Trial court decisions have exposed an Achilles' heel in licensing regimes for research method patents. In Bayer AG v. Housey Pharmaceuticals, Inc., and in Trustees of Columbia University in City of New York v. Roche Diagnostics GmbH, 150 F. Supp.2d 191 (D. Mass. 2001), trial courts have sharply limited infringement liability for offshore use of patented methods, where the products introduced as the fruits of this research are ultimately manufactured without use of the patented methods. It remains to be seen what reception the U.S. Court of Appeals for the Federal Circuit will give to this issue.
The Bayer Case
In Bayer, the German multinational Bayer AG refused to license Housey Pharmaceuticals' screening "method of determining whether a substance is an inhibitor or activator of a [target] protein * * *". Such screening methods are used by pharmaceutical companies in automated fashion to rapidly identify useful compounds, which then must be further tested in animals and eventually in humans for pharmaceutical activity. Bayer allegedly used Housey's method outside the United States to find the proverbial biological needle in a haystack (a useful drug), and then manufactured that product without further use of Housey's patented screening method. On a motion to dismiss, the court granted judgment in favor of Bayer. Key to the court's holding is that Bayer was not alleged to have used the patented method in the United States, but only to have imported products that it was able to identify through use of that method.
Theories of Infringement
Housey advanced two discrete theories to pin infringement liability on Bayer.
It first argued that Bayer had infringement liability under 35 USC § 271(g) (the section which prevents importation into the U.S. of a product made abroad by a process patented in the U.S.) by selling in the United States a drug that was determined to be an inhibitor or activator of a target protein using the patented methods. Adopting this theory would have required the court to view the Housey patent as a method of making drugs rather than as a screening method.
A second theory advanced by Housey was that Bayer had infringement liability under 35 USC § 271(g) by "import[ing] into or us[ing] in the United States knowledge and information reflecting the identification or characterization of a drug acquired from using the patented methods." Adopting this theory would have required the court to view knowledge or information as a "product" within the meaning of section 271(g).
In both cases, it is difficult to squeeze into 35 USC § 271(g) Bayer's activities of importation of a product made by a manufacturing process independent of the patented screening method. The statute provides for infringement liability where a third party "imports into the United States or * * * sells * * * within the United States a product which is made [offshore] by a process patented in the United States * * *." Denying infringement, Chief Judge Robinson ruled that "[u]pon a plain reading of the statute, the court finds that Section 271(g) addresses only products derived from patented manufacturing processes, i.e., methods of actually making or creating a product as opposed to methods of gathering information about, or identifying, a substance worthy of further development." The court bolstered its opinion by noting the narrow reading of the statute by the Federal Circuit that has limited application of section 271(g) to methods of manufacture. It cited Biotechnology Gen. Corp. v. Genentech, Inc., for the proposition that importation of hormone made by patented recombinant DNA techniques infringes under §271(g); Eli Lilly & Co. v. Am. Cyanimid Co., as holding that importation of an antibiotic does not infringe a patent claiming a method of making an intermediate compound where such compound was "materially changed", and Mars, Inc. v. Nippon Conlux Kabushiki-Kaisha, for the holding that a patent describing a process by which an electronic coin changer in vending machine analyzes coins is not a manufacturing process and, therefore, not covered by §271(g).
Applying Federal Circuit case law to the specific screening method of Housey, Chief Judge Robinson pointed out that "the asserted method claims of the [Housey] patents describe processes for recognizing substances with the potential for development into pharmaceuticals. These processes of identification and generation of data are not steps in the manufacture of final drug products. Thus, §271(g) is not applicable to the process claims in the [Housey] patents and [Housey's] infringement claim under Section 271(g) is dismissed."
In the past generation, there have been literally thousands of research method patents. Some -- like the basic Cohen-Boyer patent covering transfection methods to insert a plasmid -- give rise to new, living and self-replicating cells. Other research method patents permit finding a biological needle in a haystack: thousands of chemical or biological entities are screened to find the pharmaceutical bullet -- a new product useful to treat a disease. A common thread among these research method patents is that once the self-replicating cell is created or the screened product is identified, the research method is not needed any more to make the useful end product.
