By ALEX BERENSON
NORTH WALES, Pa. — SCIENTISTS who develop drugs are familiar with disappointment — brilliant theories that don’t pan out or promising compounds derailed by unexpected side effects. They are accustomed to small steps and wrong turns, to failure after failure — until, in a moment, with hard work, brainpower and a lot of luck, all those little failures turn into one big success.
For Darryle D. Schoepp, that moment came one evening in October 2006, while he was seated at his desk in Indianapolis. At the time, he was overseeing early-stage neuroscience research at Eli Lilly & Company and colleagues had just given him the results from a human trial of a new schizophrenia drug that worked differently than all other treatments. From the start, their work had been a long shot. Schizophrenia is notoriously difficult to treat, and Lilly’s drug — known only as LY2140023 — relied on a promising but unproved theory about how to combat the disorder.
When Dr. Schoepp saw the results, he leapt up in excitement. The drug had reduced schizophrenic symptoms, validating the efforts of hundreds of scientists, inside and outside of Lilly, who had labored together for almost two decades trying to unravel the disorder’s biological underpinnings.
The trial results were a major breakthrough in neuroscience, says Dr. Thomas R. Insel, director of the National Institute of Mental Health. For 50 years, all medicines for the disease had worked the same way — until Dr. Schoepp and other scientists took a different path.
“This drug really looks like it’s quite a different animal,” Dr. Insel says. “This is actually pretty innovative.”
Dr. Schoepp and other scientists had focused their attention on the way that glutamate, a powerful neurotransmitter, tied together the brain’s most complex circuits. Every other schizophrenia drug now on the market aims at a different neurotransmitter, dopamine.
The Lilly results have fueled a wave of pharmaceutical industry research into glutamate. Companies are searching for new treatments, not just for schizophrenia, but also for depression and Alzheimer’s disease and other unseen demons of the brain that torment tens of millions of people worldwide.
Driving the industry’s interest is the huge market for drugs for brain and psychiatric diseases. Worldwide sales total almost $50 billion annually, even though existing medicines have moderate efficacy and have side effects that range from reduced libido to diabetes.
The glutamate researchers warn that their quest for new treatments for schizophrenia is far from complete. The results of the Lilly trial covered only 196 patients and must be validated by much larger trials, the last of which may not be finished until at least 2011. Other glutamate drugs are even further away from approval. And even if the drugs win that approval, they may be viewed skeptically by doctors who have been disappointed by side effects in other drugs that were once been hailed as breakthroughs.
Still, for Dr. Schoepp, the drug’s progress so far is cause for celebration — and relief.
“I don’t think people appreciate how much money, time and good technical research goes into what we do,” he says. “Sometimes, people think the idea is the thing. I think the idea can be the easy part.”
LILLY continues to develop LY2140023 and has begun a trial of 870 patients that is scheduled to be completed in January 2009. But Dr. Schoepp is no longer involved in its development. He left Lilly in April to become senior vice president and head of neuroscience research at Merck, where he oversees a division of 300 researchers and support staff members.
Dr. Schoepp’s new base is a modest office on the top floor of a four-story Merck building here in North Wales, north of Philadelphia. He has a view of the building’s big front lawn and a busy two-lane road called the Sumneytown Pike. The huge Merck research complex called West Point, where 4,000 scientists and support staff members work, is less than a mile to the north.
For Dr. Schoepp, 52, the Merck job is the latest stop in a research career that began at Osco Drug’s store No. 807 in downtown Bismarck, N.D. He grew up in Bismarck in a working-class family; at 16, he started working at the Osco, which has since closed. He quickly decided to become a scientist.
“I just found it fascinating,” he says. “I was hungry for science.” While reading a magazine for pharmacists, he noticed an ad for a free pamphlet published by Merck called “Pharmacists in Industry.” He wrote away for the pamphlet, which convinced him that he could have a career developing medicines.
He applied to North Dakota State University, where he focused on psychopharmacology, a discipline that studies the way chemicals affect the brain. “I was really interested in psychiatric disorders,” he says. “I fell in love with dopamine.”
His love affair was so consuming that his wife joked that “dopamine” would be his daughter’s first word.
Although scientists sometimes decide to study a disease because of problems it has caused among family members, Dr. Schoepp says his fascination with mental illness has been purely academic. “My family has more heart disease than anything else,” he says.
After graduating from North Dakota State, he received a scholarship to a doctoral program in pharmacology and toxicology at West Virginia University. He graduated in 1982. Nearly five years later, he joined Lilly, which was about to introduce Prozac, the first modern antidepressant — a drug that changed both psychiatry and the public perception of depression and mental illness.
