Genomics and intellectual property: considering alternatives to traditional patenting.

• Author: Marden, Emily

1. Introduction

   In recent years, significant issues have been raised about the impacts of patenting practices in genomics. Specifically, critics have voiced concern about the possibility of creating an anti-commons where upstream--or early stage--genomic research is patented. (1) Others have suggested that the net effect of wide-scale patenting is to create patent thickets, wherein awareness and access to patents become an issue. (2) Still others point to the fact that undue concern with patenting and property by practitioners of basic research is contrary to the culture of the scientific community. (3) Some have identified the Myriad Genetics case as emblematic of the negative result for patients (4) (and health care systems) when basic genetic information is protected by patents. Other issues exist as well, such as the applicability of proprietary rights to the type of information generated by genomics research, including sequencing data and databases. (5)

   The degree, extent, and even existence of these issues as significant problems for science and society are the subject of ongoing debate. (6) As the Intellectual Property Policy and Research Group (IPPRG) at the W. Maurice Center for Applied Ethics at the University of British Columbia, we are working to better understand the potential of different alternatives specifically for upstream genomics research. (7) Thus far, we have focused our efforts on patent pools (8) and open source (OS)-like licensing (9).

   With respect to these alternatives, we propose to answer the following questions: Do these alternative approaches address the issues of anti-commons, patent thickets and open science norms, as noted above? In what circumstances and for which specific characteristics might patent pools or OS-like licensing be appropriate? Do alternative upstream regimes create new issues for downstream genomics research and commercialization? Are alternative IP regimes realistic options that are attractive to researchers, academic institutions, industry, policy-makers, or the public?

   In an attempt to foster robust discussion of these issues and to begin to test our interim answers to these questions, we held a workshop in March 2007 where we convened a small group of scientists, scholars, legal practitioners and industry representatives. (10) Our discussions there suggest that there is a growing interest in the potential benefits of alternative IP mechanisms, even while the debate over open source and patent pools for genomics continues. Further, there is recognition that utilization of these alternative IP regimes would impact the development of downstream health products and that any such impacts need to be better understood. These interim findings are discussed in more detail here in the interest of continuing to promote further understanding of how and where alternative IP might be applicable to upstream genomics research.

   1. Open source/Open science continues to be debated as a model for genomics.

      The debate over OS-like models for genomics continues at the granular level of specific research projects. Our efforts at the IPPRG, for example, are focused on the GE (3) LS portion of the "Dissecting Gene Expression Networks in Mammalian Organogenesis Project" (the MORGEN project). MORGEN is a study mapping organogenesis and gene expression of the mouse genome. It is an upstream research project which generates experimental results and bioinformatic tools related to organogenesis, drug discovery, and stem cells. Our research suggests that some forms of open source may be suitable for MORGEN's research results - specifically the publication of some data and of bioinformatics software. The latter is a candidate because it is the most analogous to OS uses in Information Technology. (11) For example, MORGEN implemented a license based on an open source philosophy known as a Creative Commons license. (12)

      One possibility that we are developing is that OS-like licenses could be implemented using a conceptual approach developed by the Biological Innovation for Open Society (BiOS) initiative. (13) In the BiOS model, standard patents are obtained, but licenses to that IP are granted utilizing OS principles that allow access to enabling technology for development. Those who agree to the terms of sharing have protected access and can make and commercialize products, royalty-free, without the need to renegotiate a commercial license. This approach is potentially appropriate for research results that are candidates for patenting (e.g., information that has some potential value to other researchers or downstream drug development). The BiOS license that we envision might be applied here is the licensing of a patented technology via open source and open access principles. (14) In BiOS' model, "licensees and those who have used the technology under MTAs may not assert rights to exclude others from use of improvements, even patented improvements, against the licensor and other licensees within the protected commons." (15) While the BiOS license is conceptually appealing, we recognize that such licenses are difficult to draft, are specifically tailored to the technology, and require a significant investment of time, resources and expertise to develop.

      Our theory is that the BiOS model could be applied in genomics and could guarantee researchers access to information through royalty-free licenses while providing access to a protected commons. (16) As Per the BiOS model, this OS license could promote liberal uptake of MORGEN data and innovations, but would also ensure the sharing of any further improvements. One of the most challenging issues, of course, is what impacts such an OS approach would have on valuable downstream developments and whether incentives are needed to ensure downstream adoption of this approach. (17) In a separate effort, we are developing a proposal that would combine: (1) OS-like licenses with a "tag back" (18) model of royalties based on total sales for upstream developers and, perhaps, (2) a new form of regulatory exclusivity (19) rewarding use of technologies covered by OS-like licenses. The aim would be to encourage use of the OS-like licenses both by upstream genomics researchers and by downstream commercial developers. This is the subject of current work by our team.

      An antithetical perspective on the potential application of the open science movement to genomics was set forth at our workshop by Tina Piper, an Assistant Professor of Law at McGill University. In Piper's view, OS licenses make sense with respect to copyright matters encountered in free software (20) and "copyleft" situations (21) - namely, in the Information Technology (IT) realm where they first arose - but she questioned their applicability to genomics. (22) Indeed, Piper suggested that the very fact that no one had yet developed a workable Open Source license for genomics demonstrated the unwieldiness of the approach. She pointed out that there were some instances of OS biotechnology, such as BioForge, (23) but reflected that this success was more about the success of open source software than about the life science aspect.

      Instead of forcing the biotechnology situation to fit the IT mould with respect to licenses, Piper argued that many of the worthwhile goals of OS could be achieved by focusing on the education aspect of what she termed IP governance. That is, she recommended building on and extending the work of existent organizations by developing tools and strategies to improve the organization and interpretation of existing patent information, ensuring open access to research publications and focusing on community building. Clearly, the question whether OS-like...