Rimonabant: a Cannabinoid Receptor Type 1 Blocker for Management of Multiple Cardiometabolic Risk Factors Eli V

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector Journal of the American College of Cardiology Vol. 47, No. 10, 2006 © 2006 by the American College of Cardiology Foundation ISSN 0735-1097/06/$32.00 Published by Elsevier Inc. doi:10.1016/j.jacc.2005.12.067 STATE-OF-THE-ART PAPER

Rimonabant: A Cannabinoid Receptor Type 1 Blocker for Management of Multiple Cardiometabolic Risk Factors Eli V. Gelfand, MD,* Christopher P. Cannon, MD† Boston, Massachusetts

Rimonabant is a first selective blocker of the cannabinoid receptor type 1 (CB1) being developed for the treatment of multiple cardiometabolic risk factors, including abdominal obesity and smoking. In four large trials, after one year of treatment, rimonabant 20 mg led to greater weight loss and reduction in waist circumference compared with placebo. Therapy with rimonabant is also associated with favorable changes in serum lipid levels and an improvement in glycemic control in prediabetes patients and in type 2 diabetic patients. At the same dose, rimonabant significantly increased cigarette smoking quit rates as compared with placebo. Rimonabant seems to be well tolerated, with a primary side effect of mild nausea. As an agent with a novel mechanism of action, rimonabant has a potential to be a useful adjunct to lifestyle and behavior modification in treatment of multiple cardiometabolic risk factors, including abdominal obesity and smoking. (J Am Coll Cardiol 2006;47: 1919–26) © 2006 by the American College of Cardiology Foundation

OBESITY, METABOLIC SYNDROME, (Adult Treatment Panel III) put forward specific guidelines AND SMOKING: THE SCOPE OF THE PROBLEM that define a population with metabolic syndrome (Table 1) (6), and describe individuals with a markedly elevated risk of Cardiovascular disease remains by far the leading cause of developing diabetes and clinically significant atherosclerosis. morbidity and mortality in the developed world, accounting Indeed, the presence of metabolic syndrome may be asso- for almost one million deaths annually in the U.S. alone. ciated with a significant excess burden of myocardial infarc- Tobacco smoking remains the primary preventable cause of tion, stroke, and overall cardiovascular mortality (7). death, contributing to nearly 20% of all deaths worldwide. Modification of cardiometabolic risk factors, including Recent startling reports project a decline in life expectancy smoking and abdominal obesity, has a well-documented in the U.S. during the 21st century (1). These reports warn favorable effect on clinical outcomes. Smoking cessation is of a catastrophic impact of the global obesity epidemic on rates of diabetes mellitus and cardiovascular disease in the associated with a rapid decrease in the risk of developing upcoming years. Currently in the U.S., 28% of men and adverse cardiovascular events, including myocardial infarc- 34% of women are obese, and the largest increases in obesity tion, stroke, and sudden death (8,9). A decrease in body rates have affected children and minorities (2). In this weight and reduction in waist circumference is associated country, obesity is estimated to be associated with over with favorable changes in the lipid profile and C-reactive 100,000 excess deaths every year (3). protein and a decrease in mortality (10,11). However, The old view of adipose tissue as an inert storage depot success with behavior modification and currently available was supplanted more recently by its depiction as a dynamic medications in achieving reliable abstinence from tobacco endocrine organ. Adipose tissue secretes a variety of factors, and with sustained weight loss is quite limited. Surgical or adipokines, that contribute to insulin resistance, vascular procedures, such as gastric banding or gastric bypass, are endothelial dysfunction, and atherogenesis (4,5). It is there- effective for sustained weight loss, but are clearly invasive and fore not surprising that abdominal obesity is not an isolated are associated with nontrivial morbidity and mortality (12). pathophysiological entity, but often coexists with hyperten- Therefore, these are generally used as a last resort for patients sion, glucose intolerance, and dyslipidemia. Recognizing with morbid obesity (body mass index [BMI] Ͼ40 kg/m2) and this, the 2001 National Cholesterol Education Program those with multiple obesity-related complications (13). With regard to smoking cessation, behavioral therapy, nicotine replacement therapy, and pharmacologic approaches (bupro- From the *Cardiovascular Division, Department of Medicine, Beth Israel Deacon- ess Medical Center, and the †TIMI Study Group, Cardiovascular Division, Depart- prion) have generally yielded only modest quit rates (14), and ment of Medicine, Brigham and Women’s Hospital, Harvard Medical School, many individuals resume smoking within one year. Addition- Boston, Massachusetts. Dr. Cannon serves on advisory boards and consult for Sanofi-Aventis and has spoken at medical symposia sponsored by this company. The ally, many who successfully quit smoking gain a considerable authors also recently completed a trial (CLARITY-TIMI 28) sponsored by Sanofi- amount of weight, which serves as a barrier for those who are Aventis, but do not currently have ongoing research grants. Dr. Cannon serves as an considering tobacco cessation. unpaid member of the STRADIVARIUS trial Steering Committee, led by Dr. Steven E. Nissen. Despite disappointing results for treatment of obesity and Manuscript received November 29, 2005, accepted December 14, 2005. for smoking cessation, both are mandated as part of the 1920 Gelfand and Cannon JACC Vol. 47, No. 10, 2006 Rimonabant May 16, 2006:1919–26

