One entered the gleaming new headquarters of Cogent Neuroscience by climbing a broad, burnished stainless-steel spiral staircase, ascending each step as a scientist might trace the individual units of a gene along the spiraling DNA double helix. Cogent's metaphorical staircase was apt. The company's ambitious aim was to understand the genetics underlying malfunctioning neural machinery and to develop drugs for a stunning array of neurological disorders, including stroke, traumatic brain injury, and Huntington's, Parkinson's, and Alzheimer's diseases.
But today, Cogent's spacious state-of-the-art laboratories are empty, its sophisticated robot manipulators for genetic experiments sit inert. In developments that exemplify both the pitfalls and the continuing promise of applying Duke discoveries, Cogent closed its doors last October when its funding consortium collapsed.
The company's failure is not one of science but of changing priorities in the venture-capital markets in a weak economy. Founded five years ago, Cogent was based on a startling achievement of Duke neurobiologists Lawrence Katz and Donald Lo. They discovered a way to insert genes efficiently into living slices of rat brain tissue and to persuade the genes actually to switch on. The technique suddenly made possible a drug-screening system that the scientists believe is nothing short of amazing, says Lo. "By putting compounds into the brain cells that would interact with the genes you had introduced, you could make the equivalent of hundreds of thousands of transgenic animals per day, because you are looking at intact brain tissue. So, you retain all the normal physiology and, importantly, you could examine the pathophysiology of neurons in specific disease states, but with all the power of molecular biology and biotechnology behind it."
The catch was that when Katz and Lo invented their technique, they realized that a university was not an appropriate place to make use of it. "About a split-second after we understood the importance of the technique, we also realized that there was no academic funding structure or even laboratory structure to exploit it, because so much had to do with scale and production," says Lo. "Thus, it was not compatible with the educational and mentorship mission of the university. You couldn't in clear conscience try to have fifty people in your laboratory working on a project whose intellectual content is relatively low--different than the mentorship you want to provide a graduate student or a postdoctoral fellow."
Katz and Lo enlisted veteran biotech entrepreneur Max Wallace '74 as president and chief executive officer, and together they launched Cogent. Cogent joined an alphabet-soup of other Duke-spawned startups with names like Merix, Norak, StemCo, and Trimeris. In fact, according to Lo, Cogent may soon rise out of its financial ashes, transformed, at least in part, into a nonprofit research institute aimed at continuing its efforts to develop new drugs for neurological disorders.
Among the nation's research universities, Duke is a relative latecomer to the arena of high-tech startups; in fact, it has only recently begun to make a concerted effort to commercialize its discoveries. "One reason we haven't had a lot of spectacular commercial successes is that venture capital is relatively new to our geographical area," says Robert Taber, director of Duke's Office of Science and Technology, which was established in 1993. According to Taber, the local venture-capital scene has changed drastically over the past five years, with funds under management growing from about $50 million to more than $400 million today. "Such venture capital leads to companies," he says, "and you need a home-grown source, since people are not going to come here from California to start a company."Such high-energy, high-promise companies represent a new era of possibilities for Duke research, say administrators. These startups, and the other burgeoning corporate collaborations with faculty scientists and engineers, constitute the only way that the university can ensure that its basic discoveries and knowledge benefit society. In fact, despite the perils facing such startups as Cogent, Duke leaders see this growing entrepreneurial spirit as critical for the continued vitality of the university's research and teaching. Yet they also caution that the boom in partnerships between corporations and Duke academics could present problems that, if not wisely managed, could compromise the quality and integrity of the university's research mission and programs.
" The second reason for our relatively few commercial successes to date is that Duke's research mandate has been largely biomedical, and developing a product from basic biomedical research takes longer," says Taber. "A decade between discovery and product is not unusual, given the lengthy approval process required by the Food and Drug Administration for drugs and medical devices."
However, Duke has become an entrepreneurial player, as evidenced by both licensing income and startup companies. Licensing revenue from Duke patents has climbed from $1.5 million in 1999 to $4.1 million in 2000 to $5.7 million in 2001. Duke discoveries, mainly biomedical, have spawned some twenty-two companies, and eleven others are now in development. Duke holds an equity position in all of them.
Amidst this explosive growth of startups, collaborations, and corporate consulting, Duke's leaders remain acutely aware of the potential for problems. Chief among them is conflict of interest, as evidenced by recent developments at other institutions. In 1999, for example, a University of Pennsylvania trial of gene therapy for a rare metabolic disorder went tragically awry when teenager Jesse Gelisinger died after the experimental treatment. The government ruled that federal research guidelines were not followed. Critics of the study worry that the trial might have been influenced by the fact that the lead investigator was a co-founder and major stockholder in the company that supported his laboratory. Also, Penn held an equity stake in the company.
