Breaking The Mold

SIX YEARS AGO, while heading a neuroscience research program at Merck & Co., P. Jeffrey Conn became concerned about the state of the drug discovery enterprise. Fewer drugs were coming out of the pipeline, and the pharmaceutical industry increasingly seemed focused on the bottom line rather than on innovative science.

He began to form a plan for what a well-oiled drug discovery operation could look like. He imagined an organization that would closely integrate basic science and applied research while taking on the high-risk ideas that he thought were too often overlooked by big drug companies.

Conn is now trying to bring that vision to life as the director of Vanderbilt University's Program in Drug Discovery, a group of 60-odd scientists who are shaking up conventional ideas of what a research institution can—or should—do.

The Vanderbilt program is a forerunner of what seems to be a growing trend for research universities to push further into applied science. For example, Emory University and Temple University both have nascent centers for drug discovery, and others are forming closer ties with industry to beef up their capabilities (C&EN, Nov. 10, 2008, page 13).

In the past, promising pharmaceutical molecules discovered by academic researchers would usually be licensed to industry at an early stage or spun off into a company to attract outside funding that could advance their development. Now, Vanderbilt has created the infrastructure to conduct the gamut of studies needed to bring a drug compound into human clinical trials.

At first glance, the Vanderbilt operation looks a lot like a biotech company operating within the walls of a university. Although students and postdocs populate the labs, they work side by side with ex-pharmaceutical industry scientists who have extensive drug discovery experience. The units one would expect to find in a company—high-throughput screening, medicinal chemistry, and pharmacology—are there in the university.

Furthermore, the lab has already secured a major partnership with Johnson & Johnson to develop schizophrenia drugs. With its up-front fee, milestone payments, and royalties, the deal looks awfully similar to the kind of agreement a biotech firm would seek.

But the Vanderbilt program has important differences that Conn believes will improve its chances of finding good drug candidates yet also maintain the university's mission to educate and advance scientific knowledge. Basic research is not abandoned in favor of an applied approach but rather is seen as a critical component of the drug discovery process. And publishing is just as important to this quasi-drug company as it is to any other academic lab.

A mild-mannered, quintessential southern gentleman, Conn does not seem a likely candidate to shake up an industry. But even in his soft-spoken tone, it's clear Conn feels strongly about one thing: The old ways of finding drugs at big pharma aren't working.

While working in industry, Conn became worried that the only projects supported are those with a direct impact on the bottom line. "Even at a place like Merck, where we had a major commitment to discovery, the ability to do the kind of basic science that informs the discovery—and is critical for choosing the best and most innovative targets—was becoming more difficult," he says.

Yet as Conn saw it, the other participants in the discovery process—academia and biotech firms—also had weaknesses. In his role at Merck, Conn collaborated with academic labs, where he saw interesting ideas about new drug targets. "But they were always half-baked," he says. Academic institutions lacked the infrastructure and experience, not to mention the cultural mind-set, to conduct the next set of studies that could validate their ideas.

BIOTECH COMPANIES, meanwhile, could explore interesting ideas that big pharma often ignored. But they were constrained by the quick turnaround expected by investors, who often forced companies to cut corners or abandon candidates that had promise but needed more work.

"In that whole thinking process, I came to the conclusion that the institutions that were most well equipped to fill this gap were academic institutions," Conn says. "You have this large, very developed research infrastructure that you could never find in a biotech setting."

Yet academia lacked the culture to support team science and the infrastructure to enable early-stage drug development—namely, high-throughput screening, medicinal chemistry, and pharmacology.

If a university would be willing to think outside of the box, Conn thought, he could add those missing pieces. The resulting organization would have the freedom to mix basic science and drug research in a way that big pharma seemed incapable of doing. And the program would be more stable than a biotech firm because instead of seeking venture capital, it would rely on the National Institutes of Health and other sources for funding that do not demand quick returns on investment.

Vanderbilt, where Conn received his Ph.D., seemed like a good candidate for such a venture. The university had just established the Vanderbilt Institute of Chemical Biology, an interdisciplinary initiative that brings together resources from the chemistry department and the medical school. It seemed primed to push the boundaries of drug discovery within academia.

The first year was tough. Conn's first few grant applications were rejected, and he began to wonder whether he had made the right decision. After all, he had turned down two solid industry job offers to embark on this experiment. "I felt like I'd stepped off the edge of a cliff, and I wouldn't be honest if I didn't say I would lie awake some nights and wonder if this was a wise move," he says. "I went from a big department and a big budget to myself alone in a lab, and I hadn't held a pipette in many years."

Part of the challenge was that at the time, NIH had no mechanism to fund straight drug discovery research. Conn's response was to instead apply for basic science grants that could support his drug discovery efforts.

The Vanderbilt Institute of Chemical Biology also played a critical role by helping to get some of the first high-throughput screening efforts going. Conn added staff incrementally. He brought in industry scientists with a range of drug discovery experience. "I do think there's a craft and a science in drug discovery that you see in pharma that you traditionally don't see in academia," he says.