Table 2. ACC/AHA Unstable Angina and Non–ST-Segment Abbreviations and Acronyms Elevation Myocardial Infarction Guidelines: Risk Factor Modification BMI ϭ body mass index ϭ CB1 cannabinoid receptor type 1 Smoking cessation* HDL ϭ high-density lipoprotein Achievement of optimal body weight* LDL ϭ low-density lipoprotein Daily exercise RIO ϭ Rimonabant in Obesity AHA diet STRATUS ϭ Studies with Rimonabant Control of hypertension to BP Ͻ130/85 mm Hg and Tobacco Use Tight control of hyperglycemia in diabetic patients* THC ϭ tetrahydrocannabinol Lipid-lowering agents for LDL Ͼ130 mg/dl Lipid-lowering agent in LDL Ͼ100 mg/dl after diet Drug therapy if HDL Ͻ40 mg/dl* overall cardiometabolic risk factor reduction for primary and *Can be impacted by rimonabant. Data from Braunwald et al. (16). ACC ϭ American College of Cardiology; AHA ϭ American Heart Associ- secondary prevention of heart disease (Table 2)(15,16). An ation; BP ϭ blood pressure; HDL ϭ high-density lipoprotein; LDL ϭ low-density intriguing new combined approach to treating the obesity lipoprotein. and glucose intolerance features of metabolic syndrome, as anolamide (anandamide), and 2-arachidonoylglycerol (2- well as aiding smoking cessation, involves manipulation of AG). Under normal conditions, the endocannabinoid sys- the endogenous cannabinoid system, specifically with the tem is not tonically active, rather endocannabinoids are cannabinoid receptor type 1 (CB ) antagonist rimonabant. 1 produced on demand, act locally, and are rapidly inactivated THE ENDOCANNABINOID SYSTEM via cellular uptake and enzymatic hydrolysis (26). More recently, cannabinoid antagonists were developed, Cannabis. Hemp (Cannabis sativa) has been cultivated in of which rimonabant has been the most extensively studied. many parts of the world for over 4,500 years. The plant is It has a high affinity for the central CB1 receptors (27), and used for its fiber and oil, and abused as a recreational drug its potential clinical uses will be discussed later in this (marijuana, hashish). The main psychoactive alkaloid in review. cannabis is ⌬-9-tetrahydrocannabinol (THC), but the plant contains more than 60 other cannabinoids, some of which PHYSIOLOGY OF THE CANNABINOID SYSTEM modulate the actions of THC. Synthetic THC (dronabinol) Cardiovascular effects. Cannabinoids have long been is used to treat post-chemotherapy nausea and emesis, as known to have potent psychotropic actions, but their wide- well as anorexia associated with human immunodeficiency ranging effects on the cardiovascular system are just begin- virus infection. ning to be unraveled. Cannabinoid receptors and their ligands. The cannabi- In anesthetized rat models, intravenously administered noids exert their pharmacologic action through the interac- anandamide produces a triphasic hemodynamic response tion with the specific receptors CB and CB , which were 1 2 (28): a brief period of vagally mediated bradycardia and described in the late 1980s and later were cloned (17,18) hypotension, followed by a transitory pressor reaction, and a (Table 3). The CB receptors are primarily distributed to 1 relatively prolonged vasodepressor response. The latter is the brain (19) and adipose tissue (20), but are also found in the dominant effect of anandamide in animal models, and it the myocardium (21), vascular endothelium (22), and sympathetic nerve terminals (23). The CB2 receptors are pri- Table 3. History of Endocannabinoid System Research marily located in the lymphoid tissue and peripheral mac- and Rimonabant Trials rophages (24). Both receptors function as transmembrane 1964 Isolation of ⌬-9 THC, the active constituent of Cannabis G-proteins. Existence of CB3 receptors has been postulated sativa (25), but the receptor itself has not yet been cloned. 1988 High-affinity cannabinoid binding site discovered in rat brain Cannabinoid receptors have affinity for at least two 1990 Cloning of the rat G-protein–coupled CB1 receptor 1991 Cloning of the human CB1 receptor endogenous ligands: small lipid molecules arachidonyleth- 1992 Discovery of anandamide, the first endogenous cannabinoid 1993 Cloning of the peripheral CB receptor Table 1. Components of Metabolic Syndrome, as Defined 2 1994 Characterization of the first selective CB1 receptor blocker, by the Adult Treatment Panel III rimonabant At least three of the following traits 1995 Isolation of a second cannabinoid, 2-AG, in brain Abdominal obesity 2004 RIO-Lipids, RIO-Europe 1 year, STRATUS-US, and RIO- Men, Ͼ102 cm (40 inches) NA studies presented Women, Ͼ88 cm (35 inches) 2005 RIO-Europe and RIO-Lipids published Serum triglycerides 150 mg/dl (1.7 mmol/l) RIO-Diabetes study presented Serum high-density lipoprotein cholesterol Ͻ40 mg/dl (1 mmol/l) 2006 RIO-NA study published Ͻ in men and 50 mg/dl (1.3 mmol/l) in women Adapted with permission from a presentation by Dr. J. P. Despres (Quebec Heart Blood pressure Ͼ130/85 mm Hg Institute, Ste. Foy, Quebec, Canada), and updated. ϭ ϭ ϭ Fasting plasma glucose 110 mg/dl (6.1 mmol/l) 2-AG 2-arachidonoylglycerol; CB1 cannabinoid receptor type 1; NA North America; RIO ϭ Rimonabant in Obesity; STRATUS ϭ Studies with Adapted with permission (6). Rimonabant and Tobacco Use; THC ϭ tetrahydrocannabinol; US ϭ United States. JACC Vol. 47, No. 10, 2006 Gelfand and Cannon 1921 May 16, 2006:1919–26 Rimonabant

