The patent law Duchy of Grand Fenwick: a comment on The mouse that trolled: the long and tortuous history of a gene mutation patent that became an expensive impediment to Alzheimer's research

Abstract

This comment on The mouse that trolled by Tania Bubela, Saurabh Vishnubhakat, and Robert Cook-Deegan discusses the authors’ description of how patents relating to the gene coding for a protein involved in Alzheimer's disease were obtained and used to sue scientific researchers, of how conflicts can arise among groups of researchers, and of how important research can be delayed or foregone by the exercise of patent rights. The authors sought to counter with evidence the optimistic views that patents on such research tools are not used to sue researchers and thus do not impede sequential innovation. The authors proposed legal reforms to avoid such impediments, particularly adoption of a broader experimental use limit to the patent infringement right. This comment agrees with the authors' analysis, but suggests that their descriptions have more substantial implications that extend beyond their example and that their recommendation for limited expansion of the experimental use doctrine does not go far enough.

The mouse that trolled by Tania Bubela, Saurabh Vishnubhakat, and Robert Cook-Deegan (‘the authors’)¹ describes a recent analysis of the effects of four patents on subsequent scientific and medical research and development (R&D). The patents were for nucleic acids, purified proteins, isolated cells, and transgenic mice and screening methods relating to the ‘Swedish mutation’. The Swedish mutation is a rare, double mutation—identified from two Swedish families—in the gene coding for the ‘amyloid precursor protein’ (APP)² that is involved in Alzheimer's disease. The authors explain how a particular, non-practicing licensee [the Alzheimer's Institute of America, Inc. (AIA)] created by the owner of the patents (Michael Mullan) sued pharmaceutical and biotechnology companies as well as public and private research organizations, notwithstanding the patent owners’ contentions to the contrary.³ The patents were ultimately found to be invalid for improper inventorship and assignment, although the patents also may have claimed the disclosed inventions too broadly and covered ineligible subject matter.⁴

As the authors discuss, the law suits caused significant costs and delays to researchers in addressing Alzheimer's disease.⁵ Specifically:

[c]orporate strategies to block access to research tools were [adopted] to protect in-house pharmaceutical R&D, all of which failed to product therapeutics. In other words, eggs were forced into a limited number of baskets to the detriment of scientific and clinical progress…. It seems likely that the bulk of licensing revenue went to support the litigation much more than research. There is little public record to support AIA's contribution to research, and much legal documentation that indicates that it imposed enormous costs on the research institutions and private companies it sued. AIA's contention that it did not hamper research is not credible.⁶

Although the authors’ descriptions are supported with evidence, it would have been helpful for them to provide more documentation and discussion of the actual costs imposed and research that was foregone because of the patent threats. Of course, opportunity costs are hard to observe. They occur in a counterfactual world, and they require for their documentation interviews or other information in the form: ‘I had planned to do x but instead did y, which led to opportunity costs z’. Further, for many readers the added details might have detracted from the main point of the authors’ story. That is, contrary to the conventional wisdom articulated by many interested parties and some academics, patent rights in some fields are not needed and instead diminish research and innovation. As the authors state in regard to one example, ‘aggressive enforcement of patents over research tools stifled research by increasing its costs while there was little evidence that patents created the key incentives to develop such tools’.⁷

To tell this story, the authors trace the complex history from the race to discover the genes and the decisions of various researchers to obtain patents on the Swedish mutation to decisions to exclude other researchers from publication and patent rights and the assertion of those patent rights.⁸ The story thus brings to mind the much more publicized account of the patents on the breast cancer genes BRCA1 and BRCA2, which were obtained by an individual scientist who broke away from a research consortium that had agreed to place its results in the public domain, transferred the patent rights to a commercial (but not a non-practicing) entity, and threatened to sue a wide range of entities including researchers. When the patents ultimately were challenged, the claims to the isolated gene sequences were found to be invalid as not constituting patent eligible subject matter.⁹

By itself, the authors’ careful description of the history adds substantially to our understanding of this particular case, of how conflicts can arise among groups of researchers, and of how important research can be delayed or foregone by the exercise of intellectual property rights—whether or not those rights are validly granted and should be enforceable. But the authors’ tale is even more cautionary, as its implications extend far beyond the particular case they describe. Just as the power obtained by the Duchy of Grand Fenwick in The Mouse That Roared led to radical restructuring of the world political order when the Duchy's minions accidentally obtained the ‘quadium bomb’ (a prototype doomsday device) by walking into a government laboratory,¹⁰ the mere threat to exercise patent power against the scientific enterprise if extended more broadly than the authors’ example could radically disrupt the existing institutional order and seriously damage the scientific research enterprise. Hence, The mouse that trolled also recalls another classic, ironic challenge to social order's faith in scientific control and bureaucratic rationality—in the context of nuclear weapons and deterrence theory. That is, of course, Dr. Strangelove, or: How I Learned to Stop Worrying and Love the Bomb.¹¹ But unlike the implications of Dr. Strangelove and its subtitle, the authors of The mouse that trolled ultimately suggest much less radical (and thus much more feasible) reforms than abolition or loving the current, untenable situation.

