Recalibrating Intellectual Property Rights to Enhance Translational Research Collaborations

+ See all authors and affiliations

Science Translational Medicine  22 Feb 2012:
Vol. 4, Issue 122, pp. 122cm3
DOI: 10.1126/scitranslmed.3003490


Multisectoral collaborative models for drug and therapeutic research and development (R&D) are emerging, requiring a recalibration of how intellectual property rights (IPRs) are used. Although these models appear promising, little study has been conducted on the optimal blend of sharing and exclusion as mediated through the proactive use or nonuse of IPRs. This Commentary is a call for a combination of theoretical and empirical analyses to build a comprehensive understanding of the interplay between formal IP laws, institutions that administer and manage IPRs, and the use of IPRs in practice to better construct and manage collaborations. Such analyses require outcome metrics formulated to measure the success of therapeutic outcomes and to capture the complexity of a highly networked R&D environment.

New innovation models for drug discovery are emerging in response to high costs, duplication of effort, and diminishing levels of product development (13). Many of these new models emphasize collaboration between academia, government, industry, nongovernmental organizations, and patient organizations (47) on the basis of the principle that no one entity can itself do most of the research and development (R&D) needed to develop a new drug or therapy (8). Such collaborations, drawing on the respective strengths of different sectors, require active governance—including that of intellectual property rights (IPRs)—to enable the sharing of informal and formal knowledge, data, and reagents. Although IPRs are theorized to promote the introduction of new and useful products and processes, there is increasing evidence that they may, in certain settings, hinder collaborations that are critical to achieving those ends. What is required is a recalibration of how IPRs are used, supported by evidence based on measures that capture the complexity of a highly networked research and development environment (Fig. 1).

Fig. 1. Boosting collaboration.

The sharing of data, reagents, and know-how enhances the efficiency of research and knowledge translation to improve clinical practice. Especially important are failed clinical trial results. Enhancing the precompetitive research environment requires creative governance mechanisms and innovative rules for sharing IP.


Here, we discuss the need for creative governance of IPRs to facilitate multisectoral collaborations. This initiative involves greater selectivity in obtaining patents, more open licensing of patents, and better-designed metrics that not only track the effectiveness of collaborations but also, once adopted by research institutions and funders, serve as an incentive to construct and maintain effective collaborations. Although we acknowledge the important role that other incentives within the broader regulatory and funding environments provide (9), we argue that the creative management of IPRs and shared knowledge creation play a critical role in promoting the development of innovative drugs and therapies. A focus on the management of IPRs is more likely to advance translational science—a current priority for the U.S. National Institutes of Health (NIH) through the proposed National Center for Advancing Translational Sciences (NCATS) (10)—than are unproductive and stale debates over enhancing IPR protection on the assumption that doing so would necessarily stimulate pharmaceutical innovation. Creative management of IPRs facilitates the development of malleable structures responsive to the pace of change in industry, including the downsizing of R&D in large pharmaceutical companies; increasing translational activity in academia; the outsourcing of R&D to India, Brazil, and China; and the expansion of those countries’ endogenous R&D capacity.


Recent scholarship and practice are converging on the opinion that new multi-stakeholder collaborative models will reverse the stagnation of the current drug discovery model. Collaborations are theorized as leading to a more efficient deployment of resources through the reduction of negotiating costs, the avoidance of duplicative research, and the use of standard research tools and software. Supporting new collaborative models are innovative forms of pooling and sharing of data and reagents between and within the public and private sectors (Table 1) (6, 7). Analysis of the first wave of collaborative models in the 2000s, most notably public-private partnerships (PPPs) funded largely by government, philanthropic foundations, and large multinational firms (11, 12), suggests that they are more efficient, quicker, and achieve better outcomes than R&D centered in pharmaceutical companies, at least in the area of neglected diseases (11, 12); evidence to date suggests that it is possible to reduce the costs of developing drugs through new strategies to bring in necessary technologies (technology sourcing) and for project management. These new strategies rely heavily both on external knowledge and expertise and on strict specifications about the characteristics and costs of the product under development. Whether this holds true in more traditional markets has not yet been assessed.

Table 1. Managing IP for innovation.

Shown are innovation models for pharmaceutical R&D. These categories are not mutually exclusive and may be used effectively in combination.

