nutrition, increased physical activity and other favorable lifestyle interventions;

• • Monotherapy of any single therapeutic agent is often insufficient to achieve desired treatment goals;

• • Compared with either agent alone, combining different therapeutic agents with complementary mechanisms of action may have greater efficacy at lower doses of each;

• • Compared with either agent alone, lower doses of the constituent drug therapies may improve upon the intolerances and toxicities associated with higher doses.

The phentermine and topiramate controlled-release (PHEN/TPM CR; VI-0521; Qnexa^®, Vivus, Inc., CA, USA ) development program was consistent with this approach. PHEN/TPM CR clinical trials were designed to evaluate the efficacy and safety of the combined use of lower doses of two different drugs with differing mechanisms of action for the purpose of weight reduction (e.g., reducing appetite and increasing satiety) over a 24-h period. By utilizing a more-than-one treatment target approach, the hope was to counteract redundant physiologic mechanisms that often limit fat weight loss and frequently promote fat weight regain in overweight patients.

The PHEN/TPM CR development program involved a once-a-day, co-formulated combination single capsule of phentermine HCl (which is readily dissolved and absorbed in the GI tract to provide therapeutic effects early in the day), added to controlled-release topiramate (to provide persistent therapeutic weight loss effects throughout the day). As noted before, the maximum approved dose of phentermine monotherapy, as found with the slower release resin formulation, is 30 mg per day, which is roughly equivalent to 37.5 mg of the immediate-release phentermine HCl. The maximum approved dose of topiramate monotherapy is 400 mg per day. In its development program, PHEN/TPM CR evaluated three combination doses. ‘Low dose’ was defined as phentermine HCl 3.75 mg/topiramate CR 23 mg; ‘mid dose’ was phentermine HCl 7.5 mg/topiramate CR 46 mg; and ‘full dose’ was phentermine HCl 15 mg/topiramate CR 92 mg. Thus, across the dose ranges, the PHEN/TPM CR combination agent contained 40% or less of the maximum approved dose of phentermine HCl (maximum dose phentermine HCl monotherapy = 37.5 mg per day), and less than 25% of the maximum approved dose of topiramate (maximum topiramate monotherapy dose = 400 mg per day). Finally, the clinical trial development program utilized a titration dosing schedule for the first 4 weeks.

A full discussion of the results of the clinical trials in the PHEN/TPM CR development program awaits peer-reviewed published data. However, much of the PHEN/TPM CR clinical trial data were presented as poster and podium abstracts at major scientific meetings. The study sponsor also issued press releases supplementing the scientifically presented data. Study sponsor press releases are generally good at reporting basic study details (as described herein), at least in part due to security regulations and requirements [158]. However, press releases may be incomplete regarding nuanced study limitations that may be more fully vetted through peer review during the publication process [158]. The same might be said about scientific abstracts. With these caveats in mind, Table 2 summarizes preliminary disclosed data regarding some of the sentinel Phase III clinical trials of the PHEN/TPM CR combination agent.

Mechanism of action

The mechanism of action of PHEN/TPM CR would be expected to reflect the mechanism of action of the individual components previously described.

Adverse experiences

Table 3 lists adverse experiences (regardless of cause) that occurred in over 5% of study participants in the two largest trials to date. Dry mouth and paresthesias (‘tingling’) occurred in approximately 20% of PHEN/TPM CR full-dose-administered subjects, with other adverse experiences occurring in less than 20% of study participants (e.g., constipation, altered taste, insomnia, headache, dizziness, nausea and blurred vision). However, from a persistence-of-use standpoint, it is often not the quantity or frequency of adverse experiences associated with a pharmaceutical, but rather the quality (degree) or severity of adverse experiences that is most clinically relevant. Many commonly prescribed drugs have a high frequency of adverse experiences that are not severe enough to negate their clinical utilization. The product information (PI) of amoxicillin/clavulanate potassium [202] describes a 9% frequency of diarrhea/loose stools, with a warning of: “serious and occasionally fatal hypersensitivity (anaphylactic) reactions” in patients treated with penicillin. The PI of ibuprofen [203] describes a 4–16% frequency of gastrointestinal complaints with a warning that: “serious gastrointestinal toxicity such as bleeding, ulceration and perforation can occur at any time, with or without warning symptoms.” The PI of metformin [204] describes a 53% frequency of diarrhea, 26% frequency of nausea/vomiting, 12% frequency of flatulence and a warning of potential lactic acidosis that: “when it occurs, it is fatal in approximately 50% of cases.” The point is that simply having associated adverse experiences is not enough to prevent the clinical use of a drug. Instead, the degree of severity should be considered when determining the clinical importance of the adverse experiences associated with a pharmaceutical treatment.

Table 4 provides data that somewhat reflect the severity of these adverse experiences through representing the percent of adverse experiences resulting in study discontinuation. The discontinuation rate due to adverse experiences associated with placebo was 9%, while the discontinuation rate due to PHEN/TPM CR low, mid and full dose was 12, 12 and 18%, respectively, with no particular adverse experience being an outlier causing disproportionate study discontinuation. Despite the modest increase in discontinuations due to adverse experiences in the PHEN/TPM CR group, it is interesting that Table 2 and Table 4 report a higher study completion rate in those in the PHEN/TPM CR group compared with the placebo group, particularly with the mid-to-full PHEN/TPM CR dose. This suggests that overall, more in the placebo group discontinued from the trials than the PHEN/TPM CR groups, due to reasons other than adverse experiences. One might speculate that the higher completion rate in the PHEN/TPM CR group might, at least in part, be due to the greater weight loss efficacy of PHEN/TPM CR over placebo, resulting in a greater inclination to remain in the studies. It is also possible that a topiramate-induced reduction in food-addictive behavior (as previously discussed) may have enhanced adherence and completion of study protocols. Finally, CNS stimulant medications may improve retention of previously required information, facilitate memory consolidation and conceivably have motivational effects that might affect clinical trial completion. However, previously published data that might support such effects typically did not include phentermine, are inconsistent and mostly studied in patients with attention-deficit hyperactivity disorder. Furthermore, while CNS stimulants might conceivably have the aforementioned effects, the evidence does not always support stimulants as being cognitive ‘enhancers’ [159].