results from CB1-mediated inhibition of norepinephrine trations during diet-induced weight loss (41). Leptin, an release from presynaptic nerve terminals (29). In humans, endogenous hormone, can reduce the food intake. Serum acute administration of the cannabinoids produces vasodi- concentration of leptin is directly proportional to the degree lation and tachycardia with a variable net effect on systemic of adiposity, but obese individuals have lower sensitivity to blood pressure (30), but long-term use of THC results in leptin (42). An adipose tissue-speciﬁc protein, adiponectin,

CB1-mediated hypotension and bradycardia (31,32). stimulates fatty acid oxidation and a decrease in body Although CB1 receptors are mostly expressed on the weight. Its levels are reduced in obesity (43). neuronal terminals, there is evidence showing that other cell Both cannabinoid receptors and their endocannabinoid types express these receptors and participate in cannabinoid ligands are present in all of the tissues that play an important physiology. Endocannabinoids induce vasodilation by acting role in regulation of food intake. Levels of endocannabi- directly on the CB1 receptors in the arterial smooth muscle noids in the hypothalamus are decreased after administra- in the brain (33). These compounds also induce vasodilation tion of leptin (44). The CB1 agonists are potent, dose- in a variety of vascular beds through an endothelium- dependent inducers of hyperphagia in rodents (45–47), and dependent increase in nitric oxide synthesis (34), but at least antagonism of CB1 receptors prevents hyperphagia in a some of the vasodilation is probably independent of the CB1 starvation model (44). Knockout mice lacking CB1 recep- receptor system. tors show a lean phenotype, primarily as a result of spon- Endocannabinoid systems seem to be involved in regula- taneously reduced caloric intake (40). Indeed, when such tion of vascular tone in hepatic disease, hypertension, and animals are fed a high-fat, obesity-promoting diet, they other disorders. In advanced cirrhosis, endocannabinoids remain lean, and compared with wild-type animals, show mediate the vasodilatory state through their interaction with lower plasma insulin levels and a higher sensitivity to leptin the CB1 receptors. In spontaneously hypertensive rats, the (48). In the liver, endocannabinoids, acting via CB1 recep- cardiac and vascular endothelial CB1 system becomes ton- tors, act to induce lipogenic gene expression and stimulate ically active, and such animals show a more pronounced de novo synthesis of fatty acids (49). vasodepressor/hypotensive response to anandamide than do Endocannabinoids and addiction. Regions of the brain wild types (35). Rimonabant blocks the vasodepressor effect thought to be involved in drug relapse behavior contain high of anandamide in hypertensive animals, but not in normal levels of CB1 receptors (19), and compelling evidence animals, indicating that the CB1 system is largely inactive suggests a role for the endocannabinoid system in formula- under normal hemodynamic conditions. tion and propagation of addiction to psychoactive sub- Recent work shows that the endocannabinoid system also stances. Speciﬁcally, endocannabinoids seem to modulate plays a role in hemodynamics of shock states. Indeed, under cue reactivity and conditioned reinforcement after pro- conditions of experimental hemorrhage (36), myocardial longed abstinence of drug and natural reinforcers (50). infarction (37), or endotoxemia (38,39), macrophages and These effects have been shown for a wide range of addictive circulating platelets elaborate anandamide, which contrib- substances, including cocaine (51), heroin (52), amphet- utes to the onset of hypotension and shock. Blockade of amines (53), and alcohol (54). Studies have shown an

CB1 receptors with rimonabant attenuates these effects. important role for the endocannabinoid system in the Metabolic effects. There is increasing evidence showing modulation of nicotine addiction. Indeed, the rewarding that the endocannabinoid system plays a central role in effects of nicotine were abolished in knockout mice lacking regulating metabolism and body composition by enhancing CB1 receptors (55), and as described further below, admin- the central orexigenic drive and increasing peripheral lipo- istration of the selective CB1 antagonist rimonabant de- genesis (Table 4)(40). creases nicotine-seeking behaviors (56). Interestingly, endo- Control of food intake and body composition is the result cannabinoid involvement in nicotine dependence seems to of a series of complex interactions between the adipocytes, be limited to its psychological aspects, as the physical the mesolimbic system, the hypothalamus, and the gastro- aspects of nicotine withdrawal are not attenuated in CB1- intestinal tract. A sense of hunger is mediated in part by the deficient mice (57). gut hormone ghrelin, which is produced in higher concen- RIMONABANT Table 4. The Effects of CB1 Blockade on Food Intake and Cardiometabolic Risk Factors Rimonabant was first described in 1994 by Rinaldi- Central blockade Decreased food intake Carmona et al. (27). At lower concentrations, it blocks the (hypothalamus) CB receptors. At very high concentrations, rimonabant Peripheral blockade Decreased abdominal fat (waist circumference) 1 (adipose tissue) 1 Adiponectin behaves as a CB2 receptor antagonist (27), blocks calcium 2 Triglycerides and potassium channels (58), and may directly affect cellular 1 High-density lipoprotein gap junctions (59). 