Medical-Journals.com
UK
Europe
USA&Canada
Aust.&NZ
Asia
|
|
NEJM
David Blumenthal, M.D., M.P.P. Relationships between academic institutions and industry in the life sciences continue to fascinate, perplex, trouble, and embarrass academic investigators, university administrators, industrial companies, and government policymakers. No matter how many solemn conferences are convened to address the issues raised by academic�industrial relationships, the controversy these relationships cause seems to persist and even grow. This situation reflects both the periodic eruption of scandal, such as that following the death in 1999 of a young volunteer during an industry-related clinical trial of gene therapy at the University of Pennsylvania,1 and the increasing complexity of the relationship. Recently, academic and government officials recognized that the tendency of universities as institutions to acquire financial interests as a result of academic�industrial relationships raises new questions that require careful scrutiny.2 The purpose of this report is to bring readers up to date on the status of relationships between academic institutions and industry and to provide a perspective on them that will enhance the understanding of future developments in this field. The stakes in these relationships are substantial. On the one hand, they promise substantial benefits to the nation's health and economy. On the other hand, they pose risks to human subjects of research and to the integrity and openness of the academic research environment. For the average clinician, this seemingly abstruse topic has implications for the availability of new tests and treatments for patients and for the likelihood that the clinical trials in which a clinician's patients may enroll will be, and will appear to be, as safe and unbiased as it is humanly possible to make them. This report explores the history of the relationships between academic institutions and industry in the biomedical sciences; the known effects of these relationships on science, medicine, and society; current efforts to manage these relationships; and likely future developments. A definition of academic�industrial relationships may be useful at the outset. For the purposes of this report, this term encompasses relationships between academic institutions (including major affiliated hospitals and research institutes) and for-profit corporations in which something of value is exchanged.3 The relationship may take the form of industrial support for research in universities, consulting by members of the faculty to companies, equity holding in companies by universities or their faculty members, support of university trainees (graduate students or fellows) by companies, the patenting and licensing of intellectual property by universities, and gifts from companies to universities, as well as combinations of and variations on all of the above. The History of Academic�Industrial Relationships Ups and Downs Relationships between academic institutions and industry in the life sciences first flourished in the United States between World War I and World War II. During this period, the pharmaceutical industry developed an independent research capability, and with that an appreciation of and need for the scientific expertise that universities could offer.4 Emblematic of the academic�industrial relationships between the wars were the agreements between Eli Lilly and the University of Toronto in 1922 and between Eli Lilly and the University of Rochester in 1931 to commercialize and distribute, respectively, insulin and the extract of liver that was used to cure pernicious anemia. By 1940, a survey by the National Research Council showed that 50 U.S. companies were supporting 270 biomedical research projects in 70 universities.4 After World War II, however, relationships between academic institutions and industry in the life sciences flagged. The decline reflected, to a large degree, the surge in federal support for biomedical research in universities, which greatly decreased the need for academic researchers to seek support from industry. In the mid-1970s, however, the budgets of the National Institutes of Health (NIH) temporarily ceased their dramatic growth, and inflation eroded the purchasing power of federal grants.4 For the first time in decades, academic investigators became interested in industrial support for their research. The passage of the Bayh�Dole Act in 1980 (named for its cosponsors, Senators Birch Bayh [D-Ind.] and Robert Dole [R-Kans.]), which permits universities to own intellectual property that is developed as a result of federally funded research, encouraged universities to work with industry. Universities could now claim a share of any profits resulting from the sale of the products of federally sponsored academic research, and industry had the expertise needed to commercialize such products. Simultaneously, the revolution in biotechnology sparked industry's interest in reestablishing relations with universities. Between 1980 and 1983, large pharmaceutical companies poured $140 million into research conducted at 13 universities.4,5 The resurgence of academic�industrial relationships in biomedical research caused considerable soul-searching among university administrators and faculty members6,7,8,9 as well as some outside scrutiny. In 1982, a then little-known Democratic congressman from Tennessee, Albert Gore, chaired hearings on the subject in which he questioned whether academic�industrial relationships in support of biomedical research would reduce both the openness of universities and their ability to provide policymakers with neutral advice.10 Additional congressional hearings in 1987 and 1990 were also critical of these relationships.11,12 At first glance, the controversy about the revival of these relationships in the life sciences may seem strange, because in the years after World War II, universities had continued to have close alliances with industry in other fields without causing such debate.13,14,15 If anything, the involvement of universities with industry during the 1980s was less intense in the life sciences than in the physical sciences, where such relationships had attracted little attention. In 1983, 43 percent of the faculty members in departments of chemistry and engineering at major research universities received support from industry for their research, as compared with only 17 percent of the faculty members in departments of the life sciences.16 However, the academic�industrial relationships in biomedicine aroused special concern, then as now, for one important reason. They had the potential in the short term to affect the welfare of human subjects of research and in the long term to affect the medical care of the public. This potential caught the attention of both the public and the media in ways that did not seem true for areas other than the life sciences. A dramatic example is the case in 1985 of Scheffer Tseng, a young faculty member at Harvard Medical School. Tseng was found to have made a substantial profit by selling equity in a company created to market a product that he had been testing in human research subjects at the Harvard-affiliated Massachusetts Eye and Ear Infirmary. Later, when the research underpinning the product was found to be flawed,17 questions arose in the media about whether the patients had been subjected to an experiment for commercial rather than scientific purposes. Such concern, however, did not prevent academic�industrial relationships from becoming an accepted and pervasive feature of American biomedicine during the last two decades of the 20th century. Studies have shown that by the mid-to-late 1990s, over 90 percent of firms that were conducting research in the life sciences had some type of relationship with a university and that about one quarter of the faculty members in the life sciences at major U.S. universities had received support for research from industry.18 The same studies showed that over 50 percent of the faculty members in the life sciences had consulted for industry and that about 7 percent of them had held equity in a company that was performing work related to the faculty member's research. A survey conducted in 1999 showed that 68 percent of the universities in the United States and Canada held equity in companies that sponsored research conducted by members of their own faculty.19 Public Promotion of Academic�Industrial Relationships The growth of academic�industrial relationships has resulted, in part, from continuing efforts by government, at all levels and across political parties, to stimulate such relationships. After World War II ended, the federal government recruited many American universities to a war against disease,20 and the growth of academic�industrial relationships can be seen, in part, as an inevitable consequence of that effort. The public invested in the biomedical research that was conducted at universities in the expectation that the resultant knowledge would lead to practical applications in hospitals, pharmacies, and physicians' offices. In a capitalist economy, to achieve this result required cooperation of some kind between the producers of knowledge (universities) and the users of knowledge (industry). In the late 1970s, senior officials in the administration of President Jimmy Carter concluded that such cooperation was needed21,22 and began the discussions with Congress that resulted in the Bayh�Dole Act of 1980. The act enshrined in federal policy both the opportunity for and the obligation of universities to commercialize their federally funded research in order to apply it.23 During the 1980s, another view of the potential return on the public investment in universities and their research took shape. At the time, international economic competition was intensifying, and U.S. economic growth and productivity were anemic.22,24 Policymakers began to see universities as vital contributors to this country's economic recovery and to its economic security in a global marketplace. The proliferation of biotechnology start-up companies in the mid-1980s seemed to validate this view, and state governments, especially in the slumping Rust Belt � including Ohio and western Pennsylvania � began to promote partnerships between academic institutions and industry in the area of biotechnology. The National Science Foundation did the same, through a series of programs to encourage such relationships in biology and other fields.25 Henry Etzkowitz argues that the enlistment of universities to promote national economic growth has resulted in a fundamental change in the role of the university in society. Academic institutions have become engines of entrepreneurship. According to Etzkowitz, "The idea of the university as an active force in firm formation and regional economic development has spread widely . . . throughout the U.S. and worldwide. . . . In the entrepreneurial academic model the university educates and graduates organizations as well as individuals."26 The implication of this model is that the university now has the explicit purpose of producing not only educated citizens and professionals but also new firms that will capitalize the knowledge they produce. As if to confirm this view, Charles M. Vest, the president of the Massachusetts Institute of Technology, said that Boston's universities were preparing to nurture the next round of entrepreneurship in their region. The proliferation of academic�industrial relationships in the life sciences exemplifies the American university's response to the needs and demands of the public and private constituencies that see academic institutions as a means to the ends of health and prosperity. In retrospect, the university would have had difficulty rejecting these demands. Nevertheless, academic�industrial relationships come with risks to the traditional values and practices of the university, which have had an important role in creating the very strengths and core values on which society has come to rely. To understand this tension, a brief review of the documented risks and benefits of these relationships is useful. The Consequences of Academic�Industrial Relationships An ample literature and a store of anecdotes attest to the potential scientific, health-related, and economic benefits of the relationships between academic institutions and industry in biomedicine and other fields. In the life sciences, these relationships provide additional support for academic research. In 1994, which was the last year for which representative data were collected, companies were found to be spending about $1.5 billion to support an estimated 6000 projects in the life sciences in American universities.27 This amount constituted about 14 percent of the total funding for academic research in the life sciences.28 The recipients reported that industrial support required less red tape than government funding, created scientific opportunities for their students, and even increased the prestige of their departments.27 From the standpoint of benefits to health and medicine, academic�industrial relationships have appeared to be effective in assisting with technology transfer. A study conducted in the early 1990s showed that the close collaboration between academic and industrial researchers was more effective in transferring technology than were the traditional forms of disseminating information, such as publication and presentations at scientific meetings.29 Researchers who have industrial support are more likely than those without it to participate in technology-transfer activities, such as patenting and licensing.27 An indirect indicator of the contribution of academic�industrial relationships to technology transfer is data on the rates of patenting and licensing of products and royalty receipts in universities. From 1991 to 2000, the number of patents granted to universities increased by 131 percent, and the number of licenses granted increased by 158 percent. The income to universities from licensing grew from $121 million in 1991 to $997 million in 2000. Because royalties accrue only from sales and because they constitute only a small fraction of the proceeds, the implication of this 724 percent increase in royalties is that commercial products developed in association with university research sold widely.30 These numbers suggest the potential economic benefits of the relationships between academic institutions and industry to the universities, the companies, and the larger economy. Studies of local markets in which there are vibrant biotechnology industries, such as Boston, the San Francisco Bay area, and San Diego, California, support the view that these relationships help to stimulate the development of "spinoffs" in the form of new companies and new patents that are filed by existing companies.31,32,33,34 Dazzling as those benefits may seem, the risks created by academic�industrial relationships have proved to be just as eye-catching. The risks take two forms and have somewhat different ethical and practical implications: on the one hand, the threats to human subjects of research, and on the other hand, the threats to the integrity of the research process and, thus, to the efficiency and objectivity of teaching and research in the biomedical sciences in universities. These risks arise from conflicts of interest. The traditional commitment of the biomedical researcher and of the academic institution to the welfare of patients and to both the integrity of scientific research and the values of the university comes into conflict with the new financial interests created by connections with for-profit companies. The interests that cause the greatest concern are those that result in economic gains to investigators and universities above and beyond reimbursement for the costs of research, such as the income that results from consulting, equity-holding, patenting and licensing, and combinations of these activities. Therefore, much of the public discourse about how to manage or regulate the relationships between academic institutions and industry has focused on the management and regulation of the conflicts of interest that these relationships commonly create. Concern for the welfare of human subjects of the research conducted by faculty members and by universities that have relationships with industry date back at least to the controversy at Harvard surrounding Tseng two decades ago. Such concern was dramatically revived in 1999, when Jesse Gelsinger, an 18-year-old man with an inherited enzyme deficiency, died after having volunteered to participate in a gene-therapy experiment conducted at the University of Pennsylvania. It turned out that the university and the director of the laboratory in which the study was being conducted held equity in a company that stood to benefit from the therapy being tested and that the study had violated several safety protocols.1 In February 2000, when opening the Senate hearings on the gene-therapy research in which Gelsinger had participated, Senator William Frist (R-Tenn.), then chairing the Senate Subcommittee on Public Health, noted that "these events pose serious ethical questions regarding the risks to patients who participate in these trials and the high financial interests at stake."35 Academic�industrial relationships create threats to the integrity of the research process through two mechanisms. First, research support from industry or through other compensated ties to companies may lead some academic investigators wittingly or unwittingly to bias their findings in the companies' favor. Several recent studies have provided evidence of such bias in biomedical research.36 Second, academic�industrial relationships seem to reduce the openness of communication within the research environment. In a survey of executives of biomedical companies, more than half admitted that their research agreements with universities included restrictions on communicating results.37 Academic investigators who are involved in commercializing their own work are far more likely than those not so involved to report that they withhold data from their colleagues and engage in other forms of secrecy.38 The media and medical journals have reported instances at the University of California at San Francisco, the University of Toronto, and elsewhere in which companies sought, with some success, to suppress or delay the publication of unfavorable results of academic research that the companies had supported.39,40 There is concern that, taken together, the risks that academic�industrial relationships pose to human subjects of research and to the integrity of academic investigation could undermine the public's faith in the university research enterprise, government funding for this research, and the willingness of patients to enroll in clinical trials. Managing the Relationships The apparent inevitability of academic�industrial relationships, together with their documented risks, has led to a consensus among representatives of universities and government officials that these relationships and the attendant conflicts of interest must be managed more carefully than they were in the past. However, moving beyond the general discussion to broad implementation of specific policies and procedures has proved to be challenging. Acting on their own, a number of universities have aggressively managed their relationships with industry for some time. Notable in this regard has been the development by Harvard Medical School of rules (now under review) that were formulated after the Tseng episode. Harvard Medical School now requires universal disclosure of relationships between its faculty members and industrial organizations and prohibits faculty members from having "significant" financial interests in companies that support clinical or nonclinical research within the university. The federal government has also addressed academic�industrial relationships in the life sciences in several sets of rules. Since 1995, the NIH has required academic investigators who are receiving funds from the NIH to disclose certain potential conflicts of interest to their home institutions, and it has required those institutions to ensure that such potential conflicts of interest are reduced, managed, or eliminated. Since 1998, the Food and Drug Administration (FDA) has required researchers to disclose potential financial conflicts of interest to industrial organizations that sponsor research, which then must share this information with the FDA at the time that the research data are submitted for regulatory review.41 However, studies of academic�industrial relationships have shown that current private and public strategies for dealing with the risks resulting from these relationships have been inadequate in several respects. A 1998 survey42 of conflict-of-interest policies in 100 U.S. universities found both a lack of specificity about the types of academic�industrial relationships that were permissible and wide variation in the management approaches used to deal with potential conflicts of interest. A study of five major universities conducted by the General Accounting Office and reported in 200141 concluded that universities were confused about what they needed to report to the federal government according to the rules established by the NIH, that they could not reliably locate the data they had collected from members of their faculties, and that they varied considerably in the types of interests that were considered acceptable. Spurred by these findings and by the Gelsinger case, Senator Edward Kennedy (D-Mass.) introduced Senate Bill 3060, the Research Revitalization Act, in 2002. The act would have substantially enhanced the role of institutional review boards and strengthened their authority to screen for and monitor potential conflicts of interest on the part of university faculty members who conduct research projects in which human subjects are involved. As is so often the case, scandal, media attention, and the specter of impending legislation galvanized both the federal government and the private sector into action. In January 2001, the Department of Health and Human Services had issued "draft interim guidance" with regard to the way research sponsors, academic investigators, and academic institutions and their internal review boards should manage conflicts of interest in clinical research. Between 2000 and 2002, both the Association of American Medical Colleges (AAMC) and the Association of American Universities (AAU) issued new recommendations for the protection of human subjects from the effects of conflicts of interest on the part of academic investigators and their universities.2,43,44 In April 2001, the AAMC, the AAU, and five other private, nonprofit, research-related organizations established the Association for the Accreditation of Human Research Protection Programs (AAHRPP), which offers academic institutions and industrial organizations that conduct research involving human subjects the opportunity to undergo voluntary accreditation. The association's executive director, Marjorie Speers, said in an interview, "It would be very difficult for an institution to be accredited if it didn't have some action in motion on investigator and institutional conflict of interest." To understand the effects of all this activity on academic�industrial relationships, it is useful to categorize the relationships as either clinical or nonclinical. Discussions about whether and how to manage academic�industrial relationships that involve clinical research have progressed much further than is the case for nonclinical research. The ethical responsibility of physicians and universities to put the welfare of their patients and of human subjects enrolled in clinical research ahead of all competing interests, financial or any other, has provided a compelling justification for managing the academic�industrial relationships that create substantial financial conflicts of interest in clinical research. Politically, Congress has signaled � through recent hearings and threats of legislation � that it has a deep, bipartisan concern about the academic�industrial relationships that create conflicts of interest that either jeopardize or appear to jeopardize the health of identifiable humans. David Korn, senior vice president of the AAMC, speaking in an interview, said he sees a "short grace period" in which the academic community will be able to address these issues before the government takes action. This sense of urgency is reflected in two recent AAMC reports on conflicts of interest in clinical research.2,43 The first report, which is concerned with individual academic investigators, recommends "full reporting" of the financial interests of individual academic investigators and "external disclosure." It states that "institutions should regard all significant financial interests in human subjects research as potentially problematic, and therefore, as requiring close scrutiny. Institutional policies should establish the rebuttable presumption that an individual who holds a significant financial interest may not conduct such research unless there are compelling circumstances for allowing the conflicts to persist." In cases in which investigators who have conflicts of interest are allowed to proceed, the AAMC recommends "rigorous, effective and disinterested monitoring" of the work.43 The second AAMC report, which is concerned with institutions' conflicts of interest, calls for a "credible separation" between decisions about technology transfer and investment, on the one hand, and decisions about the administration of research programs, on the other.2 The report states that "even when separation of function has been achieved, certain financial relationships with commercial sponsors should be examined closely for institutional conflict of interest . . . where such relationships exist, institutions should presume that absent compelling circumstances and careful management . . . the research . . . should not be conducted at . . . the conflicted institution." In marked contrast, the management of academic�industrial relationships that affect nonclinical research has received much less recent attention. In the absence of the ethical clarity and searing publicity that are associated with conflicts of interest in academic�industrial relationships that affect clinical research, both the academic community and outside observers seem far less certain about approaches to such relationships with regard to nonclinical research. The Future The distinction between academic�industrial relationships involving clinical research and those involving nonclinical research is likely to characterize debate about policymaking for such relationships in the life sciences for the foreseeable future. In the realm of clinical research, it seems almost inevitable that restrictions on these relationships will increase. Two developments are notable in this regard. First, the AAMC's recommendations, if widely adopted in the university community, would result in much more intense management of the academic�industrial relationships affecting clinical research than there ever was before. The second development is the formation of the AAHRPP.45 Greg Koski, who recently stepped down as the first director of the Office of Human Research Protections, which oversees federal efforts to protect human research subjects, sees accreditation as a critical mechanism for encouraging tighter voluntary management of the academic�industrial relationships in the area of clinical research. The influence of the AAHRPP will depend, of course, on whether research programs decide to seek accreditation (a decision that is considered likely), and on whether the AAHRPP adopts specific conflict-of-interest standards as part of its accreditation process. Nevertheless, the present accreditation process provides at least a path in the private sector toward propagation of "the rebuttable presumption" approach (i.e., the presumption that individual investigators may not conduct research that is related to their own financial interests). There remain many uncertainties on the road to the realization of this vision of a voluntary, private-sector solution to the problems associated with the academic�industrial relationship in clinical research. For one thing, the pressure from government, which was instrumental in pushing the AAMC and other groups to act, appears to have diminished, at least temporarily. On March 31, 2003, after two years of deliberation, the Department of Health and Human Services finally released a second draft of its "draft interim guidance" with regard to conflicts of interest. The document is notable for the qualified nature of its recommendations, which are not backed by any regulatory authority.46 With Senator Frist now the majority leader and Senator Kennedy now in the minority on the Health, Education, Labor, and Pensions Committee, legislation to protect human subjects seems to have a lower priority in the Senate. In addition, even if the draft interim guidance were to be widely adopted, the AAMC's rebuttable-presumption standard remains intentionally ambiguous, reflecting continuing divisions within the academic community. As noted earlier, the AAMC's initial report asserted that the presumption against conflict of interest in clinical research is rebutted whenever an academic institution's committee on conflicts of interest finds "compelling circumstances" for allowing such a conflict to persist. Joseph Martin, dean of Harvard Medical School, believes that the rebuttable-presumption standard should be narrowly construed to permit academic investigators who are found to have a conflict of interest and who have developed new devices to test these devices in a small number of human subjects (three to five) in cases in which the technical expertise of the researcher is vital to proving that the device works. Other universities, however, could interpret the standard more liberally. Nevertheless, private-sector initiatives, government pressure, and the ever-looming threat of scandal appear to have launched the university community on a path toward increased disclosure and closer management and monitoring of academic�industrial relationships in clinical research. In nonclinical research, however, there is no comparable prospect. Even at the University of Pennsylvania, which, in the death of Jesse Gelsinger, directly experienced the risks of academic�industrial relationships, Arthur Rubenstein, dean of the School of Medicine, says he believes "there can be much more flexibility . . . in the basic science area." The experience of academic institutions and industry in relationships that involve nonclinical research is in some ways more revealing of how these relationships are evolving in the life sciences. Recruited to the battle against serious diseases, the university has become a partner of society in efforts to achieve health and economic objectives. This partnership has led the university community to open its doors wider than some had anticipated. In situations in which human subjects of research are not put at risk, academic institutions, for better or worse, have accommodated, and even embraced, the resultant traffic in knowledge and personnel. Even when human subjects are put at risk, the traffic flows briskly, but in narrower lanes and with more yellow caution lights. The final tally of the benefits and costs associated with the integration of academic research institutions into the American economy remains to be seen. Dr. Blumenthal reports having served on an advisory board for
SRS Capital.
From the Institute for Health Policy, Massachusetts General Hospital�Partners Health Care System, Boston. References
|