Context Rimonabant, a selective cannabinoid-1 receptor blocker, may reduce body weight and improve cardiometabolic risk factors in patients who are overweight or obese.
Objective To compare the efficacy and safety of rimonabant with placebo each in conjunction with diet and exercise for sustained changes in weight and cardiometabolic risk factors over 2 years.
Design, Setting, and Participants Randomized, double-blind, placebo-controlled trial of 3045 obese (body mass index ≥30) or overweight (body mass index >27 and treated or untreated hypertension or dyslipidemia) adult patients at 64 US and 8 Canadian clinical research centers from August 2001 to April 2004.
Intervention After a 4-week single-blind placebo plus diet (600 kcal/d deficit) run-in period, patients were randomized to receive placebo, 5 mg/d of rimonabant, or 20 mg/d of rimonabant for 1 year. Rimonabant-treated patients were rerandomized to receive placebo or continued to receive the same rimonabant dose while the placebo group continued to receive placebo during year 2.
Main Outcome Measures Body weight change over year 1 and prevention of weight regain during year 2. Additional efficacy measures included changes in waist circumference, plasma lipid levels, and other cardiometabolic risk factors.
Results At year 1, the completion rate was 309 (51%) patients in the placebo group, 620 (51%) patients in the 5 mg of rimonabant group, and 673 (55%) patients in the 20 mg of rimonabant group. Compared with the placebo group, the 20 mg of rimonabant group produced greater mean (SEM) reductions in weight (−6.3 [0.2] kg vs −1.6 [0.2] kg; P<.001), waist circumference (−6.1 [0.2] cm vs −2.5 [0.3] cm; P<.001), and level of triglycerides (percentage change, −5.3 [1.2] vs 7.9 [2.0]; P<.001) and a greater increase in level of high-density lipoprotein cholesterol (percentage change, 12.6 [0.5] vs 5.4 [0.7]; P<.001). Patients who were switched from the 20 mg of rimonabant group to the placebo group during year 2 experienced weight regain while those who continued to receive 20 mg of rimonabant maintained their weight loss and favorable changes in cardiometabolic risk factors. Use of different imputation methods to account for the high rate of dropouts in all 3 groups yielded similar results. Rimonabant was generally well tolerated; the most common drug-related adverse event was nausea (11.2% for the 20 mg of rimonabant group vs 5.8% for the placebo group).
Conclusions In this multicenter trial, treatment with 20 mg/d of rimonabant plus diet for 2 years promoted modest but sustained reductions in weight and waist circumference and favorable changes in cardiometabolic risk factors. However, the trial was limited by a high drop-out rate and longer-term effects of the drug require further study.
Clinical Trials Registration ClinicalTrials.gov Identifier: NCT00029861
Roughly two thirds of US adults meet the criteria for overweight or obesity,1 which greatly increases the risk of developing diabetes mellitus and cardiovascular disease2 and related mortality.3 In addition to weight loss, obesity management should target reduction in the cardiometabolic risk factors of atherogenic dyslipidemia, excess abdominal obesity, and elevated glucose. Modest (approximately 5% to 10% of body weight) intentional nonpharmacological weight loss improves obesity-related cardiovascular and metabolic abnormalities4 but diet and exercise interventions have limited long-term success. As a result, long-term weight management remains a challenge for patients and clinicians.
The endocannabinoid system regulates energy homeostasis through G protein–coupled cannabinoid-1 receptors5,6 located in the central nervous system and in various peripheral tissues, including adipose tissue, muscle, the gastrointestinal tract, and the liver.7 While peripheral cannabinoid-1 receptor activation decreases adiponectin production in adipocytes,8 central cannabinoid-1 receptor activation in preclinical studies stimulates eating, decreases muscle, and stimulates hepatic and adipose tissue lipogenic pathways in animal models of obesity.9 In genetic and diet-induced obesity, rimonabant, a selective cannabinoid-1 receptor blocker, reduces overactivation of the central8,10 and peripheral11,12 endocannabinoid system8,10,13 and prevents weight gain and associated metabolic disorders, thus revealing a novel strategy for the treatment of obesity and related cardiometabolic disorders.
The RIO-North America trial evaluated the efficacy and safety of rimonabant in conjunction with a hypocaloric diet in promoting reductions in body weight and waist circumference, long-term weight maintenance, and amelioration of cardiometabolic risk factors in obese and higher-risk overweight patients.
Men and women aged 18 years or older were recruited at 64 US and 8 Canadian clinical research centers between September 2001 and April 2002 (Figure 1 and Table 1). Entry criteria included body mass index (calculated as weight in kilograms divided by the square of height in meters) of 30 or greater (obese) or body mass index of higher than 27 (overweight and treated or untreated dyslipidemia or hypertension). Patients were excluded if they had a body weight fluctuation of more than 5 kg in the previous 3 months; clinically significant cardiac, renal, hepatic, gastrointestinal tract, neuropsychiatric, or endocrine disorders; drug-treated or diagnosed type 1 or type 2 diabetes; use of medications that alter body weight or appetite; a history or current substance abuse; or changes in smoking habits or smoking cessation within the past 6 months. Women with childbearing potential were required to use medically approved contraception. Determination of race, a US Food and Drug Administration requirement, was by patient self-identification.