2 Small, dense low-density lipoprotein Rimonabant for treatment of obesity and cardiometa- 2 C-reactive protein bolic risk factors: animal data. The impact of rimonabant 2 Insulin resistance therapy on metabolism, food intake, and body composition 1922 Gelfand and Cannon JACC Vol. 47, No. 10, 2006 Rimonabant May 16, 2006:1919–26 was first investigated by several groups in standard rodent models. In an important study, Di Marzo et al. (44) showed that treatment with rimonabant was associated with a reduction in food intake and a 4% loss of body weight in wild-type mice, but not in CB1 receptor-deficient mice. Decrease in adiposity accounted for most of the rimonabant-induced weight loss, because muscle mass re- mained unchanged (60). Ravinet-Trillou et al. (61) showed that in mice with diet-induced obesity, rimonabant therapy was associated with only a transient reduction in food intake, but a marked and sustained weight reduction (20%) and a depletion of fat stores (ϳ50%). In that study, rimonabant-treated animals showed lower plasma glucose and insulin levels, as well as improved insulin resistance. Notable recent findings by the same group suggest that decreased food intake alone cannot account for the sus- Figure 1. Change in body weight among subjects in the Rimonabant in tained weight loss during rimonabant treatment. In fact, Obesity (RIO)-Lipids study. Reprinted, with permission, from Despres et after the first week of treatment, a mild increase in food al. (62). intake ensues, yet steady weight loss continues (60). Expla- nation may lie with the evidence that rimonabant induces Rimonabant for treatment of obesity: trials in humans. changes in the adipose tissue both at the cellular and at the Based on the animal data, the Rimonabant in Obesity molecular levels. Grossly, adipocytes in rimonabant-treated (RIO) phase 3 program of four randomized double-blind placebo-controlled clinical trials in humans was initiated animals are smaller and reflect a marked decrease in fat (Table 5). stores rather than adipocyte apoptosis (60). Using deoxyri- bonucleic acid chip technology, Jbilo et al. (60) showed that THE RIO-LIPIDS TRIAL. The RIO-Lipids trial enrolled gene modulations induced by rimonabant treatment were 1,036 patients with mild or moderate obesity (mean BMI, opposite to those effected by a high-fat diet, and were very 34 kg/m2) and untreated hyperlipidemia (62). Patients were similar to those in CB1 knockout mice. Rimonabant was randomized in a parallel fashion to receive rimonabant 20 also shown to increase adiponectin levels by stimulating mg/day, rimonabant 5 mg/day, or matching placebo for one adiponectin messenger ribonucleic acid expression in the year, and weight loss was assessed at the end of treatment. adipocytes (20). These findings lend strong support to the At the end of one year, treatment with rimonabant was associated with significantly greater weight loss compared CB1-mediated mechanism of the anti-obesity action of with placebo (Fig. 1). Indeed, 58.4% of subjects in the rimonabant. In addition, treatment with rimonabant was Ն associated with an induction of several glycolytic enzymes, high-dose rimonabant group had sustained a loss of 5% body weight, compared with 30.0% in the low-dose rimon- which could explain the glucose-lowering effect of rimonabant group and 19.5% in the placebo group (p Ͻ 0.001 for abant. Finally, there was a reduction in the expression of high-dose rimonabant vs. placebo). A more substantial multiple pro-inflammatory proteins, known to be upregu- Ն10% weight loss was sustained by 32.6% of subjects in the lated in obesity (60). high-dose rimonabant group, compared with 10.6% in the low-dose group and 7.2% in the placebo group (p Ͻ 0.001 Table 5. Clinical Trials of Rimonabant (as of December 2005) for high-dose rimonabant vs. placebo) (Fig. 2). Among Name (Ref. No.) Status n Sponsor patients who completed a full one-year course of treatment, 72.9% of patients in the high-dose group lost Ն5% body RIO-Lipids (62) Published 1,036 Sanofi-Aventis Ͻ RIO-Europe (64) Published 1,507 Sanofi-Aventis weight, compared with 27.6% for placebo (p 0.001). RIO-NA (65) Published 3,040 Sanofi-Aventis With regard to the lipid parameters, at the end of treatment RIO-Diabetes (66) Presented 1,045 Sanofi-Aventis the subjects in the high-dose rimonabant group had a 23% STRATUS-US (68) Presented 787 Sanofi-Aventis increase in high-density lipoprotein (HDL) levels and a STRATUS-EU Completed 787 Sanofi-Aventis 15% decrease in triglyceride levels. Both were different from STRATUS-WW Presented 5,055 Sanofi-Aventis Ͻ STRADIVARIUS (67) Enrolling 800 (projected) Sanofi-Aventis placebo (p 0.001 for both). The C-reactive protein levels Rimonabant to Enrolling 40 (projected) National Institutes were lower in the high-dose rimonabant group (27% reduc- Reduce Alcohol of Health tion vs. 11% for placebo, p ϭ 0.007), and the low-density Consumption (69) lipoprotein (LDL) levels were not significantly affected by EU ϭ Europe; NA ϭ North America; RIO ϭ Rimonabant in Obesity; treatment. Rimonabant 20 mg also increased adiponectin STRADIVARIUS ϭ Strategy to Reduce Atherosclerosis Development Involving Ͻ Administration of Rimonabant—The Intravascular Ultrasound Study; STRATUS ϭ levels by 57.7% (p 0.001), a change that was partly Studies with Rimonabant and Tobacco Use; US ϭ United States; WW ϭ worldwide. independent of weight loss alone. An important final JACC Vol. 47, No. 10, 2006 Gelfand and Cannon 1923 May 16, 2006:1919–26 Rimonabant