One of the main challenges of anti-obesity drug development, particularly from a regulatory perspective, is the effect of CNS-acting anti-obesity agents upon adverse experiences such as anxiety and depression. As opposed to several other development programs, study subjects in some of the PHEN/TPM CR trials were permitted to have mild depression, as long as their antidepressant therapy was stable prior to study entry. Preliminary reports suggest that PHEN/TPM CR does not significantly increase moderate or severe depression as measured by depression scales such as the Patient Health Questionnaire (PHQ) – 9. Other scales, such as the Columbia Suicide Severity Rating Scale reportedly demonstrated no increase in suicidality [205].

PHEN/TPM CR, adiposopathy & metabolic disease

Table 1 lists the effects of PHEN/TPM CR upon adipocyte and/or adipose tissue factors thought to contribute to metabolic disease. Most have yet to be evaluated or reported. Table 2 describes the weight loss and improvement in metabolic diseases of PHEN/TPM CR in three sentinel clinical trials.

The 6-month EQUATE trial evaluated 756 generally healthy obese subjects with a BMI between 30 and 45 kg/m². PHEN/TPM CR mid- and full-dose combination therapies significantly reduced body weight over the same doses of phentermine and topiramate monotherapies [160]. Diabetes mellitus was an exclusion criterion. Nonetheless, a subset analysis demonstrated that both mid- and full-dose PHEN/TPM CR reduced hemoglobin A1c (HbA1c) [6,161,206]. The most common adverse experiences for placebo, PHEN/TPM CR mid dose and PHEN/TPM CR full dose, respectively, were tingling (3, 16 and 23%), dry mouth (0, 13 and 18%), headache (13, 15 and 16%) and constipation (8, 7 and 15%) [207].

The 1-year EQUIP study evaluated 1267 generally healthy obese subjects with a BMI of 35 kg/m² or more. PHEN/TPM CR low and full dose significantly reduced body weight compared with placebo. Despite having minimal or drug-controlled metabolic abnormalities at baseline, PHEN/TPM CR low dose significantly reduced systolic BP and total cholesterol. PHEN/TPM CR full dose significantly reduced systolic BP, reduced diastolic BP, reduced triglycerides (TGs), reduced total cholesterol, reduced LDL cholesterol (LDL-C), and increased HDL (HDL-C) levels [205]. The significant reduction in total body weight, the reduction in waist circumference, and the significant improvement in metabolic parameters compared with placebo supported PHEN/TPM CR as being effective in treating adiposopathy and its adverse metabolic consequences.

The 1-year CONQUER trial included 2487 obese patients with a BMI of 27 kg/m² or more and common metabolic comorbidities associated with adiposopathy. This study demonstrated that PHEN/TPM CR significantly reduced body weight, and reduced waist circumference (an anatomical manifestation of adiposopathy) compared with placebo. In addition, PHEN/TPM CR mid dose significantly reduced BP, TG and total cholesterol, and increased HDL-C levels. PHEN/TPM CR full dose significantly reduced systolic BP, reduced diastolic BP, reduced TG, reduced total cholesterol, reduced LDL-C and increased HDL-C levels. Among the subset of patients with Type 2 diabetes mellitus upon study entry, both the mid- and full-dose PHEN/TPM CR significantly reduced HbA1c, fasting glucose, insulin levels (after an oral glucose tolerance test), and the homeostatic model assessment of insulin resistance [162,205]. Among the subset of patients with high BP at study entry, both the mid- and full-dose PHEN/TPM CR significantly reduced systolic BP. Among the subset of patients with hypertriglyceridemia at study entry, both mid- and high-dose PHEN/TPM CR significantly reduced TG levels [205].

A trial not listed in Table 2 was a 56-week, randomized, double-blind, placebo-controlled trial of PHEN/TPM CR full dose in 130 obese subjects (BMI 27–45 kg/m²) with drug-naive Type 2 diabetes mellitus, or Type 2 diabetes mellitus treated with oral antihyperglycemic drugs [163,208]. Patients with stable depression were eligible, as were those treated with selective serotonin reuptake inhibitors (SSRIs) or serotonin/norepinephrine reuptake inhibitors (SNRIs). Baseline mean age was 50 years, 90 were women, 40 were men, and over 90% of subjects completed the trial. PHEN/TPM CR produced a 9.4% reduction in body weight (vs a reduction of 2.7% with placebo), and a reduction in HbA1c of 1.6% (vs reduction of 1.1% with placebo). The high degree of HbA1c reduction with placebo was due to the increase in anti-diabetes mellitus therapies utilized in the placebo group during the trial versus a net decrease in anti-diabetes mellitus therapies utilized in the PHEN/TPM CR group. In addition to improvements in fasting and postprandial glucose levels, PHEN/TPM CR also resulted in a decrease in waist circumference, BP and triglyceride levels.

Taken in totality, these studies support PHEN/TPM CR as not only effective in improving the weight of patients, but also effective in improving the metabolic health of patients.

Conclusion

The PHEN/TPM CR development program involved the combined use of two pharmaceuticals known to have weight loss properties and complementary mechanisms of action. Its components were at doses lower than approved for their monotherapy treatment indications. This helps account for why the reported adverse experiences with PHEN/TPM CR were not severe enough to result in substantial discontinuance of drug therapy. In fact, the PHEN/TPM CR had higher persistence of use than the placebo group. Most importantly, PHEN/TPM CR improved adiposopathy, as evidenced by the significant loss of body weight, reduction in waist circumference and improvement in multiple metabolic parameters. This all supports PHEN/TPM CR as not only improving the weight of patients, but also improving the health of patients.

Expert commentary & five-year view

In early to mid-2010, only two anti-obesity drugs were approved for long-term use in the USA (i.e., orlistat and sibutramine). In early 2010, sibutramine was no longer approved in Europex [3,23,164], leaving only orlistat. Obesity is the greatest epidemic in human existence, based upon the number of lives affected [23]. Thus far, regulatory agencies have determined that the risks of all but one or two anti-obesity drug therapies exceed their potential benefits in treating adiposity, adiposopathy and metabolic disease.

The history of anti-obesity drug development has often been ‘challenging’ at best, and disastrous at worst [23,87]. Adverse safety experiences with previous anti-obesity drugs may have tempered, and perhaps even jaded regulatory enthusiasm for approving future anti-obesity agents [87]. Hopefully, a ‘new age’ in anti-obesity drug development will result from the increased recognition of adipose tissue being far more than simply an inert storage organ. Adipose tissue is an active endocrine and immune organ whose dysfunction promotes the most common metabolic diseases encountered in clinical practice (e.g., diabetes mellitus, high BP and dyslipidemia) [5]. As such, the next 5–10 years may represent a renaissance in what was known in the 1940s to 1970s, which is that adipocytes [14,15] and adipose tissue [28] are potentially pathogenic. In other words, recognizing adipocyte hypertrophy and visceral adiposity as contributing, or in fact ‘causing’, metabolic disease is not a novel concept [6]. It is certainly not novel to clinicians who, in the day-to-day management of patients, frequently observe onset or worsening of glucose, BP and lipid levels in their patients who gain body fat.