RIO-North America was a 2-year, randomized, double-blind, placebo-controlled trial. The institutional review boards at each center reviewed and approved the study protocol and patients provided written informed consent before entry into the trial. Following a 1-week screening period, patients were instructed to follow a hypocaloric diet (approximately 600 kcal/d deficit) that was continued during a 4-week placebo, single-blind, run-in period and then throughout the double-blind treatment period. The diet prescription was adjusted to each patient's basal metabolic rate estimated by the Harris-Benedict equation14 and self-reported physical activity at screening and at weeks 24, 52, and 76. Patients also were instructed to increase their level of physical activity throughout the study.
Patients who completed the run-in period were randomly allocated to 1 of 3 double-blind treatment groups for 1 year: placebo, 5 mg/d of rimonabant, or 20 mg/d of rimonabant. A predefined randomization schedule assigned patients using a block size of 5 and a randomization ratio of 1:2:2 to ensure sufficient numbers of rimonabant-treated patients for a rerandomization (1:1) for year 2. Rimonabant-treated patients were rerandomized to receive placebo or continued to receive the same rimonabant dose while the placebo group continued to receive placebo for year 2. Randomization was balanced within each center and stratified by weight loss (≤2 kg or >2 kg) during the run-in period. Medication compliance, defined as consumption of 80% or greater of tablets, was assessed by tablet counting at each specified visit.
Initial screening included a medical history, physical examination, electrocardiography, clinical chemistry, thyroid function, hematology, and urinalysis. Body weight was measured using a calibrated digital or balance scale at screening, biweekly during the run-in period, baseline (randomization), weeks 2 and 4, and then every 4 weeks. Waist circumference was measured using a spring-loaded measuring tape midway between the lower rib and iliac crest and followed the same measurement schedule as body weight.
Fasting serum glucose and insulin levels were measured at screening, baseline, every 12 weeks until week 36, at week 52, every 12 weeks between week 52 and week 88, and at week 104. Serum glucose, insulin, and lipids were assayed according to standard procedures.15,16 Low-density lipoprotein cholesterol was measured directly by ultracentrifugation. Metabolic syndrome status was assessed according to the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) criteria17 at baseline, year 1, and year 2.
Each safety evaluation included a physical examination with collection of vital signs and recording of adverse events. Hematology and serum chemistry were evaluated every 3 months. The hospital anxiety and depression scale,18 a validated tool for the evaluation of mood and psychological traits that includes depression and anxiety subscales, was assessed at screening, baseline, and at weeks 24, 52, 76, and 104. Electrocardiography screening was performed every 3 months. Adverse events were assessed by spontaneous report at each visit.
Sample Size. The sample size was calculated based on the assumption that the SD of weight change at year 1 would be 10 kg. Thus 2800 randomized patients (560 patients in the placebo group and 1120 patients in each rimonabant dose group [to ensure sufficient patients for rerandomization at the end of year 1]) provided 99% power to detect a 3-kg difference between 1 dose of rimonabant and placebo after 1 year. We chose an α level of .025 to ensure an overall type I error rate of .05 according to a modified Bonferroni procedure.
Analysis Populations. Efficacy analyses were performed at the end of years 1 and 2. The 1-year modified intent-to-treat (ITT) population was defined as all randomized patients who received at least 1 dose of the double-blind study drug during the first year and had at least 1 postbaseline trial assessment (Table 2 and Figure 2). The 2-year modified ITT population for the analysis of the prevention of weight regain was composed of all randomized patients who completed year 1, received at least 1 dose of study drug in year 2, and had at least 1 weight assessment after rerandomization (Table 3 and Figure 2). The modified ITT population for the analysis of efficacy over 2 years included patients who received the same double-blind study drug for the entire study (including those who discontinued study participation during the first year).
Primary and Secondary Efficacy Analyses
The primary efficacy variable was weight loss over year 1. The other primary efficacy variable was prevention of weight regain between the first and second years expressed as the change in weight from the end of the first year (rerandomization baseline) to the end of year 2.
Other weight-related criteria were the percentage of patients achieving weight loss of 5% or greater and weight loss of 10% or greater from baseline to years 1 and 2 and changes in waist circumference. Secondary efficacy end points were changes in level of high-density lipoprotein (HDL) cholesterol from baseline to year 1 and the prevalence of the metabolic syndrome. Additional secondary efficacy variables included changes from baseline in systolic and diastolic blood pressure, levels of fasting glucose and insulin, lipids, and insulin resistance measured by homeostasis model assessment19 (HOMA-IR), which is calculated by multiplying fasting insulin by fasting glucose and dividing by 22.5.
