Sponsored by the U.S. Department of Energy Human Genome Program
Human Genome News Archive Edition
Human Genome News, July-September 1996; 8:(1)
Earlier this year, HGN 7(5) carried an article describing the collaboration between the pharmaceutical giant Merck & Co. and the Genome Sequence Center at Washington University (St. Louis) to develop a publicly available data set of partial cDNA sequences (called expressed sequence tags, or ESTs).
Because cDNAs represent genome coding areas (i.e., genes), databases containing these sequences offer researchers a way to speed-read through the genome in the hunt for disease genes, bypassing billions of base pairs of noncoding genomic DNA. In the following guest article, intellectual-property lawyer Rebecca Eisenberg (University of Michigan Law School, email@example.com), a DOE-funded researcher, analyzes the different cDNA database-usage strategies undertaken by Merck and two major genome-sequencing companies.
Intellectual-property issues have been unusually conspicuous in the recent history of genomic advances, even by the standards of the patent-weary genetics and molecular biology communities. Controversy has been particularly acute over intellectual-property rights in the results of large-scale cDNA sequencing.
Beginning in 1991 with NIH's filing of patent applications on the first batch of ESTs from Craig Venter's laboratory, each new development has been met with lively speculation about its strategic significance from an intellectual-property perspective. Are cDNA fragments of unknown function patentable, or must they undergo further research or characterization before they satisfy patent-law standards? Will patents on such fragments promote commercial investment in product development or interfere with scientific communication and collaboration and retard the overall research effort?
In the absence of patent rights, how might the owners of private cDNA sequence databases earn a return on their investment while still permitting other investigators to obtain information access on reasonable terms? What are the rights of those who contribute such resources as the cDNA libraries that are used to create the databases and of those who formulate appropriate queries to identify interesting sequences from the morass of information? Will the disclosure of ESTs in the public domain preclude patenting of subsequently characterized full-length genes and gene products? And why would a commercial firm invest its own resources in generating an EST database for the public domain?
Two factors have contributed to the fascination with intellectual-property issues in this setting. First is a perception that some pioneers in large-scale cDNA sequencing have sought to claim intellectual-property rights that reach far beyond their own actual achievements to cover the future discoveries of others. For example, the controversial NIH patent applications claimed not only the ESTs for specified sequences but also the corresponding full-length cDNAs and smaller portions that might not even include the disclosed ESTs. More recently, private owners of cDNA sequence databases have conditioned data access on advance agreements offering either a license or right of first refusal to any resulting intellectual property. These actions raise questions about the fairness and efficiency of the system to protect intellectual property. Such concerns are particularly compelling to research scientists, who have more than commercial interests at stake.
Second is the surprising alignment of interests in the data. NIH, a public institution, initially took an aggressive position in favor of patenting discoveries that some industry representatives thought were unpatentable and should remain unpatented. Merck & Co. ultimately took on the quasigovernmental function of sponsoring a university-based effort to place comparable information in the public domain. These topsy-turvy positions raise intriguing questions about the proper roles of government and industry in genomics research and about who stands to benefit and to lose from the private appropriation of genomic information.
Promoting R&D Through Exclusive Rights
Research scientists who work in public institutions often are troubled by the concept of intellectual property because their norms tell them that science will advance more rapidly if researchers enjoy free access to knowledge. By contrast, the law of intellectual property rests on an assumption that, without exclusive rights, no one will be willing to invest in research and development (R&D).
Patenting provides a strategy for protecting inventions without secrecy. A patent grants the right to exclude others from making, using, and selling the invention for a limited term, 20 years from application filing date in most of the world. To get a patent, an inventor must disclose the invention fully so as to enable others to make and use it. Within the realm of industrial research, the patent system promotes more disclosure than would occur if secrecy were the only means of excluding competitors. This is less clear in the case of public-sector research, which typically is published with or without patent protection.
The argument for patenting public-sector inventions is a variation on the standard justification for patents in commercial settings. The argument is that postinvention development costs typically far exceed preinvention research outlays, and firms are unwilling to make this substantial investment without protection from competition. Patents thus facilitate transfer of technology to the private sector by providing exclusive rights to preserve the profit incentives of innovating firms.
Nonpatent Strategies for Commercial Exploitation
No patents have been issued so far on cDNA fragments of unknown function, although a number of private firms have pending patent - applications that claim thousands of such fragments. Meanwhile, three firms Human Genome Sciences (HGS), Incyte Pharmaceuticals, and Merck are pursuing different nonpatent strategies for exploiting the value of these sequences as unpatented information. These strategies are exclusive licensing, nonexclusive licensing, and dedication to the public domain, and it is still too early to tell how each will pay off. We can see, however, how different firms are placing their bets, and we also have some idea of the sizes of those bets.
HGS and Incyte are exploiting their databases commercially by controlling access to them, in effect using contracts and trade secrecy to protect their intellectual property. The viability of these strategies may be limited by Merck's sponsorship of a competing cDNA sequencing effort at Washington University dedicated to the public domain. The commercial value of the private databases is likely to decline as public-domain information increases. Although public-domain databases are growing rapidly, the private ones remain larger at this point and claim to offer superior products. These products include longer sequences of contiguous cDNA fragments; more complete sequence annotations, including information about expression in different types of tissues; high-powered bioinformatics capabilities; and user-friendly software.
A significant limitation on the value of public-domain databases is the pending patent applications of private database owners. If these applications ripen into issued patents, they could preempt the use of any covered sequences, even if those sequences were disclosed publicly before the patent was issued, as long as the patent applicants are able to establish their priority.
