Intellectual-Property Issues in Biological Research
“Many professors in the period from the late nineteenth century through World War II had contended that university science should be unadulterated by commercial considerations. . . . Yet, it is evident that taking patents on the results of faculty research is not new in the annals of American academia.”–Daniel Kevles
PATENTING AND OTHER KINDS OF PROTECTION OF INTELLECTUAL PROPERTY
People with backgrounds in scientific research are often deeply troubled by patents and other means of protecting intellectual property because they find it counterintuitive to try to promote scientific progress through exclusive rights to prior discoveries. They believe that science will advance most rapidly if researchers enjoy free access to prior knowledge.
In contrast, the working assumption behind intellectual-property law is that exclusive rights promote technology because without the promise of exclusivity, incentives to invest in research and development would be inadequate. This body of law has its origins in disputes between commercial competitors. The theory is that inventions and discoveries are often expensive to make, but inexpensive and easy to copy. The public benefits from new inventions and discoveries, so it is desirable to encourage firms to invest in research and development. But firms might not be willing to do so unless they have some means of preventing competitors from reaping the benefits of their investment without sharing in the initial risk and cost. Without such means, the competitors–or misappropriators, as they are sometimes termed in intellectual-property jargon–would be able to undercut the price of the original innovators and make innovation unprofitable.
One way of keeping inventions and discoveries out of the hands of competitors is to keep them secret. If only Coca-Cola knows the Coca-Cola formula, the Coca-Cola Company does not have to worry about competition from outsiders who did not share in the cost of developing it.
Secrecy works only for some types of inventions, such as manufacturing processes that can be exploited commercially without disclosure. Many
inventions and discoveries are self-disclosing once the products that incorporate them are sold.
Even if secrecy is feasible, it might not be socially desirable. Disclosure of underlying technology might help in marketing a product, for example. One might also want to promote disclosure in the interest of furthering technological progress in the field. This could be of particular relevance to science.
Patent protection is an alternative strategy for protecting rights in some inventions. It does not require secrecy. A patent gives an inventor the right to exclude others from making, using, and selling the patent invention for a limited term: 17 years from the date that the patent is issued under current U.S. law and 20 years from the application-filing date in most of the world. The inventor may choose to make, use and sell the patented invention or to license others to do so on an exclusive or nonexclusive basis, or even to suppress the use of the invention entirely. At least under U.S. law, there is no compulsory licensing provision. One can use a patent to block the use of an invention entirely.
One thing that cannot be done when one obtains a patent is to keep the invention secret. To get a patent, one files an application that fully discloses the invention, including how to make and use it. In Europe, this disclosure is made public 18 months after a patent application is filed. In the United States, it is made public as soon as the patent is issued. (One can disclose an invention as soon as the application is filed without jeopardizing the likelihood of getting a patent.) Patents compel disclosure. They also promote it by providing a system of rights that survive disclosure.
In industrial research, it is fairly uncontroversial that patents promote greater disclosure of research results than would the absence of patent rights. That is less clear in academic research, which would otherwise presumably be freely disclosed (rather than maintained as a trade secret), and that is one reason why some scientists are hostile to the patent system: exclusive patent rights promote disclosure in the long run, but they entail some measure of secrecy in the short term.
Because the patent system protects useful inventions, rather than basic knowledge, an investigator might learn something that is worth publishing before the research has ripened into a patentable invention. The delay need not be great, once the investigator makes a patentable invention and has access to a patent lawyer. It could be longer if the research has not yet yielded a patentable invention. Disclosure at this early stage, when one has interesting scientific information but not a patentable invention, might limit what can be patented later, because the scientific information will become “prior art” that will limit what can be patented. U.S. law provides a one-year grace period under that allows an inventor to file a patent application within a year of
publishing a paper on the discovery. However, if one wants to preserve patent rights outside the United States, one cannot publish research results until after the patent application has been filed.
Thus, an investigator who is concerned with patent rights has an incentive to delay disclosure–which means delaying publication–until he or she is ready to file a patent application. That will undoubtedly delay disclosure of research results in circumstances in which a patent will be sought.
A less tractable problem than the issue of secrecy or delay is that patents can restrict access to inventions. Indeed, that is what makes patents valuable. By restricting access to inventions, patentholders are able to charge royalties and perhaps earn higher profits.
But restricting access to discoveries is highly problematic in research science. It is directly antithetical to traditional norms. It prevents other scientists from verifying results through replication of experiments. It impedes gaining new perspectives on prior inventions from people with different backgrounds and different ideas. At a more practical level, it can add substantially to the financial cost and administrative burden of research or even foreclose avenues of research entirely if researchers are unable to obtain the necessary licenses from patentholders.
