Rimonabant Improves Glycemic Control in Insulin-Treated Type 2 Diabetes - ARPEGGIO


The goal of this trial was to evaluate the safety and efficacy of rimonabant on glycemic control in patients with type 2 diabetes and cardiovascular disease, and who are not adequately controlled with insulin.


Rimonabant is superior to placebo in improving diabetes control in patients with type 2 diabetes and cardiovascular disease, who need better glycemic control despite being on insulin.

Study Design

Patients Screened: 570
Patients Enrolled: 360
Mean Follow Up: 48 weeks
Mean Patient Age: 57.8 years
Female: 57

Patient Populations:

• Age >18 years
• Diagnosis of type 2 diabetes as defined by World Health Organization criteria
• Type 2 diabetes treated with insulin for at least 3 months (insulin dose of at least 30 U/day for at least 4 weeks)
• HbA1c ≥7%


• Weight loss >5 kg within 3 months prior to screening visit
• Pregnancy or lactation
• Multiple allergies
• Prior participation in rimonabant study
• Presence of any clinically significant endocrine disease
• Fasting C-peptide <1.0 ng/ml
• Presence or history of cancer within the past 5 years
• Positive test for hepatitis B surface antigen and/or hepatitis C antibody
• Abnormal thyroid stimulating hormone (TSH) level (TSH greater than upper limit of normal or less than lower limit of normal)
• Antidiabetic drugs other than insulin within 3 months prior to screening visit
• Drugs affecting weight (e.g., sibutramine, orlistat, herbal preparations, etc.)

Primary Endpoints:

Change from baseline to week 48 in HbA1c

Secondary Endpoints:

• Change in glucose levels
• Changes in total insulin dose
• Change in body weight and waist circumference
• Change in lipid parameters
• Safety and tolerability

Drug/Procedures Used:

Rimonabant 20 mg daily or matching placebo

Concomitant Medications:


Principal Findings:

ARPEGGIO was performed in 60 centers across 12 countries. A total of 368 patients were randomized, 179 to rimonabant, and 187 to placebo. The baseline glycated hemoglobin (HbA1c) was 9.1%, and the baseline fasting glucose (FSBG) was about 202 mg/dl. The mean insulin dose was about 83.6 U. Baseline lipids: high-density lipoprotein (HDL) 51 mg/dl, low-density lipoprotein (LDL) 114 mg/dl, triglycerides 233 mg/dl. Baseline body weight was about 94.4 kg.

Efficacy: Rimonabant was associated with a significant improvement in HbA1c from baseline compared with placebo (-0.89 vs. -0.24, p < 0.0001); the percentage of patients with adequate glycemic control (HbA1c <7%) was also better with rimonabant (18.4% vs. 6.7%, p < 0.0012). Reduction in FSBG was better with rimonabant (11 vs. 33 mg/dl, p = 0.019). The mean insulin dose was reduced by 2.9 U in the rimonabant arm compared with placebo (-2.71 vs. -0.21, p = 0.0004). There was also a 2.56 kg increased weight loss with rimonabant compared with placebo (-2.49 vs. 0.13, p < 0.0001). HDL was increased with rimonabant (3.14 vs. -7.14, p < 0.0001), whereas triglycerides were reduced (-3.99 vs. 7.64, p = 0.03).

Safety: Serious side effects were similar between rimonabant and placebo (16.8% vs. 19.3%), including mortality (0 vs. 1.1%). Psychiatric side effects were more common in the rimonabant arm (31.3% vs. 18.7%), as was the incidence of polyneuropathy (2.8% vs. 0). Hypoglycemia was also more common in the rimonabant arm, both symptomatic (54.7% vs. 47.1%) and severe (4.5% vs. 3.7%).


ARPEGGIO is part of a series of trials using rimonabant in patients with diabetes, including SYMPHONY (with sulfonylureas) and ALLEGRO (with metformin). This is the first trial done in patients who are also on insulin. The results of ARPEGGIO indicate that rimonabant, a CB1 receptor antagonist, is superior to placebo in improving glycemic control in patients with type 2 diabetes and cardiovascular disease, who are on insulin. In addition, there is a small reduction in mean insulin requirements, modest weight loss, as well as some improvement in lipid parameters, with an increase in neuropsychiatric side effects and hypoglycemia.

The reductions in HbA1c noted in this trial are similar to those noted with thiazolidinedione, vildagliptin, and orlistat. A head-to-head comparison between rimonabant and these agents is necessary to assess relative safety and cost-effectiveness profiles.


Rimonabant Improves Glycemic Control in Insulin-Treated Type 2 Diabetes: The ARPEGGIO Trial. Presented by Dr. Priscilla Hollander at the 68th Scientific Sessions of the American Diabetes Association, San Francisco, CA, June 2008.

Hollander PA, Amod A, Litwak LE, Chaudhari U; ARPEGGIO Study Group. Effect of rimonabant on glycemic control in insulin-treated type 2 diabetes: the ARPEGGIO trial. Diabetes Care. 2010 Mar;33(3):605-7.

Clinical Topics: Dyslipidemia, Heart Failure and Cardiomyopathies, Prevention, Lipid Metabolism, Statins, Diet

Keywords: Polyneuropathies, Insulin, Tryptamines, Weight Loss, Diabetes Mellitus, Type 2, Lipids, Coronary Disease, Pyrazoles, Hypoglycemia, Hemoglobin A, Glycosylated, Piperidines, Cannabinoid Receptor Antagonists, Carbazoles, Nitriles, Blood Glucose, Metformin, Lactones, Pyrrolidines, Receptor, Cannabinoid, CB1, Thiazolidinediones, Fasting

< Back to Listings