View this table:

There are several types of existing PPPs. The aim of product development PPPs is to develop drugs to treat particular diseases. Other collaborations include consortia for the development and distribution of reagents, such as mouse models for human diseases as well as translational medicine and therapeutics development initiatives within academic centers and through funding agencies such as NIH. Although these emerging models appear promising, little study has been conducted on the optimal blend of sharing and exclusion as mediated through the proactive use or nonuse of IPRs.

Direct funding from public and philanthropic sources may be sufficient to further intersectoral collaboration in an altruistic setting, but different incentives and structures will be required to encourage collaboration in a for-profit setting. A combination of theoretical and empirical analyses is required to build on the early evaluations conducted in the more limited context of the global health field. Theory will lead to a comprehensive understanding of the interplay between formal IP laws, of institutions that administer and manage IPRs [for example, patent offices, regulatory agencies, and technology transfer offices (TTOs)], and of the use of IPRs in practice to predict which models are likely to function best in which circumstances (5). Empirical studies will allow researchers to test and refine those predictions. Both sets of studies will require outcome metrics that actually measure successful therapeutic outcomes and enhancement of scientific knowledge. Such new metrics may be contrasted to the current reliance on input-output measures along a linear innovation pipeline, such as the number of patents and start-up companies and the amount of licensing revenue generated. The pipeline model of innovation assumes that reliance on high IPR enforcement and limited sharing are the best ways to facilitate further pharmaceutical innovation. However, as we discuss below, new metrics based on a networked and collaborative model of innovation are difficult to develop and implement and will require policy and funding commitment (13).


The dominant forms of IPR in the pharmaceutical sector are a blend of patents, trade secrets, and legislative protection of data—in particular, data submitted to regulatory agencies during the drug or device approval processes. Patents have the goal of stimulating the creation of new technologies with practical applications through the grant of exclusive rights to make, use, sell, and import inventions for a period of 20 years (and longer in some jurisdictions and under some circumstances). However, the economic impact of patents in the life sciences and their role in stimulating innovation and attracting industry investment in R&D, especially in smaller markets outside of the United States, have been hotly debated (14).

Experience shows that enhanced IPR protection does not necessarily equate with increased R&D or innovative new products. For example, Canada enhanced the level of IPR protection for pharmaceutical innovation in the late 1980s and again in the 2000s, but after an initial slight increase in R&D investment in the 1990s (mostly at the clinical stage rather than at the basic discovery and mechanistic research stage), industry investment has declined to levels before those changes in IPRs (15). Similarly, despite increases in IPR protection in the UK, Pfizer closed its drug-development laboratory in Sandwich, UK, in February 2011 and downsized its Connecticut R&D headquarters. Underperformance is not limited to the pharmaceutical sector; globally, the biotechnology sector overall has yet to show a profit in any year since the birth of the industry, although a few companies, such as Amgen and Genentech, have been profitable (16, 17).

There is a substantial amount of systemic inertia in the management of IPRs that undermines any impact caused by attempts to tinker with formal IP laws. Rather than address fundamental problems in the pharmaceutical innovation system, including out-of-date IPR practices, the industry has, at least until recently, called for greater levels of IPR protection despite the industry’s decline in productivity. Some industry leaders as well as international organizations such as the Organisation for Economic Cooperation and Development (OECD) have come to recognize that the problem with declining levels of productivity can only be solved in the long term by developing creative mechanisms through which to enhance the flow and use of knowledge rather than through the creation of greater IPR fortresses (18).

Several aspects of current IPR management create these fortresses. For example, there is some evidence that patents may block downstream innovation when a key patent, such as one over a platform technology, is held by an actor that restricts access to the technology, often by charging high royalties for licenses (permission to use). Unfortunately, those situations often originate when a research tool or platform technology developed at a university is exclusively licensed to a single industrial actor that does not make it available to others at affordable prices. This occurred, for example, with respect to cre-lox technology (19). Second, patent thickets—in which there are a large number of overlapping patents that cover a research area—may make it impossible to identify all the patent holders and negotiate licenses with all of them. Patent thickets appear to be emerging, for example, around human embryonic and induced pluripotent stem cell lines (20), and this concern is of particular relevance with respect to microarray DNA testing and whole-genome sequencing, in which technology platforms aggregate data across large numbers of genes, many of which are patented (21). Although academic institutions are often the victim of the resulting high licensing costs, they also contribute to thickets by patenting products and processes with little or no commercial value (13).


