Experts Discuss Policy On Patent Landscapes For Life SciencesPublished on 22 April 2008 @ 2:18 pm
Intellectual Property Watch
By Kaitlin Mara
Access to quality data in intellectual property policy in the life sciences is as necessary as it is hard to obtain, said a recent meeting of technical experts, called to discuss challenges in patents and innovation in medical, agricultural, and other biological technologies.
Patents on life sciences figure into the “key debates of the day,” including agriculture and food security, public health and pharmaceuticals, and environmental protection, said Antony Taubman, head of the global IP issues division at the World Intellectual Property Organization.
Shakeel Bhatti, executive secretary of the International Treaty on Plant Genetic Resources for Food and Agriculture, spoke of the current social unrest in places around the world over affordable access to basic foodstuffs. The paramount importance of food security underscores efforts to keep food prices down by ensuring access to wide varieties of plant genetic resources, he said. The treaty creates a global gene pool of the most important food crops that will be accessible to everyone, and then includes plans to allow for standard material transfer agreements.
Access to medication also is at issue. Precious Matsoso, who directs the Department of Technical Cooperation for Essential Medicines at the World Health Organization (WHO) said that access to current and accurate “information on the patent status of essential medicines” often is limited in developing countries, which can affect the ability to public health needs.
Recurrent themes of the 7-8 April event at WIPO entitled “Symposium on Public Policy Patent Landscaping in the Life Sciences” were: the difficulty of finding complete, up-to-date patent data in easily searchable formats and the complexity of an increasingly multilingual and multilateral patent system.
Patent landscapes are data collections showing ownership, claims and validity of patents on a particular technology or component parts of a technology. Particularly intransigent challenges are the difficulty of keeping a patent landscape current, given the dynamism of the patent system; variant capacities to access information in different parts of the world; and the ability of a dishonest patent applicant to deliberately obfuscate claims using ambiguous language, making information-finding difficult for busy patent examiners and public interest researchers lacking adequate ways to parse data.
What Goes into a Patent Landscape
Those attending the symposium generally agreed on a need for accurate, complete patent data, but debated over what comprised a full patent landscape, and discussed the best way to frame and analyse data.
There is “a big haze” when one talks about patent landscaping, said Edward Hammond, former director of the Sunshine Project and the Rural Advancement Foundation International. He said a patent landscape had to be understandable to those not necessarily well-versed in patent law, such as soil scientists, or medical doctors: people who have technical knowledge but lack an intellectual property background.
Victoria Henson-Apollonio, from the Consultative Group on International Agricultural Research’s (CGIAR) Central Advisory Service on Intellectual Property, pointed out that analysis of information in patent databases is only part of the story for policymakers.
Citing a well-known case involving “Enola beans” in which a man named Larry Proctor patented a bean he had selectively bred to exhibit a specific shade of yellow, she noted that it is difficult to find a common language that might describe prior art in this case, as there is no standardised colour reference for plant breeders. The patent claims include any common bean “that is completely yellow in colour” as defined by the “Munsell Book of Colour,” which Henson-Apollonio pointed out was not a widely used reference among plant breeders. There is also a need to include “grey” literature – works in print but not officially published, for example academic working papers, newsletters, brochures, government briefings, blog entries or business manuals – as much as possible in databases used by patent examiners, she said.
Kerry Fluhr, an IP analyst from the Centre for the Application of Molecular Biology in International Agriculture (CAMBIA), noted the tension between the drive for a broad, but simplified, patent landscape – able to be used by policymakers but lacking some technical detail – and the drive for very narrow, specific, and technical landscapes that could be used only by a smaller readership.
Keeping Landscapes Current; Complete
Ben Prickril from Public Interest Intellectual Property Advisors (PIIPA) said the main challenge is to keep patent landscapes regularly updated and able to provide guidance to stakeholders from government, research and health professionals, and civil society groups in the case of a pandemic, and discussed PIIPA’s efforts to map patents on avian influenza.
Tahir Amin of Harvard Medical School echoed this, noting that patent landscapes go out of date very quickly. He also said that in many countries simply figuring out where to find data is very difficult.
Even once data is obtained, there is often difficulty in searching and sorting it. Fluhr presented a solution for creating effective patent landscapes. She talked about the organisation’s PatentLens search tool, which uses a system called BLAST to align sequences of nucleotides in a genome with relevant patents and patent applications. CAMBIA’s goal is to “counter perception with facts, increase efficiency of innovation, and transform information into knowledge,” Fluhr said.
CAMBIA has undertaken a rice genome analysis that illustrates the importance of complete patent landscapes: when fully mapped, CAMBIA found that only 0.26 percent of the genome was covered by granted patent claims in the United States – significantly lower than the human genome, which is 20 percent covered by granted claims, and lower than suspected before the research was undertaken. The 20 percent figure was the result of a report co-authored by Kyle Jensen of the Public Intellectual Property Resource for Agriculture, who also spoke at the event.
There are, however, important caveats to this figure: 74 percent of the rice genome is covered by US patent applications, though Fluhr noted that the majority of these are in bulk sequence applications – in which large sets of nucleotides are placed in early patent applications by researchers who will then usually have to narrow their claims during the application process.
An audience member raised the concern that US patent examiners might be more selective – and that these kinds of bulk sequence claims could become a problem in developing countries where the capacity for true patent examination is lacking and their offices may just approve whatever they come across.
