The road to a sleeping pill that offers the snooze-inducing benefits of Ambien and Lunesta, while avoiding next-day grogginess and disorientation, has been paved with failure, tripping up GlaxoSmithKline Plc and Takeda Pharmaceutical Co.

Now, Merck & Co. may have found a solution.

The drugmaker is in final testing on an insomnia product, named suvorexant, that works differently than the market leaders. It blocks a tiny group of receptors that keep the body alert, rather than boosting the brain’s complex sleep system. The target is more precise, offering less room for side effects to occur, the Whitehouse Station, New Jersey-based company said.

The strategy, which could also allow people to use the treatments over longer periods, may help the pill generate $900 million annually if it becomes the first of its class approved by the Food and Drug Administration, according to Tony Butler, an analyst with Barclays Plc in New York.

“It’s not whether you put the patient to sleep or not, it’s how you feel the next day,” said Rafael Pelayo, a professor at Stanford School of Medicine’s Division of Sleep Medicine, in a telephone interview.

About 70 million, or one-third of American adults, have trouble falling or staying asleep, according to the American Psychiatric Association. Pelayo predicts that sleep issues will grow with the stress of modern life and the disruption from ever-present personal electronic devices, such as Apple Inc.’s iPhone, boosting the market for insomnia treatments.

“That’s why I think the pharma industry and investors look at sleep as a good place to put their money,” Pelayo said.
Merck shares declined 1.2 percent to $43.41 at the close in New York.

Recent Failures

History, though, isn’t on Merck’s side. London-based Glaxo last year ended development of its leading sleep drug, almorexant, after spending as much as 4 billion Swiss francs ($4.07 billion) to buy it from Actelion Ltd. The problem: patients couldn’t tolerate its side effects.

Sanofi’s bid to replace its own Ambien, which faced generic competition for the first time in 2006, ended when the Paris- based drugmaker failed to get U.S. approval for the treatment, called Ciltyri, in 2009.

Meanwhile, the two insomnia drugs that have been marketed since Ambien and Lunesta have failed to take hold with consumers.

Takeda’s Rozerem, approved in the U.S. in 2005, generated just $11.7 million in sales last year for the Osaka, Japan-based drugmaker, while Somaxon Pharmaceuticals Inc.’s Silenor had only $16.2 million in sales last year.

Disappointed Clinicians

The newer therapies failed because they’ve haven’t reduced side effects, disappointing clinicians eager for new medicines, said Stanford’s Pelayo.

“People said, ‘Here’s the next big thing,’ and they turned out to be duds,” he said. Along with grogginess and driving troubles, Ambien can have side effects including stomach upset, diarrhea, nausea and headaches.

The failures have helped clear the field for Merck’s suvorexant, part of a potential new class of drugs known as orexin receptor antagonists.

The medicines may provide the same or better effectiveness against insomnia with fewer side effects, said Thomas Roth, director of the Sleep Disorders and Research Center at Henry Ford Health System in Detroit, by telephone.

“The therapy of insomnia has changed from trying to augment the power of the sleep system to trying to inhibit the wake system,” Roth said, explaining that orexins are a more focused target than the older class of drugs, including Ambien and Lunesta, barbiturates and benzodiazepines.

GABA Agonists

Those medicines called GABA agonists target billions of neuron receptors, affecting a wide range of body activities to achieve their goals. For instance, the therapies can be used as muscle relaxants and anesthetics, along with treating anxiety and convulsion, said Darryle Schoepp, Merck’s head of neuroscience development.

Orexin receptors targeted by the Merck drug are far fewer, with tens of thousands located in the brain’s hypothalamus. That makes for a more focused drug, Schoepp said.

In a study sponsored by Merck, patients on the company’s suvorexant performed better at a driving test the next day, staying in their lane more than patients on eszopiclone, the generic name for Lunesta from Osaka, Japan-based Dainippon Sumitomo Pharma Co.

Merck’s drug is the only one of the new class in the final of three stages of clinical trials usually required for regulatory approval.

“There’s no second at this point,” said Roth, who worked with Merck on the study and consults other drugmakers. The company also has a back-up orexin molecule, MK-6096, which it is also studying for migraines and depression.

Catching Up

Others are trying to catch up. Glaxo, for instance, is developing its own orexin drug, GSK-649868, after failing with almorexant. The U.K.’s largest drugmaker completed a 48-person trial of the drug in May 2007, though it has yet to report data. The company is “looking at the pharmacology of GSK-649868 to determine the next best steps,” Melinda Stubbee, a spokeswoman for the company, said in an e-mail.

Takeda is also sticking a toe into the orexigen pond, according a scientific study describing the work.

Merck’s clinical trial has lasted a year, more than any other sleep drug, which the company says may mean it’s possible for patients to use the drug for longer than current therapies. It has yet to do a head-to-head trial against Ambien and Lunesta.

Pelayo predicts that Merck’s drug, if approved, will lead to more effective combinations of orexin drugs with the old GABA drugs.

“The 1.0 version of these anti-orexin drugs will have some issues, and they’ll be fine-tuned down the road,” he said.