Towards the goal of ‘rediscovering’ the pathogenic metabolic potential of adipose tissue in human disease, the terms ‘adiposopathy’ and ‘sick fat’ as scientific and clinical terms, respectively, may afford advantages over older, less meaningful terms. ‘Obesity’, ‘metabolic syndrome’ and ‘ectopic fat’ are examples of terms that may soon become obsolete, or at least acquire different meanings than intended today. For example, if it is to continue to describe a ‘disease’ at all, the term ‘obesity’ may eventually refer to body-mass pathology alone [23]. The potential of adipose tissue to contribute to metabolic disease is better assessed based upon its pathogenic endocrine and immunological responses, relative to a simple measurement of body weight [17]. Likewise, the term ‘metabolic syndrome’ remains problematic for many reasons; most of all because it was never intended to represent or describe a unified pathophysiologic process [1,165]. Broken down to its basic components, ‘metabolic syndrome’ is nothing more than a collection of common metabolic abnormalities that tend to cluster and which are CHD risk factors [6]. The National Education Program, Adult Treatment Panel III includes increased waist circumference as one of the five components of defining ‘metabolic syndrome’. The International Diabetes Federation goes further in recognizing the pathogenic importance of pathogenic adipose tissue by requiring increased waist circumference as being present to meet the definition of the ‘metabolic syndrome’, which must then be accompanied by other metabolic abnormalities. These definitions provide implicit recognition of the pathogenic potential of adipose tissue with respect to promoting metabolic abnormalities and increasing CHD risk. Thus, one might argue that the term ‘metabolic syndrome’ might best be replaced by a term that better describes the underlying cause of the clustering of common atherosclerotic risk factors. ‘Adiposopathy’ is such a term [1,2].

A third term that may be destined for evolution or dissolution is ‘ectopic fat’. Adipose tissue has biologic and physiologic importance, such as by providing heat insulation, mechanical cushion, endocrine functions, immune functions and storage of energy. Thus, the presence of adipose tissue is important for metabolic health; the absence of body fat is pathologic [6]. Ectopic fat describes the situation wherein existing adipose tissue is no longer able to efficiently store excess energy. Fats (free fatty acids, triglycerides, etc.) are inappropriately transported and stored in body organs, increasing the risk of metabolic disease. Some of this excessive energy may be stored through a pathologic transfer of free fatty acids to non-adipose tissue organs (e.g., the liver and muscle), which contributes to metabolic disease [6]. However, most excessive energy is stored in various adipose tissue depots. This results in the sometimes curious characterization of additional adipose tissue (such as increased visceral adipose tissue) as representing ‘ectopic fat’ [166]. In fact, any excessive fat in any location is considered by some as representing ‘ectopic fat’. However, if all excessive fat storage is ‘ectopic’, then this raises the question as to the clinical utility of this term. One could argue that ‘adiposity’ is a term sufficient in describing excessive body fat, and the term ‘adiposopathy’ is more precise in describing when excessive body fat is pathogenic in causing or promoting adverse metabolic consequences leading to metabolic disease.

As the pathogenic potential of adipose tissue becomes more widely acknowledged, specific diagnostic criteria will need to be developed for adiposopathy or ‘sick fat’ [1]. Once this process is complete, then perhaps regulatory agencies would be more open to establishing treatment indications to improve adiposopathy (a true disease) [167], and thus free themselves from approval processes whose only definitive metrics are focused on reductions in body weight [168,209]. In other words, the current FDA guidance for industry has comments such as how: “excessive body fat increases the risk of death and major comorbidities such as Type 2 diabetes, hypertension, dyslipidemia, cardiovascular disease, osteoarthritis of the knee, sleep apnea and some cancers” [209], suggesting a belief that adiposity is merely associated with, rather than potentially causative of, fat-mass complications and metabolic disease. Furthermore, although current guidance does stress the importance of improvement in metabolic disease, the degree by which weight loss agents safely improve the weight of patients remains the focus, as opposed to the degree by which these agents improve the metabolic health of patients.

This brings us to PHEN/TPM CR. Phentermine is an anti-obesity drug indicated for short-term treatment of obesity. It is an approved pharmaceutical that has been in clinical use for approximately 50 years, and is the most widely prescribed anti-obesity agent. Phentermine has no reported adverse drug interactions when used with topiramate, and has no product information statements warning against its use with topiramate. Topiramate is an anti-seizure drug and migraine headache prophylactic drug. It is an approved pharmaceutical that has been in development since the 1970s and in clinical use for over a decade. It is among the most commonly prescribed anticonvulsant therapies. Topiramate has no reported adverse drug interactions when used with phentermine, and has no product information statements warning against its use with phentermine. Data in this review support PHEN/TPM CR as being no less safe or no less tolerated than what would be expected from the individual components. This might be anticipated given that the PHEN/TPM CR combination agent includes topiramate in a controlled-release formulation (although this has yet to be proven to afford benefits over non-controlled-released formulations), and utilizes doses of phentermine and topiramate that are a fraction of their respective maximum approved doses. The safety dataset will be substantially enhanced by the results of the long-term OB-305 extension study (SEQUEL), due sometime in latter 2010 or early 2011. If clinicians accept PHEN/TPM CR as being as safe as the individual components, then fundamental decisions will have to be made regarding the clinical importance of adiposity and adiposopathy.

Clinicians may question whether the weight loss with PHEN/TPM CR is sufficient to improve fat-mass diseases. The OB-204 study reported a significant improvement in sleep apnea with PHEN/TPM CR, which would seem to be a significant benefit in reducing fat mass [210]. Clinicians may also question whether the statistically significant weight loss with PHEN/TPM CR represents clinically meaningful weight loss. Insight as to what may be ‘clinically meaningful’ might be derived from regulatory agency guidance. The FDA ‘Guidance for Industry: Developing Products for Weight Management’ [209] has suggested regulatory metrics for approval of weight loss drugs and therapeutic biologics (termed ‘products’). According to the FDA, Phase I and II trials should characterize the pharmacokinetics and dose response of the product, over a range of doses and across a broad range of BMI (e.g., BMI ≥30 kg/m²; or ≥27 kg/m² if with comorbidities) to determine no-effect, maximally tolerated doses and efficacy of all doses relative to placebo. The duration should be long enough to capture the product’s maximal or near-maximal weight loss effects. Primary end points should be mean absolute or percent change in body weight in the active versus placebo group, as well as the proportion of each treatment group with a weight loss of 5% or more of baseline weight. Other end points should include evaluations of weight-related comorbidities. The EQUATE study (Table 2) supports PHEN/TPM CR as fulfilling many of these criteria.

