Reflections on the value of systems models for regulation of medical research and product development.

• Author: Bouchard, Ron A.

INTRODUCTION (1)

In a recent editorial in Science, (2) Bill Wulf used a systems ecology framework to construct a model for innovation in the life sciences. He defined an "innovation ecology" as the various "interrelated institutions, laws, regulations and policies" necessary to underwrite successful commercialization of publicly funded research through an "infrastructure that entails education, research, tax policy, and intellectual property protection, among others." (3) In this formulation, private intellectual property and regulatory (IPR) rights form the linchpin between innovative publicly funded medical research, reduction to practice of basic research by firms and university technology transfer offices, product approval and marketing by government and firms as well as public consumption of approved medical products. As such, 'large scale' IPR rights-intensive translational research and technology commercialization constitute important market push and pull levers for domestic governments and provide the legal and regulatory basis for the drug development cycle writ large. Even so, and as lamented by Wulf in his editorial, a narrow "one size fits all" IPR rights framework has the potential to stifle rather than encourage innovation.

Casting the innovation landscape as an open complex organic ecology rather than a closed historical linear model of basic-to-applied research (4) is consistent with newer more open-ended analytical models such as complex adaptive systems, (5) network dynamics (6) and systems dynamics. (7) These 'systems' frameworks view and model systems as dynamic, adaptive and indeterminate networks where the behavior of the system as a whole is governed by the ever-changing and non-linear nature of the connections between actors and institutions rather than as a predictable sum of a set of linear deterministic nodes. At the heart of the functioning of a complex adaptive system is the number and nature of the interactions between network nodes, which produce novel and ever changing properties as the layers of complexity increase. This dynamic structure-function relationship of complex systems is nicely summed up by the phrase "more is different." (8)

One implication of a systems view of IPR rights-intensive innovation in the medical and life sciences is that local innovation ecologies are collapsing globally. (9) This is due, among other things, to the global reach of patent decisions of first instance such as that in KSR International Co. v. Teleflex Inc., (10) harmonization of regulatory processes and standards, such as those relating to biomedical product approval, marketing and patenting, adoption of international IPR rights-sensitive instruments such as the WTO's Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) and, less obvious, the convergence of national science and technology (S&T) policies and normative behaviors aimed at commercialization of publicly funded medical research. Within the larger political and legal cultures of participating nations, there is an increasing space being carved out for translational research and commercialization.

Indeed many nations, including Canada, are in the process of implementing strong IPR rights regimes that explicitly encompass publicly funded research efforts in order to reproduce the phenomenal success of university technology transfer and commercialization in the United States. This effort is hardly unique to Canada. Not only are other jurisdictions attempting to emulate U.S. translational research, but the United States itself, self-reflective after 25 years of Bayh-Dole, (11) is seeking to identify new and improved ways of commercializing public research in the context of its public health mandate. In the context of this debate, one hears increasingly vocal deliberation over the value of closed IPR rights models.

PURPOSIVE POLICY

Despite the growing visibility of network (12) and other "systems" theories, (13) linear models of organizations and organizational change have and continue to dominate analyses of the behavior of individuals, groups and institutions and to provide the benchmarks by which both public and private ordering are gauged. (14) One of the major differences of linear and non-linear models is the narrow range and simple nature of the assumptions and operating conditions that characterize linear models, including the desired outcome of maximizing certainty and predictability. The most important limitation of linear models is that, despite the good intentions behind them and the accrual of knowledge in relation to discrete silos, they often inhibit or even prevent the very thing they seek to facilitate through their unintended consequences. This outcome, referred to as "policy resistance" by Sterman, (15) is only just getting onto the radar of key decision-makers. Opposition to the novel claims of systems frameworks may have occurred due in some part to the fact that experts are immersed (and therefore have a stake) in their own specialties and the fear that systems work lacks the required degree of scientific rigor in the face of "real life" complexity. A range of potential examples of policy resistance in the public health sphere include clinical trial design, (16) disease management, (17) responses to acute public health crises, (18) as well as a host of broader state endeavors involving policies relating to health economics, (19) innovation, (20) public health (21) and drug regulation. (22)

Nevertheless, there is growing acknowledgement of the drawback of these linear models, and the narrow range of assumptions that underpin them, due partly to a number of controversies surrounding therapeutic products, (23) food, (24) food containers, (25) and children's toys (26) that have been approved for marketing then later found to be unsafe. In addition, a spate of popular books describing the emerging importance of complexity and networks in the realms of medicine and biology, (27) physics, (28) mathematics, (29) the Internet, (30) business organizations, (31) branded products, (32) public policy, (33) and economics (34) have raised the level of public discourse on systems models considerably. Decision-makers are coming closer to accepting that the cost of studying risk management using novel, albeit confusing systems tools far outweighs the costs to society of failing to understand them. (35) Tellingly, the problem of policy resistance, and the potential of novel systems-based approaches to answer it, has been the subject of recent editorials in prominent journals such as Science (36) and Nature. (37)

As I will discuss today, a useful metaphor for a purposive systems-based policy development process is that of setting an Origami sculpture (representing purposive public policy) into a river and then "letting it go" without touching it versus letting it go then walking beside it with the intent of making sure it neither gets caught in various eddies and oxbows nor goes over Niagara Falls (system collapse or breakdown (38)). In the former scenario, control is past-oriented and the performance of the system is seen to be at the mercy of future events which may be either exogenous or endogenous to the system and thus seen to be either outside or within control of the system. By contrast, control in the "walking by" example is future-oriented in the context of certain system constraints (legal-democratic, economic, technological) and responsive to endogenous effects due among other things to various feedback loops, stocks and flows, time delays and non-linearities inherent to the system. Robust adaptive planning not only allows us to walk by our policy creations, but also helps us to be prepared no matter which way the river turns.

The goal of my research program is to determine how innovative therapeutic product development and regulation, and the national S&T policies that drive these processes, are linked and operate as part of a complex innovation ecology. We are particularly interested in the notion of a systems-based regulated Therapeutic Product Lifecycle (rTPL) as it is embedded within a larger public health discourse that in turn is constrained by prevailing legal and democratic norms. This presentation is intended to take a high level view of whether newer systems models, including network theories and complex adaptive systems theory, can be of use as a form of "alternative intellectual property" to policy-makers in their attempts to enhance translational research and technology commercialization. (39)

THE REACH OF UNCERTAINTY

The implications of complex adaptive systems theory for law, particularly areas of law that are strongly contingent on science, was first recognized by J.B. Ruhl. (40) Reading this and related work from the perspective of someone having spent nearly 20 years at the bench (41) prior to entering legal scholarship got me thinking about the value of systems approaches to the interface between medical science and the law. Particularly valuable areas of debate include the creative nature of breakthroughs in the life sciences and how these are parsed in innovation and litigation discourses, the methods and evidence with which novel therapeutic products are approved and regulated, how and why approval processes are becoming increasingly harmonized over time internationally, and the nature of national S&T policies being advanced for innovation and drug regulation in the context of globalization.

My own experience in the lab, likely little different from that of others, underscores the inherent uncertainty involved in "doing science." As discussed in the context of evidence given by expert witnesses in pharmaceutical litigation, (42) this uncertainty encompasses all phases of experimentation no matter how planned the work may be from a grant application or project management perspective. A great deal of effort is required to accept and mitigate the "voodoo" inherent in...