Primary efficacy analyses were applied to the ITT population with the last observation carried forward (LOCF; Figure 3). Comparisons of the primary efficacy end point (weight change from baseline) were conducted using analysis of variance with the modified Bonferroni procedure20 to adjust for multiple comparisons. The analysis of variance model included treatment and randomization stratum (weight loss of ≤2 kg or >2 kg during the run-in period) as fixed effects. Similar analyses were applied to the secondary efficacy variables. Because this analysis excluded repeated measurements made over the course of the study, a post-hoc repeated-measures approach was applied to changes in weight from baseline using a model that included fixed effects (randomization stratum, treatment, days after randomization, and treatment × days interaction) and a random effect for patients. Similar methods were applied to other efficacy end points.
The prevention of weight regain during year 2 was analyzed using a 2-way analysis of covariance model including rerandomization treatment sequence and randomization stratum as fixed effects and weight loss during year 1 as the covariate. Each rerandomized dose group (ie, 20 mg of rimonabant during year 1 and then 20 mg of rimonabant during year 2) was compared with the same dose group that was switched to placebo (ie, 20 mg of rimonabant during year 1 and then placebo during year 2).
As an assessment of sensitivity, a more conservative imputation method than LOCF for handling missing data was applied. For study dropouts with efficacy data and improvement in an end point, the imputed last value was defined as a weighted average of the baseline and LOCF values in which the weights were defined as the proportion of treatment duration during the trial. If a study dropout had efficacy data and showed no improvement in an end point, the last value was not imputed because the imputation method would have provided a better result than the LOCF. If a study dropout had no efficacy data, the imputed value was set to the baseline value and the change from baseline was set to zero.
The percentage of patients losing at least 5% or 10% of their baseline body weight in the groups receiving either 20 mg or 5 mg of rimonabant and the percentage of patients meeting the Adult Treatment Panel III criteria17 for the metabolic syndrome were compared with the percentage of patients receiving placebo using logistic regression.
The estimates of responses in secondary end points to treatment that could not be attributed to weight loss alone were based on standard regression methods in which weight loss (change in weight from baseline to 1 year) was introduced as a covariate (analysis of covariance). The weight-adjusted analysis of covariance model for treatment effect was: Y = a + βT + γW + e where Y is the efficacy variable, T is the treatment indicator, and W is weight loss. The weight-independent portion of the total treatment effect was calculated as the ratio of the weight-adjusted treatment effect β to the treatment effect β1 in the overall unadjusted analysis of variance model: Y = a + β1T + e1.21 This ratio reflects the proportion of the total effect size that cannot be explained by weight loss. All statistical tests were 2-sided at the .05 significance level except as noted; all P values presented herein are unadjusted. Unless otherwise noted, results are for the ITT population. Statistical analyses were performed using SAS software version 8.2 (SAS Institute Inc, Cary, NC).
A total of 3045 patients completed the 4-week, placebo run-in and were randomized to double-blind treatment with placebo (n = 607), 5 mg of rimonabant (n = 1216), or 20 mg of rimonabant (n = 1222). Five randomized patients (2 in the group who received 5 mg of rimonabant and 3 in the group who received 20 mg of rimonabant) did not receive double-blind study medication. The disposition of patients over 2 years appears in Figure 1. The characteristics of the study population at randomization were similar in the 3 treatment groups (Table 1).
Year 1 was completed by 51% of patients (n = 309) in the placebo group, 51% (n =620) in the 5 mg of rimonabant group, and 55% (n = 673) in the 20 mg of rimonabant group. More than 98% of patients receiving rimonabant took more than 80% of the prescribed study medication. There were no differences in compliance between completers and noncompleters. The completion rates for rerandomized patients in year 2 were 72% for patients who received placebo both years, 70% for patients who received 5 mg of rimonabant in year 1 and placebo in year 2, 69% for patients who received 20 mg of rimonabant in year 1 and placebo in year 2, 71% for patients who received 5 mg of rimonabant in both years, and 77% for patients who received 20 mg of rimonabant in both years.
Weight Loss During Year 1
During the 4-week placebo plus diet run-in period, body weight decreased by a mean (SEM) of 1.9 (0.04) kg and waist circumference by 2.1 (0.08) cm. Also during this period, level of HDL cholesterol decreased by a mean (SEM) of 5.8% (0.2%) and level of triglycerides decreased by 1.2% (0.7%). After randomization, weight loss from baseline in the 1-year modified ITT population (Figure 2 and Figure 3) was significantly greater in patients receiving 20 mg or 5 mg of rimonabant than in patients receiving placebo. Similar results were seen when the data were expressed as a placebo-subtracted change from baseline (Table 2). The percentage of patients achieving a 5% or greater weight loss at 1 year was 26.1% for patients receiving 5 mg of rimonabant (odds ratio [OR], 1.4; 95% confidence interval [CI], 1.1-1.8; P = .004), 48.6% for patients receiving 20 mg of rimonabant (OR, 4.1; 95% CI, 3.2-5.2; P<.001), and 20.0% for patients receiving placebo. The percentage of patients achieving a 10% or greater weight loss was 25.2% for patients receiving 20 mg of rimonabant and 8.5% for patients receiving placebo (OR, 4.0; 95% CI, 2.9-5.5; P<.001). However, only 10.6% of patients receiving 5 mg of rimonabant achieved a 10% or greater weight loss (OR, 1.3; 95% CI, 0.9-1.8). Compared with the patients receiving placebo, waist circumference decreased more in the patients receiving 20 mg of rimonabant (Figure 2 and Figure 3).