Phase III trials should be randomized, double-blind, placebo-controlled 1-year studies in patients with a BMI of 30 kg/m² or more, or 27 kg/m² or more if with comorbidities such as Type 2 diabetes mellitus, hypertension, dyslipidemia, sleep apnea and cardiovascular disease (with some having a BMI >40 kg/m²), and with representation from demographic, ethnic and racial groups having a high prevalence of obesity. The concomitant lifestyle modification program: “should strike an appropriate balance between effectiveness and simplicity.” A reasonable safety set would include 3000 subjects randomized to active drug and no fewer than 1500 randomized to comparator placebo. Recommended primary end points would be mean difference in percent loss of baseline body weight (active treatment vs placebo) and categorical (proportional) analysis of study participants who lost at least 5% of baseline body weight (active treatment vs placebo). Secondary end points should include changes in fasting glucose, insulin and HbA1c levels, BP, pulse and lipid levels. Waist circumference should also be measured; however, while acceptable as a surrogate in clinical practice, waist circumference is not considered an acceptable surrogate for visceral fat in the context of a clinical trial. More direct measures for visceral adiposity would include dual energy x-ray absorptiometry composition analysis, which is also important to ensure that the weight loss is primarily fat and not loss of lean body mass. Concomitant metabolic drug therapies should be assessed by protocol-defined algorithms to determine their initiation, withdrawal, increase in dose and decrease in dose during the course of the trials. ‘Effectiveness’ of a weight loss product is achieved if after 1 year of treatment, one of the two following benchmarks are met:

• • The difference in mean weight loss between the active product and placebo-treated groups is at least 5%, and the difference is statistically significant;

• • The proportion of subjects who lose 5% of baseline body weight or more in the active product group is at least 35%, is approximately double the proportion in the placebo-treated group, and the difference between groups is statistically significant.

Although overweight and obese study participants are expected to receive standard of care for metabolic comorbidities, and although no specific criteria are given, glycemic, BP and lipid measurements are expected to improve as would be anticipated with the degree of weight loss.

Other guidance recommendations are specific for individual weight loss agents. Weight management products that act upon the CNS should include an assessment of neuropsychiatric function through validated scales, as well as potentially additional studies to evaluate abuse liability. Weight management products that are therapeutic proteins should undergo assessment of immunogenic potential for 6–12 months. Combination weight management products should first have data comparing the combination product to its individual components of sufficient duration to capture their respective maximal or near-maximal weight effects. The combination weight management product should have greater weight loss efficacy compared with its individual components, with weight loss twice the individual components being preferred.

Special guidance is made regarding the evaluation of overweight and obese patients with Type 2 diabetes mellitus. Weight loss with bariatric surgery can substantially improve many metabolic parameters. Its favorable effects upon Type 2 diabetes mellitus are particularly noteworthy, with remission of Type 2 diabetes mellitus being as high as 80% [50]. Conversely, several prior anti-obesity drug trials have often resulted in a less than robust response in weight loss and in improving diabetes mellitus, possibly because dramatic weight loss may be required to significantly improve fat function (particularly in older patients). This suggests that the degree of success of weight-loss interventions in improving diabetes mellitus may be dependent upon the age of the patient, the pathology of the diabetes mellitus, the duration of the diabetes mellitus, and the degree of potential improvement in the functionality of adipocytes and adipose tissue [17], as well as other body organs. In their guidance, the FDA reflected this viewpoint by stating: “compared with nondiabetic patients, overweight and obese patients with Type 2 diabetes often respond less favorably to weight-management products.”

In consideration of all the above, Table 2 supports PHEN/TPM CR as meeting many of the FDA weight loss guidance criteria. Table 2 also supports PHEN/TPM CR as providing clinically meaningful improvements in metabolic diseases such as diabetes mellitus, hypertension and dyslipidemia. PHEN/TPM CR does not appear to significantly adversely increase depression, as assessed by common neuropsychiatric scales [205]. Finally, as noted before, in a trial of overweight or obese patients with Type 2 diabetes mellitus, PHEN/TPM CR produced a robust 9.4% reduction in body weight (vs a reduction of 2.7% with placebo), and a substantial reduction in HbA1c of 1.6% (vs reduction of 1.1% with placebo) [163,208]. The high degree of HbA1c reduction with placebo was at least partially due to an increase in anti-diabetes mellitus therapies. The even greater HbA1c reduction with PHEN/TPM CR was despite a net decrease in anti-diabetes mellitus therapies. It is unclear that these data suggest PHEN/TPM CR as having unique weight loss and/or antiglycemic properties that make it superior to other anti-obesity drug therapies in patients with Type 2 diabetes mellitus. To prove this, head-to-head controlled clinical trials would need to be performed. However, it is interesting to revisit the fact that topiramate was originally developed as an agent to treat Type 2 diabetes mellitus [60,107–109].

Having met efficacy criteria, the main issue regarding the regulatory approval of PHEN/TPM CR would seem to be (not surprisingly) matters of safety [211]. On 15 July 2010, a FDA advisory committee issued a mixed decision of 10 to 6 against endorsing PHEN/TPM CR as an approved pharmaceutical at that time. Reports suggested that: “the FDA and most panel members did not question the effectiveness of” PHEN/TPM CR; “several panel members said they could just as easily have voted the other way.” However, the main potential safety concerns noted were an increase in heart rate (“…in the 1-year cohort, small mean increases in heart rate were observed in the QNEXA treatment groups [0.6–1.6 beats per minute at the top dose] compared with placebo treatment groups…”) [211], possible birth defects, impaired memory and concentration, metabolic acidosis with kidney stones and psychiatric problems (e.g., depression and suicidal ideations) [212].

All of these potential adverse experiences were discussed previously in this review, except for the potential for birth defects. The pregnancy risks of pharmaceuticals are often characterized by the FDA according to categories. The safest are pregnancy categories A (e.g., thyroid medication at replacement doses) and B (e.g., amoxicillin). The least safe are pregnancy categories D (e.g., tetracycline) and X (e.g., statins). Phentermine and topiramate are both pregnancy category C [211], which means that animal reproduction studies have shown an adverse effect on the fetus, but no adequate and well-controlled studies exist in humans, and the potential benefits may warrant use of the drug in pregnant women despite potential risks.

