PHILADELPHIA— Several therapeutic agents aimed at novel targets in the brain are visible on the far horizon for the treatment of depression, including one drug with an infamous history in a much different application. But expectant clinicians and their patients will likely have to endure many ups and downs throughout the development process. Easily half of the trials necessary for a drug to be approved will fail, according to the prognostication of at least one expert on new treatments for depression.

A review of the promising but “too-soon-to-call” formative periods of these new agents illustrates just how much antidepressant drug development is not only subject to trial and error but fraught with chance and dependent on the instincts of investigators.

FOLLOWING A HUNCH AND FINDING A NEW ROLE FOR A NOTABLE COMPOUND

One agent, a compound called MK-869 (aprepitant), acts on the NK-1 receptor in the brain. It is one among a family of such targets known as neurokinins that were discovered in the brain and first shown to have pharmacologic activity in 1934. They are also known under the collective name Substance P.

Substance P was originally thought to be a target for the treatment of pain, although therapeutic agents developed to work via this mechanism have never been effective as pain relievers. It took a hunch by researchers, impressed by suggestive but inconclusive animal data, to see if the NK-1 mechanism would work for depression. There are now at least several compounds based on this mechanism in development.

A second agent, which is at an earlier stage of development than the NK-1 compounds, is derived from the corticotropin-releasing hormone (CRH) that regulates adrenal access. CRH is a stress hormone that is elevated in human depression and in animals known to mimic the symptoms of depression. To date, of the several CRH antagonists in development, only one compound has gotten as far as an open-label clinical trial, after which it was discontinued due to an association with increased liver problems.

A third compound with a novel mechanism is being investigated for use in the treatment of psychotic depression. In psychotic depression, there is excess production of the hormone cortisol, and this compound is therefore being developed as a glucocorticoid receptor antagonist. Its generic name is mifepristone, but it is better known by the original number for the compound—RU-486.

Most recently, mifepristone was shown to be effective in psychotic depression in a trial in which investigators compared a low dose with a high dose of the drug. “In a very sick, depressed population, you cannot use placebo,” observed Ranga K. Krishnan, MD, in his presentation at the 2002 Annual Meeting of the American Psychiatric Association. “The high dose worked better than the low dose.” He finds mifepristone’s efficacy in depression to be the bright side of a rather mixed picture. “It’s a two-edged sword, when people connote that it is a drug for abortion, although its benefits are likely to be for other things,” he said. A Professor of Psychiatry and Behavioral Sciences at Duke University in Durham, North Carolina, Dr. Krishnan is recognized for his contributions to psychiatry’s knowledge base on new forms of depression, including vascular depression in the elderly.

HURDLES TO OVERCOME IN ANTIDEPRESSANT DRUG DEVELOPMENT

In addition to the three new agents that he focused on, Dr. Krishnan offered an incisive overview of the problems inherent in developing antidepressant drugs based on new mechanisms. For one, researchers do not have useful preclinical models in this area. “The animal models we have are very good at identifying the drugs that we are already using,” he said, “but how well do they predict a drug we want to develop?”

The science behind antidepressant therapy is entering a new era in that all of these novel compounds under investigation work via a mechanism of action different from that of most of the agents currently available. Such antidepressants as those in the class of drugs called selective serotonin reuptake inhibitors (SSRIs), for example, work via a monoamine mechanism of action. This difference helps to explain why novel antidepressants are hard to develop, since our understanding of the pathophysiology of the disease remains limited, according to Dr. Krishnan. “We can speculate a lot about it and we can talk a lot about it, but we still do not have clearly identifiable targets with regard to the pathophysiology of the disease,” he said.

There is also a lack of hard outcome measures for determining whether a patient is getting better. “The best measure is still the clinical response that a patient tells you or that you can observe,” Dr. Krishnan said. But these are highly subjective and therefore less reliable measures that make it difficult to develop initial, proof-of-concept trials. What’s more, the odds of achieving a successful outcome in a clinical trial in antidepression are greater the further one goes in the developmental process, Dr. Krishnan observed. He cited the excessive number of trials needed to launch such successful SSRIs as paroxetine and sertraline, including several failed trials. “With later antidepressants, even though the drugs worked using the exact same mechanism, the number of failed trials increased,” he said.

Of the drugs currently on the market, Dr. Krishnan suggested, on average, at least 50% of the trials required for their development failed. NK-1 is a case in point. After a successful proof-of-concept trial, the compound MK-869 failed in a second, fixed-dose, multi-arm trial that is required for regulatory approval.

“These trials probably have less than a third of a chance of succeeding, even if the drug actually works,” he explained. “When you tell people they only have a one in five chance of getting placebo, more people think they’re going to get better. The power of expectancy is very high, and it plays a role.” Additionally, “these trials get to be very, very large and therefore are more error prone.”

A subsequent NK-1 trial, using a different compound but the same mechanism, was positive, and the drug is now in phase III clinical trials. If it continues to be successful, the agent is likely to be on the market by 2005, Dr. Krishnan projected. He concluded his presentation with advice to investigators on the pitfalls of designing such trials as those conducted so far in NK-1. “The greater the number of arms that you put in,” he cautioned, “the more likely you are to waste your money.”

—Fred Balzac

Return to table of contents