Weight Loss During Year 2
The 2-year modified ITT population that was previously treated with 20 mg of rimonabant continued treatment with 20 mg of rimonabant and maintained a mean (SEM) weight loss from baseline of 7.4 (0.4) kg whereas the participants who were rerandomized to placebo regained most of their previous weight loss (Figure 2 and Figure 3). A similar pattern was seen for waist circumference.
Weight Loss in Patients Receiving the Same Treatment for 2 Years
Compared with patients receiving placebo, cumulative weight loss was significantly greater in patients receiving 20 mg of rimonabant in both years (but not in those receiving 5 mg of rimonabant in both years) (Table 3). A greater percentage of patients receiving 20 mg of rimonabant achieved a weight loss of 5% or greater (40% vs 19% of patients receiving placebo; OR, 2.9 [95% CI, 2.3-3.7]; P<.001) and 10% or greater (17% vs 8% of patients receiving placebo; OR, 2.3 [95% CI, 2.1-3.3]; P<.001). The 2-year mean (SEM) change from baseline in waist circumference was also significantly greater in patients receiving 20 mg of rimonabant (−5.0 [0.2] cm) compared with patients receiving placebo (−2.2 [0.3] cm; P<.001).
Cardiometabolic Risk Factors During Year 1
Levels of HDL cholesterol increased and fasting insulin levels decreased in patients receiving either 5 mg or 20 mg of rimonabant. Levels of triglycerides decreased in patients receiving 20 mg of rimonabant but not in patients receiving 5 mg of rimonabant (Table 2 and Figure 4). The prevalence of the metabolic syndrome according to Adult Treatment Panel III criteria significantly declined in patients receiving 20 mg of rimonabant (from 34.8% to 21.2%) compared with patients receiving placebo (31.7% to 29.2%; P<.001). Levels of total cholesterol and low-density lipoprotein cholesterol were not significantly different among the 3 groups (data available on request). Insulin resistance estimated by the HOMA-IR increased in patients receiving placebo but not in patients receiving 20 mg of rimonabant. Systolic and diastolic blood pressures tended to decrease slightly but not significantly in patients receiving either 5 mg or 20 mg of rimonabant (Table 2).
In patients receiving 20 mg of rimonabant, the observed effects at 1 year in levels of HDL cholesterol, triglycerides, fasting insulin, and in HOMA-IR were approximately twice that attributable to the concurrent weight loss alone using analysis of covariance. For example, of the observed 7.2% increase in level of HDL cholesterol in patients receiving 20 mg of rimonabant, there was only a 4.2% increase in level of HDL cholesterol after adjustment for weight loss (P<.001). The residual effects after weight-loss adjustment in patients receiving 20 mg of rimonabant were 47% of observed effects for triglycerides (P = .008); 50% of observed effects for fasting insulin (P = .04); and 51% of observed effects for HOMA-IR (P = .07).
Cardiometabolic Risk Factors in Year 2
Compared with patients who continued to receive 5 mg or 20 mg of rimonabant, patients who were rerandomized to placebo in year 2 had increased levels of triglycerides and decreased levels of HDL cholesterol (data available on request). In the patients who completed the study and who were treated with either placebo or 20 mg of rimonabant for 2 years, levels of HDL cholesterol continued to increase from baseline during year 2 but significantly so in patients who were treated with 20 mg
of rimonabant (P<.001; Figure 5). Compared with patients receiving placebo, both levels of triglycerides and the prevalence of the metabolic syndrome declined more from baseline in patients receiving 20 mg of rimonabant (P<.001).
The percentage of patients reporting at least 1 adverse event was similar across treatment groups (85.5% for patients receiving 20 mg of rimonabant, 83.4% for patients receiving 5 mg of rimonabant, and 82.0% for patients receiving placebo; Table 4). Compared with patients receiving placebo, the overall incidence of adverse events leading to study withdrawal in year 1 was slightly higher in patients receiving 5 mg of rimonabant and even greater in patients receiving 20 mg of rimonabant, mainly due to psychiatric, nervous system, and gastrointestinal tract adverse events. Compared with patients receiving placebo, adverse events (upper respiratory tract infection, nasopharyngitis, nausea, influenza, diarrhea, arthralgia, anxiety, insomnia, viral gastroenteritis, dizziness, depressed mood, and fatigue) were reported in 5% or greater of patients receiving 20 mg of rimonabant. There were no differences among the treatment groups in changes over time in corrected QT interval and either the anxiety or depression subscales of the hospital anxiety and depression scale.