And Hammond, referencing the Enola bean case, said that it was dangerous to assume broad patent applications necessarily narrow during the approval process. Anyone looking at the Enola bean patent, Hammond said, would have said “it’s crazy that any patent examiner would issue a patent on all yellow beans of this type,” but then it happened.
Pascale Boulet of Médicins Sans Frontières echoed this, and added that the difficulty in patent searching could exacerbate the situation. Patents are usually presumed valid until challenged, she said, which is a public policy problem in many developing countries that lack adequate prior art searches. She added that “counting patents” does not necessarily equate to new innovation.
Diversity and Difficulty in Data Collection
William Meredith, head of the Patent Information and IP Statistics Service at WIPO, said that despite appearances of increasing patent activity, “the rate of growth is about 4 percent,” not surprising in the context of GDP and trade growth. What is changing, he said, is that there “is an increasing diversity in the use of the patent system,” with the current top three patent offices now being the United States, Japan, and China, followed by Korea and the European Patent Office. This diversification of originating countries, and with that, language, of patent applications, adds further challenges to the act of patent searching and landscaping.
WIPO’s patent search tool, Patentscope, now has over 30 patent offices collaborating, according to Meredith, and is undertaking projects to integrate more, particularly from Latin America, Africa, and Asia. There is also a translation project, as searching English-only key words leaves out 35 percent of the database. Further challenges arise from the fact that searching by keywords can be ambiguous – viruses, for example, can refer to both human health and information technology problems – semantic differences, and deliberate obfuscation of claims.
There also was discussion as to whether the distinction between public ownership and private ownership of patents hinted inappropriately that data owned by, for example, universities was more likely to reach the general public than that owned by companies. Jensen said that the lack of incentive to commercialise in the public sector means privately owned IP is likely to be in public hands faster.
Also a problem, said Pierre Roger, IP manager of Limagrain, a French seed company, is that while tools are available to find patent data, the quality and reliability of that data, as well as the cost of access to it, is very unequal. Fred Abbott, professor of international law at Florida State University College of Law, said that even major information portals have their shortcomings. The Orange Book, the US Food and Drug Administration’s index of patents on approved drugs, is “an invaluable source of research leads” said Abbott, but lacks data on medicines patented outside the United States.
Another problem is the intricacy of life science patents in general. Roger said that the patent thicket on plant varieties is particularly complex: patents can apply to a plant variety: everything from lab techniques used to develop it – including molecular markers, DNA extraction and cloning – to the addition of genetically engineered compounds, the use of microbiological processes such as agrobacterium, and finally to the product itself.
Martin Friede, of the WHO Initiative for Vaccine Research, Product Research, and Development, Immunization, Vaccines and Biologicals, explained that there are “no true generic” vaccines as drug equivalence with vaccines is not easily determined and requires full clinical trials each time. Patents on a single vaccine can be incredibly arcane, encompassing several levels of biological science owned by different players, said Friede. A single vaccine might include patents on: the antigen needed to produce the proper immune response, including its DNA sequence and particular expression; the adjuvant, which is used to facilitate a person’s response to an antigen, the excipient, which is the substance and antigen and adjuvant are stored in; the vaccine itself; and finally, its method of delivery.
Henson-Apollonio said the complexity of some patent landscapes has the effect of rendering them “unintelligible, and it’s scary” for those trying to work functionally within the confines of patents on life sciences. She said later that one great challenge is to spread understanding of patent searching to a broader audience, and spoke of wanting the average reader of science journal Nature to be interested and able to comprehend the patent world.
Looking Beyond Patent Landscapes
Corrina Moucheraud Vickery from the Program on Global Health and Technology Access at Duke University said that policymakers “have to look at patent landscapes as though they are flawed,” and suggested that efforts to increase disclosure and transparency of patent systems might help. She also raised the idea of a technology trust, a project she acknowledged would require substantial norm shifting in the way people think about ownership over life sciences innovations. Such a trust would be much more than a “patent pool” she said, as IP goes far beyond patents; it could also be linked with funding, prizes, or advanced market agreements for important innovations as incentive to pull innovators into the scheme.
The implementation process of the International Treaty on Plant Genetic Resources included the creation of a tool kit for standard genetic material transfers, which will eventually have an online portal to facilitate transfers, Bhatti said.
WIPO and Landscaping
WIPO’s newly approved budget (IPW, WIPO, 1 April 2008) calls for work on patent landscaping issues in the life sciences, Taubman added, and this meeting is the start of future planned work on the topic.
Also commenting at the event were: Maria Jose Amstalden Sampaio, policy and governmental affairs officer, EMBRAPA, who talked about soybean patent landscapes; Thomas Cottier of the World Trade Institute, who talked about methodological issues in patent landscaping; Muriel Lightbourne, a visiting scholar from the University of Illinois at Urbana-Champaign, who spoke on the implementation of the International Treaty on Plant Genetic Resources; Raj Hirwani, of the Unit for Research and Development of Information Products (URDIP) in India, who presented the results of an extensive patent landscaping report on several neglected diseases, and Paul Oldham of the Economic and Social Research Council’s Centre for Economic and Social Aspects of Genomics, who spoke on biodiversity, biopiracy, and traditional knowledge in the patent system.
Kaitlin Mara may be reached at firstname.lastname@example.org.