September 18, 2024

Where are the new drugs?

The push to improve the pipeline

Editor’s Note:  This story, originally published in 2005, has been updated.

Tomorrow's medicine chest?  Three-dimensional model of a heterotrimeric G protein, pursued as a possible target for drug therapy.

Photo illustration by Dean Dixon

In 2004, only 23 truly new drugs, called “new molecular entities,” were approved in the United States.

That’s less than half of the number approved in 1996, even though annual research-and-development spending by the pharmaceutical industry more than doubled—to nearly $40 billion—during the same eight-year period.

With the sequencing of the human genome has come a plethora of new technologies to mine it. Yet this new wealth of biological understanding, coupled with the growing demand for drugs that can treat and prevent chronic disease, has raised the bar for proving safety and efficacy to unprecedented heights. Consequently the search for new drugs has become more complicated—and much more expensive.

Depending on the calculations, the journey of a single pill through the convoluted development pipeline can take 15 years and cost more than $1 billion. That’s before any money is spent on marketing.

Much has been written lately about the perceived excesses of drug marketing and inadequate efforts to ensure drug safety. This issue of Lens magazine begins with a look at the top of the pipeline, and how academic medical centers are partnering with industry and the federal government to replenish the shelves of society’s medicine cabinet.

“Drug companies realize the need to cover a broader range of biology. They just can’t do it all and never have,” says Lawrence J. Marnett, Ph.D., director of the Vanderbilt Institute of Chemical Biology. “And so partnerships with universities, with academic health centers make a lot of sense.”

The institute exemplifies the growth of translational research programs at universities around the country. Aided and encouraged by the federal government, these efforts are designed to develop the tools and the knowledge base needed to meet today’s drug-development challenges.

“Our goal is to take those very early stage discoveries around drug targets and lead compounds, and go another step toward handing that information off to biotechnology and pharmaceutical companies and other organizations that can hopefully translate our discoveries into new drugs for patients,” says Jeffrey Balser, M.D., Ph.D., dean of the Vanderbilt University School of Medicine and associate vice chancellor for Research.

Toward that end, the VICB in 2005 opened a high-throughput screening facility to help search for small molecules (a class of organic chemicals) with drug-like activity. Vanderbilt also has signed a master research agreement with biotechnology giant Amgen to conduct an array of collaborative research projects.

The intent of these efforts is to encourage Vanderbilt researchers to pursue the therapeutic potential of their discoveries.

Discovering a potential drug target is not enough, explains Jeffrey Conn, Ph.D., director of the Vanderbilt Program in Drug Discovery. If academic scientists took the next steps—identifying a compound that acted on the target, and conducting the laboratory and animal tests necessary to validate its therapeutic potential—you can then justify a company really locking into a full-scale drug discovery program, he contends.

Jennifer Washburn, author of University, Inc., is more than skeptical. In the February 2005 issue of The American Prospect magazine, she wrote: “Instead of honoring their traditional commitment to teaching, disinterested research, and the broad dissemination of knowledge, universities are aggressively striving to become research arms of private industry.”

Gordon R. Bernard, M.D., assistant vice chancellor for Research at Vanderbilt, disputes that contention.

While collaborations with industry can result in conflicts of interest, many universities, including Vanderbilt, have implemented procedural and contractual safeguards to identify and manage such conflicts. These safeguards protect the academic mission while permitting opportunities to transfer new information and technologies for the benefit of society, he says.

Marnett agrees. “We are not going to be drug companies. But we can advance the field,” he says. We can identify new therapeutic concepts, new drug-design concepts. That’s what we should be doing.”

Turn up the light

Where will the new drugs come from?

One area to watch: G protein-coupled receptors (GPCRs).

GPCRs are embedded in the membranes of nearly every cell and are the most common conduit for signaling pathways found in nature.

Two-thirds of all drugs target these receptors. The beta-blocker drug propranolol lowers blood pressure by preventing adrenaline from binding to its GPCR. Drugs that are given to relieve symptoms of Parkinson’s disease act through a GPCR that binds dopamine.

Parkinson’s disease illustrates the complexity of the signaling pathways that utilize GPCRs. Characterized by tremors, difficulty walking and muscle weakness, the disease is caused by the progressive loss of dopamine-producing nerve cells and the resulting lack of dopamine, a neurotransmitter involved in the coordination of muscle movement.