For some uses of patented inventions in research, the theoretical conflict between the scientific imperative for free access to prior discoveries and the commercial imperative for restricting access to those discoveries might make little practical difference. Such uses include ones that either do not come to the attention of patentholders or, for whatever reason, do not provoke their objection. In most instances, no serious problems have arisen.
Making and using a patented invention within a research laboratory can be fairly inconspicuous. The patentholder might never notice. Even if the patentholder knows about the use, it might not be worth the trouble, expense and adverse publicity of pursuing a lawsuit against a researcher who does not represent an important threat to the patentholder's commercial interest.
Some patentholders might not object to the unlicensed use of their inventions in purely academic research that has no commercial application – at least as long as the lost royalties from those uses are minor. Moreover, some patentholders might not object to the unlicensed use of their inventions even in their own fields if they think that the research might open up new markets for their inventions or improve on them in ways that increase the value of the patent rights.
Patent protection will not substantially restrict later researchers ' access to inventions that are readily available on the market from patentholders or their licensees at reasonable prices. For example, scientists who wish to use a patented chemical in their research might find that they can buy it from the patentholder less expensively than they can make it themselves. If the
patentholder is charging a high price for the product, the researcher might find it less expensive to make the product. However, that situation is the same for the researcher and for other consumers of patented inventions or users of copyrighted software. The other persons would find it less expensive to use someone else's copy of a copyrighted software product, such as a word processor or spreadsheet, but must pay to use it legally. No compelling argument can be made that a researcher should be exempt from paying a reasonable royalty for the use of a patented laboratory resource.
In some contexts, the use of patented inventions in research might be exempt from infringement liability, but the boundaries of this exemption are unclear, particularly under U.S. law. No broadly applicable, statutory exemption for experimental or research use exists under U.S. patent law, although some narrow exemptions are available in some contexts. Many judicial decisions have stated that the unlicensed use of patented inventions in pure research does not constitute patent infringement, but almost no decisions have found that this exemption applies to the facts of the cases in question. There is a good reason for that: no one is going to bother to file a patent infringement suit unless the activities of the infringer threaten commercial interests of the filer. When people have bothered to file patent-infringement suits in such cases, the research-use exemption has often been raised but almost never sustained.
Some authorities have suggested that the rationale behind the research exemption is that the use of patented inventions in pure research does not harm the financial interest of the patentholder. Under that reasoning, the case for research exemption is weakest when the patented invention is essentially a tool for which researchers provide a primary or even exclusive market. If researchers could not be compelled to pay royalties for the use of such inventions, the patents would be virtually worthless.
In contrast, if there is a substantial market for the invention outside the research community, the unlicensed use of the invention by researchers might have little impact on the profits of the patentholder. Consequently, research use of a mouse that makes profits for its inventor by, for example, being the producer of a valuable protein, might go uncontested by the patentholder. Thus, in the specific context of mouse patents, there might be a distinction between mice that are primarily research tools–which probably is the case for all the mice that are of interest to academic investigators today–and mice that are genetically altered and are primarily hosts for the production of proteins.
Researchers might legitimately ask what would be lost by eviscerating the value of patents on research tools such as transgenic mice, or even by eliminating these patents entirely. The question might be more pointedly asked in cases where the inventions were made through the use of public funds. Some might question the justification for allowing patent rights in government-
sponsored research at all. It might be argued that patent protection is not needed to promote research and development when the government is paying for the research. In addition, downstream development might be impeded by patent that reach upstream into basic research and by exclusive licensing.
A BRIEF HISTORY OF ACADEMIC PATENTING IN BIOLOGY
Today, many scientists are concerned that the dynamics driving the commercialization of the products of laboratories might damage the time-honored tradition of sharing among biological scientists–both the exchange of resources and the free interchange of information. The principles that university research should be unadulterated by commercial interests and that university faculty should forswear patenting discoveries were well established in academic circles from the beginning of such research. Nevertheless, it is evident that taking patents on the results of faculty research is not new in the annals of American academe.
The justification for taking patents in earlier times was primarily altruistic–to ensure that an invention was directed to the public good or encouraged further research. It had also been argued at least since the turn of this century that patenting could foster university research by producing income that could be plowed back into the laboratories. In 1912, for example, Frederick G. Catrell, a chemist at the University of California, Berkeley, invested the argument with institutional life by forming an entity called the Research Corporation. It was designed as an independent, nonprofit organization to manage patents arising from academic research (including the starter patent that Catrell himself had obtained on an electrical method of recovering valuable materials from smokestack emissions) and ultimately to use the income from such patents held by the Research Corporation for grants in aid of university research.
