The 339 participants in RIO-Diabetes who were randomized to receive rimonabant 20 mg/day lost significantly more weight than the 348 who received placebo and also showed significant improvements in levels of high-density-lipoprotein (HDL) cholesterol, triglycerides, glycosylated hemoglobin (HbA1_C), C-reactive protein, and other metabolic markers. Much of the metabolic improvement was independent of weight loss, consistent with the idea that rimonabant, a selective inhibitor of the endocannabinoid system, has a direct modulating effect on such risk markers, according to author Dr André J Scheen (University of Liege, Belgium) and colleagues.

It's right to be very conservative. So many of us remember the Redux era and then its hasty withdrawal from the market.

Rimonabant has been shown in several trials to promote weight loss and improve components of the metabolic syndrome in overweight or obese individuals, making it seem like a natural for type 2 diabetics. Even with its limitations, RIO-Diabetes, which was presented in preliminary form at the June 2005 annual meeting of the American Diabetes Association and covered by heartwire then, suggests the drug might well improve many of such people's most prominent CV risk factors.

"So far this looks like the best chance that we've had for pharmacologic therapy to deal with some of the root causes of the metabolic syndrome, mainly the abdominal obesity," Dr Roger Blumenthal (John Hopkins School of Medicine, Baltimore, MD), who wasn't involved in RIO-Diabetes, told heartwire. "The differences observed aren't dramatic, but nevertheless we have a lot of data suggesting that if people are able to decrease their weight 10 to 15 pounds and keep it off, they can markedly improve their risk-factor profile. And realistically, that's what we're hoping for, that we'll have a good safe medication that can be used in selective instances [to] achieve that."

RIO-Diabetes: Weight and waist-circumference changes at one year from baseline (all differences, p<0.0001)

A third randomization group of 358 people treated with rimonabant 5 mg/day showed significant but less-pronounced reductions in weight and waist circumference and almost no significant changes in metabolic parameters.

All participants in the trial, conducted in North America, Europe, and South America, had type 2 diabetes that was "inadequately" controlled with either metformin or a sulfonylurea over at least six months and a body mass index of 27 to 40 kg/m² and were free of other clinically significant conditions, including severe vascular complications of diabetes.

Dropout rates were about the same in the three groups, the outcomes of which were analyzed on an intention-to-treat basis. More people taking rimonabant than taking placebo withdrew because of adverse events, most often "depressed mood disorders, nausea, and dizziness," the authors write.

The accompanying editorial by Drs Stephen J Cleland (Stobhill Hospital, Glasgow, Scotland) and Naveed Sattar (University of Glasgow, Scotland), which Blumenthal called "thoughtful" and "very well done," describes the limitations of the trial and its findings—most acknowledged in the RIO-Diabetes publication—that point to shortfalls in what is known for sure about rimonabant as a weight-loss drug.

The suggestion that rimonabant increases depression and anxiety is concerning and something that should be formally assessed before consideration of this pharmacologic approach.

Cleland and Sattar point out that only two thirds of study participants actually completed one year of treatment, "with over half dropping out for reasons other than adverse events." They also question the underpinnings of the contention that metabolic improvements were partially independent of weight loss. They further observe that the report included no information on the dietary advice participants received and how much it was heeded. Nor is it known how much participants exercised, they note, which could have influenced metabolic parameters independent of weight loss.

In addition, the editorialists write, the prevalence of psychological disorders is high in obese people, so "in this context, the suggestion that rimonabant increases depression and anxiety is concerning" and something that "should be formally assessed before consideration of this pharmacologic approach."

Rimonabant was approved in Europe for weight loss in June 2006. In the US, the FDA is expected to eventually give the drug a similar thumbs-up, despite the conspicuously long time that has elapsed since it declared to Sanofi-Aventis that it was "approvable."

"It's right to be very conservative. So many of us remember the Redux [dexfenfluramine] era and then its hasty withdrawal from the market," Blumenthal said, referring to a drug that, along with the fenfluramine-phentermine combination popularly called fen-phen that was taken off the market, was blamed during the 1990s for causing valvular dysfunction. "I think the company and the investigators have been appropriately cautiously optimistic."

RIO-Diabetes: Metabolic parameters, changes at one year from baseline

1. Scheen AJ, Finer N, Hollander P, et al. Efficacy and tolerability of rimonabant in overweight or obese patients with type 2 diabetes: A randomised controlled study. Lancet 2006;DOI:10.1016/S0140-6736(06)69571-8. Available at: http://www.thelancet.com.
2. Cleland SJ, Sattar N. Does rimonabant pull its weight for type 2 diabetes?Lancet 2006;DOI:10.1016/S0140-6736(06)69572-X. Available at: http://www.thelancet.com.
   
   
   heartwire

   ---