One solution to the problems caused by patents is to enhance the “precompetitive” environment. Precompetitive refers to the time during R&D in which there is collaboration but no competition. Traditionally, pharmaceutical companies have been very conservative in defining the scope of the precompetitive space (13, 22). During this stage of discovery and development, IPRs lead not only to the direct cost of obtaining and maintaining IPRs but also to substantial transaction costs associated with the negotiation of licenses to protected technology—most of which turns out to have little or no commercial value—and the high costs of duplicating the research of one’s competitors because of the lack of access to their results. Given the high failure rate in drug discovery research, these costs represent a real drag on the innovation system.

On the other hand, the cost of managing knowledge within consortia is relatively modest and includes the funds required to establish the consortium and maintain its ongoing governance. Although there is a theoretical risk of free-riding on the knowledge so produced, this is fairly minimal in cases in which the major players—funders and companies alike—form part of the consortium and exercise a high degree of normative influence on the behavior of others (4). This is especially true when the precompetitive collaboration is structured so as to permit actors to appropriate knowledge and innovation that is closer to practical application (6, 23). In these cases, economic theory would argue that collective management of precompetitive knowledge is more efficient than are proprietary models (24).

The line between precompetitive and competitive research is in constant flux and has shown a tendency to move increasingly downstream toward clinical or therapeutic application up to and including proof of concept (22, 25), driven in part by the current trend for venture capital firms to invest later in the development process, after proof of concept in humans. This may mean that certain actors, whose business models rely on claimed proprietary rights in emerging precompetitive activities, may be left out. From the perspective of these actors, the space is competitive, not precompetitive; losing proprietary rights may well mean that these actors will need to find creative ways of becoming partners in what has become precompetitive, will need to change the focus of their activities, or may even go out of business. However, unless the policy of keeping competitively focused actors—most often small biotechnology companies—alive despite the inefficiencies of doing so is more important than increasing the efficiency of drug development (and few but the most hard-lined would so argue), then the loss of these actors is not only inevitable but, in the long term, beneficial.

Despite the challenges, some consortia have developed around the concept of the precompetitive environment (Table 1). Although most are relatively recent, some lessons may be gleaned from their experience. In general, these consortia reduce transaction costs for acquiring access to trained personnel, data, reagents, and technology and for sharing knowledge among members. To achieve these cost reductions, consortia require smoothly operating governance structures on the basis of trust among participants. Trust may be difficult to accomplish when dealing with a heterogeneous set of stakeholders because of the range of their incentives and motivations to participate (4). For example, small biotechnology companies may be reluctant to share or give up exclusive IPRs, as these often form their only major asset, whereas large pharmaceutical companies may be willing to do so because they rely on downstream patents for their livelihood. Academic researchers may be reluctant to have their research confined to the specific goals of the consortium or participate in activities that may have a negative impact on key performance metrics that are publication- and increasingly IPR-related. In addition, few institutions have reward structures that recognize the substantial time and energy commitments that are necessary to create, maintain, and govern consortia.

Precompetitive research collaborations require clear negotiation around IPRs as both inputs and outputs of the collaboration. Models range from open, with virtually no IPRs, to differential IPRs for participants versus external actors, to highly proprietary models with one actor holding control (Table 1). Some research collaborations, such as the Structural Genomics Consortium (SGC), are set up on an open-access model through contract and social norms. Inputs are shared within the consortium, and all outputs are placed in the public domain concurrently with delivery to consortium members (4). Evidence suggests that the SGC model of providing research tools (chemical and biological probes such as cell-permeable inhibitors of protein function and antibodies) to the community expands research to new therapeutic targets, at least in one family of proteins (human protein kinases) (26).

