Merck & Co. on Thursday announced its decision to cancel further investigation into its experimental obesity drug called taranabant because of side effects uncovered in clinical trials.

The company said that both effectiveness and side effects were dependent on dose levels; at more effective dosages, the side effects climb too much. The side effects associated with the drug taranabant were psychiatric, including anxiety and depression, Merck spokeswoman Amy Rose said.

“Available Phase III data showed that both efficacy and adverse events were dose related, with greater efficacy and more adverse events in the higher doses,” John Amatruda, Merck’s senior vice president and research head for diabetes and obesity, said in a statement.

“Therefore, after careful consideration, we determined that the overall profile of taranabant does not support further development for obesity.”

The company plans to present data from tarabant clinical studies at a major obesity conference which is scheduled to begin this weekend.

The drug maker’s shares were down $0.25, or 0.78 percent, to $31.84 in early afternoon trading, near the bottom of the stock’s 52-week range between $30.34 and $61.62.

In 2007, Sanofi-Aventis’ anti-obesity drug, generically known as rimonabant (Acomplia) was also rejected by the Food and Drug Administration in the United States due to concerns about the risk of suicidal thoughts and depression among some users. Sanofi subsequently withdrew its application.

Taranabant is a cannabinoid-1 (CB-1) receptor inverse agonist, which means it works by blocking cannabinoid receptors in the brain and it suppresses appetite. Cannabinoid receptors are the same receptors that make people hungry when smoking marijuana.

The experimental therapy was in a phase III clinical trial and just one step away from being submitted for regulatory approval. In a first study including 553 obese patients, who were divided in more groups, which were administered either 0.5,2,4 or 6 milligrams daily, or a placebo, the researchers found that those who took the drug taranabant ate 20 percent fewer calories than those who were given placebo. The drug also increased resting energy expenditure or metabolism by 5 percent. But the negative consequence of the treatment was that 30 percent of people who received taranabant reported psychiatric-related adverse events including depression and anxiety, compared with 18 percent in the placebo group. Larger amounts also led to stomach upsets, nausea and vomiting and caused patients to become more irritable.

After one year, a study involving over 2500 obese patients found that, in combination with diet and exercise, patients given a 2 mg dose lost an average of 14.5 pounds (6.6 kg), compared with 5.7 pounds for the placebo group. With regards to some of the negative side effects, 28.4% of those taking taranabant experienced some type of psychiatric side-effects, versus 20.3% in the placebo group. There were also some psychiatric problems reported such as depression, anxiety and irritability. The company continued to study taranabant even after it became aware of the psychiatric side effects in midstage studies. Merck has said the high rate of obesity justified the continuation of studies. The company had planned to bring the drug before the FDA approval by the 2nd half of 2008.