And in 2002, it was reported that M.D. Anderson Center in Houston failed to inform patients enrolled in trials of an experimental cancer drug that the institution's president held a major financial interest in the product. While there was no evidence that the trials were affected, the center acknowledged that it should have informed patients. It has adopted strict new policies to avoid such problems in the future.
Multiple surveys of clinical trial results over the last few years have found that those sponsored by companies producing the drug being tested tend to show more favorable results than independently funded studies. Reviewing these surveys, Thomas Bodenheimer, a clinical professor at the University of California at San Francisco, said, "The evidence I have...makes a reasonable case that scientific misconduct does take place in clinical drug trials, that conflict of interest is a risk factor for scientific misconduct, and that something must be done about it."
Quoted in The Washington Post, Bodenheimer told an August 2000 National Institutes of Health conference on academic-corporate ties, "Pharmaceutical companies--with a little help from their friends in academia and the NIH--have created products of great benefit to the world. This activity must continue. But to reduce the risk of scientific misconduct, investigators and authors need greater independence from their funders."
Bodenheimer cites as an example a 1999 study in the Journal of the American Medical Association analyzing forty-four articles that explored the economics of cancer drugs. According to the analysis, only 5 percent of the drug-company-funded studies offered conclusions unfavorable to the companies' products; 38 percent of those with no industry funding reached unfavorable conclusions.
And in a 1996 study, Mildred Cho and colleagues at the Stanford Center for Biomedical Ethics found that 98 percent of company-sponsored drug studies published in peer-reviewed journals or in symposium proceedings between 1980 and 1989 reported that new therapies were more effective than standard drugs. In comparison, only 79 percent of studies without industry financing found that the new drug was more effective.
" Can you imagine an election in which someone gets 98 percent of the vote?" Bodenheimer asked in an article he wrote on conflict of interest. "I guess if you pay for the votes, it's possible. It seems likely that some of these trials were designed to favor the sponsor, or the data were analyzed to favor the sponsor, or the trials were written to favor the sponsor, or the trials that didn't favor the sponsor were not published."
Some policy experts would ban any financial involvement of researchers with companies whose drugs they are testing. "I'm not comfortable with scientists' owning substantial equity in small firms that are sponsoring their clinical research," David Blumenthal, director of the Institute for Health Policy at Massachusetts General Hospital, told The New York Times. "I think it creates a conflict of interest, and the conflict is particularly difficult to justify in cases where patients' welfare may be affected."
To Sandy Williams M.D. '74, dean of Duke's medical school, the key to avoiding conflict-of-interest problems is preparation. "My view is that conflicts of interest are inevitable," he says. "Powerful conflicts potentially exist even in a school that has no contact with industry. Our academic reward system creates conflicts of interest even when it has nothing to do with personal profit.
" These conflicts need to be managed, and the first step is to require complete disclosure," he continues. "I believe we do have policies in place that protect us from the dark side of conflict-of-interest problems."
Nevertheless, Williams and his colleagues are all too aware of the professional pitfalls of corporate involvement. "I know of sad circumstances where excellent scientists have wrecked their careers by the failed pursuit of commercial interests that stemmed from their discoveries," he says. "What's needed is knowledge and good counseling, particularly at the early stages of their lives, for faculty members who make discoveries they believe have practical value."
On the other hand, as Lo emphasizes, ties between universities and startups can foster extraordinary scientific synergies that benefit the universities and society. "One reason I retained my faculty position at Duke is because in this new era I think there is a third place to do science," he says. "Historically, there has been either the academic environment or the pharmaceutical industry. But now the biotechnology industry has matured over the past decade, so that the best science in some cases can be done right at the interface of biotechnology and academics.
" In a perfect world, people could do some phases of their work in the private sector because they need the scale and capitalization, and then other phases in academe, where you need the intellectual freedom to establish a scientific basis for what you're studying. And that freedom is really only available in the academic environment."
Williams stresses the practical benefits of such academic-industry partnerships: "Greater flow between industry and Duke will help us have better research infrastructure. Corporate support brings access to technology, either financially or [by providing] a better ability to outsource certain capital-intensive things we want to do."
