January 15, 2008 (Vol. 28, No. 2)
Jonathan E. Rackoff
Proposed Ethics Organization Would Beef-up Oversight of a Critical Area Neglected by FDA
These are difficult times for the FDA. At the close of 2007, a major controversy swirled around one of the agency’s core oversight functions: the protection of human subjects of clinical research. In September, the DHHS Office of Inspector General (OIG) released a report that was sharply critical of the FDA’s Bioresearch Monitoring (BiMo) program.
The BiMo program audits sponsors, clinical investigators, and institutional review boards (IRBs) to confirm data quality. Onsite BiMo inspections are also FDA’s principle mechanism for ensuring that human subjects are protected once clinical trials are under way. Unfortunately, BiMo is underperforming—badly.
During the fiscal 2000–2005 period, the FDA inspected fewer than 1% of clinical trial sites, and of those, 75% were surveillance inspections, targeting previously completed trials and generally without benefit to subjects of ongoing research.
The OIG report explained that the FDA lacks a comprehensive and accurate information system for tracking clinical research activity involving drugs and medical devices registered with the agency. As a result, FDA is unable to identify ongoing clinical trials, the number of associated trial sites, or the IRBs charged with overseeing the research.
These data limitations combined with chronic scarcities of funding and inspectors hamstring FDA’s inspection capability and hamper its ability to detect violations of and ensure compliance with federal regulations designed to protect human subjects.
Yet, these charges are merely the latest in a series of blows to FDA’s credibility. Well-publicized contamination scares involving E. coli recently led to a firestorm of criticism over food safety that continues to rage; a fragmented domestic inspection regime marked by overlapping authority and inadequate resources were blamed. High-profile market withdrawals of profitable prescription drugs such as Vioxx, due to deadly complications from prolonged use, have intensified similar criticisms of FDA’s drug safety sytem.
A March 2006 Government Accountability Office (GAO) report asserted that the FDA “lacks a clear and effective process for making decisions about and providing management oversight of postmarket drug safety.” Warranted or not, critics loudly complain that the agency’s inability to fulfill these core oversight missions signifies endemic incompetence. How, then, facing such weighty challenges on other fronts, should the FDA and the broader research community respond to the OIG’s findings on human subjects’ protections?
At the outset, it is important to note what the OIG has not found: signs of true scandal. Of roughly 15,000 clinical studies at some 350,000 trial sites under FDA jurisdiction from 2000 to 2005, the OIG discovered no evidence that research subjects’ safety was in jeopardy.
In fact, rates of research injury remain impressively low, and overlapping IRB oversight at the protocol level makes widespread violation of FDA rules unlikely. