Nature Reviews Drug Discovery 4, 891-897 (2005); doi:10.1038/nrd1879
Outlook FINDING IMPROVED MEDICINES: THE ROLE OF ACADEMIC–INDUSTRIAL COLLABORATION Jaye Chin-Dusting1, Jacques Mizrahi2, Garry Jennings1 & Desmond Fitzgerald3 1 Jaye Chin-Dusting and Garry Jennings are at the Alfred and Baker Medical Unit, Wynn Domain, Baker Heart Research Institute, Commercial Road, Melbourne 3004, Australia. 2 Jacques Mizrahi is at F. Hoffmann-La Roche Ltd, Grenzacherstrasse 124, Basel CH-4070, Switzerland. 3 Desmond Fitzgerald is at Materia Medica, Mere Croft, Chester Road, Mere Knutsford, Cheshire WA16 6LG, UK. correspondence to: Jaye Chin-Dusting [EMAIL PROTECTED] This paper reviews models of academic–pharmaceutical industry collaboration and debates the value of such partnerships so that those contemplating an alliance can reflect a priori on the purpose, nature and process that will provide a constructive outcome. The scope is confined to the biomedical discipline, because collaborations in other fields, such as physics and engineering, have not suffered as a result of the concerns associated with biomedicine. The scientific community has, by nature, a strongly competitive culture. Competition for peer recognition, funding and professional preferment is embedded in academia, yet progress in modern scientific research is achieved predominantly by collaborative teamwork. In the early nineteenth century, the concept of fundamental research was exemplified, especially in Germany, as "a dedicated scientific pursuit of natural phenomena without seeking any practical application." This was the implication of the scientific philosophy of German Wissenschaft — that is, 'pure research'. This scientific philosophy is still widely held today. For example, Dean Ross of John Hopkins University, in reflecting on academic and pharmaceutical industry relationships, emphasized that academic scientists have as their goals the acquisition and dissemination of knowledge as full-time independent 'scholar scientists'. He contrasted this role with the pharmaceutical industry's goal of profit arising from the full-time employment of scientists1. Given these differing goals and cultures, it is obvious that the possibility for misunderstandings as to the nature of a specific collaborative venture is high. Advances in biochemistry and pharmacology from the 1930s onwards, and the creation of research departments within pharmaceutical companies, radically changed the balance of competence and scientific credibility, so that by the 1950s productive scientific research collaboration became feasible. However, during that decade, collaboration within the United States between industry and academia seems to have fluctuated in relation to the availability of public funding for biomedical research2. Blumenthal speculated that the increase in this type of collaboration was due to a reduction in grant-sponsored research. If this interpretation is true, it raises the important issue of what the real motivations were for such collaborations. If the motivation of academic researchers is a mixture of survival, desperation or even greed, then this does not bode well for the quality or outcome of the collaboration. Similarly, the motivation for industry to sponsor external research efforts, ranging from tax-benefits to marketing and public relations to genuine knowledge exchange, needs to be explored. It is therefore timely to reflect on the nature and purpose of scientific research collaboration between academia and the pharmaceutical industry, particularly in light of current concerns about its purpose and even validity3, 4, including views that the academic, government and industrial complex has a detrimental effect on the culture and motivation of academic research5-8. In this review we focus on the biomedical discipline. It would seem that academic–industrial collaborations in the fields of physics and engineering have generally enjoyed a constructive and fruitful relationship over many years, without raising the concerns associated with biomedicine9. Even this restricted focus provides a broad topic. Given the breadth of the subject, it is necessary to apply some definitions and constraints. We largely confine ourselves to collaborations in the preclinical, discovery phase rather than clinical development. By collaboration we mean "to work together, especially in a joint intellectual effort" or, less elegantly, "to team up"10. The terms 'relationship'3 and 'alliance', on the other hand, although suggestive of a certain connectivity, do not necessarily imply "a joint intellectual effort". It is therefore important to reflect on what the reasons for collaboration are, because these will usually determine the motivation and structure of any collaboration. We come back to this fundamental question in the section on areas of potential discord. A question for the future is how academic–pharmaceutical industry research collaborations should be arranged in order both to meet society's needs for advancing knowledge towards novel and improved medicine and to reassure those constituencies that decry such arrangements? Such issues are explored in greater detail in the later sections of this review. Before that, however, we outline some common academic–industrial relationships. Types of relationships There are many variations of academic–industrial relationships, ranging from expert individual consultation to academic–industry–government liaisons. Models which fall under the 'collaboration' rather than 'relationship/alliance' description are asterisked. Consultations and fee for service, including contract research outsourcing. Under this model, academics/clinicians consult for industry for a stated term on a particular project. The expertise delivered varies from technical expertise and clinical know-how to the execution of clinical trails, and in principle does not affect core research activities, although there is potential for these relationships to be a strong motivator of curiosity-driven research. The benefit for industry is the accruement of data/expertise and, for the academic, supplemental funding. Publications may or may not be generated. The relationship is usually considered a personal one between the academic/clinician and industry, as there may be major risks if the academic employer is also dealing with the industry sponsor at other levels in the organization. For example, academic employers may perceive these arrangements as a source of leakage of institutional intellectual property. The potential for conflict of interest under this model is quite high and pertains to both the clinical and preclinical sectors, but is particularly high with academics, who may also be prescribers. Transparency, clarity and formal documentation are crucial. Competitive grants sponsored by industry. Competitive grants sponsored by industry are similar to grants from other funding bodies in many respects. Conditions do, however, vary on a case-by-case basis. An example of such a grant is The International HDL Research Awards programme from Pfizer11, under which the scope of the research to be funded is clearly stated and usually aligns with the interest of the company. This can bring the advantage to the company of generating awareness about new, as yet unestablished, areas of therapeutics. Under the Pfizer programme, publication and scientific presentations of result findings are encouraged, with the caveat that the awardee and host institution pre-discloses any invention from the work conducted at least 30 days before publication or public disclosure. In the event of a new discovery (not involving the company's pharmaceutical tools), the host institution is generally responsible for applying for a patent on behalf of the investigator and the company has right of first refusal. Industrial sponsorship of training and education programmes. These range from sponsorship of postgraduate students to postdoctoral fellows. There are generally two types, those in which the students/postdoctorates work within industry and those in which the students/post-doctorates work in an academic environment but whose stipend comes from industry. The latter is usually a good-will mechanism with few caveats. The former offers invaluable insight and training to the young scientist and is exemplified by the Roche Palo Alto post-doctoral programme12. Interestingly, some companies view this as an excellent tool for future recruitment of their in-house scientists, whereas others have a firm policy not to recruit these trainees into permanent employment. An innovative programme initiated in 1999 in Roche13, and still on-going, uses an intense 3-day symposium to provide selected promising early-stage chemists and biologists14 with direct insights into modern pharmaceutical research. At the same time as promoting networks and providing a forum for idea-exchange, these symposia have also been highly successful in recruiting talent into the multi-national company. Industrial sponsorship of investigator-led research*. A company sponsors a part or a whole of an academic's research (usually on a project-by-project basis but can be a wider programme). The academic institution negotiates, administers and supports the partnership. Conditions vary widely as such partnerships are individually tailored. Some institutions have now set up processes that are geared towards 'match-making' academics within their walls with the appropriate companies/executives. Such examples include the Massachusetts Institute of Technology's (MIT) industrial liaison programme15 and the Strathclyde Institute for Drug Research16, which presents opportunities for industrial liaison with the University of Strathclyde. Institute–institute liaisons*. These partnerships develop on the basis of joint key strategic areas. They work best when longer-term negotiations are agreed on and develop from the bottom up, with genuine intellectual interest emanating from both sides and developed over many years, perhaps via smaller-scale industrial sponsorship/alliances. Equity is shared, trust developed and enough time allowed to pass to develop intimate know-how of the partner's strengths and strategic plans. The 1994 alliance between MIT and Amgen17 is often touted as the benchmark example of such an alliance. Under the terms, Amgen had certain patent and technology-licensing rights to findings resulting from the collaboration in exchange for funding up to US$3 million a year for up to 10 years. Academia, industry and the government*. The establishment of joint academic–industrial alliances that use their collaborative efforts to leverage funding from government has been actively encouraged by many government-directed initiatives. Examples range from funding targeted at exposing young scientists to both academia and the industrial sector from the British Biotechnology and Biological Sciences Research Council (BBSRC)18 and Australian National Health and Medical Research Council (NHMRC)19 to large-scale academic–industry–government initiatives. A recent example of the latter is an award recently granted to the Scripps Research Institute, La Jolla, USA, in collaboration with Novartis Pharma AG from National Institutes of Health (NIH) funds20. The programme is targeted towards developing new treatments for depression and nicotine addiction such that the work is divided between Novartis and Scripps Research. Novartis conducts the molecular biology studies and medicinal chemistry, as well as the traditional aspects of drug discovery and design. Scripps focuses on behavioural testing of the compounds. In general, the strategic logic is that if promising leads are identified, then the industrial partner will devote more resources towards getting them more rapidly into clinical care. Academic spin-out companies Companies spinning out from academia (spin-out companies) are a phenomenon of the past two decades and are a completely different species of the academic–industrial alliance. The concept is best exemplified by the origin of Genentech21, founded in 1976 by University of California biochemist Herbert Boyer (who together with Stanford University geneticist Stanley Cohen pioneered the field of recombinant DNA technology) and the 28-year-old venture capitalist Robert Swanson. Genentech went public in 1980, raising US$35 million with an offering that leapt from $35 a share to $88 after less than an hour on the market. The first recombinant DNA drug they marketed (human insulin) was licensed to Eli Lilly in 1982, and in 1985 they were the first biotechnology company to manufacture and market the first recombinant pharmaceutical product, somatrem (a growth hormone for children). In 1995, Genentech merged with Roche Holdings Ltd. Genentech paved the way for the current global explosion of academic spin-out companies. The estimated number of new gene patents granted alone increased from 400 in 1990 to 2,800 in 1999, with the universities' share increasing from 55% to 73%22; 5,500 thousand new US patents were filed between 1991 and 1999, and in 1999 alone 344 new companies were formed from these licensed technologies. A key enabling factor was the Bayh–Dole Act of US congress23, which enabled universities to capture and develop intellectual property irrespective of whether the research leading to its development had been government funded. Companies spinning out from academia theoretically allow the academic scientist to maintain (some) ownership and management over their intellectual property. The relationship between the academic and the spin-out company is completely separate from that between a spin-out company and a bigger (industrial) partner. As such, the spin-out company can be viewed as the middle man, at best, bridging the gap between academia and industry, and at the very least, acting as stop-gate for the flood of 'therapeutically applicable' information reaching the market. Who really pays for this sifting of information can be debated. Attitudes to biomedical collaboration Perceptions of the desirability of collaboration between academia and industry are as varied and numerous as the individuals who hold them. Most of the assumptions are long-standing and include views that it is the business of academia to make original discoveries, irrespective of their utility, and that of industry to make medicines. Another common view is that the driver for academia is the 'publish or perish' ethos, whereas that of industry is to maintain competitive advantage, which in turn necessitates secrecy. A key concern is that any industrial support necessarily limits academic freedom and provokes conflicts of interest that are detrimental to the individual or society at large. Opinions on the management of academic associations with industry are well debated and articulated. Some hold the view that a properly managed, symbiotic, seamless relationship between the two sectors is a vital and necessary component of producing new medicines and other advances24-27. Others espouse a model in which there is no industrial partnership: academia forges on independently to translate basic findings into new drugs28. Still others call for a clear and separate divide so that academics, and clinical scientists in particular, can reclaim their ethical priorities29. As with all perceptions and opinions, no one is necessarily right or wrong. However, the reality is that industrial–academic alliances are extensive and likely to increase steeply. In 1940 in the US alone, 50 companies were reported to support 270 biomedical research projects. In 1994 (the most recent year for which such data is available), companies were found to be supporting an estimated 6,000 projects in the biomedical sciences in American universities3. In a 2001 review22 it was calculated that a substantial portion of the total US$55–60 billion spent on industrial R&D goes towards supporting research in academia (typical figures ranging between one-fifth of a large company's R&D funds to more than half for a small company), compared with the total US federal spending (2000) of $25 billion and $8–10 billion of private foundation funds. Unless a large increase in governmental funding occurs, as was observed after the Second World War3, this extensive relationship between the two sectors is likely to remain. In this section we review the facts behind some of the perceptions relating specifically to the sources of improved medicines and issues surrounding the secrecy of scientific observations. Who really contributes to the creation of 'innovative' drugs? In 2001, the consumer watchdog 'Public Citizen' released a report which, together with a separate report from the Joint Economic Committee (JEC) of the US congress, created a maelstrom of controversy amid claims that tax-payer funds (nominally via the NIH) support the discovery of the highest-impact medicines, absorb much of the risk and costs, yet the profits from drugs remain in the hands of industry. In contrast, the NIH in 2001 reported that of FDA-approved medicines with >US$500 million per year sales in the US (47 drugs), only 4 were developed with patented technologies for which the government has use or ownership rights. These issues are well described by Reichert and Milne30. So who really contributes to, funds and/or can take intellectual credit for the making of high-impact and/or innovative medicines? In one of the most comprehensive analysis of its kind, Maxwell and Erkhardt31 traced the histories of 30 lines of research that led to the discovery of 32 drugs deemed innovative by a poll of clinician–scientists who were experts in the areas covered (including cardiovascular, pulmonary, gastrointestinal and renal). Figure 1 summarizes the contribution of the type of institutions at which the pivotal work was performed. By this account, industry (53%) and non-industry (47%) contributions are roughly equal. The authors concluded that industrial contribution (which included unaided discovery of 12 of the 32 drugs) was substantial, and was involved in the development of at least 75% of the drugs. The contribution of non-industrial entries was also highly significant: universities alone made a major contribution towards half of the drugs discovered. We suggest that the majority of high-impact, innovative drugs come about because of significant, synergistic efforts from both sectors. This finding was recently supported by a separate analysis by Kneller32 who, in an attempt to determine the origin of all new molecular and biological entities approved by the FDA from 1998 to 2003, concluded that linkages between universities, biotech companies and pharmaceutical companies are important for the discovery and development of new drugs. As Reichert and Milne30 demonstrated in their re-analysis of the relationship between the private and public sectors in drug discovery and development, of 21 'impact' drugs selected by economists Cockburn and Henderson (many of which align with those discussed by Maxwell and Erkhardt31) the process "still starts with good science and ends with good medicine." Although the public sector still largely provides the key enabling discovery, the actual drug is, more often than not, first synthesized or isolated by industrial scientists. Does industry inhibit dissemination of information? An often voiced and highly significant obstacle for many academics in its alliance with industry is that the latter is cloaked with secrecy, which severely limits academic freedom and, in the worse-case scenario, creates ethical misalignments. An example of the potentially harmful effects of academic–industrial alliances was the disclosure dispute in the Olivieri/Apotex debacle33. In this case, the clinical investigator (Olivieri) was sued by the company (Apotex) for premature disclosure of adverse events, which raises valid concerns about the interface between the public and private sector and the constraints one culture can impose on the other. A more common concern is that alliances with industry severely limit the ability to publish research findings in a timely manner. Given that scientific freedom and dissemination of new knowledge is of paramount importance to most academic scientists, this obstacle is often perceived as insurmountable. Although perhaps true up to the mid-1980s, this perception is, however, naive in light of the passing of the Bayh–Dole Act in the US23, which allows the transfer of ownership and patent rights to universities on inventions that were discovered or launched on federally funded schemes. In essence, with or without an alliance to industry, the academic scientist now has a responsibility to address disclosure issues prior to any public dissemination of information. The topic of secrecy in medical research has been explored by Rosenberg34, who describes the pervasively negative effect of secrecy even within academia itself. Rosenberg's view is that it is not uncommon for scientists to "maintain their competitive edge over other workers in the field or to protect future patent subjects" and he concludes that "Secrecy in science is understandable, but it is not justifiable." A current positive example in which an academic–industrial collaboration is trying to re-dress the problem of secrecy and confidentiality in a collaboration involving industry-funded research is the Broad–Novartis initiative, which provides a novel and enlightened joint approach to address a pressing public-health problem (Box 1). This not withstanding, in the interest of examining industry's contribution to high-impact publications, we conducted a snapshot analysis examining the contribution of industrial (that is, any profit-making institution) and non-industrial support of publications in two of the top (based on current impact factors) international medical (general) journals: the New England Journal of Medicine (NEJM) and the Lancet (which was included at the expense of the second-rank US journal, the Journal of the American Medical Association, to remove a US bias) and the top medical (research and experimental) journal, Nature Medicine. All original papers published in these journals in 2003 and 2004 were scored according to reported funding and/or support source (Fig. 2). Although the contribution of industry to publications in Nature Medicine is insubstantial, this sector does contribute noticeably to publications reported in the Lancet and the NEJM. The former had an average industrial contribution of 21% in 2003 and 2004. It is timely at this point to make the distinction between basic discovery research and clinical trials, in that the latter arguably pertains more to drug evaluation rather than drug discovery. It is likely that publications in both the Lancet and NEJM demonstrate a larger impact and influence of industrial support because both are general medical journals directed at the more 'applied' end of therapeutics. One interpretation of the data presented above is that industry contributes more to the development and evaluation end of drug discovery. On the other hand, it might be that greater secrecy shrouds the basic end of drug discovery when industry is involved. Bekelman et al.35 report that there is a statistically significant association between industry sponsorship and published pro-industry conclusions, which is relevant in this regard. Whether this reflects the broader issue of positive outcome bias in publications36 or that industry uses publications to self-promote is difficult to quantify. It is also difficult to quantify the delay in publishing new findings, and the nature of restrictions placed on publications, resulting from industrial associations. Areas of potential discord Mainschein10 suggests that a collaboration is when "individuals have come together in pursuit of a shared goal." She subsequently suggests that "co-labouring towards a common goal is not necessarily sufficient for all to be equal partners or full collaborators and indicates that a distinction should be drawn between primary collaborators, that is, those who define the goal and secondary collaborators who contribute to the work." Such an analysis leads to the creation of a hierarchy of workers with differing degrees of responsibility and recognition of credit. Failure to articulate these subtle distinctions in roles could lie at the root of many subsequent failures of collaboration. In this section we reflect on certain principles of cooperation between individuals or groups in academia and collaboration. The advice for those about to enter into this (and any other) form of research collaboration is based on Axelrod's39 seminal book The Evolution of Cooperation. On the basis of an analysis of a computer tournament based on the Prisoner's Dilemma, the advice is to not be envious nor too clever, and do not be the first to defect. Instead, reciprocate both cooperation and defection. Axelrod's results suggest that the bases of rewarding research collaboration should be fairness and trust. The common sources of discord that apply to any form of basic research collaboration, including misunderstandings/arguments over interpretation of scientific data, misuse of research funding and arguments over authorship, priority or sequence, also apply in the context of academic–industry collaboration involving preclinical research. In addition, there are a number of issues that pertain particularly to the academic-industrial partnership: misalignments in time-lines; poor communication between two sectors which can, and often does, use different terminology; withholding of information, resulting in unnecessary duplication; uncertainties over the coverage of the intellectual property agreement; failure to declare or recognize competing interests; and lack of clarity in the termination process. Biomedical research collaborations involving human subjects pose additional difficulties. The involvement of patients and other families in the various forms of research can be a source of serious dispute, particularly when they have not been adequately informed and consulted. Studies involving genetic screening and other uses of human tissue with commercial potential are a specific cause for concern and a frequent source of misunderstanding. The prospect of significant financial rewards, either for clinical scientists and/or their institutions, has to be balanced against the primary purpose of academic institutions, which is teaching and research in the public interest. A fine balance, therefore, needs to be struck between legitimate financial rewards and the value system of academic institutions. This sets the scene for much potential discord between the institutes' and individual investigator's value systems and the attitude of the industrial company. In a recent paper, Cho40, in summarizing the results of a survey of 100 universities in the United States, detected a marked lack of uniformity concerning the types of academic–industrial relationships that were considered acceptable. As a consequence, a series of initiatives have been developed, including the Research Revitalization Act of 2002, to improve the detection and monitoring of potential conflicts of interest relating to academic staff involved in research on human subjects. Creating a cooperative environment The key elements promoting effective collaboration are trust, mutual scientific respect and concomitant scientific goals. However, there is a long tradition of distrust about the motivations for collaborating on the part of academia and industry. The distinguished American pharmacologist J. J. Abel wrote that "a pharmacologist of any training or ability should have so many problems of his own awaiting solutions that he should not spend time on matters of little theoretical importance for his science." He went on to spurn the idea of patents, a view shared by many medical scientists at that time. A consequence of such views was that industrial pharmacologists were barred from membership of the American Society for Pharmacology in experimental therapeutics until the early 1940s5. Any scientist considering initiating an academic collaboration with the pharmaceutical industry should understand that human factors and behaviour are the major determinants of a satisfactory and constructive outcome. In this section we consider situations in which the potential participants are not familiar with each other. As in so many social aspects of relationships, the initial phase is characterized by enthusiasm and optimism about the prospect of joint intellectual endeavour. Although such attitudes are meritorious, they can lead to erroneous assumptions on either side as to the core motivation underlying the collaboration. At the initial discussion a practical approach should be taken, which may help to identify potential problems or misconceptions. At such a meeting, the following points should be addressed: define the scientific question being asked; identify the respective tasks and roles; agree on the possible structure of the experimental protocols; agree on the policy on authorship sequence and the timing of any publications; the equitable sharing of data; intellectual property and royalties; the criteria for terminating the project; and the possibility of obtaining an unfavourable result. Although such an approach may seem pedestrian and an inhibitor of creativity, it is designed to seek out those human factors and assumptions which, if not identified at an early stage, can lead to conflict of varying severity. In exploring the expectations and support of the private industry, Mueller41 further stresses the importance for a recognition of the intrinsic differences in industrial and academic agendas. First, the curiosity-driven component of academia and the purpose-driven aspect of industry must be mutually recognized and respected. This requires compromises to be made on both sides — for example, academic groups should be free to follow-up scientifically rewarding side-tracks in parallel to agreed tasks. Second, particularly when young investigators (undergraduate, graduate or postgraduate) are involved, it should be recognized that such investigators will need to be educated to the fruition of their project, and that co-supervision from both sides needs to be encouraged to foster a balanced mentorship. As such, when a project needs to be terminated prematurely following rational assessment of results and goals, it is imperative that industry is sensitive to the vital element of the education-through-research mission of academia and therefore tolerate a mid-term, rather than immediate, smooth conclusion of the project. Perhaps one of the more difficult aspects of establishing cooperation is the effectiveness of a university's technology transfer office and its interaction with the company's legal staff. It is essential that the academic researcher works closely with those involved with the technology transfer group so that the science underpinning the collaboration is fully understood. The involvement of accountants and lawyers from both sides of the divide can lead to protracted and demotivating delays. Such delays can lead to a rapid loss of interest by either or both parties. Participants in such negotiations could learn much by studying the process and outcomes involved in the Prisoner's Dilemma. Case studies The following case studies provide examples of a more constructive outcome. The first example typifies a bottom-up approach initially based on mutual intellectual interest and recognition of each partner's ability to fill a gap in the strategy of the other (Box 2). The second example (Box 3) was initiated by the Strathclyde Institute of Drug Research (SIDR) in 1988 and has generated a positive financial income every year. A recent attitudinal survey of companies expressed their strong approval of the SIDR process and functioning. Discussion This review has so far focused entirely on the relationship between academia and industry in the context of biology and clinical science. In our experience, the interface between academic and industrial departments of chemistry seems to be qualitatively different — in general, the attitude towards industry is very positive. For example, professors of chemistry often propose their best graduates for positions in industry and such graduates tend to have a genuine motivation for such roles, in contrast, it may be said, to biologists (and particularly clinicians), who tend to view a role in industry as a second-best career option. As a consequence, many in industry describe collaboration with chemistry departments as synergistic — from the perspective of chemistry departments industry is not principally a source of research funding, but rather a true collaborator trying to resolve different facets of intellectually demanding problems. In their analysis of university–industry interactions in Germany, Meyer-Krahmer and Schmoch42 report that the fields of mechanical engineering and chemistry track best in terms of patent applications, an indication of successful industrial linkage. Of interest is their interpretation that chemistry performs well due to its explicit focus on basic science, whereas mechanical engineering lies at the other end of the spectrum in that it is problem-oriented and less scientifically or theoretically based. A common denominator, however, is that both are old industries in terms of industrial life-cycles, with many mature subsections. Perhaps it is only a matter time before the difficulties faced by the biological sciences and industry are resolved. The future challenge is to create an environment in which the balance between the need for academic science that is appropriately detached and unbiased, and the longer-term goals of society for the development of improved diagnostics and medicines requiring corporate resources, can be organized and managed to mutual and community benefit. Writing about research, innovation and university industry linkages in the early 1980s, Prager and Omenn43, based in the Office of the President (US), concluded that "although substantial institutional and attitudinal barriers to such relationships exist, the potential benefits are sufficiently compelling to engender confidence that those barriers can be surmounted." Twenty-five years later, these optimistic views have yet to be fully realized and many might say today that there are more problems than solutions. The basic issue is whether academic institutes should become more 'engines to drive the economy' or retain the original values postulated by Merton44, who defined the norms of universalism, communalism, disinterestedness and organized scepticism. The outcome of such behaviour amongst the scholars was the creation of a freely available pool of knowledge. The wish now to fully exploit such knowledge, particularly in the field of molecular biology, is posing a serious challenge to these established principles. The lure of potential additional income for the university is legitimate, but should be kept in proportion. The Association of University Technology Managers in the United States showed in their survey that some US$1 billion of royalty income was generated in 2002, but this is modest compared with the total revenue of more than $200 billion in grant income and more than $30 billion in contract income. In that same year, there were 6,500 patent applications and 3,700 licensing deals. Paradoxically, Levine and Boldrin45, in an article on the economics of ideas of intellectual property, conclude that "it would be best to eliminate patents and copyright altogether." They propose that governments should provide extra incentives for invention and creation by means other than the granting of patent monopoly. Even if such policy is not feasible, academic institutions should be much more self-critical in the patents they seek because of the very significant costs involved. Perhaps such institutions should create an internal audit or 'due diligence' analysis of proposed patents in order to improve the value of such activities. Although the conflicting views on the future strategy for patenting are beyond the scope of this article, such considerations do directly influence how future academic–industrial collaboration will develop. This is particularly important in the context of the need of drugs for neglected diseases. Groups such as Global Alliance for TB development and the Medicines for Malaria Venture have been set up to address this worldwide need. The creation of public–private research partnership in antimalarial research has been impressive, as exemplified by the collaboration between GlaxoSmithKline and a special programme for research and training in tropical diseases (TDR)46, 47. Progress in identifying new chemical leads is encouraging. Nevertheless, major hurdles remain in the form of lack of professional resources for managing novel drug development, which requires toxicological evaluation, dose-ranging