Limited Access To cDNA Database Has Drug Manufacturer Up In Arms

Date : December 12, 1994 Incensed that the terms exclude industrial scientists, Merck announces sponsorship of competing project A powerful new, privately held database of human complementary DNA (cDNA) sequences has become available to researchers eager to test its capabilities to rapidly identify new genes. The proposed terms of access to the database, however, exclude pharmaceutical and other industry-affiliated scientists, angering some companies, including drug giant Merck and Co. Inc. The conditions have prompted the Whitehouse Station, N.J.-based firm to sponsor another, largely duplicative project that will deposit data into a publicly accessible database. Other terms of access have raised concerns among potential noncommercial users of the database developed by J. Craig Venter, president of The Institute of Genomic Research (TIGR), Gaithersburg, Md., a nonprofit research institute. These include the right to preview papers on resulting discoveries and to retain control of patents. The database, two years in development, currently consists of about 150,000 expressed sequence tags, or ESTs, which are partial sequences of human cDNA obtained from various tissues. Because many genes are expressed in multiple tissues, the data represent about 35,000 unique human genes, more than a third of the estimated 100,000 human genes. TIGR has a contract worth about $100 million with biotech firm Human Genome Sciences Inc. (HGS), Rockville, Md., for patent options on its discoveries; in turn, Philadelphia-based SmithKline Beecham has agreed to pay HGS about $125 million for development rights to the discoveries. Merck claims its objection to the terms is based purely in science. "We see human genes [sequences] as being essential research tools for all of drug research in the future," says Alan R. Williamson, vice president for research strategy worldwide at Merck. "And we've taken the position that research tools should be shared amongst everyone who wants to make an impact on drug research or any therapeutic application to benefit mankind. They should not be proprietary. We could have taken a private position on this [new data], too, but we decided to take a public position." The competing sequencing project being backed by Merck will be conducted at Washington University School of Medicine in St. Louis. The aim of the effort, expected to take roughly 18 months, is to sequence about 200,000 cDNA segments, with the results deposited immediately into GenBank, a publicly accessible database of sequences managed by the National Center for Biotechnology Information at the National Library of Medicine in Bethesda, Md. Venter predicts the protests will lessen over time. He notes that the Howard Hughes Medical Institute has agreed to the terms, as have a number of universities. "As people look at it closely, the issues go away, and the initial emotional responses from the publicity that's been generated by Merck and others totally vanishes, because all the data's available to academic scientists anywhere in the world," Venter says. "Over half the data has no strings attached to it at all, except that somebody be at a not-for-profit institution and that they cite the database when they publish their data." University technology-transfer officers say that the terms for use of the database by academic scientists are comparable to those of companies providing proprietary re-search materials. Genetics investigators with no plans to commercialize their research can use the database without restriction, and those who have used it praise its power and do not anticipate any limitations on their work. One point of contention underlying the picture of dueling databases is a sharp difference of opinion as to the intellectual-property value of the sequences that constitute Venter's database. The National Institutes of Health attempted to patent, and thereby control all future uses of, Venter's ESTs several years ago, when he was an intramural researcher at the agency. The move was widely criticized at the time because no basic biological knowledge of the genes' functions accompanied the partial sequences. Some critics charged that such patents, if allowed, would have foreclosed future research. Venter, however, claimed that placing the partial sequences into the public domain would have rendered full genes developed from them unpatentable, removing the incentive to work on them (R. Eisner, The Scientist, Dec. 9, 1991, page 1). After an initial rejection of the patent filings, NIH chose not to pursue appeals. "It's only the added biology that makes the DNA sequences of any value at all as far as intellectual property goes," Williamson contends. Many scientists expected that Venter's database, when completed, would be made publicly available because he developed the EST technology in 1991 while he was a government researcher or because TIGR is a nonprofit organization. Venter acknowledges that because of the contractual ties established among TIGR, HGS, and Smith-Kline Beecham, the database is being made available with restrictions to protect and capitalize on its value as intellectual property. "I think that everyone assumed that you only needed to do this once, that it was being done by HGS and TIGR--and particularly by TIGR, which was a non-profit-making institution--and that it would be made [publicly] available," says Williamson. "The academic community backed off from generating this sort of data, because they were under the impression that this would all be made public. Once we realized that wasn't going to be the situation, then the need for this sort of database became apparent to us, and we decided to commission the production of it." According to Williamson, "many academics have said they wanted to do this sort of thing, but they couldn't get funded to do it because they were told it's being done." Officials at the National Center for Human Genome Research (NCHGR) at NIH confirm that, because of statements made by Venter early in his development of the database at TIGR, they expected the data to be publicly available and did not finance what they thought would be redundant work. "There was the anticipation on our part and the part of people who are advising us that that information would come into the public domain," recounts Mark Guyer, assistant director for program coordination at NCHGR. "And I think Craig would argue that he's putting it into the public domain now and is fulfilling those statements. So, the idea from our advisers was: This is being done, the information is going to get out, and therefore there's no point in the federal project duplicating it." Widespread expectations that the database would be made public, Guyer adds, probably were "not entirely consistent with what a private organization has to do to