In fact, Duke administrators say, allowing such flexibility and collaboration may well be the only way universities can attract and hold talented scientists and engineers. According to Vice Provost for Research James Siedow, "Nobody's naÔve enough to say we're not involved in commercial ventures for the money. Of course, we'd like to hit it big with a breast-cancer drug such as Taxol [which used Duke research], or with a Gatorade [developed at the University of Florida]. But more importantly, we have to recognize that younger faculty have been deeply immersed in an environment of entrepreneurship. So, unless we make Duke a very nurturing place with respect to technology transfer, we're going to lose the very kinds of talented faculty that we want to attract and retain."
Siedow says also that learning entrepreneurship is a significant part of an academic's professional development and the university's advance. "My observation from the other schools that are really active with spin-offs is that once one faculty member succeeds, then other faculty begin to realize, 'You know, I can do that, too, and I have ideas that are worthwhile.' And entrepreneurship becomes a self-generating system."
He also emphasizes that--beyond the financial and academic-quality factors spurring corporate collaborations--the university is obligated under federal law to try to commercialize discoveries made by federally funded research. The 1980 Bayh-Dole Act requires universities to make every effort to patent and license discoveries, so that they will be applied for the public good.
Entrepreneurial drive is especially important in the Pratt School of Engineering, says Dean Kristina Johnson. "The kind of faculty members we are looking for are excellent scholars and excellent teachers, and a good majority of them are interested in translating their ideas and products and processes into new companies. So, when we look at faculty members, that interest is certainly a plus."
Johnson's extensive experience with entrepreneurship in her former post at the University of Colorado has convinced her that faculty entrepreneurship--even if the faculty member leaves--redounds to the ultimate benefit of the university. "I have observed that faculty members who have gone out and started companies--and that includes my students and me--when they make it, they're very grateful that they've been treated well. And ultimately, they can be some of our best supporters in the future for encouraging the next generation of students and faculty members." As a result, she says, "I would love to have all our faculty members rich and have stock in startup companies and serve on boards, because I think they bring value to the students and the educational process as well as help us retain them."
What's more, involvement with industry invariably improves a faculty member's understanding of his or her field, Johnson says. She cites her own experience with a startup in the field of electronic image displays. "When I worked with the company, one year I went to a different customer every week for fifty-two weeks--in Japan, in Europe, in the U.S. And as a result, I learned what was state-of-the-art across the whole display industry.
" I did sign nondisclosure agreements, but even if I don't communicate what a specific company is doing, I know what's important and not to waste my time even going down that alley. So, sometimes it's just as important to shut down pathways that aren't important [as to] have to explore them all."
Finally, says Johnson, the energy required to commercialize a discovery, and thus benefit society, is most likely to come from the discoverer. "I've found that the folks who are most wildly enthusiastic about the ideas and most willing to license them are the inventors themselves," she says. "So, allowing an easy way for faculty members, their postdocs, or graduating graduate students to start companies is really important." Johnson favors a university policy that allows faculty members to found companies and serve on their boards and as consultants, yet remain at the university. Such an arrangement enables the university to structure venture-capital and licensing arrangements carefully to benefit both the faculty member and the university, she says.
The ultimate result of such encouraging policies, says Williams, is that an active involvement with industry has become the hallmark of a high-quality research institution. "As I look at programs in molecular biology and biomedical science around the country, some of the people I admire and respect the most as great academicians are also the ones who have substantial interactions with industry," he says. "So, I believe that a more proactive stance on Duke's part to foster entrepreneurship will help us recruit and retain the best faculty."
Provost Peter Lange emphasizes, however, that the university does not recruit faculty members based on the economic potential of their research, but rather on the importance of those ideas to the advancement of knowledge, the university's teaching mission, and society. "We should not be in the business of hiring faculty because we think they have ideas that will be profitable," he says. "We need to hire faculty because they have ideas that are at the cutting edge of their field and that really will advance knowledge. But at the same time, when those researchers develop ideas that could allow us to improve society and/or profit from those ideas, we do need to take advantage of those opportunities."
The university has launched several initiatives to ensure that faculty members like Lo, who are interested in applying their discoveries, are given the greatest encouragement and aid. Williams and his colleagues have proposed a for-profit venture, Translational Medicine Inc. (TMI), as a way to give faculty access to corporate expertise in technology transfer, technology development funding, and consulting opportunities. "Our plan is to move this initiative outside the university's walls, both to attract investors and to create a user-friendly gateway to the university," says Williams.