loss, this 3,045-subject trial had evaluated whether weight loss achieved with rimonabant could be maintained after withdrawal of the drug. As in prior RIO trials, subjects were initially randomized to rimonabant 20 mg/day, rimonabant 5 mg/day, or placebo for one year. However, after the completion of one year of treatment, subjects in the two rimonabant groups underwent a second randomization, either to continue receiving their previously assigned dose of rimonabant or to be switched to a matching placebo. The first-year outcomes were similar to those of the other RIO trials: weight loss was significantly greater in the rimonabant treatment groups (Ϫ6.3 kg for 20 mg/day, Ϫ4.4 kg for 5 mg/day vs. Ϫ1.6 kg for placebo, p Ͻ 0.01 for both comparisons). After two years, subjects re-randomized to placebo after the end of one year have regained much of their weight (overall loss of 3.2 kg vs. 2.3 kg for patients on placebo for two years), whereas those who were treated with Figure 2. Percentage of subjects achieving Ն10% weight loss at one year with rimonabant (solid bars) 20 mg/day versus placebo (open bars) in the rimonabant 20 mg/day for the full two years lost an average first two Rimonabant in Obesity (RIO) trials among those completing the of 7.4 kg (p Ͻ 0.01 for placebo comparison). Rates of study. metabolic syndrome were improved with rimonabant, with which a significant reduction was seen in the 20-mg dose finding was that at the end of treatment, the proportion of patients satisfying the National Cholesterol Education group (34.8% to 21.1%) compared with placebo (31.7% to Program-Adult Treatment Panel III criteria for metabolic 29.2%). An increase in HDL of 24.5% was seen in the syndrome was significantly lower in the high-dose rimonabant- rimonabant 20 mg/day group, compared with 13.8% in the treated group compared with placebo (25.8% vs. 41.0%, p Ͻ placebo group. 0.001) (62). THE RIO-DIABETES TRIAL. The most recent study reported THE RIO-EUROPE TRIAL. Results of RIO-Lipids were con- from the RIO phase 3 program was the RIO-Diabetes trial firmed in a similar-sized RIO-Europe trial (63,64) that (66), which enrolled 1,047 patients with type 2 diabetes 2 enrolled obese subjects with BMI Ն30 kg/m2,orϾ27 mellitus and a BMI 27 to 40 kg/m . Again, subjects were kg/m2 with a comorbidity, defined as hypertension or randomized to receive rimonabant 20 mg/day, rimonabant 5 dyslipidemia. In this trial, 1,507 subjects were enrolled and mg/day, or placebo for one year. All patients were also assigned randomly to receive rimonabant 20 mg/day, treated with an oral hypoglycemic drug, as prescribed by rimonabant 5 mg/day, or placebo. Subjects were also given their treatment physician, with the majority receiving met- instructions for moderate physical exercise and a mild formin. At the end of one year, therapy with rimonabant 20 hypocaloric diet. Among patients completing one full year mg was associated with an average weight loss of 5.3 kg, of treatment (61%), loss of Ն5% body weight was achieved compared with 1.4 kg in the placebo group (p Ͻ 0.001). significantly more frequently in both rimonabant groups Average levels of glycosylated hemoglobin were decreased compared with placebo (67.4% for 20 mg, 44.2% for 5 mg by 0.6% in the rimonabant 20 mg group from a baseline vs. 30.5% for placebo, p Ͻ 0.01 for both placebo compar- level of 7.3%, but were increased in the placebo group by isons). Treatment with either dose of rimonabant was also 0.1% (p Ͻ 0.001). Effects of the rimonabant 5 mg/day were associated with significantly greater waist circumference less significant. Importantly, 43% of all subjects treated with reduction than placebo (6.5 cm for 20 mg, 3.9 cm for 5 mg, rimonabant achieved an optimal glycosylated hemoglobin Ͻ 2.4 cm for placebo, p 0.01 for both comparisons). level of Ͻ6.5%, compared with just 21% of those receiving Triglyceride levels decreased and HDL levels increased in placebo (p Ͻ 0.001). both rimonabant groups, and the investigators suggested In summary, the RIO trials showed that in patients that elevated levels of adiponectin were contributing to with obesity, including those with cardiovascular comor- these effects. Finally, treatment with high-dose rimonabant bidities, continued therapy with rimonabant as compared was associated with a significantly greater reduction in the with placebo is associated with a significant reduction in percentage of subjects with metabolic syndrome than pla- body weight and waist circumference. Such therapy is cebo: from 42.2% at baseline to 19.6% at one year (63) and also associated with other favorable changes in the 21.5% at two years (p Ͻ 0.001 compared with placebo) (64). cardiometabolic risk profile, including an improvement in THE RIO-NA TRIAL. Obese patients in North America (NA) glycemic control in type 2 diabetics, an improvement in were enrolled into the RIO-NA trial (65). In addition to the lipid profile, and an overall decrease in the prevalence evaluating the efficacy of rimonabant for primary weight of metabolic syndrome. 1924 Gelfand and Cannon JACC Vol. 47, No. 10, 2006 Rimonabant May 16, 2006:1919–26

THE STRATEGY TO REDUCE ATHEROSCLEROSIS DEVELOP- neuropsychiatric, cardiac, vascular, and metabolic patho- MENT INVOLVING ADMINISTRATION OF RIMONABANT— physiological processes, and a wealth of animal data on both THE INTRAVASCULAR ULTRASOUND STUDY (STRADIVARIUS). endogenous ligands and rimonabant, offer multiple intrigu- The ongoing STRADIVARIUS trial (67) will test whether ing possibilities for clinical use in humans. A phase 2 clinical the improvement in the cardiometabolic risk profile effected trial of rimonabant to reduce alcohol consumption is being by rimonabant translates into changes within the coronary sponsored by the National Institutes of Health (69). Animal circulation. The STRADIVARIUS trial is enrolling obese experiments showing that blockade of CB1 receptors with subjects who either are smokers or have at least two rimonabant attenuates shock caused by extreme hemorrhage additional features that fit the standard definition of meta- (36), endotoxemia (38), or myocardial infarction (37) will bolic syndrome, and in whom a clinically indicated coronary likely prompt human clinical trials in the near future. angiography reveals a 20% to 