In light of several high-profile court cases on patenting of DNA sequences, the U.S. Patent and Trademark Office (USPTO) in June issued interim guidelines for helping patent examiners determine if the so-called "written description" requirement for patent applications has been met. John J. Doll, director of biotechnology examination at USPTO, says the interim guidelines have become necessary to determine just how court decisions such as University of California Regents v. Eli Lilly and Co. will affect the patent application process.
Under U. S. patent law, inventors and scientists are required to submit a written description of the invention or discoveries so that examiners can establish that the applicant has indeed invented or discovered the subject matter claimed in the application. The written description must make it apparent to those skilled in the affected art or science that the inventor was in possession of the claimed invention when the patent application was filed. While meeting the descriptive requirement for patent applications typically is not a problem, recent court cases have highlighted the difficulty sometimes encountered in establishing claims in biotechnology.
REEXAMINATION: John J. Doll, director of biotechnology examination at USPTO, says the interim guidelines have become necessary to determine just how high-profile court decisions will affect the patent application process.
Circuit Judge Alan D. Lourie wrote in the opinion that while UC claimed to describe through a "constructive, prophetic example" a related process by which cDNA for human insulin could be produced, they did not provide a written description of that cDNA. "While the example provides a process for obtaining human insulin-encoding cDNA, there is no further information pertaining to that cDNA's relevant structural or physical characteristics; in other words, it does not describe human insulin cDNA," he noted.
Doll says the ruling highlights the need for applicants to provide as thorough a specification of the invention as possible. The new guidelines, for example, note that if the complete structure of a claimed DNA is given in the description, then the descriptive requirement has been met. But in cases where the complete structure is not disclosed, patent examiners must determine if other identifying characteristics of the invention are sufficient to identify the invention and establish that the applicant was in clear possession of the claimed invention.
UC Regents v. Lilly has become one of the most bitterly fought patent infringement cases in biotechnology. The UC regents first brought the patent infringement action against Lilly in 1979 for its process for producing human proinsulin. When Lilly refused to pay royalties--arguing that the process for encoding human proinsulin was not obvious given UC's research--the regents sued in 1990. University lawyers contended that UC had actually developed the process for expressing human proinsulin, which is converted to insulin in the body. Lilly, however, argued that Genentech Inc., of South San Francisco, Calif., had actually developed the process and UC's claim was invalid. In 1995, U.S. District Judge Hugh Dillin in Indianapolis ruled in favor of Lilly, claiming that UC had not adequately described the claimed process.
UC, which demanded a change of venue, appealed the ruling, contending that the patent described in sufficient detail how to prepare human insulin. But in July 1997, a federal appeals court rejected the appeal. Circuit Judge Lourie noted in the opinion, "The name cDNA is not itself a written description of that DNA; it conveys no information in the patent pertaining to that cDNA's relevant structural or physical characteristics; in other words, it does not describe human insulin cDNA." The court goes on to say that describing a method of preparing cDNA or even describing the protein does not necessarily describe the cDNA itself.
UC, however, did find victory in defeat. Earlier in the trial, Lilly accused UC-San Francisco researchers who filed the patent of using improper means to achieve the cloning of rat insulin. They claimed that the researchers violated federal gene-splicing rules and then attempted to hide this fact in published reports. Although Judge Dillin found UC researchers guilty of "inequitable conduct," the U.S. Court of Appeals ultimately tossed the charges. Based on this ruling, the order for UC to pay Lilly's court costs estimated at $18.5 million for the seven years of litigation was overturned. UC now only has to foot its own legal bills of about $12 million.
Jonathan A. King, a professor of biology at the Massachusetts Institute of Technology and an outspoken critic of gene patents, says the high cost of protecting gene patents ought to make researchers, universities, and companies question the value to society of turning biological entities into private property. While he admits that such patents can produce substantial revenue for universities and industry, King notes that universities and companies often have to spend nearly as much to defend a patent from infringement by other scientists.
"These kinds of tit-for-tat court battles have almost become the norm in the area of gene patents," King says. "As soon as a patent is issued, you can expect one company to sue another to defend their claims. You have to wonder if this is the best way to spend money and devote resources in scientific endeavor."
