Psychiatric Side Effects Cloud Another Promising Obesity Drug

By John Gever
RAHWAY, N.J., Jan. 8 -- Although obese patients lost weight with investigational taranabant, some also had psychiatric effects similar to those that have stymied approval of another diet drug in the same class, researchers here said.
After 12 weeks of taking taranabant, a cannabinoid-1 receptor inverse agonist, in a phase II trial, there were dose-dependent decreases of 2.9 to 5.3 body mass index points, from a baseline BMI mean of about 36, Steven B. Heymsfield, M.D., of Merck, and colleagues reported in the January issue of Cell Metabolism.
All of these decreases were significant relative to the 1.2-point decrease in BMI in placebo-treated patients (P<0.0001).
But 20.8% to 31.4% of patients taking taranabant had psychiatric adverse effects, including depressed mood, irritability, anger, mood swings, tearfulness, anxiety, insomnia, and nervousness. Those effects led some 15% of patients to drop out.
The randomized, multicenter, double-blind study tested four doses of taranabant -- 0.5, 2, 4, and 6 mg daily -- and placebo in a total of 533 evaluable patients.
The percent of patients who lost more than 5% of their baseline body weight at week 12 were 12.9% for placebo and 26.5% for 0.5 mg of taranabant, 30.3% for 2 mg, 42.7% for 2.4 mg, and 60.8% for 6 mg, respectively, with the trend test in the 5% responders being significant over the entire range of doses from placebo The corresponding results for 10% responders were 2.0%, 4.9%, 6.1%, 5.8% and 14.4%, respectively.
The mean changes from baseline in waist circumference at week 12 were -2.8 cm for placebo, and -4.4 cm for 0.5 mg of taranabant, -4.4 cm for 2 mb, -4.7 cm for 2,4 mg, and -5.1 cm for 6 mg.
Most of the anxiety events occurred in the higher-dose taranabant groups, said Dr. Heymsfield. Insomnia and mood-related effects occurred with similar frequency in all groups.
In addition to the psychiatric effects, gastrointestinal symptoms were also common, affecting 38.7% to 61.9% of patients receiving taranabant and 38.1% of the placebo group. The frequency of these effects was greater in the high-dose taranabant groups.
In about half the patients, the gastrointestinal and psychiatric effects appeared within the first three weeks of treatment.
No significant trend was detected over the range of doses from placebo through taranabant at 6 mg (P>0.20) in glycemic and lipid measures, including mean percent change from baseline LDL, HDL, total cholesterol, non-HDL-cholesterol, median triglycerides, and change in fasting plasma glucose.
The percentage of patients who discontinued during the 12-week treatment phase were 38.1%, 27.4%, 27.5%, 39.0% and 32.4% for placebo, taranabant 0.5, 2, 4 and 6 mg, respectively.
Mechanism-of-action studies suggested that engagement of the cannabinoid-1 receptor inverse agonist by taranabant leads to weight loss by reducing food intake and increasing energy expenditure and fat oxidation.
Taranabant targets the cannabinoid-1 receptor, as does rimonabant (Zimulti, Acomplia), developed by sanofi-aventis. The latter's application to the FDA for approval has been held up because of clinical reports of severe depression and suicidal ideation among some users. Rimonabant is available in some 40 other markets, including the European Union.
A Merck executive told stock analysts last month that psychiatric adverse effects had been seen in taranabant-treated patients, but details had not been disclosed until today's journal publication.
A Merck spokesperson said it now appears that psychiatric effects are intrinsic to drugs targeting the cannabinoid-1 receptor.
The exact mechanism for the effect is unclear, Dr. Heymsfield and colleagues wrote, but they believe it is related to its binding to the cannabinoid-1 receptor. They believe the psychiatric effects of taranabant are manageable at the doses studied in the phase II trial.
On the basis of earlier studies, the researchers believed that 30% occupancy of cannabinoid-1 receptors in the brain would be necessary to achieve substantial weight loss. They anticipated that doses of at least 4 mg/day would be needed to meet this target for receptor saturation. "We didn't expect weight loss at all doses," Dr. Heymsfield said.
Because the drug appears effective at the lower doses, which also produced fewer adverse events overall in the phase II study, pivotal trials now underway are using doses in the 0.5 to 2-mg range.
Dr. Heymsfield said the phase III data would cover a longer treatment period.
The authors noted as a concern the high dropout rate over the 12-week evaluation period, which they noted is common to most weight-loss drug trials.
They also noted that the brain cannabinoid-1 receptor inverse agonist occupancy was evaluated only in lean male participants, and the extent to which these data can be extrapolated to obese females, who are likely to comprise a significant proportion of the treatment population, needs to be confirmed.
Merck funded the study. Dr. Heymsfield and 22 co-authors are Merck employees. No other potential conflicts of interest were disclosed.
Primary source: Cell MetabolismSource reference:Addy C, "The acyclic CB1R inverse agonist taranabant mediates weight loss by increasing energy expenditure and decreasing caloric intake Cell Metabolism 2008; 7: 68-78.