In year 2, the overall rates of adverse events, study withdrawals, and adverse event–related study withdrawals were lower than in year 1; there were no differences in overall rates among the treatment groups (Table 5). Upper respiratory tract infection, nasopharyngitis, or influenza occurred in 5% or greater of patients receiving either 5 mg or 20 mg of rimonabant at year 2 and overall were more frequent in rimonabant-treated patients for both years.
Rimonabant, the first selective cannabinoid-1 receptor blocker to enter clinical trials, was tested in a randomized, double-blind, placebo-controlled 2-year multicenter study. The results suggest that 20 mg/d of rimonabant is effective in reducing body weight and waist circumference, while also favorably affecting several cardiometabolic risk factors. Most of these effects were dose-dependent. Furthermore, the differences in the patients receiving 20 mg of rimonabant compared with patients receiving placebo in levels of HDL cholesterol, triglycerides, fasting insulin, and HOMA-IR appeared to exceed that expected from the weight loss achieved. These findings support and extend results from other randomized controlled trials of rimonabant therapy.22,23
RIO-North America addressed the efficacy and safety profile of rimonabant over 1 year; the long-term effectiveness of rimonabant in preventing weight regain; and the efficacy and tolerability of continuous long-term rimonabant treatment over 2 years. Clinically significant weight loss achieved during year 1 was well maintained during year 2 in patients receiving 20 mg of rimonabant during both years. When patients treated with 5 mg or 20 mg of rimonabant at year 1 were rerandomized to placebo in year 2, they regained a substantial amount of the weight they had lost. However, body weight still remained slightly lower in these patients than in the patients treated with placebo for 2 years. These findings highlight the concept that sustained weight loss and associated favorable changes in cardiometabolic risk factors require continuous long-term treatment as seen in other chronic disorders, such as diabetes and hypertension in which treatment is effective only for as long as patients are receiving therapy.
Compared with patients who received placebo, patients who received 20 mg of rimonabant had favorable changes in levels of HDL cholesterol, triglycerides, and fasting insulin and in HOMA-IR that appeared to be approximately twice that expected from the achieved weight loss alone, suggesting a direct pharmacological effect of rimonabant on glucose and lipid metabolism beyond the weight loss achieved. In patients who received 20 mg of rimonabant, levels of HDL cholesterol increased continuously throughout the 2-year study whereas body weight stabilized, further supporting a direct pharmacological effect not attributable to weight loss alone. Preclinical studies indicate that rimonabant increases adiponectin gene expression and production in adipose tissue,11 increases insulin-mediated glucose uptake in isolated soleus muscle,12 and that cannabinoid-1 receptor antagonism or deletion decreases de novo hepatic fatty acid synthesis and lipid accumulation in response to the consumption of high-fat foods.9 Patients with the metabolic syndrome who have insulin resistance have multiple defects in glucose and lipid metabolism associated with excess intraabdominal fat, hypoadiponectinemia, and high levels of cytokines and adhesion molecules.24 While further study is needed to elucidate the specific mechanisms underlying the apparent direct action of rimonabant on lipid and glucose metabolism, these effects may be mediated by adiponectin and reduction of abdominal obesity.23,25
Rimonabant significantly reduced waist circumference, a measure of abdominal adiposity, and the prevalence of the metabolic syndrome. A recent study26 showed that measured intraabdominal fat was independently associated with all 5 of the metabolic syndrome criteria, suggesting that it may have a central pathophysiological role. Moreover, multivariable analyses indicated that waist circumference and level of triglycerides together might be a useful surrogate marker for measured insulin resistance and intraabdominal adiposity in individuals without diabetes. Furthermore, fasting insulin and waist circumference predicted insulin sensitivity measured directly by the hyperinsulinemic euglycemic clamp and intraabdominal fat measured by computed tomography. These results suggest that fasting insulin level and waist circumference are reliable indicators of high-risk patients in clinical practice. In the context of the current obesity epidemic and the associated burden on health care resources, clinical tools such as waist circumference may enable physicians to identify those patients at high risk for type 2 diabetes and cardiovascular disease, who may benefit from early intervention to improve their cardiometabolic risk status.
Rimonabant was generally well tolerated with adverse effects that were mostly mild and moderate. In patients receiving the same treatment for 2 years, the study withdrawal rate due to adverse events became comparable among all patients during year 2, suggesting that the adverse effects occur early and that 5 mg/d and 20 mg/d of rimonabant have a comparable safety and tolerability profile with placebo.