Similarly, community resources such as large-scale databases and repositories either facilitate the entry of data and reagents into the public domain or manage and distribute these by using the least restrictive terms possible. For example, the Jackson Laboratory (JAX) facilitates access to mouse models for human disease through prenegotiated agreements with donors of new mouse strains that protect academics from having to sign material transfer agreements (MTAs) or licenses. JAX distributes both patented and unpatented mice pursuant to these agreements to academic and not-for-profit researchers using a simple notification that the mice be used for research purposes and may not be sold or transferred to third parties without permission. Mice are distributed to industry or for commercial use if permitted by the donor via MTAs or licenses negotiated at arms’ length between the donor and industry recipient. In this way, JAX ensures that donors accept a research commons approach for academics in return for JAX acting as a bridge between donors and industry (27).

Some research collaborations are set up under open innovation strategies that rely on IPRs but facilitate knowledge exchange through broad licensing either within the consortium (“club goods”) or more generally. The Top Institute Pharma PPP is an example of a club good in which inputs are shared among consortium members, but outputs are proprietary, with ownership, royalty, and other structures prenegotiated among consortium members (4). Although less of a challenge to traditional models of IPR management, this approach nevertheless requires a shift in focus from the maximization of one stakeholder’s IPR position to maximizing value for the consortium as a whole by maintaining the free flow of knowledge within the collaboration.

The Innovative Medicines Initiative, Europe’s largest PPP for accelerating drug development and a joint undertaking between the European Union and the European Federation of Pharmaceutical Industries and Associations (EFPIA), operates under a similar model (28). Consortia members cannot enforce IPRs against other consortium members within a collaborative project. Instead, members agree to keep project-related information confidential from those outside the consortium so as not to jeopardize the grant of potential IPRs during the course of the research. In addition, consortium members must decide what to do with jointly developed IP, including whether and how potential IPRs should be disclosed to the consortium members; which consortium member(s) have priority over commercializing the IPRs; whether and how the IPRs will be licensed and to whom (for example, exclusively or nonexclusively); whether consortium members retain the rights to use the IPRs for research or other purposes; and how to allocate any revenues generated. All of this governance requires some sophistication in one’s understanding and management of IPRs that may result from collaborative research.

A hybrid strategy may be to adopt a “wait and see” approach to decisions about IPRs and release of data, reagents, or other outputs until they have proven utility for a specific project. Such a strategy would delay the seeking of IPRs but ensure that IPRs are only sought for outputs relevant to the goals of the consortium. It may shift the dominant IPR to, for example, the therapeutic product, which receives marketing approval, in contrast to current IPRs on the composition of matter that has poor odds (less than 1 in 10,000) of becoming an approved drug. Such a shift enables tools, compounds, and drugs that have failed at some point in the development process for a given indication to be rapidly released into the public domain without attendant IPRs (22). This option leads to the potential for repositories for these reagents. Nevertheless, this strategy requires a very strong governance mechanism, as disagreements over IPRs for real inventions are more difficult to resolve than are theoretical IPRs for an as-yet-undefined future invention. Projects without a strong governance structure and well-defined, project-enhancing principles underlying that structure are likely to end poorly. This IPR approach may also require incentives through the participation of diverse funders such as venture capital firms, angel investors, philanthropic foundations, and even sovereign wealth funds.

In selecting among these options, stakeholders must assess whether an elaborate IPR evaluation and management process with attendant legal and administrative costs surpasses any realistic financial returns to the consortium or its members derived from IPRs. It remains an open question which of these models will work best in which circumstances (4), and answering this question requires intermediate and outcome metrics designed to refine our understanding.


Although many of the emerging collaborative models appear promising, there are institutional impediments to rational discourse about IPRs. This is especially concerning because the innovation literature suggests that “differences in the growth rates of national economies are largely due to differences in the ‘social capability’ for institutional change, and especially to those processes of institutional change that are related to the organisation and advancement of knowledge” (29). In most developed countries, the institution that has been established to manage the shifting relationships between researchers and the private sector and to seek and manage IPRs that cover university-based, mainly publicly funded research is the TTO. TTOs worldwide are rewarded for commercialization activities as well as managing industry partnerships and sponsored research. The metrics used to assess the success of TTOs focus on intermediate input-output measures, such as number of disclosures of invention, number of patents filed and granted, number of spin-off companies created, and licensing revenues that reflect only a linear path for innovation; these metrics encourage the treatment of TTOs as profit centers for universities (30), but this proprietary model of innovation is becoming increasingly out of date. In reality, TTOs are costly: Sixty percent of American universities and half of British universities do not earn enough from their licensing activities to cover the costs of their TTOs, and revenues are highly concentrated at a few universities that have patented “blockbuster” inventions (31, 32).