studies, drug-combination formulations and drug–drug interaction studies. These are the well-established skills in the pharmaceutical industry and are mentioned to emphasize that whatever the currently vocal critics of the pharmaceutical industry may claim, there is at present no alternative effective human institution that can match their efficiency and willingness to risk the required investments. The issue of affordability in public–private partnerships was discussed at a meeting of the Geneva-based Initiatives on Public–Private Partnerships for Health (IPPH) meeting48. Sustained donations in excess of US$1–2 billion will be required for the marketing of improved treatments for tropical infectious and parasitic diseases. Although the pharmaceutical industry has lost interest in tropical diseases, perhaps this is a field in which pharmaceutical companies could re-examine their current business models and initiate large-scale academic–industrial collaborations that in the long term would result in correcting the current imbalance in developing countries between health, poverty and economic development. Kramer and Glennerster in their book Strong Medicine: Creating Incentives for Pharmaceutical Research on Neglected Diseases49, provide a realistic model based on 'push' and 'pull' research incentive schemes that would encourage manufacturers to undertake projects with seemingly poor financial rewards. Such programmes will require collaboration between academia and the pharmaceutical industry, whose success might be enhanced if applied in a considered manner. Boxes Box 1 | An enlightened approach: the way forward? In October 2004, the Broad Institute of Massachusetts Institute of Technology, Harvard and Novartis Institutes for Biomedical research (the research division of Novartis) announced a new model for public–private collaborations called the Broad–Novartis Diabetes Initiative37 aimed at placing all findings about type 2 diabetes onto the worldwide web38. With this first of its kind alliance, Mark Fishman, president of Novartis's biomedical division and a former Harvard professor, is on record as saying "I'm doing this to make a statement... that the patient should come first." The deal funnels US$4.5 million of Novartis funding over 3 years, with Broad contributing its high-end genetic equipment and the expertise of its 149 scientists. DNA samples, provided by Lund University in Sweden, are also a major component in this venture. The team forswears filing patents on the database; others, however, are allowed to patent a new therapy or diagnostic application based on the shared information. There are no restrictions or delays on publications. Box 2 | Case study: the Roche (Basel)–Baker (Melbourne) collaboration The Roche–Baker postdoctoral programme, fellowships for which were advertised in Nature in 2003, targets the recruitment of outstanding young scientists interested in exploring both the academic and industrial environments for future career options. The postdoctoral fellow is co-supervised by senior investigators from both institutes, and is allowed to work in a pre-defined area of research that is of mutual intellectual and strategic interest. The postdoc also has access to the expertise and resources of both environments, and presents regularly to teams from both sides. The co-supervision model is crucial in that it allows intellectual ownership of both parties. The slight variation of this programme from many of the other industrial postdoctoral fellowship programmes available (see text for details) is that this programme is under the stewardship of a single division within Roche and the Baker Heart Research Institute, Melbourne. This has enabled the concentration of several postdoctoral fellows under a single programme and allowed a critical mass of support and interest from a number of senior investigators from both sides. A steering committee comprising the heads of both institutes/divisions oversees the strategic direction of the programme. Aside from the obvious benefits of such a programme for all the parties involved, academics often enjoy the process of working within big pharma and with industrial scientists, and the scientific creativity demanded by such collaborations. Box 3 | Strathclyde Institute of Drug Research Pharma Links is a recently created function designed to facilitate the exploitation of the biomedical strengths of the Universities of Glasgow and Strathclyde (Scotland). Its purpose is to encourage bioscience companies access to these resources and it ensures that the interest of collaborators are delivered on time and budget. References 1. Ross, R. Academic research and industry relationships. Clin. Investigat. Med. 9, 269–272 (1986). | ISI | ChemPort | 2. Blumenthal, D. Academic–industrial relationships in the life sciences. N. Engl. J. Med. 349, 2452–2459 (2003). | Article | PubMed | ISI | ChemPort | 3. Weiner, C. Patent and academic research: historical case studies. Sci . Technol. Human Values 12, 50–62 (1987). | ISI | 4. Swann, J. Co-operative Research in the 20th Century America (Johns Hopkins Univ. Press, Baltimore, 1988). 5. Angell, M. Excess in the pharmaceutical industry. CMAJ 171, 1451–1453 (2004). | PubMed | 6. Kassirer, J. P. These two make quite a team. Washington Post B1,B5 (2004). | PubMed | 7. Triggle, D. Patenting the sun: enclosing the scientific commons and transforming the university — ethical concerns. Drug Dev. Res. 63, 139–149 (2005). | Article | ISI | 8. Washburn, J. University, Inc: The Corporate Corruption of Higher Education (Basic Books, New York, 2005). 9. Blumenthal, D., Gluck, M., Louis, K. S., Stoto, M. A. & Wise, D. University-industry research relationships in biotechnology: implications for the university. Science 232, 1361–1366 (1986). | PubMed | ISI | ChemPort | 10. Mainschein, J. Why collaborate? J. Hist. Biol. 26, 167–183 (1993). | Article | 11. Pfizer Public Medical & Academic Partnerships (MAP) Grants [online], <http://www.promisingminds.com/GrantsProgType.aspx?ProgID=405&ProgAreaID=22&ProgTypeID=4> (2005). 12. Roche Postdoctoral Fellowships [online], <http://paloalto.roche.com/careers/PostDocsFellow.html> (2001). 