50% stenosis. The volume of Rimonabant may also find use in the treatment of vasodi- atheroma will be assessed by intravascular ultrasound. Sub- latory state and chronic hypotension in patients with ad- jects will then be randomized to rimonabant 20 mg/day, vanced liver disease (70). rimonabant 5 mg/day, or placebo, and the end point will be Adverse effects in clinical trials of rimonabant. Initial change in the volume of target atheroma at the time of a experience with rimonabant shows that it is generally well mandatory repeat angiography at 18 months. There are also tolerated. In the RIO phase 3 program, the one-year plans for a large clinical outcomes trial to begin in the dropout rates were high (36% to 49%), but were typical of coming year. obesity trials and did not differ from placebo. The most Rimonabant for smoking cessation: the Studies with common adverse effect was mild nausea. Given its pharma- Rimonabant and Tobacco Use (STRATUS) trials. En- cology, there is concern regarding the neuropsychiatric rolling concurrently with the RIO program, the STRATUS effects of rimonabant, such as higher incidence of anxiety trials are examining the potential role of rimonabant as an and depressed mood disorders. Use of the Hospital Anxiety adjunct in smoking cessation (Table 5). and Depression scale in RIO-Europe showed no difference In a randomized, double-blind, placebo-controlled between the treatment groups in the average subscale scores STRATUS-United States (US) trial (68), 787 subjects were for either major depression or anxiety (63). During one year enrolled who smoked Ն10 cigarettes/day (average, 23 cig- of treatment, six subjects (1.0%) in the rimonabant 20 arettes/day) for at least two months and who were motivated mg/day group and one subject (0.3%) in the placebo group to quit. Subjects were randomly assigned to receive rimon- discontinued their study drug because of depression. For abant 20 mg/day, rimonabant 5 mg/day, or placebo for 10 depressed mood disorders, the rates were 3.7% and 3.0% for weeks, and were asked to quit smoking on day 15 of the rimonabant 20 mg/day and placebo, respectively. Similarly, study. End points included smoking abstinence rate as well in RIO-Lipids, the Hospital Anxiety and Depression scale as a change in body weight in those who were abstinent scores were similar for anxiety and depression between the from cigarettes at one year. At the end of the study, the rate two treatment groups and placebo (62). Therefore, it seems of abstinence was significantly higher in the high-dose that the percentage of patients experiencing neuropsychiat- rimonabant group compared with placebo (36.2% vs. 20.6%, ric side effects is small. Monitoring for on-treatment anxiety p Ͻ 0.001), but not in the low-dose rimonabant group and depression in the future will nonetheless be necessary to (20.2%). Among subjects with prolonged abstinence, those ensure safe use of this important new therapy. in the placebo group gained an average of 3.7 kg of body weight, compared with 0.6 kg in the high-dose rimonabant Reprint requests and correspondence: Dr. Christopher P. Cannon, group (p Ͻ 0.001), representing an impressive 84% reduc- Cardiovascular Division, Brigham and Women’s Hospital, 75 tion in weight gain for rimonabant over placebo. Subgroup Francis Street, Boston, Massachusetts 02115. E-mail: cpcannon@ analysis showed that among subjects who were initially partners.org. overweight, those who were abstinent from tobacco while receiving rimonabant 20 mg/day had not gained any weight by one year (weight change Ϫ0.1 kg vs. ϩ1.7 kg for placebo, REFERENCES Ͻ p 0.001). As with the RIO trials, no differences were 1. Olshansky SJ, Passaro DJ, Hershow RC, et al. A potential decline in noted in the rate of dropout among the treatment groups. life expectancy in the United States in the 21st century. N Engl J Med The STRATUS-Europe (EU) has a protocol identical to 2005;352:1138–45. 2. Hedley AA, Ogden CL, Johnson CL, Carroll MD, Curtin LR, Flegal STRATUS-US, and is following up 789 subjects in Europe; KM. Prevalence of overweight and obesity among US children, STRATUS-Worldwide (WW) is a large one-year mainte- adolescents, and adults, 1999–2002. JAMA 2004;291:2847–50. nance study with a treatment-free one-year follow-up that 3. Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deaths associated with underweight, overweight, and obesity. JAMA 2005; was conducted among 5,055 subjects across 54 sites world- 293:1861–7. wide. Results from both of these trials are expected within 4. Lyon CJ, Law RE, Hsueh WA. Minireview: adiposity, inflammation, one year. and atherogenesis. Endocrinology 2003;144:2195–200. 5. Eckel RH. Obesity. Circulation 2005;111:e257–9. Potential uses of rimonabant in other disorders. Involve- 6. Executive Summary of the Third Report of the National Cholesterol ment of the endocannabinoid system in a wide variety of Education Program (NCEP) Expert Panel on Detection, Evaluation, JACC Vol. 47, No. 10, 2006 Gelfand and Cannon 1925 May 16, 2006:1919–26 Rimonabant