What is of even greater concern is the reflexive way in which TTO metrics have come to dominate science policy at a broader level. This is not because these metrics accurately reflect the goals of innovation—whether economically or socially—but because they are easily synthesized and understood by institutional and governmental policy-makers (33). As a result, broad innovation policy focuses inordinately on proprietary models despite current experience and scholarship indicating that collaborations are more productive. Because we reward what we can measure, these metrics have the perverse effect of discouraging relationship building with industry, managing industry-sponsored or contract research, and collaborative research projects with industry and other partners.

However, the landscape is changing. Groups outside of traditional academic and industry research are becoming increasingly involved in translational biomedical research, which is by nature interdisciplinary. These groups include patient advocacy organizations and private foundations that provide research funding and are driving research agendas, funding translational research, and potentially building or participating in new collaborative models through innovative patient-led PPPs (34). There are strong arguments to be made that patient organizations or related not-for-profit organizations will increasingly act as intermediaries between the public and private sectors and will demand more than the status quo drug-development productivity.

At the same time, a number of policy initiatives are recognizing the need for metrics that capture the highly networked research environment. The development of such metrics to understand the emerging innovation ecosystem, supported by databases and analytical tools, should be a priority for governments. New tools for visualizing and analyzing complex systems, temporally and spatially, provide a more nuanced understanding of the networked and iterative processes of innovation (35). The new metrics are based on the wealth of data available on the Internet or in other digital records, and academic researchers as well as companies such as Thomson Reuters and Elsevier are adding analytical layers to public domain data, such as that found in patent and publication databases, which enable sophisticated analyses of innovation systems. These tools have the capacity to capture collaborations through co-authorship and co-inventorship, corporate structures, licensing deals, and history and to track tacit knowledge flows through the movement of highly skilled people (35). These tools are an excellent first step, but data quality is highly variable, and important parameters, such as co-funding and licensing data, are largely missing or limited. Initiatives to improve metrics as well as the quality and availability of publicly accessible data should be a priority for funders and governments.

Last, regulators have an important role to play in providing incentives for the collaborative use of information. Regulators such as the U.S. Food and Drug Administration house substantial information on compounds that failed in human trials—some for toxicity, some for failed efficacy with respect to the chosen indication, and some because changing business priorities caused the abandonment of further clinical development. With increasing fervor, calls are being made for placement of this information (36) in the public domain because of moral obligations to patients, especially clinical trial participants; benefits to companies wanting to repurpose drugs; or benefits to companies that seek to develop drugs in a class in which a predecessor, unknown to them, has failed. One proposed role for NCATS is to become the “honest broker” of such information alongside the compilation of information on drugs in development that is in the public domain but not aggregated for ready accessibility. Regulators can provide incentives to release information or can mandate release of data in the public interest with the support of lawmakers (37).


New collaborative models are emerging that have great potential to reduce the transaction costs for sharing of data and reagents. As these models come on stream, we need to develop a better understanding of the emerging landscape, to understand its deals and structures, its IPR management policies for inputs and outputs, and its licensing agreements. New metrics are needed that are able to capture the enhanced value of collaborative networks as well as milestones along the R&D development pathway, which is now understood to be complex and iterative.

Governments and other funders have an important role to play in facilitating the development, governance, and operation of these collaborative networks and in building the capacity, within publicly funded institutions (especially TTOs), to develop collaborations that are based on a realistic assessment of contributions and the value of IPRs. TTOs need to be supported to act as trusted intermediaries in facilitating collaborative networks. Finally, governments and funders need to support the development of the underlying evidentiary infrastructure and nuanced metrics on which to base innovation policies in an increasingly networked environment.

References and Notes

  1. Acknowledgments: We thank R. Adams and A. Zylstra for research assistance. Funding: The research of T.B. and E.R.G. in this field is supported by the Canadian Stem Cell Network, Genome Canada, and the Ontario Genomics Institute. G.A.F.’s work in this area is supported by grant UL1-RR-024134 from the U.S. National Institutes of Health. Competing interests: The authors declare no competing interests.
View Abstract

Navigate This Article