There are several limitations to our study. The low retention rates of only about 50% in all treatment groups, while consistent with previous studies in overweight or obese patients,27 present a major challenge in data analysis and interpretation. The use of the LOCF approach to impute missing values assumes that individual data at the time of dropout are representative of data at the end of the study if the participant had completed the study.28 The results of the study also may be affected by participants who derived less benefit and dropped out more frequently. Moreover, data from patients who completed the study may not be representative of the overall study population when the drop-out rate is high. However, sensitivity analyses, including a repeated-measures approach and an imputation of final values adjusted for duration of participation, supported the conclusions of the LOCF analysis. Other factors that may diminish the generalizability of the study results include the limited racial diversity and the overall predominance of white women in the study. Lastly, larger studies are necessary to assess less frequent adverse events and longer duration studies will be needed to confirm the long-term safety of rimonabant beyond 2 years.
In conclusion, in the RIO-North America trial, 20 mg of rimonabant plus a standard dietary intervention produced sustained, clinically meaningful weight loss and favorable changes in cardiometabolic risk factors over 1 year and prevented weight regain in year 2 with favorable effects compared with placebo on fasting serum levels of HDL cholesterol and triglycerides and HOMA-IR. Compared with patients who had received 20 mg of rimonabant in year 1 and were then reassigned to receive placebo in year 2, those treated with 20 mg of rimonabant for 2 years maintained weight loss and differences from patients receiving placebo in multiple cardiometabolic risk factors, reflecting the potential effectiveness of long-term rimonabant therapy. It must be acknowledged that the trial was limited by a high drop-out rate and that long-term effects of the drug require further study. Still, our observations collectively suggest that rimonabant may well represent an innovative approach to the management of multiple cardiometabolic risk factors, facilitating and maintaining improvements through weight loss–dependent and –independent pathways.
Corresponding Author: F. Xavier Pi-Sunyer, MD, Obesity Research Center, St Luke’s-Roosevelt Hospital, 1111 Amsterdam Ave, WH1020, New York, NY 10025 (fxp1@columbia.edu).
Author Contributions: Dr Pi-Sunyer had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Pi-Sunyer, Aronne, Heshmati.
Acquisition of data: Pi-Sunyer, Aronne, Heshmati, Rosenstock.
Analysis and interpretation of data: Pi-Sunyer, Aronne, Heshmati, Devin, Rosenstock.
Drafting of the manuscript: Pi-Sunyer, Aronne.
Critical revision of the manuscript for important intellectual content: Aronne, Devin, Rosenstock.
Statistical analysis: Devin.
Obtained funding: Pi-Sunyer.
Administrative, technical, or material support: Pi-Sunyer, Heshmati, Rosenstock.
Study supervision: Aronne, Heshmati, Rosenstock.
Financial Disclosures: Drs Pi-Sunyer and Aronne have received honoraria from Sanofi-Aventis for speaker's presentations. No other authors reported financial disclosures.
Funding/Support: RIO-North America was supported by a research grant from Sanofi Synthelabo Research, a division of Sanofi Synthelabo Inc, a member of the Sanofi-Aventis group.
Role of the Sponsor: Sanofi-Aventis participated in discussions regarding study design and protocol development and the sponsor provided logistical support for the trial, data collection, and data analysis, and helped in preparing the manuscript. The sponsor was permitted to review the manuscript, but the final decision on content was with the corresponding author in conjunction with the other authors.
Independent Statistical Review: All study data were transferred from Sanofi-Aventis to the Department of Medicine at St Luke's-Roosevelt Hospital Center for independent reanalysis by Stanley Heshka, PhD. Statistical reanalyses of the raw data were performed by Dr Heshka. There were no discrepancies between the reanalysis and the original interpretation of the results and conclusions. In lieu of financial compensation for Dr Heshka's time and effort in performing the statistical analyses, an unrestricted educational grant from Sanofi-Aventis was given to the Obesity Research Center at St Luke’s-Roosevelt Hospital Center in New York, NY.
Data and Safety Monitoring Board: Alain Leizorovicz, MD (chairman), Université Claude Bernard, Lyon, France; Michael Weintraub, MD, University of Rochester School of Medicine and Dentistry, Rochester, NY; Jean-Louis Imbs, MD, Hôpital Civil, Strasbourg, France; Elliot Danforth, MD, University of Vermont, Burlington; David P. L. Sachs, MD, Palo Alto Center for Pulmonary Disease, Palo Alto, Calif.