13. Roche Symposium for Leading Chemists of the Next Decade <http://euroweb.roche.com/symposium/index.htm> (2005). 14. Roche Symposium for Leading Bioscientists of the Next Decade http://euroweb.roche.com/biosymposium/ (2005). 15. MIT Industrial Liason Program [online], <http://ilp-www.mit.edu/display_page.a4d?key=H1> (2004). 16. Strathclyde Institute for Drug Research [online], <http://www.sidr.org/> (2005). 17. Massachusetts Institute of Technology. MIT and Amgen announce biological research agreement [online], <http://web.mit.edu/newsoffice/1994/amgen-0330.html> (2005). 18. Biotechnology and Biological Sciences Research Council. How to get funding. [online], <http://www.bbsrc.ac.uk/funding/fellowships/Welcome.html> (2005). 19. National Health and Medical Research Council. NHMRC Industry Fellowships [onlien], <http://www.ro.mq.edu.au/fund/nhmrc/IndustryFellow/indflyer.pdf> (2002). 20. Scripps Research Institute. NIH funds Scripps Research–Novartis collaboration to target new treatments for depression and nicotine addiction [online], <http://www.scripps.edu/newsandviews/e_20031208/markou.html> (2005). 21. Genetech, Inc. Corporate overview [online], <http://www.gene.com/gene/about/> (2005). 22. Moses, H. & Martin, J. B. Academic relationships with industry: a new model for biomedical research. JAMA 285, 933–935 (2001). | Article | PubMed | ISI | 23. Council on Governmental Relations.The Bayh–Dole Act: A Guide to the Law and Implementing Regulations [online], <http://www.cogr.edu/docs/Bayh_Dole.pdf> (1999). 24. Krimsky, S. & Baltimore, D. The ties that bind or benefit. Nature 283, 130–131 (1980). | Article | PubMed | ISI | ChemPort | 25. Bearn, A. G. The pharmaceutical industry and academe: partners in progress. Am. J. Med. 71, 81–88 (1981). | Article | PubMed | ISI | ChemPort | 26. Molinoff, P. B. Common to both academia and industry: the challenge of discovery. An interview with Perry Molinoff. Mol. Interv. 1, 78–83 (2001). | PubMed | ChemPort | 27. Gelijns, A. C. & Thier, S. O. Medical innovation and institutional interdependence: rethinking university-industry connections. JAMA 287, 72–77 (2002). | Article | PubMed | ISI | 28. Alper, J. Drug development. Biotech thinking comes to academic medical centers. Science 299, 1303–1305 (2003). | Article | PubMed | ISI | ChemPort | 29. Gale, E. A. Between two cultures: the expert clinician and the pharmaceutical industry. Clin Med 3, 538–541 (2003). | PubMed | ISI | 30. Reichert, J. M. & Milne, C. P. Public and private sector contributions to the discovery and development of 'impact' drugs. Am. J. Ther. 9, 543–55 (2002). | Article | PubMed | 31. Maxwell, R. A. & Eckhardt, S. B. Drug Discovery: A Casebook and Analysis (Humana, New Jersey, 1990). 32. Kneller, R. The origins of new drugs. Nature Biotechnol. 23, 529–530 (2005). | Article | PubMed | ISI | ChemPort | 33. Nathan, D. G. & Weatherall, D. J. Academic freedom in clinical research. N. Engl. J. Med. 347, 1368–1371 (2002). | Article | PubMed | ISI | 34. Rosenberg, S. A. Secrecy in medical research. N. Engl. J. Med. 334, 392–394 (1996). | Article | PubMed | ISI | ChemPort | 35. Bekelman, J. E., Li, Y. & Gross, C. P. Scope and impact of financial conflicts of interest in biomedical research: a systematic review. JAMA 289, 454–465 (2003). | Article | PubMed | ISI | 36. Callaham, M. L., Baxt, W. G., Waeckerle, J. F. & Wears, R. L. Reliability of editors' subjective quality ratings of peer reviews of manuscripts. JAMA 280, 229–231 (1998). | Article | PubMed | ISI | ChemPort | 37. Braod Institute. Broad and Novartis announce collaboration to uncover genetic basis for type 2 diabetes [online], <http://www.broad.mit.edu/media/2004/novartis_1027.html> (2004). 38. Lawler, A. Diabetes research. Broad–Novartis venture promises a no-strings, public gene database. Science 306, 795 (2004). 39. Axelrod, R. M. The Evolution of Cooperation (Basic Books, New York, 1984). 40. Cho, M. K., Shohara, R., Schissel, A. & Rennie, D. Policies on faculty conflicts of interest at US universities. JAMA 284, 2203–2208 (2000). | Article | PubMed | ISI | ChemPort | 41. Mueller, K. in Proceedings of the Seminar of the Magna Charta Observatory: Managing University Autonomy 115–127 (Boronia Univ. Press, 2004). 42. Meyer-Krahmer, F. & Schmoch U. Science-based technologies: university-industry interactions in four fields. Research Policy 27, 835–851 (1998). | Article | ISI | 43. Prager, D. J. & Omenn, G. S. Research, innovation, and university-industry linkages. Science 207, 379–384 (1980). | PubMed | ISI | ChemPort | 44. Merton, R. Theoretical and Empirical Investigations (Univ. Chicago Press, Chicago, 1974). 45. Boldrin, M. & Levine, D. K. The economics of ideas and intellectual property. Proc. Natl Acad. Sci. USA 102, 1252–1256 (2005). | Article | PubMed | ChemPort | 46. Ridley, R. & Toure, Y. Winning the drugs war. Nature 430, 942–943 (2004). | Article | PubMed | ISI | ChemPort | 47. Fidock, D. A., Rosenthal, P. J., Croft, S. L., Brun, R. & Nwaka, S. Antimalarial drug discovery: efficacy models for compound screening. Nature Rev. Drug Discov. 3, 509–520 (2004). | Article | PubMed | ISI | ChemPort | 48. Sharp, D. Not-for-profit drugs — no longer an oxymoron? Lancet 364, 1472–1474 (2004). | Article | PubMed | ISI | 49. Kremer, M. & Glennerster, R. Strong Meedicine: Creating Incentives for Pharmaceutical Research on Neglected Diseases (Princeton Univ. Press, Princeton, 2004). Acknowledgements The authors would like to acknowledge the invaluable discussions held with G. Gromo (Switzerland), C. Johnston (Australia), R. Flower (UK), P. Vanhoutte (Hong Kong), A. Harvey (Scotland) and B. Duncan (Scotland). We also acknowledge input from E. Jones, A.-M. Jefferies and M. Vincent from the Baker Heart Research Institute for their background research efforts. Competing interests statement. The authors declare competing financial interests. ____________________________________________________ Quipo Free Internet - 2 email, 150 Mb di spazio web e molto di pi�. ADSL, Hardware&Software Online Store: http://www.quipo.it This E-mail was scanned for viruses by Declude Virus. -- www.e-laser.org [email protected]