Regarding issues that may have influenced the split decision, some reports suggest that the FDA’s advisory committee’s decision was influenced by the controversy (just days before) involving matters of another FDA advisory committee’s deliberation regarding an otherwise unrelated anti-diabetes mellitus drug [212]. With regard to the FDA’s initial reaction to the decision:

“Eric Colman, deputy director of the FDA’s Division of Metabolism and Endocrinology Products, noted that the agency’s current draft guidelines only call for one year of safety data. He told reporters after the meeting that based on the outcome of Thursday’s committee and the upcoming panels for lorcaserin and Contrave, the FDA likely will update that guidance before releasing a final version of the document. Colman said he was ‘surprised’ by the outcome of Thursday’s vote. ‘I’ve learned that you never know what is going to happen, what they are going to say, how they are going to vote,’ he told reporters. But Colman said he thought there was ‘a little bit of hesitancy’ from those who voted against approval, adding that those panelists ‘were not strongly against the drug, but had lingering concerns that were enough to vote no.’ While the FDA is requiring makers of Type II diabetes drugs to conduct cardiovascular outcomes trials, Colman noted that the agency has yet to mandate the same for weight loss drugs. ‘We’ve had those conversations,’ he said. ‘But the discussion hasn’t evolved to the point to where we are making it formal,’ Colman said, noting the panel’s concerns about increased heart rates in Vivus’ studies of Qnexa” [213].

The reference to the regulatory requirements of cardiovascular outcomes pertains to the FDA guidance on the development of diabetes mellitus drug therapies, as issued in December 2008 (Box 1)[214]. It is therefore possible that similar regulatory recommendations may soon apply to anti-obesity drug development, with the potential requirement of a planned or partially enrolled cardiovascular disease outcome study upon approval. Alternatively or additionally, longer term clinical trials might be required (longer than 1 year) to assess the safety risks associated with chronic anti-obesity therapy.

In addition to legitimate matters of safety, unique obstacles may exist to anti-obesity drug development. These impediments may include bias against pharmaceutical therapies for obesity, especially among those who believe obesity is mainly a behavioral problem that should only be managed through appropriate nutrition and lifestyle changes. But while true that adiposity often has a strong behavioral component, medical science has not historically withheld drug development or medical care on the basis that a disease or condition is substantially due to a lack of medically favorable personal behavior [7]. As importantly, the resistance to anti-obesity drug development may also reflect medical and social biases against the obese, which have existed for centuries [169], and manifest by a sense that treating adiposity is not as important as treating other diseases. In other words, perfectly valid clinical reasons exist why medical science currently has only one or two drugs available to treat the world’s greatest epidemic [87]. But it is also reasonable to question the apparent belief among many that unless a weight-loss pharmaceutical has virtually no risk, then the benefits of weight loss are not sufficient to warrant approval. It is disturbing that adiposity and adiposopathy are on the brink of representing the first epidemics in history that medical science has implicitly decided are untreatable by pharmaceutical therapy.

As such, a most fundamental question facing researchers, regulatory agencies and clinicians is that relative to other diseases, how important is treating adiposity, adiposopathy and affiliated metabolic diseases (e.g., Type 2 diabetes mellitus, hypertension and dyslipidemia)? Through their prescribing habits, clinicians have already decided that the risk:benefit ratio of the 100 mg dose of topiramate (not controlled-release) is acceptable for prophylaxis of migraine headaches [170]. Do clinicians likewise agree that PHEN/TPM CR (which incorporates, at most, 92 mg of topiramate in a controlled-release formulation) is acceptable for treatment of adiposity, adiposopathy and metabolic disease? Is preventing migraine headaches of greater or less clinical importance than improving the numerous metabolic abnormalities associated with adiposity and adiposopathy? (Figure 2). If clinicians ultimately agree that the pathologies of adiposity and adiposopathy are equally as worthy of treatment as preventing migraine headaches, then based upon available evidence, PHEN/TPM CR appears to not only improve the weight of patients, but also appears to improve the metabolic health of patients. This meets the evolving, suggested criteria for weight loss drug development, which states: “An emerging concept is that the development of anti-obesity agents must not only reduce fat mass (adiposity) but must also correct fat dysfunction (adiposopathy)” [60].

Update

Following the online publication of this manuscript on 16 August 2010, the following updates have occurred:

• • In October 2010, sibutramine (Meridia) was withdrawn from the US market due to its potential to increase the risk for cardiovascular events. This left only one drug, orlistat, approved for long-term treatment of obesity (see the section of this manuscript entitled ‘Expert commentary & five-year view, for more information).

• • After the decision of the US FDA advisory panel (described in this manuscript), the FDA issued a Complete Response Letter to the manufacturer of phentermine HCl/topiramate controlled-release (PHEN/TPM CR), wherein additional data was requested [218], including:

  – Clinical: the FDA requested a comprehensive assessment of PHEN/TPM CR’s teratogenic potential, cardiovascular implications of the associated increase in heart rate, and data from the already completed SEQUEL study (OB-305) (see the section of this manuscript entitled ‘Expert commentary & five-year view’ for more information);

  – Labeling: the FDA reserved the right to comment further on proposed labeling;

  – Risk Evaluation and Mitigation Strategy (REMS): the FDA requested a further discussion of a previously submitted REMS plan to better determine the long-term health outcomes of PHEN/TPM CR;

  – Safety update: the FDA requested a safety update of any new adverse experiences;

  – Drug scheduling: the FDA stated that if approved, PHEN/TPM CR would be a Schedule IV drug due to the phentermine component (see the section of this manuscript entitled ‘Phentermine’ for more information).

Table 1. Examples of treatments for adiposopathy (‘sick fat’) and their effects upon illustrative and selected adipose tissue factors that may contribute to metabolic disease.