Current dopamine replacement therapy squelches the tremors and improves coordination, but prolonged use of the drugs can cause significant side effects, including involuntary muscle movements and hallucinations, and the medications become less effective as the disease progresses.

Because loss of dopamine disrupts a complex web of signaling pathways in the brain, it may be possible to restore this balance by “tweaking” pathways involving other neurotransmitters.

While at Merck Research Laboratories, where he was head of neuroscience, Conn and his colleagues found that activating a particular GPCR that bines the neurotransmitter glutamate—mGluR4—relieved symptoms of Parkinson’s disease in animals. However, they could not find a compound that binds only to mGluR4, and does not activate other glutamate receptors elsewhere in the brain.

Allosteric modulation might solve the problem.

This tongue twister refers to the ability of some compounds to bind to a secondary site on a receptor in a way that “modulates” its activation by a primary “ligand” such as a neurotransmitter or hormone. Primary ligands fit into the receptor’s main binding site like a key fitting a lock, and “turn it on.”

The modulator, on the other hand, acts like the dimmer switch in an electrical circuit, adjusting the intensity of the receptor’s activation. The anti-anxiety drugs Valium, Xanax, Librium and Ativan, for example, “potentiate” or turn up the activity of the benzodiazepine receptor when it binds to its primary ligand, the neurotransmitter gamma-aminobutyric acid (GABA).

Conn wondered whether he could find an allosteric potentiator that was specific for mGluR4. However, “my department could only handle a maximum of three programs at any given time,” he says. “And to take a kind of half-baked idea… and decide we’re going to really pull the trigger on a drug discovery program was such a high risk.”

Then, in 2003, he saw an opportunity to pursue his idea at Vanderbilt.

A generation ago, Conn might have spent his entire career searching for a compound that could modulate mGluR4 activity. Now, thanks to Vanderbilt’s high-throughput screening facility, he and his colleagues can test tens of thousands of small molecules for drug-like activity in a single day.

Ultra low volume liquid handlers squirt nanoliter amounts of the compounds into 384-well “microplates” containing their target. Reactions are detected via fluorescence or luminescence as the plates are maneuvered by articulated robots through the screening system.

Compounds that bind to the allosteric site on mGluR4 will be tested in animal models of Parkinson’s disease to see if they actually relieve muscle rigidity and restore coordination.

Conn admits that there is considerable skepticism among his colleagues in industry about “whether we can really pull it off.” But that hasn’t discouraged universities across the country from developing similar capabilities for screening compounds.

Three-dimensional crystal structure of a G protein coupled receptor (GPCR) embedded in a cell membrane, with its loosely attached heterotrimeric G protein, consisting of alpha, beta and gamma subunits, inside the cell. When a ligand, such as a neurotransmitter or hormone, binds to its GPCR, the receptor changes shape in a way that catalyzes the release of guanosine diphosphate (GDP) from the alpha subunit. GDP, an organic molecule involved in intracellular energy exchange, is replaced by the higherenergy guanine triphosphate (GTP). That, in turn, causes the alpha subunit to break apart from the beta and gamma subunits. The subunits then interact with other intracellular proteins to transmit signals down two independent pathways. Within a few seconds, GTP is converted back to GDP, the subunits recombine, and the signals are "turned off."

Illustration by William Oldham

“This is where we fill the gap,” he explains. “I think we are at a turning point in the whole drug discovery industry… We are at a point where different players in the whole therapeutic discovery arena can start to bring a lot more to bear to this process…

“I see it as a really challenging time. But mostly I see it as a very exciting time.”

Pie in the sky

Another potential source of new drugs: compounds that interact with G-proteins.

G-proteins are intracellular molecular switches, involved in nearly every physiological—and presumably, pathological—process. They translate and transmit signals from the receptor to the “response machinery” deep inside the cell.