The stated central premise of The mouse that trolled was to counter (with evidence) the optimistic views held by some jurists and in the broader community: (1) that patent holders do not sue scientific or academic researchers because there is no money in the research activity itself and therefore no damages to be obtained by suing; and (2) that patents on important biotechnological or medical discoveries therefore lead to broader publication and sharing of research:¹²

Our case study is a counterexample to temper Chief Judge Rader's sanguine assessment. While it describes an unusual outlier example and is not an argument against patents, this case study suggests that without a clear research exemption, or other mechanisms to enable access to research tools, biomedical researchers can face patent infringement litigation that imposes significant costs and slows down both academic and commercial scientific inquiry.¹³

The adverse effects from patents on research tools that the authors describe were made possible in large part by a change in the US patent law experimental use doctrine, which had previously kept scientific and medical research free from patent rights. Specifically, in Madey v. Duke University,¹⁴ the US Court of Appeals for the Federal Circuit further constricted the experimental use doctrine by holding that academic research at educational institutions is commercial in nature given the benefits it provides from grant funding, tuition dollars, etc.; thus, such research is not excluded from infringement, as was commonly believed up to that point.¹⁵ After Madey, it was clear that any unauthorized use of patented in scientific research would infringe patent rights.

As the authors note, earlier work by other academic researchers—particularly Wesley Cohen and John Walsh—had suggested that the Madey decision had not caused significant problems for scientific research endeavors, including for use of patented research tools.¹⁶ This was principally because patent rights were largely ignored by scientific researchers who used the patented inventions without licenses. Stated differently, patent holders were not enforcing their patents against widespread knowing or unknowing infringement by scientists, and thus no significant delays or foregone research and innovation had occurred. But as I have previously argued (with collaborators Christopher Holman and Henrik Holzapfel): (1) it is not an appropriate conclusion that the patent system is working well or is stable in regard to scientific research and research tools, when that result depends on foregoing exercise of rights that are the system's premise; and (2) that these concerns will continue to exert pressure to restore some of the broader, historical scope of the experimental use exception under US patent law.¹⁷

One of the principal recommendations of the authors is thus, unsurprisingly, to change the experimental use doctrine to restore a robust limit of the patent infringement right, so that it does not apply to scientific research of this kind:

In the case of [non-practicing entities] bringing suits against publicly funded researchers and their institutions, the most comprehensive solution would be a statutory research exemption. The USA has no research exemption or compulsory licensing authority, although the US Government does have use rights subject to ‘reasonable compensation’ to a patent-holder.… Indeed, the precedent set by Madey … underscores the desirability of a research exemption. However, recent debates over a research exemption in the America Invents Act further illustrate complexities of legislative reforms. Early drafts of the US patent reform bill included a research exemption, but it did not become law. One problem was in the definition—it is difficult to draw clear distinctions between non-commercial research covered by a research exemption, research that is translational, and research with commercial intent. However, Australia has recently clarified its research exemption in its ‘Raising the Bar’ Act of 2012. In line with some European jurisdictions, Australia allows research on an invention rather than research with the invention, thereby protecting the commercial interests of research reagent companies.¹⁸

Although I agree with the authors’ recommendation, I believe that some research with patented inventions (including research tools) should also be permitted, thereby restoring the historic understanding of the scope of the experimental use exception. Specifically, the experimental use doctrine arose as an interpretation of the limits of the patent infringement right, and not as a judicially imposed (common-law) defense to that right (as is commonly argued).¹⁹ Patents simply did not and should not extend to scientific research.²⁰ Accordingly, the experimental use doctrine not only should not be construed narrowly as an ‘exception’ to infringement, but also the burden of proving a non-experimental, infringing use should lie with the patent holder.²¹ More importantly, as the patent infringement right did not extend to research of any kind, even a commercial enterprise could (by itself) make or use a patented invention for research, so long as it did not actually compete in the marketplace for use of the patented invention nor commercially benefit in its own production operations from using the invention. Thus, researchers remained free to make patented inventions (including research tools) and to use them in their own research, even when seeking to develop (and when actually developing) commercially valuable products. They just could not compete in the marketplace to sell such inventions (including research tools) nor use them in commercially producing their own goods.²²

In summary, the mouse that trolled tells a cautionary tale. But the authors’ warning is not limited to important patents that ultimately are found invalid, like the Swedish mutation patents the authors describe. For important but valid patents, the downstream research concerns are simply more salient; for invalid patents that should never have been granted (under some, but not all validity doctrines), there should be no countervailing benefit to society from inducing innovation, publication, and dissemination. As I stated previously—and it is no less true for less important, valid patented research tools that rarely get scrutinized:

the famous Cohen-Boyer recombinant DNA patent was widely licensed in the United States, but was of questionable validity given unrestricted dissemination of information about the invention more than a year before filing.… Imagine what would have happened if Boyer and Cohen had decided to keep their patent limited to University of California and Stanford? … [W]e clearly need to prevent such important basic patents from being restricted from further use in experimentation.²³

Accordingly, we need to thank Tania Bubela, Saurabh Vishnubhakat, and Robert Cook-Deegan for: reminding us that our current state of affairs is unstable and fraught with peril; highlighting the opportunity costs of foregone innovation that is too often invisible in the public sphere; and repeating the call to reform our experimental use limit of the patent law infringement right.