Yet it was not only to sustain academic science that the Research Corporation was founded and that patent rights were transferred to it. The arguments for academic patents also addressed a public interest. Without patent protection, a university would have no control over who developed one of its inventions or whether it were developed at all. Thus, it would have no way to ensure the reliability of the ultimate product; for example, impure biomedical substances might pose a hazard to public health. Nor would the university have any way to prevent the monopolization of such products by developers.
With patent protection, the university could exercise power, through licensing agreements, as to who used a patent and how. The institution could also ensure that no profit-making organization patented materials or products and used them in a way contrary to general public interest. Such reasoning
figured in the decision taken in 1890 by the University of Wisconsin to seek a patent for the enormously valuable test for butter fat that Stephen M. Babcock had invented. A patent was taken and dedicated to the public according to the wishes and intent of Babcock and the university.
Similarly, in 1922, Frederick Banting and John J.R. Macleod and their colleagues at the University of Toronto patented the process for extracting insulin from animal pancreas, an achievement for which Banting and Macleod were awarded the 1926 Nobel Prize in physiology or medicine. They explained to the president of the university that “the patent would not be used for any other purpose than to prevent the taking out of a patent by other persons,” and they added that “when the details of the method of preparation are published, anyone would be free to prepare the extract, but no one could secure a profitable monopoly.”
In 1924, Harry Steenbock of the University of Wisconsin was led to the U.S. Patent and Trademark Office (PTO) out of concern for the safety of his state's dairy industry. Steenbock's research had yielded a method for enriching the vitamin D content of foods by ultraviolet irradiation. Steenbock patented his method mainly to prevent its use by producers of oleomargarine, which does not naturally contain vitamin D, so that they would not have an advantage in competition with dairy producers. Steenbock donated his patent to the University, which in 1925 established the Wisconsin Alumni Research Foundation (WARF) to administer it. The foundation declined to license his process to the oleomargarine industry, but it did license the process to the Quaker Oats Company for use with cereals and to several other firms for the manufacture of concentrated vitamin D oil.
However, commercial patenting was not the norm in universities between the late nineteenth century and World War II. Many professors in that period contended that university science should be unadulterated by commercial considerations. Both physical and biological scientists held to this claim as an article of moral faith and as affirmation of the sanctity of their academic temple, and of noble pursuit within it of knowledge for its own sake.
Academic culture's resistance to patenting was particularly strong in the biomedical sector, where it was commonly assumed that discoveries useful to health and medicine should be made freely available for public benefit. The University of Toronto scientists who were responsible for the isolation of insulin, for example, excluded themselves from shares in revenues from the insulin patent, assigning their rights to the university for $1 each. Steenbock wanted no part of the royalties from his vitamin D patent. He had to be forced by the trustees of WARF to accept some money, and he put the money into a fund for support of research in his department. In the 1920s, Harvard declined in practice to profit from faculty discoveries in health or therapeutic agents, and in 1933 it transformed practice into formal university policy. Over the
following 4 decades, Harvard took out perhaps a half-dozen patents and dedicated all of them to the public.
Apart from the goal of serving the public, professional reasons having to do with the protection and nurturing of the research enterprise gave academic scientists reason to resist the commercialization of their discoveries. If universities came to value patents highly (which was more likely if the rewards from the patent were not restricted to the inventor), they were likely to value professors and research programs that might produce patents a good deal more than professors and programs that focused merely on the advancement of knowledge. Furthermore, commitments to commercial advantage in university research could foster secretiveness. The closing off of one laboratory from another would undermine the open exchange of information and resources that has been an essential requirement of scientific progress. The attitude of openness guided the scientists who developed strains of Drosophila but did not patent them or who developed strains of bacteriophage that they freely exchanged with colleagues.
During the depression of the 1930s, universities were broadly tempted to translate the products of their laboratories into funds for further research through the use of patents. Considerable inspiration came from the very good fortune of WARF, whose revenues from licenses on the Steenbock patent proved large enough to enable the foundation to build a substantial endowment. The foundation was able to use its income for the support of the university's program of training and research. Indeed, by the end of the 1930s WARF had earned the reputation of having been the savior of the University of Wisconsin as a great university because of the money it provided to research.
By the late 1940s, following on that example, some 200 universities and colleges in this country had established procedures or agencies for patenting faculty inventions. However, the financial pressures that led so many universities to flirt with commercialization during the Depression eased during and after World War II as the federal government began to supply universities with abundant funds for scientific research and training. Federal policy on commercializing government-funded research varied from agency to agency, but generally it discouraged patenting for private gain, especially in types of work fundamental to national security, such as atomic energy. Patenting was permitted, but the government usually required the use of the patented product or process on a royalty-free license basis.