Intervention	May affect glucose metabolism, BP and lipid metabolism					May affect glucose metabolism		May affect BP		May affect lipid metabolism
	Visceral adipose tissue^†	Free fatty acids	Leptin	Adiponectin		TNF-α		Renin–angiotensin–aldosterone enzymes		Androgens	Estrogens
Weight loss interventions
Appropriate nutrition and physical activity	↓	↓	↓	↑		↓		↓		↓ (women) ↑ (men)	↓/- (men)
Orlistat	↓	↓	↓	↑		↓		?		↓ (women)	?
Sibutramine	↓	↓	↓	↑/-		?		?		↓ (women)	?
Cannabinoid receptor antagonists	↓	↓	↓	↑		↓		?		?	?
Lorcaserin	↓	?	?	?		?		?		?	?
Phentermine	↓ [105]	?	?	?		?		?		?	?
Topiramate	↓ [119]	↓ [172]	↓/- [119,130]	↑ [119]		- [119]		?		?	?
Phentermine/topiramate combination	↓	?	?	?		?		?		?	?
Laparoscopic adjustable gastric banding	↓	?	↓	↑		↓		↓		?	?
Gastric bypass	↓	↓^‡	↓	↑		↓		↓		↑ (men)	↓ (men)
Diabetes mellitus therapies^§
PPAR-γ agonists^¶(pioglitazone, rosiglitazone)	↓/-	↓	↓/-	↑		↓		-		↓	↓/- (men)
Metformin	↓	↓	↓	↑		-		↓		↓ (women)	?
Sulfonylurea	↑	↓	↑/-	↑		↑/-/↓		?		?	?
Insulin	↑/-	↓	↑	↓		↓/-		↑		↓/- (women)	?
Glucagon-like peptide-1 (GLP-1)^#	↓	↓	↓/-	↑/-		-		?		↓	?

^†Visceral adiposity may be approximated by measurements of waist circumference.

^‡Acutely (e.g., 1 month), free fatty acids may be increased during rapid weight loss.

^§Not all of the listed metabolic effects of anti-diabetes mellitus therapies are adipocyte or adipose tissue mediated.

^¶PPAR-γ agents may: (1) increase adipose tissue proliferation and differentiation, (2) favorably alter the visceral-to-subcutaneous adipose tissue deposition ratio, (3) reduce hepatic fat deposition, and (4) improve other aspects of adipose tissue function [3,23,53]. While some of the weight gain associated with PPAR-γ agents is due to fluid retention, much of the weight gain observed with these agents used to treat metabolic diseases traditionally associated with fat weight gain are (paradoxically?) due to promoting increased amounts of functional adipose tissue.

^#Not all of the listed metabolic effects described with GLP-1 have yet been proven to occur with GLP-1 agonists.

↑: Increased; ↓: Decreased; ?: Unreported; -: Neutral effect; PPAR-γ: Peroxisome proliferator-activated receptor-γ.

Data taken from [2,3,23,171].

Table 2. Preliminary data of Phase III clinical trials of the once-a-day phentermine/topiramate controlled-release combination agent.

Clinical trial	Patients	Treatments	Duration	Patients completing the study (%)	Percent weight loss (ITT)	Patients with ≥5% weight (%; ITT)	Metabolic effects compared with placebo	Ref.
EQUATE (OB-301)	n = 756 BMI = 30–45 kg/m^2 Mean BMI = 36.3 kg/m^2 Women = 79% African–American = 19% Diabetes mellitus = 0%	Placebo Phentermine HCl 7.5 mg Phentermine HCl 15 mg Topiramate CR 46 mg Topiramate CR 92 mg PHEN/TPM CR mid dose PHEN/TPM CR full dose	4-week titration followed by 24 weeks of treatment = total of 28 weeks	66% of total study population completed the study on study drug	Placebo = -1.7% Phentermine HCl 7.5 mg = -5.5% Phentermine HCl 15 mg = -6.1% Topiramate CR 46 mg = -5.1% Topiramate CR 92 mg = -6.6% PHEN/TPM CR mid dose = -8.5%* PHEN/TPM CR full dose = -9.2%*	Placebo = 16% PHEN/TPM CR mid dose = 62%** PHEN/TPM CR full dose = 66%** (other values not reported)	PHEN/TPM CR mid and full dose significantly reduced HbA1c by 0.10 and 0.11%, respectively*** (baseline HbA1c = 5.42 and 5.48%, respectively)	[160,161,207,215]
EQUIP (OB-302)	BMI >35 kg/m^2 Mean BMI = 42 kg/m^2 Women = 83% FBG <110 mg/dl TG ≤200 mg/dl BP ≤140/90 mmHg (patients could be on medication for TG and BP)	Placebo (n = 514) PHEN/TPM CR low dose (n = 241) PHEN/TPM CR full dose (n = 512)	4-week titration followed by 52 weeks of treatment = total of 56 weeks	Placebo = 47% PHEN/TPM CR low dose = 57%*** PHEN/TPM CR full dose = 59%***	Placebo = -2% PHEN/TPM CR low dose = -5%*** PHEN/TPM CR full dose = -11%***	Placebo = 17% PHEN/TPM CR low dose = 45%*** PHEN/TPM CR full dose = 67%***	PHEN/TPM CR low dose significantly reduced waist circumference, systolic BP and total cholesterol; PHEN/TPM CR full dose significantly reduced waist circumference, systolic BP, diastolic BP, TG, total cholesterol and LDL-C, and increased HDL-C	[205,216]
CONQUER (OB-303)	BMI = 27–45 kg/m^2 Mean BMI = 37% Women = 70% Two or more of the following comorbidities: high BP, high TG, hyperglycemia, increased waist circumference, Type 2 diabetes mellitus = 16%	Placebo (n = 994) PHEN/TPM CR mid dose (n = 498) PHEN/TPM CR full dose (n = 995)	4-week titration followed by 52 weeks of treatment = total of 56 weeks	Placebo = 57% PHEN/TPM CR mid dose = 69%*** PHEN/TPM CR full dose = 64%***	Placebo = -2% PHEN/TPM CR low dose = -8%*** PHEN/TPM CR full dose = -10%***	Placebo = 21% PHEN/TPM CR low dose = 62%*** PHEN/TPM CR full dose = 70%***	PHEN/TPM CR mid dose significantly reduced waist circumference, systolic BP, TG and total cholesterol, and increased HDL-C levels; PHEN/TPM CR full dose significantly reduced waist circumference, systolic BP, diastolic BP, TG, total cholesterol and LDL-C, and increased HDL-C levels	[162,205,216]

‘Low dose’ = phentermine 3.75 mg/topiramate 23 mg per day. ‘Mid dose’ = phentermine 7.5 mg/topiramate 46 mg per day. ‘Full dose’ = phentermine 15 mg/topiramate 92 mg per day.

*p < 0.001 PHEN/TPM CR versus phentermine and topiramate monotherapies.