Prozac became a blockbuster almost instantly after Lilly introduced it in 1987, making the company one of the most visible players in Big Pharma and giving it room to invest in long-shot scientific research. Ray Fuller, a Lilly scientist who was a co-discoverer of Prozac, encouraged Dr. Schoepp to focus his attention on glutamate.
Glutamate is a pivotal transmitter in the brain, the crucial link in circuits involved in memory, learning and perception. Too much glutamate leads to seizures and the death of brain cells. Excessive glutamate release is also one of the main reasons that people have brain damage after strokes. Too little glutamate can cause psychosis, coma and death.
“The main thoroughfare of communication in the brain is glutamate,” says Dr. John Krystal, a psychiatry professor at Yale and a research scientist with the VA Connecticut Health Care System.
Along with Bita Moghaddam, a neuroscientist who was at Yale and is now at the University of Pittsburgh, Dr. Krystal has been responsible for some of the fundamental research into how glutamate works in the brain and how it may be implicated in schizophrenia.
Schizophrenia affects about 2.5 million Americans, about 1 percent of the adult population, and it usually develops in the late teens or early to mid-20s. It is believed to result from a mix of causes, including genetic and environmental triggers that cause the brain to develop abnormally.
The first schizophrenia medicines were developed accidentally about a half-century ago, when Henri Laborit, a French military surgeon, noticed that an antinausea drug called chlorpromazine helped to control hallucinations in psychotic patients. Chlorpromazine, sold under the brand name Thorazine, blocks the brain’s dopamine receptors. That led the way in the 1960s for drug companies to introduce other medicines that worked the same way.
The medicines, called antipsychotics, gave many patients relief from the worst of their hallucinations and delusions. But they also can cause shaking, stiffness and facial tics, and did not help the cognitive problems or the so-called negative symptoms like social withdrawal associated with schizophrenia.
In the 1980s, drug companies looked for new ways to treat the disease with fewer side effects. By the mid-1990s, they had introduced several new schizophrenia medicines, including Zyprexa, from Lilly, and Risperdal, from Johnson & Johnson. At the time, the new medicines were hailed as a major advance — and the companies marketed them that way to doctors and patients.
In fact, the new medicines, called second-generation antipsychotics, had much in common with the older drugs. Both worked mainly by blocking dopamine and had little effect on negative or cognitive symptoms. The newer medicines caused fewer movement disorders, but had side effects of their own, including huge weight gain for many patients. Many doctors now complain that the companies oversold the second-generation compounds and that new treatments are badly needed.
“People say that there are drugs to treat schizophrenia,” says Dr. Carol A. Tamminga, professor of psychiatry at the University of Texas Southwestern, in Dallas. “In fact, the treatment for schizophrenia is at best partial and inadequate. You have a cadre of cognitively impaired people who can’t fit in.”
WHILE most of the industry focused on second-generation medicines during the 1980s and 1990s, a handful of academic and industry researchers found intriguing hints that glutamate might provide an alternative treatment pathway.
Psychiatrists and neuroscientists have wondered about a possible connection between glutamate and schizophrenia since the early ’80s, when they first learned that phencyclidine, the street drug commonly called PCP, blocks the release of glutamate.
People who use PCP often have the hallucinations, delusions, cognitive problems and emotional flatness that are characteristic of schizophrenia. Psychiatrists noted PCP’s side effects as early as the late 1950s. But they lacked the tools to determine how PCP affected the brain until 1979, when they found that it blocked a glutamate receptor, called the NMDA receptor, that is at the center of the transmission of nerve impulses in the brain.
The PCP finding led a few scientists to begin researching glutamate’s role in psychosis and other brain disorders. By the early 1990s, they discovered that besides triggering the primary glutamate receptors — NMDA and AMPA — glutamate also triggered several other receptors.
They called these newly found receptors “metabotropic,” because the receptors modified the amount of glutamate that cells released rather than simply turning circuits on or off. Because glutamate is so central to the brain’s activity, directly blocking or triggering the NMDA and AMPA receptors can be very dangerous. The metabotropic receptors appeared to be better targets for drug treatment.
“Rather than acting as an all-or-nothing signal, they fine-tune that signal and modulate that signal,” said P. Jeffrey Conn, director of a Vanderbilt University drug research program. “It’s really an attempt to be very subtle in the way that you regulate the system.”
During the 1990s, molecular biologists discovered genes for eight metabotropic glutamate receptors, which were located at different places inside nerve cells and had different structures. The finding allowed for the possibility that drug companies could create chemicals to turn them on and off selectively, rather than hitting all of them at once.