protect its own interests. That is, in part, where the current disagreements are based." "People can accept a certain amount of trade-secret mentality from a traditional company," says Joyce Brinton, director of the office for technology and trademark licensing at Harvard University, "But this company isn't a traditional company, so maybe people aren't as willing to accept the idea that they would act like a company." In defending his institute's contractual relationship with HGS, Venter draws a parallel between his situation and that of some of the scientists who now hope to access his database. "What I needed to do to get the funding was: I signed away rights to commercialize discoveries made by myself and my colleagues at the institute," Venter observes. "Now, because of drops in federal funding, more and more scientists are having to do that on a smaller scale at their universities. To get funding from a pharmaceutical company, they sign away all their rights to commercialize [discoveries]." A question confronting university technology-transfer officers and academic scientists is whether the information in the database is worth what TIGR and HGS are asking: a piece of any patents that may result. The issue becomes one of assessing relative intellectual contributions and recalls the debate over whether ESTs are tools or patentable discoveries. "It comes down to this question of how much should a provider of something get in return for it," says Brinton. "Do we--and I don't know the answer--have a distorted picture of the relative values of what our scientists would be doing vs. the [value of the] information that we got from HGS in the first place?" The agreement that HGS is asking institutions and their investigators to sign to use the database is typical, Brinton says, of so-called material-transfer agreements involving the exchange of proprietary biological reagents between companies and academic re-searchers. "But usually information has been treated in a different way than [materials], and there is a general view among people in the academic community that information should be shared," she points out. On balance, Brinton expects that arrangements will be possible allowing Harvard researchers access to Venter's cDNA sequences. The information, which is the main product of TIGR and HGS, does have value, she notes, and "a company does have obligations to its investors. It can't give away its seed corn." While the monetary worth of the database is still to be determined, the experiences of several scientists who have had the opportunity to search for genes using it suggest that its research value may be significant. "There was a yeast gene that we knew the amino acid sequence of, that was involved in a biochemical pathway of purely aca-demic interest," recalls Phillip A. Sharp, chairman of the biology department at Massachusetts Institute of Technology and a 1993 Nobel Prize winner in physiology or medicine. "And it was very laborious to think about getting the human counterpart. But by taking the amino acid sequence and looking directly in the cDNA database, we were able to find a human cDNA that encoded a related amino acid sequence, and, therefore, the equivalent protein, which we could not get by nucleic acid homology. We then just got the cDNA and cloned out the gene. So, it's going to be useful, there's no question." "The database provides a very efficient way to identify those genes for which you might be looking because you already know the sequence of some related gene or some gene with a similar function," says Bert Vogelstein, a professor of oncology at Johns Hopkins University. "Sometimes the classical approach to identifying a gene can be very difficult. It can take a long time, sometimes months, sometimes even longer, or sometimes it may not work at all. But if the gene has already been cloned and is in the database, it's a matter of, literally, a few seconds to find that gene." Vogelstein adds that, for his purposes, there were no problems with terms of access to the database. TIGR and HGS officials are confident that, as researchers become more familiar with the database and the terms of access, they will come to see the conditions as fair ones. William Haseltine, chairman and chief executive officer of HGS, notes that for most scientists who use the database for research purposes only, there is no cost. They are, however, obliged to show their scientific papers to HGS at least 30 days before publication for an assessment of patentability. Only when a discovery is deemed to be patentable and to have commercial potential does HGS seek to exercise its patent option, according to Haseltine, meaning that the company will be the sole developer of commercial products based on the discovery. The patent itself, Haseltine notes, is the property of the university, and the company and the school must still negotiate specific terms for development. If they cannot agree within six months, nor within a three-month extension period, the rights remain with the university. "If [a discovery] has commercial value, we've given them something of value, and we'd like something back," Haseltine says. "And that's reasonable. You don't give people the fruits of your labor for free. You just don't do it." Officers in TIGR and HGS note that access to the database is essentially unencumbered for most basic researchers. For those involved in mapping projects, however, rather than hunts for single genes, this may not be the case. Mapping requires the use of many sequences, and, concerned about losing control of their data, TIGR and HGS have placed restrictions on how maps derived from the human cDNA database may present information. "The people who want to construct a map have said that they want all of the data that goes into that map freely available," Haseltine explains. "Our position is we're willing to make the data available in bulk for the mapping community, provided that the points on the map get referred to the database, and the sequences go back to the database." Haseltine points out that a sequence is more than a point on a map. It is also an enabling technology, particularly for genetic medicine. Richard K. Wilson, a research associate professor at Washington University who will be directing the Merck-backed cDNA-sequencing project along with genetics department chairman Robert Waterston, cites as one of the strengths of his project the fact that the entire body of his results will be available to mappers. For these scientists, Wilson says, "our effort is not redundant, because we're going to provide a lot of sequences very quickly. These [sequences] are just going straight to GenBank from our labs, and there is absolutely no restriction on anything in GenBank."

Limited Access To cDNA Database Has Drug Manufacturer Up In Arms

Interested in reading more?

Become a Member of