**p < 0.001 PHEN/TPM CR versus placebo.

***p < 0.0001 PHEN/TPM CR versus placebo.

BMI: Body mass index; BP: Blood pressure; CONQUER: A Phase III Randomized, Double-Blind, Placebo-controlled, Multicenter Study to Determine the Safety and Efficacy of VI-0521 in the Treatment of Obesity in Adults with Obesity-related Co-morbid Conditions [217]; EQUATE: A Phase III Randomized, Double-Blind, Parallel-Design Study Comparing Multiple Doses of VI-0521 to Placebo and their Single-Agent Phentermine and Topiramate Constituents for the Treatment of Obesity in Adults [217]; EQUIP: A Phase III Randomized, Double-Blind, Placebo-controlled Multicenter Study to Determine the Safety and Efficacy of VI-0521 in the Treatment of Obesity in an Adult Population with BMI ≥35 kg/m²[217]; FBG: Fasting blood glucose; HbA1c: Hemoglobin A1c; HDL-C: High-density lipoprotein cholesterol; ITT: Intent-to-treat population defined as randomized patients with at least one dose of therapy and one post randomization assessment; LDL-C: Low-density lipoprotein cholesterol; PHEN/TPM CR: Phentermine HCl and topiramate controlled-release; TG: Triglyceride.

Table 3. Percent of adverse experiences occurring in over 5% of study participants (regardless of causality) with PHEN/TPM CR during the EQUIP (n = 1264) and CONQUER (n = 2485) trials.

Adverse experience	PHEN/TPM CR EQUIP study				PHEN/TPM CR CONQUER study
	Placebo (%)	Low dose (%)	Full dose (%)		Placebo (%)	Mid dose (%)	Full dose (%)
Dry mouth	3.7	6.7	17.0		2.4	13.5	20.8
Tingling	1.9	4.2	18.8		2.0	13.7	20.5
Constipation	6.8	7.9	14.1		5.9	15.1	17.4
Upper respiratory infection	10.9	15.8	12.3		12.9	12.2	13.4
Altered taste	1.0	1.3	8.4		1.1	7.4	10.4
Insomnia	4.9	5.0	7.8		4.7	5.8	10.3
Headache	10.1	10.4	11.9		9.1	7.0	10.2
Dizziness	4.1	2.9	5.7		3.1	7.2	10.0
Common cold	7.2	12.5	9.0		8.7	10.6	9.9
Sinus infection	5.5	7.5	7.2		6.7	6.8	8.6
Back pain	5.1	5.4	5.5		4.9	5.6	7.2
Nausea	4.7	5.8	7.2		4.2	3.6	6.8
Blurred vision	3.1	6.3	4.5		3.6	4.0	6.0
Bronchitis	4.3	6.7	5.5		4.3	4.4	5.2
Diarrhea	4.5	5.0	4.7		4.8	6.4	5.8
Urinary tract infection	3.5	3.3	4.7		3.7	5.2	5.4
Cough	3.5	3.3	5.1		3.0	3.8	4.8
Influenza	4.7	7.5	5.1		4.3	4.6	3.5

‘Low dose’ = phentermine 3.75 mg/topiramate 23 mg per day. ‘Mid dose’ = phentermine 7.5 mg/topiramate 46 mg per day. ‘Full dose’ = phentermine 15 mg/topiramate 92 mg per day.

CONQUER: A Phase III Randomized, Double-Blind, Placebo-controlled, Multicenter Study to Determine the Safety and Efficacy of VI-0521 in the Treatment of Obesity in Adults with Obesity-related Co-morbid Conditions [217]; EQUIP: A Phase III Randomized, Double-Blind, Placebo-controlled Multicenter Study to Determine the Safety and Efficacy of VI-0521 in the Treatment of Obesity in an Adult Population with BMI ≥35 kg/m²[217]; PHEN/TPM CR: Phentermine HCl and topiramate controlled-release.

Preliminary data taken from [205].

Table 4. Discontinuations due to adverse experiences in the EQUIP and CONQUER studies.

Parameter	Placebo	PHEN/TPM CR low dose	PHEN/TPM CR mid dose	PHEN/TPM CR full dose
Subjects (n)	1508	241	498	1507
Study completion (%)	53	57	69	62
Total discontinuations due to AEs (%)	9	12	12	18
Blurred vision (%)	0.5	2.1	0.8	0.7
Headache (%)	0.7	1.7	0.2	0.9
Insomnia (%)	0.4	0.0	0.4	1.7
Depression (%)	0.2	0.0	0.8	1.4
Tingling (%)	0.0	0.4	1.0	1.2
Irritability (%)	0.1	0.8	0.8	1.2
Anxiety (%)	0.3	0.0	0.2	1.1
Dizziness (%)	0.2	0.4	1.2	0.8

‘Low dose’ = phentermine 3.75 mg/topiramate 23 mg per day. ‘Mid dose’ = phentermine 7.5 mg/topiramate 46 mg per day. ‘Full dose’ = phentermine 15 mg/topiramate 92 mg per day.

AE: Adverse experience; CONQUER: A Phase III Randomized, Double-Blind, Placebo-controlled, Multicenter Study to Determine the Safety and Efficacy of VI-0521 in the Treatment of Obesity in Adults with Obesity-related Co-morbid Conditions [217]; EQUIP: A Phase III Randomized, Double-Blind, Placebo-controlled Multicenter Study to Determine the Safety and Efficacy of VI-0521 in the Treatment of Obesity in an Adult Population with BMI ≥35 kg/m²[217]; PHEN/TPM CR: Phentermine HCl and topiramate controlled-release.

Preliminary data taken from [205].

Box 1. US FDA: 2008 guidance for evaluating cardiovascular risk associated with new anti-diabetes mellitus drug therapies.

“To establish the safety of a new antidiabetic therapy to treat Type 2 diabetes, sponsors should demonstrate that the therapy will not result in an unacceptable increase in cardiovascular risk. To ensure that a new therapy does not increase cardiovascular risk to an unacceptable extent, the development program for a new Type 2 antidiabetic therapy should include the following.

For new clinical studies in the planning stage:

• • Sponsors should establish an independent cardiovascular end points committee to prospectively adjudicate, in a blinded fashion, cardiovascular events during all Phase II and Phase III trials. These events should include cardiovascular mortality, myocardial infarction, and stroke, and can include hospitalization for acute coronary syndrome, urgent revascularization procedures and possibly other end points.