" There are two schools of thought in fostering relationships with industry," he says. "One is that the role of the central administration is to be a passive conduit, simply to regulate it, to keep faculty from doing too much, to manage conflicts of interests and consulting time, and so forth. But I think the new era will give competitive advantage to those institutions that turn the gateway to and from industry into a point of real creative thought and development, and become active in fostering the right kinds of relationships between industry and academia." TMI's structure, Williams says, will actually help the university avoid ethical problems. "By having this gateway outside the university, we can better manage conflict-of-interest issues and foster the relationships we want with industry, because contractual agreements and confidentiality issues will be easier."
Similarly, the Pratt School has launched the Duke Engineering Venture Investment Limited (DEVIL) Fund. This fund raises support from venture capitalists, enabling the engineering school to invest in promising startups and to use the revenue to support new faculty and programs. Johnson and her colleagues are also seeking outside support for a venture-fund capability--dubbed Duke Engineering Research, Inc.--to fund faculty members and graduate students in the early stages of developing an idea, which the students can transform into a startup company once they graduate.
In fact, Johnson sees little problem with carefully monitored corporate research going on within the walls of the university. "One advantage the university has that startup companies don't is facilities and resources," she says. "Certainly we don't want to see faculty members who are thinking about starting a company being approached by venture capitalists [VCs] who tell them, 'If you could just do a proof-of-principle experiment in your lab, we could fund your company.' But I don't have a problem with the work going on in the lab if the VC puts up money to pay for the lab space, time, and other resources."
The emphasis in all these entrepreneurial efforts, says Lange, is outside support. While some universities--including the University of North Carolina at Chapel Hill and North Carolina State--have launched venture-capital funds with their own money, Duke will not be among them. "It's not the university's resources that we want to commit," he says. "We bring the ideas and the ability to develop them. We also market our ideas to those in the business of supporting ideas."
Those universities that forego investing their own money often cite the infamous experience of Boston University, when in 1987 it invested $120 million, much of it from the university's endowment, in the biotechnology company Seragen. Amid conflict-of-interest allegations and complaints of incompetence, Seragen's stock plummeted, and the university recovered only a fraction of its investment.
Besides conflict of interest, says Lange, Duke also guards carefully against "conflict of commitment," in which commercial involvement compromises a faculty member's commitment to academic research and teaching. "The entrepreneurial spirit is a double-edged sword," he says. "To a great extent, it serves our broader mission, but push it too far, and it can compromise that mission."
Contrary to popular belief, he says, the growing interest among faculty members in commercializing their discoveries does not constitute a basic change in their operational style. "We have always had a very entrepreneurial faculty in the sense of starting new programs, going out and getting grants, and being interdisciplinary," he says. "I think what's changed now is that there's an added interest in taking advantage of intellectual property and new ideas we develop on campus and converting them into products that improve society and that have the potential for financial return to the researchers and to the university itself."
Neurobiologist Lo discovered that the difference between being an academic entrepreneur and a commercial one is not all that great. "When we first started Cogent, I had the assumption that academics were unworldly, nerdy people who were shut off in an ivory tower, and that the world of business was quite a different world. What I didn't realize was that, basically, all research faculty are entrepreneurs. All of us are CEOs and chief scientific officers when we start our own labs with university seed funding. From that funding, we have to get some tangible results that allow us to get our first real round of funding, typically from the National Institutes of Health. And then you have to quickly build your little entrepreneurial startup in the academic environment to get enough traction to establish the proof of principles that you need to get that next round of funding.
" So, in a sense, what I find myself doing here in the corporate context is everything that I used to do as an academic at Duke. Everything has a slightly different name, so the language I use is different, but it's more of a difference in dialect than in language."
In developing its policies, Duke will take a page from the books of other major research universities, says Lange. "Even in the case of universities with the most entrepreneurial cultures, such as Stanford and the Massachusetts Institute of Technology, nobody has questioned their commitment to basic research," he says, "because they have been vigilant about conflict of commitment and not letting what might be called a 'lucre philosophy' take over."
Another worry of academics is that corporate relationships might compromise the free flow of scientific information traditional in academe. The concern is not unwarranted, say those who have investigated the effects of corporate relationships. Mass General's David Blumenthal and his colleagues did a 1996 survey of life-science-company executives that led him to conclude that corporate relationships "may pose greater threats to the openness of scientific communication than universities generally acknowledge." More than 80 percent of executives said they had required researchers to keep some information secret, pending a patent application.
In another survey of academic researchers in the biological sciences, Blumenthal and his colleagues found that nearly a tenth reported withholding research results from other university scientists. A fifth of the respondents reported delaying the publication of results for more than six months, to maintain a lead over competitors or to give them time to obtain a patent. Especially significant, he discovered, is that professors who engaged in industry-supported research with commercial applications were more than three times as likely to delay publication as those who had no industry support.