Although patenting was not unknown in academe before the 1980s, most university research, especially in the basic life sciences, had yielded little that was patentable and much less that had market value. To a large extent, the products were research tools (such as Drosophila or bacteriophage) that were of interest only to other investigators. University patent revenues were for the most part insignificant in the post-World War II period and even before the
war. Between 1932 and 1963, the patent income at the Massachusetts Institute of Technology, which was by far the most aggressive academic institution in attempting to exploit the progress of its own university research in its own laboratories, totaled only $2 million. That pattern is consistent with the general history of the U.S. patent system: over 5 million patents have been granted since Thomas Jefferson founded the system, and perhaps fewer than 10% of them have ever earned a profit for anyone.
Thus, although in the postwar era scores of universities had established the means to patent the results of university research, the procedures served mainly to regulate the use of the patents rather than to earn substantial income. Many universities turned their patents over to the Research Corporation, not wanting to be bothered with administering or enforcing them, or they assigned the patents to independent entities modeled after WARF. That strategy provided them with only a fraction of revenues from the patents, if revenues were to be had, but it insulated them from the commercial and legal complications associated with exploiting patents directly.
In sum, the richness of federal support and the constraints on patentability in Federal grants and contracts, the traditional culture and values of research and the remoteness of most such work from commercial value combined to create an environment that for many years kept basic biomedical research generally free from patents and commercialization and, by extension, helped to maintain the open exchange of information and resources.
THE CHANGING ENVIRONMENT OF THE 1980s
The passage of the Bayh-Dole Act and other events in the 1980s changed the way patenting affects research. The underlying premise of the statutory scheme is that the public benefits more from inventions that are made in the course of federally sponsored research if the inventions are patented and exploited commercially in the private sector. Inventions left in the public domain are presumed to languish in government and university archives rather than being freely and widely exploited.
The act encourages the patenting and commercialization of inventions made in the course of federally sponsored research as a means of promoting the transfer of technology to the private sector and later product development. It permits recipient institutions to retain title to their inventions if they agree to file patent applications in a timely manner and to ensure that the inventions are used. If a recipient does not elect to retain title to the invention, the sponsoring agency may pursue patent rights itself. Under the Stevenson-Wydler Act, an inventor who is a federal employee may retain the rights to the invention and either exploit the patents directly or license someone else to exploit them. If
the inventor chooses not to exploit or license a patent, the government reserves the right to grant licenses to ensure practical application of the inventions.
Originally, this pro-patent policy for the results of federally sponsored research was limited to universities and small businesses. However, it has been steadily expanding in scope to apply to inventions made in the course of sponsored research by all types of institutions. It has also been expanded to apply to the results of intramural research in government agencies.
Bayh-Dole was an added incentive to many universities to seek to commercialize the products of their investigators' laboratories. Incentive already existed in that federal support for biomedical research in the 1970s and 1980s was increasing at a rate insufficient to meet the needs of increasing numbers of well-trained scientists and the rich opportunities opening in many fields of biology. Many universities had been active in making arrangements with companies for the support of biological research. With Bayh-Dole, many universities took a fresh look at the possibility that patenting of products might be a source of income.
Other events of the 1980s contributed to the developments that changed academe's view of commercialization. Before 1980, living things were generally considered unpatentable either because they were thought to be products of nature or because they were not amenable to written description as required by the patent statutes.
In 1980 the U.S. Supreme Court considered the case of Diamond v. Chakrabarty. The court was asked to determine whether patent protection could be extended to a living bacterium that had been genetically modified in a laboratory. The bacterium had been transformed with DNA plasmids to give it the capacity to break down multiple components of crude oil.
The PTO, on the basis of a long-standing prohibition against the patenting of natural products, had long taken the position that living organisms were not patentable. The briefs and PTO papers in the case of the bacterium also reveal a concern about the morality of patenting living organisms.
By a vote of five to four, the Supreme Court declared not only that one could patent “anything under the sun that is made by man,” but it also said whether the product was alive or dead was irrelevant. All that counted was that it conform to the fundamental patent law in language written by Thomas Jefferson in 1793–that one can get a patent for anything that is made by humans and is a “manufacture or new composition of matter.”
With that broad directive from the Supreme Court, the PTO quickly expanded the categories of living subject matter that it considered eligible for patent protection. In 1985, the office held that corn plants were eligible for standard utility patents as opposed to the more limited plant-variety protection for which they had been eligible previously. Two years later, the office held that oysters fell within the range of patentable subject matter. Shortly