Here’s how they work:

"There are real benefits… to the scientists doing the research. Even if only 10 percent of these compounds were picked up by industry, the scientific programs would benefit from having potent new tools to probe the biology in cells and even in animals more deeply, leading to new discoveries."
Heidi Hamm, Ph.D., Earl W. Sutherland Jr. Professor and Chair of the Department of Pharmacology at Vanderbilt

Photo by Dean Dixon

When a neurotransmitter or hormone binds to its G protein-coupled receptor on the surface of a cell, the receptor, in turn, activates G proteins that bind to it inside the cell. The proteins actually split into two active parts—alpha subunits and beta/gamma subunits—both of which can stimulate independent signaling pathways.

Drugs that target GPCRs are rather blunt instruments; they can trigger far-ranging side effects. Is it possible to develop drugs that can be delivered—with “nano-surgical” precision—to the G protein of a specific receptor inside a particular type of cell? Could that achieve the therapeutic manipulation of a unique signaling pathway without affecting physiology anywhere else?

That prospect has tantalized Heidi Hamm, Ph.D., for more than two decades. But until recently the idea was, as Hamm puts it, “total pie in the sky.”

In 1993, Hamm helped solve the structure of the alpha subunit with the late Paul Sigler, M.D., Ph.D., and his colleagues at Yale.

More recently, she and colleagues at the University of Illinois at Chicago and the University of Wisconsin-Madison showed how the beta/gamma subunit of an inhibitory G protein controls the release of neurotransmitters and hormones. It prevents vesicles containing these chemical messengers from fusing to the cell membrane and spilling their contents outside the cell.

The discovery, reported in 2005 in the journal Nature Neuroscience, could lead to new ways to treat conditions as diverse as pain and diabetes.

Hamm admits that G protein “therapy” is unlikely to attract major drug company investment—at least not yet. So several years ago, about the time she was moving from Northwestern University to Vanderbilt, she and her colleagues formed their own drug discovery company in Evanston, called cue BIOtech.

They chose to study a receptor embedded in the membrane of clot-forming platelets that binds the coagulation factor thrombin.

Blood clotting is essential for wound healing, but too much thrombin in the wrong place can trigger a heart attack. Blood-thinning drugs like Coumadin can prevent platelets from forming clots, but—unless the dose is carefully monitored—they can cause uncontrollable bleeding.

It has been difficult to block thrombin, which actually is an enzyme that activates its receptor by chopping it in half. So Hamm and her colleagues are trying to tackle the problem from inside the cell, by blocking receptor action instead of receptor binding.

So far they’ve been able to make “very potent” small molecules that prevent the thrombin receptor from binding to or activating its G protein. “In cells—we haven’t gotten to animals yet—they do exactly what we want them to do,” she says. “They’re inhibitors of platelet aggregation.”

Drug companies are still skeptical, but now at least Hamm’s idea doesn’t seem so pie in the sky.

The role of government

The onerously high cost of making new drugs has not escaped the attention of federal health officials.

In a 2004 report entitled “Innovation or Stagnation: Challenge and Opportunity on the Critical Path to New Medical Products,” the U.S. Food and Drug Administration called for increased public-private collaboration to boost drug development through the application of new technologies.

“We must modernize the critical development path that leads from scientific discovery to the patient,” the report urged.

Developing new tools to aid drug discovery also is the goal of the Molecular Libraries Screening Center Network, established in 2004 by the National Institutes of Health as part of its Roadmap initiative to help translate new scientific knowledge into “tangible benefits for people.”

The aim is harness the fruits of the genomic revolution, make them available to scientists in universities and industry alike, and encourage them to work together as never before, explains Christopher Austin, M.D., senior advisor for translational research at the National Human Genome Research Institute.

“What we hope to do… is the high capital investment… take the assay, do the robotic screening on a big library, do some initial chemistry, and give (scientists) back a small molecule compound which allows them to query the function of that gene or pathway—to test a hypothesis,” Austin says.

The federal efforts have their share of skeptics, including Steven Paul, M.D., president of Lilly Research Laboratories. “I am worried that obtaining the kind of molecular probes required for even in vivo testing may prove to be too time-consuming and expensive,” Paul says, “and may divert precious NIH funds away from basic or clinical biomedical research.”

The federal initiatives in no way are meant to diminish government’s role in supporting fundamental discovery, Austin responds. Tools developed by the public sector, however, can help establish the therapeutic potential of new compounds, and encourage industry to push them through the pipeline.

“As long as… we’re all aware of what we can do and can’t do, I think we’ll be fine,” he says.