Most recently, a Duke Medical Center study found that research agreements between industry and academic medical centers frequently do not adhere to international guidelines to protect the integrity of research and the rights and interests of academic investigators. The study, published in the New England Journal of Medicine, consisted of a survey of the contracts with clinical trial sponsors of 108 academic medical centers.
Duke leaders emphasize that they have adopted policies aimed at protecting academic freedom from compromise by corporate agreements. According to Vice Provost Siedow, the university accepts no corporate agreements that restrict scientific publication. The university does allow corporate partners to review scientific papers before publication, but only to enable them to determine whether any of the corporation's proprietary information is about to be published. And, like other universities, Duke limits faculty consulting during the academic year to the equivalent of one day a week.
Given Duke's rapid rise in corporate relationships, administrators do express concern over the gaps that remain in the university's technology transfer policies. President Nannerl O. Keohane has commissioned a task force of senior administrators to recommend ways to strengthen, and, in some cases, establish policies on technology transfer. For example, says Siedow, the university needs to develop an explicit policy governing whether faculty members can serve on the boards of or hold equity positions in their startup companies. Also, technology transfer policy must be fair to all faculty members.
" We need to set up a culture that supports faculty who want to maintain a more purely academic approach and may not want to think about commercial implications of their work," Siedow says. "By the same token, we don't want to penalize someone who is interested in the commercial implications."
Despite a university's best efforts, developing its own technology transfer policy is a difficult, onerous job, given that no standards exist, he says. "There's just no silver bullet. In fact, as we survey other universities, we see everything from the policy equivalent of muzzle loaders to automatic weapons. So, basically, Duke will need to generate its own policy, based on its own culture. We certainly recognize that we are now changing our culture. But we will not stray from what makes Duke the kind of university it is."
Besides carefully formulating its policy, Duke faculty members and administrators are turning to medical ethicists such as Jeremy Sugarman '82, M.D. '86, who also worries that conflicts of interest and commitment inherent in an entrepreneurial culture will affect academic medicine. As director of the Center for the Study of Medical Ethics and Humanities, Sugarical school in exploring such questions.
The center--and its collaborating physicians and scholars--has plenty of work ahead, says Sugarman. "We don't currently understand the implications of concentrating more on areas that are thought to bring profit in its commercial sense, rather than profit in its intellectual sense. And, we need to consider how we assess what scientific questions get asked within the clinical or academic community. How will we ensure that we continue to address important questions, whatever they are, regardless of their profitability and their ability for translation to commercial products?"
" With respect to education, one concern is whether we are diverting students to projects for which the investigator-investor has a primary concern," Sugarman continues. "How do we provide appropriate protections for our students, including those who are working closely with a faculty mentor? With respect to clinical care, we need to raise hard questions about the appropriateness of enrolling patients seeking care here in clinical trials related to a project in which we are investing.
" It's always exciting to talk about the prospects without talking about the gritty mechanisms we use to deal with the issues involved. We just have to handle them explicitly and use our ingenuity to deliberate about the best ways of coming up with necessary protections."
Fortunately, Sugarman says, he is finding a new climate of ethical consciousness among faculty members. "I've been receiving more calls from faculty about a variety of ethical issues, especially in research ethics. They want to know how to design a particular research protocol so that it is ethically appropriate. And to me, this 'preventive ethics' approach is much more appropriate than the Monday-morning quarterbacking that some of my colleagues have found themselves having to do."
Even with the thorny issues of ethics and policy, medical dean Williams emphasizes that the benefits of entrepreneurship and commercial collaboration far outweigh the problems. "Twenty years ago when I was a young scientist, I never even thought about patenting my discoveries," Williams says. "It just wasn't on my radar screen. And I think there were some sad, wasted opportunities in that era, where many people, if they had been clued into this world, would have gained great value for themselves, for the institution, for society."
"The older school of thought was that the academy should keep itself free of encumbering contacts with industry, because that was a distraction from our core missions of education and fundamental research," says Williams. "No one values the core mission of the academy more than I do, but I believe that a practical, realistic view today is that we not only should, but we must have many forms of contact with industry."
"It is, I think, fundamental to our core mission that we not only discover new principles that underlie human health and disease," he says, "but we also work to foster the movement of those new ideas into clinical practice. And, by and large, that doesn't happen without innovation moving into industry."