U.S. patent applications are confidential until a patent is issued, so determining which sequences are the subject of patent applications is impossible. Those who use sequences from public databases today risk facing a future injunction if those sequences turn out to be patented by HGS or Incyte on the basis of previously filed patent applications. The same uncertainty applies to sequences obtained from private databases; for example, sequences that are obtained from the Incyte database may turn out to be covered by a previously filed HGS patent. Because the Merck initiative got off to a late start, its sequences are more likely to be covered by other firms' prior patent applications.
Exclusive Licensing. For $125 million over a 3-year period plus royalties on product sales, HGS has licensed exclusive rights to access its database to SmithKline Beecham (SB). SB also gained the right of first refusal to develop and market protein therapeutic and diagnostic products from information in the database. HGS has entered into separate collaborative agreements with other research partners for gene-therapy and other DNA-based product development.
During the period of SB's exclusive license, investigators in academic and nonprofit institutions may obtain access to some of the same sequence information through a separate database maintained by The Institute for Genomic Research (TIGR) under the terms of a Database Access Agreement. The TIGR database includes sequences that are similar to previously published sequences and accessible to nonprofit investigators with minimal restrictions on use. It also includes proprietary sequences that are accessible only to those who sign more restrictive agreements giving HGS rights to prepublication review and an option to negotiate a license to any resulting inventions. Some academic investigators also have obtained access to sequences in the separate HGS proprietary database by signing a Materials Transfer Agreement granting HGS "a sole and exclusive worldwide right and license" to develop any resulting products on terms to be negotiated in the future.
An obvious advantage of this exclusive licensing strategy for HGS is that it has generated a lot of revenue; SB placed what appeared to be a very large bet 3 years ago. An obvious concern is that restricting database access to such a degree may limit the value that can be extracted during the term of the license. Perhaps this concern motivated SB and HGS to enter into collaborative agreements announced this past summer to share the database with four additional pharmaceutical firms [Takeda Chemical Industries, Merck KGaA (not related to Merck & Co.), Schering Plough, and Synthelabo SA]. With the signing of these agreements, SB appears to have made its money back even before bringing any new products to market the agreements call for payments totaling $140 million plus royalties on product sales.
Nonexclusive Licensing. Incyte has offered nonexclusive licenses to as many firms as will take them, at a much cheaper price than SB paid for its exclusive deal with HGS. So far ten pharmaceutical firms have signed on as subscribers, including Pfizer, Pharmacia & Upjohn, Novo Nordisk, Hoechst, Abbott Laboratories, Johnson & Johnson, BASF AG, Hoffmann-La Roche, Zeneca, and Schering AG Berlin. Financial terms for most of these agreements have not been disclosed, but press accounts report that they total more than $160 million, excluding contingent payments such as milestones and product royalties.
Although each Incyte subscriber has placed a smaller bet than SB did, in the aggregate they may well provide more funds for the development of Incyte's genomic databases. From a broader social standpoint, of course, the more interesting question is not the size of the bets but the ultimate payoffs. Which approach will yield more discoveries or more commercial products?
Public Domain. The Merck strategy of putting sequence information into the public domain is the newest approach and, at first glance, the most puzzling. How does this strategy advance Merck's own interests? By placing data in the public domain, Merck can generate the sequence information more cheaply indeed, almost unbelievably cheaply. Merck is placing a very small bet, somewhere under $10 million, but by positioning itself as a public benefactor, the company is able to take advantage of existing infrastructure at Washington University, put in place with public funds, for its sequencing efforts.
Apart from generating sequence information more cheaply, Merck expects to promote research and derive more benefit by distributing the data widely. As Merck sees it, sequence information will not yield products for commercial development until further fundamental research is done to understand functions and biological pathways associated with the partially sequenced genes. Merck's interest is in developing specific drugs at a later stage in the R&D process. Nothing obligates researchers to bring any potential products to Merck for commercial development, but Merck is confident that it can capture an adequate share of resulting products to justify the company's modest investment in generating the database.
Some observers have suggested the more cynical possibility that Merck may seek to undermine the value of its commercial competitors' investments in existing sequence databases. HGS and Incyte will be dependent on patents to protect their proprietary positions in the long run, and Merck may be betting that the two companies will not obtain much in the way of patent rights.
Preliminary indications suggest that the public data is generating considerable interest, with EST-database accessions showing a dramatic increase. A big part of the increase has come in daily anonymous FTP downloads of the entire database, a form of query likely to be popular with commercial users who do not want to leave an electronic record of what they are looking for.
The most obvious benefit of disseminating information in the public domain is that free availability encourages widespread use of information, minimizes transaction costs, and makes R&D cheaper and faster. Of particular relevance to research science, a vigorous public domain can supply a meeting place for people, information, and ideas that might not find each other in the course of more organized, licensed encounters.
Finally, information in the public domain is accessible to users who otherwise would be priced out of the market. In emphasizing intellectual-property rights in the past, we may have underestimated the value of a rich public domain to private as well as public sectors. We may need now to reconsider the limits of private appropriation of new information as a means of promoting commercial development.
A similar article by Eisenberg was published in Elsevier's Trends in Biotechnology [Vol. 14, 302-7 (August 1996)].
The electronic form of the newsletter may be cited in the following style:
Human Genome Program, U.S. Department of Energy, Human Genome News (v8n1).
The Human Genome Project (HGP) was an international 13-year effort, 1990 to 2003. Primary goals were to discover the complete set of human genes and make them accessible for further biological study, and determine the complete sequence of DNA bases in the human genome. See Timeline for more HGP history.
Published from 1989 until 2002, this newsletter facilitated HGP communication, helped prevent duplication of research effort, and informed persons interested in genome research.