King also is concerned about the effect of patents and other legal agreements on the free exchange of scientific information and resources between investigators. He says many researchers now find themselves unable to discuss research or publish papers until legal steps have been taken to assure future protection of the studies. "These issues have not been discussed often in public forums," King adds. "But I can assure you that it is a hot topic and constantly under discussion by researchers gathered around the water cooler in the laboratory."
NIH recently acknowledged some of these concerns over the effects of patents in a report to the director on research tools. The report, issued in June, details the worries of scientists in academia and industry over the potential of patents and other legal agreements to burden the process of discovery in science, especially in biotechnology. For example, the report notes that:
- Limitations on publication and discussion of research due to patents and legal agreements have become a major concern at universities. Confidentiality agreements are often far-reaching and can interfere with the ability of researchers to publish.
- Researchers in industry report difficulty accessing research tools created in NIH-funded research. They believe that it is unfair that they must undergo drawn-out negotiations for access to research tools created through public funds.
The report also notes that patents and licenses for research tools such as expressed sequence tags (ESTs), or partially mapped sequences of DNA, could slow down and burden the process of research. More than six years ago, NIH was at the center of this controversy. In 1991 and 1992, NIH touched a nerve in the biotechnology community when it filed patent applications on more than 2,700 partial DNA sequences (S. Veggeberg, The Scientist, 6:1, 1992). Although patenting fully characterized DNA sequences was an accepted practice, no one had yet attempted to gain property rights to ESTs.
Back then, NIH thought the patents were necessary given several unresolved legal issues. Officials at the patenting and licensing program of the NIH division of technology development and transfer, note that it was unclear if placing ESTs into the public domain through publishing or a database would jeopardize future patent claims from companies and universities utilizing this information. The major concern was that placing ESTs into the public domain would seriously discourage researchers from completing and finding medical utility for the sequences, they explain. Under the orders of NIH Director Harold Varmus , NIH abandoned patent applications on ESTs in 1995. Around this time, courts made it apparent that ESTs did not render all subsequent discoveries obvious and thus, nonpatentable.
BEST DEFENSE: Gerry J. Elman, a Media, Pa.-based attorney who specializes in biotechnology law, says that under current law, patents may be the best way to protect claims to genes.
But some large pharmaceutical companies have taken a different stance. In collaboration with Washington University and the University of Pennsylvania, Merck and Co. Inc., based in Whitehouse Station, N.J., created a public database for DNA sequences. The database represents more than 75 percent of publicly available EST information. "Even before NIH made the controversial move to file patents for ESTs and stirred debate, we viewed partial DNA gene sequences as research tools," C. Thomas Caskey, vice president for research at Merck, says. "As such, we never regarded them as patentable material and saw their value as a tool in developing downstream products."
"Many patent [applications] for ESTs are filed at this point. This is a strategy that has been taken by many companies to secure primary possession and ownership of genes. My main concern is if these property rights could create a very serious limitation on the freedom of researchers to determine the functionality of genes. I think this will be the next focus of discussion and debate on gene patents."
Currently, no patents for ESTs have been issued, according to Doll at USPTO. But he notes that this is not because such patents are not permitted under law. "Our position is that ESTs are clearly patentable subject matter," Doll says. "They are nonobvious and have novelty and utility. So, there is nothing in principle that rules out such patents. However, we've faced a tremendous backlog of patent applications, and claims for ESTs are still working their way up the examination gauntlet."
Patent applications for biotechnology in general have created a challenge for USPTO. In October 1996, the office announced that it was revising the application process due to a huge backlog of applications still awaiting review. The annual cost of reviewing the applications. which included more than 500,000 sequences, was estimated at more than $35 million dollars and would take one examiner about 200 years to process.
To speed the review process, PTO limited claims to ten sequences per application and required that the sequences be submitted in a computer-readable format for examination. Dolls reports that PTO has made substantial progress in catching up on applications. "Within the fiscal year, we expect to have worked off the backlog and then all pending EST patents would be current," he says.