And Treatment of High Blood Cholesterol in Adults (Adult 30. Huestis MA, Sampson AH, Holicky BJ, Henningfield JE, Cone EJ. Treatment Panel III). JAMA 2001;285:2486–97. Characterization of the absorption phase of marijuana smoking. Clin 7. Isomaa B, Almgren P, Tuomi T, et al. Cardiovascular morbidity and Pharmacol Ther 1992;52:31–41. mortality associated with the metabolic syndrome. Diabetes Care 31. Benowitz NL, Jones RT. Cardiovascular effects of prolonged delta-9- 2001;24:683–9. tetrahydrocannabinol ingestion. Clin Pharmacol Ther 1975;18:287–97. 8. Iso H, Date C, Yamamoto A, et al. Smoking cessation and mortality 32. Lake KD, Compton DR, Varga K, Martin BR, Kunos G. from cardiovascular disease among Japanese men and women: the Cannabinoid-induced hypotension and bradycardia in rats mediated JACC Study. Am J Epidemiol 2005;161:170–9. by CB1-like cannabinoid receptors. J Pharmacol Exp Ther 1997;281: 9. Goldenberg I, Jonas M, Tenenbaum A, et al. Current smoking, 1030–7. smoking cessation, and the risk of sudden cardiac death in patients 33. Gebremedhin D, Lange AR, Campbell WB, Hillard CJ, Harder DR. with coronary artery disease. Arch Intern Med 2003;163:2301–5. Cannabinoid CB1 receptor of cat cerebral arterial muscle functions 10. Eriksson KF, Lindgarde F. No excess 12-year mortality in men with to inhibit L-type Ca2ϩ channel current. Am J Physiol 1999;276: impaired glucose tolerance who participated in the Malmo Preventive H2085–93. Trial with diet and exercise. Diabetologia 1998;41:1010–6. 34. Deutsch DG, Goligorsky MS, Schmid PC, et al. Production and 11. Wadden TA, Anderson DA, Foster GD. Two-year changes in lipids physiological actions of anandamide in the vasculature of the rat and lipoproteins associated with the maintenance of a 5% to 10% kidney. J Clin Invest 1997;100:1538–46. reduction in initial weight: some findings and some questions. Obes 35. Batkai S, Pacher P, Osei-Hyiaman D, et al. Endocannabinoids acting Res 1999;7:170–8. at cannabinoid-1 receptors regulate cardiovascular function in hyper- 12. Flum DR, Salem L, Elrod JA, Dellinger EP, Cheadle A, Chan L. tension. Circulation 2004;110:1996–2002. Early mortality among Medicare beneficiaries undergoing bariatric 36. Wagner JA, Varga K, Ellis EF, Rzigalinski BA, Martin BR, Kunos G. surgical procedures. JAMA 2005;294:1903–8. Activation of peripheral CB1 cannabinoid receptors in haemorrhagic 13. Buchwald H, Avidor Y, Braunwald E, et al. Bariatric surgery: a shock. Nature 1997;390:518–21. systematic review and meta-analysis. JAMA 2004;292:1724–37. 37. Wagner JA, Hu K, Bauersachs J, et al. Endogenous cannabinoids 14. Karnath B. Smoking cessation. Am J Med 2002;112:399–405. mediate hypotension after experimental myocardial infarction. J Am 15. Stampfer MJ, Hu FB, Manson JE, Rimm EB, Willett WC. Primary Coll Cardiol 2001;38:2048–54. prevention of coronary heart disease in women through diet and 38. Varga K, Wagner JA, Bridgen DT, Kunos G. Platelet- and lifestyle. N Engl J Med 2000;343:16–22. macrophage-derived endogenous cannabinoids are involved in 16. Braunwald E, Antman EM, Beasley JW, et al. ACC/AHA 2002 endotoxin-induced hypotension. FASEB J 1998;12:1035–44. guideline update for the management of patients with unstable angina 39. Godlewski G, Malinowska B, Schlicker E. Presynaptic cannabinoid and non–ST-segment elevation myocardial infarction—summary arti- CB(1) receptors are involved in the inhibition of the neurogenic cle: a report of the American College of Cardiology/American Heart vasopressor response during septic shock in pithed rats. Br J Pharmacol Association Task Force on Practice Guidelines (Committee on the 2004;142:701–8. Management of Patients With Unstable Angina). J Am Coll Cardiol 40. Cota D, Marsicano G, Tschop M, et al. The endogenous cannabinoid 2002;40:1366–74. system affects energy balance via central orexigenic drive and peripheral 17. Matsuda LA, Lolait SJ, Brownstein MJ, Young AC, Bonner TI. lipogenesis. J Clin Invest 2003;112:423–31. Structure of a cannabinoid receptor and functional expression of the 41. Cummings DE, Weigle DS, Frayo RS, et al. Plasma ghrelin levels cloned cDNA. Nature 1990;346:561–4. after diet-induced weight loss or gastric bypass surgery. N Engl J Med 18. Munro S, Thomas KL, Abu-Shaar M. Molecular characterization of a 2002;346:1623–30. peripheral receptor for cannabinoids. Nature 1993;365:61–5. 42. Considine RV, Sinha MK, Heiman ML, et al. Serum immunoreactive- 19. Herkenham M, Lynn AB, Little MD, et al. Cannabinoid receptor leptin concentrations in normal-weight and obese humans. N Engl localization in brain. Proc Natl Acad SciUSA1990;87:1932–6. J Med 1996;334:292–5. 20. Bensaid M, Gary-Bobo M, Esclangon A, et al. The cannabinoid CB1 43. Fruebis J, Tsao TS, Javorschi S, et al. Proteolytic cleavage product of receptor antagonist SR141716 increases Acrp30 mRNA expression in 30-kDa adipocyte complement-related protein increases fatty acid adipose tissue of obese fa/fa rats and in cultured adipocyte cells. Mol oxidation in muscle and causes weight loss in mice. Proc Natl Acad Sci Pharmacol 2003;63:908–14. U S A 2001;98:2005–10. 21. Bonz A, Laser M, Kullmer S, et al. Cannabinoids acting on CB1 44. Di Marzo V, Goparaju SK, Wang L, et al. Leptin-regulated endo- receptors decrease contractile performance in human atrial muscle. cannabinoids are involved in maintaining food intake. Nature 2001; J Cardiovasc Pharmacol 2003;41:657–64. 410:822–5. 22. Liu J, Gao B, Mirshahi F, et al. Functional CB1 cannabinoid receptors 45. Williams CM, Kirkham TC. Anandamide induces overeating: medi- in human vascular endothelial cells. Biochem J 2000;346:835–40. ation by central cannabinoid (CB1) receptors. Psychopharmacology 23. Ishac EJ, Jiang L, Lake KD, Varga K, Abood ME, Kunos G. (Berl) 1999;143:315–7. Inhibition of exocytotic noradrenaline release by presynaptic cannabi- 46. Jamshidi N, Taylor DA. Anandamide administration into the ventro- noid CB1 receptors on peripheral sympathetic nerves. Br J Pharmacol medial hypothalamus stimulates appetite in rats. Br J Pharmacol 1996;118:2023–8. 2001;134:1151–4. 24. Hanus L, Breuer A, Tchilibon S, et al. HU-308: a specific agonist for 47. Kirkham TC, Williams CM, Fezza F, Di Marzo V. Endocannabinoid CB(2), a peripheral cannabinoid receptor. Proc Natl Acad SciUSA levels in rat limbic forebrain and hypothalamus in relation to fasting, 1999;96:14228–33. feeding and satiation: stimulation of eating by 2-arachidonoyl glycerol. 