Dave Weaver was Conn's first hire. A former Bristol-Myers Squibb scientist, he is now head of Vanderbilt's high-throughput screening center. Not long afterward, Conn hired Colleen Niswender, a molecular pharmacologist, who works with Weaver to develop assays and direct lead optimization efforts. Carrie Jones, a behavioral pharmacologist in Eli Lilly & Co.'s central nervous system division, came aboard to develop better animal models for complex psychological disorders. Craig Lindsley, a medicinal chemist who had worked with Conn at Merck, joined the team in 2006.

It is exactly the kind of discovery team industry would put together, Weaver says. There are differences, though, both practical and philosophical.

For example, the ex-pharma folks who signed on in the early days walked into an empty lab with not much more than the standard start-up package for a research professor. But they've built up a steady revenue stream from grants and other pacts and have been supported by the university when they needed key equipment. Lindsley notes that students working in his lab are now exposed to the same instruments they'd see in industry.

More fundamentally different is the focus on basic science, which both advances the goals of the university and enables scientists to do better informed drug discovery. According to Weaver, he and his colleagues want to "make sure this doesn't become a basic analog of industry in a different building."

Indeed, the drug industry refugees working at Vanderbilt quickly learned that within a university there are many levels of success short of a drug candidate. "You immediately feel the impact of a different reward system," Conn says. "Here, a publication really means something, and it really is important for the institution and for the group. Preliminary data that help us get a grant are critical."

Dennis C. Liotta, a chemistry professor at Emory who has been instrumental in starting up a drug discovery center there, agrees. "In pharma, if you're given a year to do a project and you're not successful, that's a failure," he says. "If we do the same thing and we publish a paper in Science, that's a success. Our markers are different."

And because the reward system is different, scientists have freedom to explore the risky ideas that big pharma often ignores. While at Merck in the early 2000s, Conn proposed developing allosteric modulators of G-protein-coupled glutamate receptors, thinking that molecules acting on a site different from the primary ligand-bonding site would be safer and more effective. Preliminary work was done at Merck, but it was an unusual concept and the company soon abandoned it. As Lindsley explains, no one else was working on the target, a situation that tends to make big drug companies nervous.

At Vanderbilt, Conn was able to return to the concept, which has since led to the candidates licensed by J&J. Programs such as Vanderbilt's actually open opportunities for big pharma, he argues. His group published more than 50 papers during the past six years that cumulatively demonstrate the allosteric modulator approach could in fact work, effectively lowering the risk for big pharma. "I think it's fair to say that most major pharma companies now have programs in allosteric GPCR modulators," he says.

THE SUCCESS with allosteric modulators is not the only sign the new model is flourishing. The Program in Drug Discovery has gone from having zero outside funding when Conn started writing grants in 2003 to $12 million in 2008. This year, the projection is for $16 million in outside support. The program now boasts some 60 scientists—including former researchers from GlaxoSmithKline, Pfizer, Roche, Sanofi-Aventis, and Merck—nearly half of whom are in the medicinal chemistry group. Lab space has grown from about 2,000 sq ft to roughly 15,000 sq ft.

In addition to the deal with J&J, the program has a partnership with Seaside Therapeutics to develop compounds that suppress fragile X syndrome, an inherited disease of the brain. The program also has received several grants from the Michael J. Fox Foundation to support research in Parkinson's disease.

Vanderbilt isn't the only institution trying its hand at drug discovery. Under the guidance of Liotta, who invented several HIV drugs, Emory has set up the Institute for Drug Discovery. Its goal is to discover a compound, take it through early-stage development, and then on to an approved Investigational New Drug Application.

So far, Liotta has hired 14 ex-pharma people, including medicinal chemists, molecular modeling experts, animal pharmacologists, and pharmacology-metabolism scientists. The center works closely with the university's high-throughput-screening facility to search for leads in the facility's library of 300,000 compounds. Starting next month, the center will have two fully equipped vivaria—one rodent and one nonhuman primate—in the Yerkes National Primate Research Center at Emory.

Temple University is also pursuing drug discovery. Magid Abou-Gharbia, who spent 26 years at Wyeth, where he was senior vice president of chemical and screening sciences, joined Temple last year to direct its Center for Drug Discovery Research. Abou-Gharbia has hired one former Merck chemist with expertise in parallel synthesis and expects to hire three more scientists from industry or academia. The plan at Temple is to make small-molecule probes to explore basic biological questions while also potentially serving as therapeutics. The center has already established a collaboration in neuroscience, Abou-Gharbia says.

Universities' adoption of this new model also coincides with a critical attitude shift in industry. When Conn left Merck six years ago, big pharma wasn't interested in licensing early-stage drug candidates from the outside. "Now, there's a tremendous demand," he says. Whereas biotech firms typically try to develop a drug candidate as far as they can before partnering and are a source of clinical-stage compounds, universities could be the source of compounds that are still at the preclinical stage.

Whether industry will bite is another question. "Ten years from now, when we look at the whole drug discovery process, it'll be so different that it won't be recognizable," Conn says. "It is left to be seen whether academics will play a role in that."