For Dr. Schoepp and others, finding the receptors was only the first part of the struggle. They also had to find chemicals that would either block or trigger the receptors selectively. At the same time, the chemicals had to be relatively easy to formulate and capable of crossing the blood-brain barrier, which protects the brain from being easily penetrated by outside agents.
The work was arduous, but the Lilly scientists made slow progress. In 1999, Dr. Schoepp and two other scientists published a 46-page research paper that detailed scores of different chemicals that produced reactions at the glutamate sites.
At about the same time, scientists at Yale, led by Dr. Moghaddam, were demonstrating that activating metabotropic glutamate receptors in rats could reverse the effects of PCP — a seminal finding, providing the first proof that altering the path of glutamate transmission in the brain might help relieve the symptoms of psychosis.
Although the finding in rats was promising, developing animal models for schizophrenia and other brain diseases is extremely difficult, said Paul Greengard, professor of molecular and cellular neuroscience at Rockefeller University.
Even when compared with diseases like cancer, brain disorders are notoriously complex. Scientists have only a limited understanding of the chemistry of consciousness, or of how problems in the brain’s electrical circuitry affect the ability to form memories, learn or think.
“We do not know with any of these neuropsychiatric disorders what the ultimate basis is,” Dr. Greengard says. “Let’s say you could find that too much of protein X was involved in schizophrenia. Would you then know what schizophrenia is? You would not.”
Nonetheless, the findings in rats were promising. Those studies, as well as Dr. Krystal’s tests in 2001 of volunteers given ketamine, a drug that has effects similar to PCP, hinted that the glutamate drugs might help to treat the cognitive and negative symptoms of schizophrenia. Drugs currently on the market do little to treat those symptoms.
Even before the findings at Yale, Lilly had put its first metabotropic glutamate receptor compound into human testing. Researchers initially tested the drug on patients with panic disorder, and it showed some positive results. But Lilly stopped human testing of the drug in 2001 when long-term testing in animals showed that it caused seizures.
Even so, Lilly decided that it had enough evidence to justify tests of another chemical compound, LY404039, that affected the same receptors.
“They had to take a risk on letting these drugs be tested on models or for disorders that were justified purely on pretty basic science,” Dr. Krystal says. “There is nothing with these drugs that is straightforward or makes developing them a basic path.”
When it tried to test LY404039 in humans, the company ran into yet another hurdle. The human body didn’t easily absorb it. So Lilly created a drug that the body could absorb, LY2140023, which is metabolized into LY404039 in the body.
Bingo. LY2140023 was the drug that got Dr. Schoepp jumping out of his office chair in 2006, nearly three years after the first trials in humans began. In the Lilly test, the drug was slightly less effective over all than Zyprexa, which is considered the most effective among the widely used schizophrenia treatments.
But LY2140023 also appeared to have fewer side effects than Zyprexa, which can cause severe weight gain and diabetes. The new drug also appeared to improve cognition, something that existing treatments don’t do, said Dr. Insel of the National Institute of Mental Health.
IF Lilly’s new round of tests confirms the drug’s efficacy by early next year, the company is likely to move ahead to an even larger clinical trial, involving thousands of patients, that could lead to federal approval for the compound. Still, approval is at least three to four years away, and other big drug makers are already scrambling to compete with Lilly.
In January, Pfizer agreed to pay Taisho Pharmaceutical, a Japanese company, $22 million for the rights to develop Taisho’s glutamate drug for schizophrenia. Taisho will receive more payments if the drug moves forward in development.
Since it hired Dr. Schoepp, Merck has also been moving aggressively. It has struck two deals since December to work with Addex Pharmaceuticals, a Swiss company, to develop glutamate drugs for schizophrenia, Parkinson’s and other diseases. Merck has paid Addex $25 million so far, with more payments to come if the drugs move forward.
Another glutamate drug, meanwhile, has been shown in preclinical studies to reverse mental retardation in adult rats, a finding that previously appeared impossible, Dr. Insel said.
Dr. Steven M. Paul, the president of Lilly Research Laboratories, says Lilly expects competition in glutamate research to intensify. “We’d like to believe we have a head start here, and hopefully a good head start,” he says. “But this area will heat up here; this will be an area where there will be a lot of investment.”
For Dr. Schoepp, the sudden interest in glutamate is exciting, and he acknowledges that he eagerly awaits the results of the large Lilly trial early next year. And what if the drug fails in that trial, after all the work that he and scientists around the world have put in?
“I would probably go out and have a beer,” he says. “You have to define failure. If you collect information and it tells you what you need to know, you’re not a failure.”