Critics charge that these patents drive up the prices of pharmaceuticals by creating royalty "taxes" paid by pharmaceutical companies to the owners of the screening method patents. In many instances, taxpayer money in the form of government-sponsored research grants was used to discover the research methods in the first place, resulting in a double tax on consumers. See Rebecca S. Eisenberg, Public Research and Private Development: Patents and Technology Transfer in Government-Sponsored Research, 82 Va. L. Rev. 1663, 1710 (1996).
Insofar as a purely domestic patent licensing and infringement model has been concerned, the universities and small startups have seized the opportunity to license their research methods keyed to sales of final products the licensee may develop. These so-called "reach-through" licenses key licenses to sales of the products that are ultimately manufactured without use of the research method patent, but which could not have been identified or created in the first instance without the patented research method. Professor Eisenberg of the University of Michigan Law School has been critical of the Cohen-Boyer licensing effort, pointing out that it "involve[s] patents that yield substantial royalty payments through nonexclusive licenses, although nonexclusive licenses do little or nothing to give licensees an advantage over their competitors and thus are unlikely to enhance the profitability of product development." Id. She further points out that "[t]he reason universities count these patents as successes is not that they have helped move the technology out to the private sector for commercial development, but rather that they have generated a lot of revenue for the institutions that own them." Id.
Echoing the criticisms of Professor Eisenberg, academics from the two institutions that own the Cohen-Boyer patents have joined the debate. Professor Barton of Stanford has considered the antitrust implications inherent in the research method patents which key royalties to a "reach-through" basis to nonpatented technology. John H. Barton, Patents and Antitrust: A Rethinking in Light of Patent Breadth and Sequential Innovation, 65 ANTITRUST L.J. 449 (1997). Professor Merges of the University of California has also considered the competitive aspects of a nonexclusive license for research tool patents. Robert P. Merges, Intellectual Property Rights and Bargaining Breakdown: The Case of Blocking Patents, 62 TENN. L. REV. 75 (1994).
Supporters point out that many research method patents do provide a competitive advantage in drug discovery and in other fields by identifying new drug targets and by accelerating the development of new drugs. Moreover, they point out that such patents have propelled the U.S. biotechnology industry to new heights, giving investors the security needed to fund startups.
While the debates have largely focused upon the domestic considerations, often lost has been the principle of territoriality. If the patent tax of a nonexclusive license imposed upon a pharmaceutical house is too great, the prospective licensee may simply seek to avoid the patent altogether by taking his research from Boston to Bangalore or from California to Kyoto. Under the principle of patent territoriality, research methods patented in the United States may be used with impunity in India, Japan or any other country of the world where there is no counterpart patent. There simply can be no patent infringement because the research method is not used "within the United States", the necessary condition precedent for direct patent infringement under 35 USC § 271(a).
Proponents of reach-through licensing programs have argued that, if one benefits from a patented research method abroad, one should be liable for infringement royalties for use of the product in the U.S. made possible only through the research method. A theory of liability based on the "fruit of the poisonous tree" was rejected for foreign use of an invention where the products imported into the United States were outside the scope of the claim. Denying this theory in Columbia v. Roche Diagnostics, 150 F.Supp. at 203, Judge Gertner said that "Columbia's suggestion that [35 USC §] 271(a) liability extends to 'fruits of the poisonous tree' simply does not comport with the unequivocal terms of the statute [that] makes reference to patented invention, not to by-products that derive from that invention" (emphasis in original).
As noted above, the Court of Appeals for the Federal Circuit has yet to weigh in on these "fruit of the poisonous tree" challenges that attempt to provide an extraterritorial reach - and thereby justify - reach-through royalty payments on end products derived from patented research methods. In the meantime, the challenge for biotechnology companies is to strengthen their position by putting research methods to work in their own house and obtaining product patents directly covering the valuable end products. And many are doing just that. Aided by their own research methods which can be carried out on automated biochips, they are rapidly identifying biologically active molecules and racing to patent them as products. Here, too, there are challenges, for the patents must contain an adequate description and enough data to support the claimed utilities if they are to survive judicial scrutiny.