• • Sponsors should ensure that Phase II and Phase III clinical trials are appropriately designed and conducted so that a meta-analysis can be performed at the time of completion of these studies that appropriately accounts for important study design features and patient or study level covariates. To obtain sufficient end points to allow a meaningful estimate of risk, the Phase II and Phase III programs should include patients at higher risk of cardiovascular events, such as patients with relatively advanced disease, elderly patients and patients with some degree of renal impairment. Because these types of patients are likely to be treated with the antidiabetic agent, if approved, this population is more appropriate than a younger and healthier population for assessment of other aspects of the test drug’s safety.

• • Sponsors also should provide a protocol describing the statistical methods for the proposed meta-analysis, including the end points that will be assessed. At this time, we believe it would be reasonable to include in a meta-analysis all placebo-controlled trials, add-on trials (i.e., drug versus placebo, each added to standard therapy), and active-controlled trials, and to preserve the study level randomized comparison but include, when possible in the meta-analysis, important identifiers of study differences or other factors (e.g., dose, duration of exposure, add-on drugs). It is likely that the controlled trials will need to last more than the typical 3–6 months duration to obtain enough events and to provide data on longer-term cardiovascular risk (e.g., minimum 2 years) for these chronically used therapies.

• • Sponsors should perform a meta-analysis of the important cardiovascular events across Phase II and Phase III controlled clinical trials and explore similarities and/or differences in subgroups (e.g., age, sex, race), if possible.

For completed studies, before submission of the new drug application (NDA)/biologics license application (BLA):

• • Sponsors should compare the incidence of important cardiovascular events occurring with the investigational agent to the incidence of the same types of events occurring with the control group to show that the upper bound of the two-sided 95 percent confidence interval for the estimated risk ratio is less than 1.8. This can be accomplished in several ways. The integrated analysis (meta-analysis) of the Phase II and Phase III clinical trials described above can be used. Or, if the data from all the studies that are part of the meta-analysis will not by itself be able to show that the upper bound of the two-sided 95 percent confidence interval for the estimated risk ratio is less than 1.8, then an additional single, large safety trial should be conducted that alone, or added to other trials, would be able to satisfy this upper bound before NDA/BLA submission. Regardless of the method used, sponsors should consider the entire range of possible increased risk consistent with the confidence interval and the point estimate of the risk increase. For example, it would not be reassuring to find a point estimate of 1.5 (a nominally significant increase), even if the 95 percent upper bound was less than 1.8.

• • If the premarketing application contains clinical data that show that the upper bound of the two-sided 95 percent confidence interval for the estimated increased risk (i.e., risk ratio) is between 1.3 and 1.8, and the overall risk-benefit analysis supports approval, a postmarketing trial generally will be necessary to definitively show that the upper bound of the two-sided 95 percent confidence interval for the estimated risk ratio is less than 1.3. This can be achieved by conducting a single trial that is adequately powered or by combining the results from a premarketing safety trial with a similarly designed postmarketing safety trial. This clinical trial will be a required postmarketing safety trial.

• • If the premarketing application contains clinical data that show that the upper bound of the two-sided 95 percent confidence interval for the estimated increased risk (i.e., risk ratio) is less than 1.3 and the overall risk-benefit analysis supports approval, a postmarketing cardiovascular trial generally may not be necessary.

• • The report of this meta-analysis should contain sufficient detail for all the analyses; conventional graphical plots for meta-analysis finding by study, subgroup, and overall risk ratio; and all the analysis data sets that would allow a verification of the findings.”

Key issues

• • Adipose tissue is an active endocrine and immune organ.

• • An increase in fat storage in adipocytes can result in pathologic anatomical, mechanistic, endocrinologic and immunologic adipose tissue responses that may contribute to metabolic disease.

• • Adiposopathy (‘sick fat’) is defined as pathogenic adipose tissue anatomical changes that often occur with positive caloric balance, particularly among those who are environmentally or genetically predisposed, which results in pathological endocrine and immune responses that are important contributors to metabolic disease.

• • Phentermine is a noradrenergic sympathomimetic amine approved for short-term treatment of obesity.

• • Topiramate is a sulfamate-substituted monosaccharide derivative of the naturally occurring sugar monosaccharide D-fructose approved as treatment for migraine headaches and seizure disorders, which also has weight loss effects.

• • Phentermine HCl/topiramate controlled-release is a combination agent containing immediate-release phentermine and controlled-release topiramate that may not only improve the weight of patients, but may also improve adiposopathy-associated metabolic diseases.

Financial & competing interests disclosure

Harold Bays has served as a Clinical Investigator for (and has received research grants from) pharmaceutical companies such as Abbott, Aegerion, Akros, Amarin, Amgen, Amylin, Alteon, Arena, Arete, AstraZeneca, Aventis, Bayer, Boehringer, Bristol-Myers Squibb, Cargill, Ciba Geigy, Daiichi Sankyo, Eli Lilly, Esperion, Essentialis, Fujisawa, GelTex, Genentech, GlaxoSmithKline, Hoechst Roussel, Hoffman LaRoche, Home Access, InterMune, Intekrin, Ironwood Pharmaceuticals, ISIS, Johnson & Johnson, KOS, Kowa, Kyorin, Lederle, Marion Merrell Dow, Merck, Merck Schering Plough, Metabolex, Microbia, Miles, Neuromed, Nicox, Novartis, NovoNordisk, Obecure, Orexigen, Parke Davis, Pfizer, Pliva, Posen, Purdue, Reliant, Roche, Rorer, Regeneron, Sandoz, Sanofi, Sciele, Searle, Shionogi, Schering Plough, SmithKline Beecham, Surface Logix, Takeda, TAP, UpJohn, Upsher Smith, Warner Lambert, Vivus and Wyeth-Ayerst. Harold Bays has served as a consultant, speaker and/or advisor to and for pharmaceutical companies such as Abbott, Amarin, Arena, AstraZeneca, Aventis, Bayer, Bristol-Myers Squibb, Cerenis, Daiichi Sankyo, DSM Nutritional Products, Inc., Essentialis Therapeutics, GlaxoSmithKline, Ironwood Pharmaceuticals, Johnson & Johnson, KOS, Merck, Merck Schering Plough, Metabasis Therapeutics, Microbia, Novartis, Nicox, Ortho-McNeil, Parke Davis, Perenis Therapeutics, Pfizer, Reliant, Roche, Sandoz, Sanofi Aventis, Schering Plough, SmithKline Beecham, Surface Logix, Takeda, UpJohn, Vivus and Warner Lambert. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No funding or writing assistance was utilized in the production of this manuscript.

Notes

Data taken from [214].