RIO North America Investigators:Canada (Ontario): Denis Callaghan, Hamilton; Jeff Daiter, Richmond Hill; John Nuttall, Kingston; William O’Mahony, Corunna; Duncan Sinclair, Aylmer; Donald Spink, Peterborough; Paul Willoughby, Woodstock; Paul Ziter, Windsor. United States: Andrew Ahmann, Portland, Ore; James Anderson, Lexington, Ky; Louis Aronne, New York, NY; Richard Atkinson, Madison, Wis; Michael Basista, Bloomfield, NJ; Kathleen Baskett, Missoula, Mont; Harold Bays, Louisville, Ky; Gregory Bishop, San Diego, Calif; Scott Bleser, Bellbrook, Ohio; Marshall Block, Phoenix, Ariz; Stephen Brady, Naples, Fla; Ronald Brazg, Renton, Wash; Robert Call, Richmond, Va; Antonio Caos, Ocoee, Fla; Harry Collins, South Plainfield, NJ; Gordon Connor, Birmingham, Ala; Martin Conway, Albuquerque, NM; Lydia Corn, Sarasota, Fla; Walter Dunbar, Atlanta, Ga; Ronald Emkey, Wyomissing, Pa; James Ferguson, Salt Lake City, Utah; Harold Fleming, Spartanburg, SC; Nicholas Fleming, Spartanburg, SC; Arthur Frank, Washington, DC; Ken Fujioka, San Diego, Calif; Sidney Funk, Atlanta, Ga; Elizabeth Gallup, Overland Park, Kan; W. Thomas Garland, Lawrenceville, NJ; Jeffrey Geohas, Chicago, Ill; Eric Goldberg, West Palm Beach, Fla; Frank Greenway, Baton Rouge, La; Paula Hall, Indianapolis, Ind; Wayne Harper, Raleigh, NC; Scott Horn, San Antonio, Tex; Roy Kaplan, Concord, Calif; Richard Krause, Chattanooga, Tenn; Diane Krieger, Miami, Fla; Roberta Loeffler, Wichita, Kan; Barry Lubin, Norfolk, Va; Thomas Marbury, Orlando, Fla; Clark McKeever, Houston, Tex; James McKenney, Richmond, Va; Curtis Mello, Swansea, Mass; Neerja Misra, Lawrenceville, NJ; Patrick O’Neil, Charleston, SC; F. Xavier Pi-Sunyer, New York, NY; David Podlecki, Longmont, Colo; Gary Post, Highlands Ranch, Colo; R. Walter Powell, Newark, Del; Stephen Rafelson, Langhorne, Pa; Kenneth Rictor, Scotland, Pa; Julio Rosenstock, Dallas, Tex; Daniel Rowe, West Palm Beach, Fla; John Rubino, Raleigh, NC; Donald Schumacher, Charlotte, NC; Douglas Schumacher, Columbus, Ohio; Howard Schwartz, Miami, Fla; Simona Scumpia, Austin, Tex; Stephan Sharp, Nashville, Tenn; Earl Shrago, Madison, Wis; Diane Smith, Augusta, Ga; Norman Soler, Springfield, Ill; Paul Tung, Dover, NH; Greggory Volk, Beavercreek, Ohio; Ralph Wade, Salt Lake City, Utah; Richard Weinstein, Walnut Creek, Calif; Todd Wine, Oceanside, Calif; Lisa Wright, Birmingham, Ala; John Zerbe, Cincinnati, Ohio; Douglas Zmolek, Manlius, NY.
Previous Presentation: Presented in part at the American Heart Association's Scientific Sessions, New Orleans, La, November 7-10, 2004.
Acknowledgment: We appreciate the expert assistance of Stanley Heshka, PhD (St Luke’s-Roosevelt Hospital Center, Columbia University, New York, NY) in conducting an independent statistical analysis of the study data and in providing guidance on appropriate statistical methods.
1.Hedley AA, Ogden CL, Johnson CL, Carroll MD, Curtin LR, Flegal KM. Prevalence of overweight and obesity among US children, adolescents, and adults, 1999-2002.
JAMA. 2004;291:2847-285015199035
Google ScholarCrossref 2.Klein S, Burke LE, Bray GA.
et al. Clinical implications of obesity with specific focus on cardiovascular disease: a statement for professionals from the American Heart Association Council on Nutrition, Physical Activity, and Metabolism: endorsed by the American College of Cardiology Foundation.
Circulation. 2004;110:2952-296715509809
Google ScholarCrossref 3.Solomon CG, Manson JE. Obesity and mortality: a review of the epidemiologic data.
Am J Clin Nutr. 1997;66:1044S-1050S9322585
Google Scholar 4.National Institutes of Health Expert Panel on the Identification Evaluation and Treatment of Overweight in Adults. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: the evidence report.
Obes Res. 1998;6:(suppl 2)
51S-209S9813653
Google ScholarCrossref 5.Matsuda LA, Lolait SJ, Brownstein MJ, Young AC, Bonner TI. Structure of a cannabinoid receptor and functional expression of the cloned cDNA.
Nature. 1990;346:561-5642165569
Google ScholarCrossref 6.Munro S, Thomas KL, Abu-Shaar M. Molecular characterization of a peripheral receptor for cannabinoids.
Nature. 1993;365:61-657689702
Google ScholarCrossref 7.Howlett AC. The cannabinoid receptors.