25. Fride E, Foox A, Rosenberg E, et al. Milk intake and survival in Br J Pharmacol 2002;136:550–7. newborn cannabinoid CB1 receptor knockout mice: evidence for a 48. Ravinet Trillou C, Delgorge C, Menet C, Arnone M, Soubrie P. CB1 “CB3” receptor. Eur J Pharmacol 2003;461:27–34. cannabinoid receptor knockout in mice leads to leanness, resistance to 26. Giuffrida A, Beltramo M, Piomelli D. Mechanisms of endocannabi- diet-induced obesity and enhanced leptin sensitivity. Int J Obes Relat noid inactivation: biochemistry and pharmacology. J Pharmacol Exp Metab Disord 2004;28:640–8. Ther 2001;298:7–14. 49. Osei-Hyiaman D, DePetrillo M, Pacher P, et al. Endocannabinoid 27. Rinaldi-Carmona M, Barth F, Heaulme M, et al. SR141716A, a activation at hepatic CB1 receptors stimulates fatty acid synthesis and potent and selective antagonist of the brain cannabinoid receptor. contributes to diet-induced obesity. J Clin Invest 2005;115:1298–305. FEBS Lett 1994;350:240–4. 50. De Vries TJ, de Vries W, Janssen MC, Schoffelmeer AN. Suppression 28. Varga K, Lake K, Martin BR, Kunos G. Novel antagonist implicates of conditioned nicotine and sucrose seeking by the cannabinoid-1 the CB1 cannabinoid receptor in the hypotensive action of anandam- receptor antagonist SR141716A. Behav Brain Res 2005;161:164–8. ide. Eur J Pharmacol 1995;278:279–83. 51. De Vries TJ, Shaham Y, Homberg JR, et al. A cannabinoid mecha- 29. Niederhoffer N, Schmid K, Szabo B. The peripheral sympathetic nism in relapse to cocaine seeking. Nat Med 2001;7:1151–4. nervous system is the major target of cannabinoids in eliciting 52. Fattore L, Spano MS, Cossu G, Deiana S, Fratta W. Cannabinoid cardiovascular depression. Naunyn Schmiedebergs Arch Pharmacol mechanism in reinstatement of heroin-seeking after a long period of 2003;367:434–43. abstinence in rats. Eur J Neurosci 2003;17:1723–6. 1926 Gelfand and Cannon JACC Vol. 47, No. 10, 2006 Rimonabant May 16, 2006:1919–26

53. Anggadiredja K, Nakamichi M, Hiranita T, et al. Endocannabinoid 63. VanGaal LF, Rissanen AM, Scheen AJ, Ziegler O, Rossner S. system modulates relapse to methamphetamine seeking: possible Effects of the cannabinoid-1 receptor blocker rimonabant on mediation by the arachidonic acid cascade. Neuropsychopharmacology weight reduction and cardiovascular risk factors in overweight 2004;29:1470–8. patients: 1-year experience from the RIO-Europe study. Lancet 54. Gallate JE, Saharov T, Mallet PE, McGregor IS. Increased motivation 2005;365:1389–97. for beer in rats following administration of a cannabinoid CB1 64. VanGaal LF. 2-year data from the RIO-Europe study: metabolic effects receptor agonist. Eur J Pharmacol 1999;370:233–40. of rimonabant in overweight/obese patients. Presented at: American 55. Castane A, Valjent E, Ledent C, Parmentier M, Maldonado R, Valverde College of Cardiology Scientific Sessions; Orlando, FL: 2005. O. Lack of CB1 cannabinoid receptors modifies nicotine behavioural 65. Pi-Sunyer FX, Aronne LJ, Heshmati HM, et al. Effect of rimonabant, responses, but not nicotine abstinence. Neuropharmacology 2002;43:857– a cannabinoid-1 receptor blocker, on weight and cardiometabolic risk 67. factors in overweight or obese patients: RIO-North America: a 56. LeFoll B, Goldberg SR. Rimonabant, a CB1 antagonist, blocks nicotine- randomized controlled trial. JAMA 2006;295:761–75. conditioned place preferences. Neuroreport 2004;15:2139–43. 66. Scheen AJ. Effects of rimonabant in patients with type 2 diabetes 57. Balerio GN, Aso E, Berrendero F, Murtra P, Maldonado R. Delta9- mellitus. Results of the RIO-DIABETES trial. Presented at: tetrahydrocannabinol decreases somatic and motivational manifestations American Diabetes Association Scientific Sessions; San Diego, CA: of nicotine withdrawal in mice. Eur J Neurosci 2004;20:2737–48. 2005. 58. White R, Hiley CR. The actions of the cannabinoid receptor antag- 67. STRADIVARIUS (Strategy to Reduce Atherosclerosis Development onist, SR 141716A, in the rat isolated mesenteric artery. Br J Pharmacol 1998;125:689–96. Involving Administration of Rimonabant—the Intravascular Ultra- sound Study). Available at: http://www.clinicaltrials.gov/ct/gui/show/ 59. Chaytor AT, Martin PE, Evans WH, Randall MD, Griffith TM. The ϭ endothelial component of cannabinoid-induced relaxation in rabbit NCT00124332?order 2. Accessed September 26, 2005. mesenteric artery depends on gap junctional communication. J Physiol 68. Anthenelli RM. Effects of rimonabant in the reduction of major 1999;520:539–50. cardiovascular risk factors. Results from the STRATUS-US trial 60. Jbilo O, Ravinet-Trillou C, Arnone M, et al. The CB1 receptor antagonist (Smoking Cessation in Smokers Motivated to Quit). Presented at: rimonabant reverses the diet-induced obesity phenotype through the regula- American College of Cardiology Scientific Sessions; New Orleans, tion of lipolysis and energy balance. FASEB J 2005;19:1567–9. LA: 2004. 61. Ravinet-Trillou C, Arnone M, Delgorge C, et al. Anti-obesity effect of 69. Rimonabant to Reduce Alcohol Consumption—A Phase II Clini- SR141716, a CB1 receptor antagonist, in diet-induced obese mice. cal Trial. Available at: http://www.clinicaltrials.gov/ct/gui/show/ Am J Physiol Regul Integr Comp Physiol 2003;284:R345–53. NCT00075205?orderϭ1. Accessed September 26, 2005. 62. Despres JP, Golay A, Sjostrom L. Effects of rimonabant on metabolic 70. Batkai S, Jarai Z, Wagner JA, et al. Endocannabinoids acting at risk factors in overweight patients with dyslipidemia. N Engl J Med vascular CB1 receptors mediate the vasodilated state in advanced liver 2005;353:2121–34. cirrhosis. Nat Med 2001;7:827–32.