Prostaglandins Other Lipid Mediat. 2002;68:619-63112432948
Google ScholarCrossref 8.Cota D, Marsicano G, Tschop M.
et al. The endogenous cannabinoid system affects energy balance via central orexigenic drive and peripheral lipogenesis.
J Clin Invest. 2003;112:423-43112897210
Google Scholar 9.Osei-Hyiaman D, DePetrillo M, Pacher P.
et al. Endocannabinoid activation at hepatic CB1 receptors stimulates fatty acid synthesis and contributes to diet-induced obesity.
J Clin Invest. 2005;115:1298-130515864349
Google Scholar 10.Di Marzo V, Goparaju SK, Wang L.
et al. Leptin-regulated endocannabinoids are involved in maintaining food intake.
Nature. 2001;410:822-82511298451
Google ScholarCrossref 11.Bensaid M, Gary-Bobo M, Esclangon A.
et al. The cannabinoid CB1 receptor antagonist SR141716 increases Acrp30 mRNA expression in adipose tissue of obese fa/fa rats and in cultured adipocyte cells.
Mol Pharmacol. 2003;63:908-91412644592
Google ScholarCrossref 12.Liu YL, Connoley IP, Wilson CA, Stock MJ. Effects of the cannabinoid CB1 receptor antagonist SR141716 on oxygen consumption and soleus muscle glucose uptake in Lep(ob)/Lep(ob) mice.
Int J Obes (Lond). 2005;29:183-187
Google ScholarCrossref 13.Ravinet Trillou C, Arnone M, Delgorge C.
et al. Anti-obesity effect of SR141716, a CB1 receptor antagonist, in diet-induced obese mice.
Am J Physiol Regul Integr Comp Physiol. 2003;284:R345-R35312399252
Google Scholar 14.Harris JA, Benedict FG. A Biometric Study of Basal Metabolism in Man. Washington, DC: Carnegie Institute of Washington; 1919. Publication 279
15.Jacobs D, , DeMott W, , Grady H, , Horvat R, , Huestis D, , Kasten B, . Laboratory Test Handbook. 4th ed. Cleveland, Ohio: Lexi-Comp Inc; 1996
16.Rifai N, , Warnick G, . Laboratory Measurement of Lipids, Lipoproteins, and Apolipoproteins. Washington, DC: AACC Press; 1994
17.National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Executive summary of the third report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III).
JAMA. 2001;285:2486-2497
Google ScholarCrossref 18.Zigmond AS, Snaith RP. The hospital anxiety and depression scale.
Acta Psychiatr Scand. 1983;67:361-3706880820
Google ScholarCrossref 19.Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man.
Diabetologia. 1985;28:412-4193899825
Google ScholarCrossref 20.Hochberg Y. A sharper Bonferonni procedure for multiple tests of significance.
Biometrika. 1988;75:800-802
Google ScholarCrossref 21.Buyse M, Molenberghs G. Criteria for the validation of surrogate endpoints in randomized experiments.
Biometrics. 1998;54:1014-10299840970
Google ScholarCrossref 22.Van Gaal LF, Rissanen AM, Scheen AJ, Ziegler O, Rossner S. Effects of the cannabinoid-1 receptor blocker rimonabant on weight reduction and cardiovascular risk factors in overweight patients: 1-year experience from the RIO-Europe study.
Lancet. 2005;365:1389-139715836887
Google ScholarCrossref 23.Despres JP, Golay A, Sjostrom L.Rimonabant in Obesity-Lipids Study Group. Effects of rimonabant on metabolic risk factors in overweight patients with dyslipidemia.
N Engl J Med. 2005;353:2121-213416291982
Google ScholarCrossref 24.Salmenniemi U, Ruotsalainen E, Pihlajamaki J.
et al. Multiple abnormalities in glucose and energy metabolism and coordinated changes in levels of adiponectin, cytokines, and adhesion molecules in subjects with metabolic syndrome.
Circulation. 2004;110:3842-384815596567
Google ScholarCrossref 25.Berg AH, Combs TP, Scherer PE. ACRP30/adiponectin: an adipokine regulating glucose and lipid metabolism.
Trends Endocrinol Metab. 2002;13:84-8911854024
Google ScholarCrossref 26.Carr DB, Utzschneider KM, Hull RL.
et al. Intra-abdominal fat is a major determinant of the National Cholesterol Education Program Adult Treatment Panel III criteria for the metabolic syndrome.
Diabetes. 2004;53:2087-209415277390
Google ScholarCrossref 27.Padwal R, Li S, Lau D. Long-term pharmacotherapy for overweight and obesity: a systematic review and meta-analysis of randomized controlled trials.
Int J Obes Relat Metab Disord. 2003;27:1437-144612975638
Google ScholarCrossref 28.Gadbury GL, Coffey CS, Allison DB. Modern statistical methods for handling missing repeated measurements in obesity trial data: beyond LOCF.
Obes Rev. 2003;4:175-18412916818
Google ScholarCrossref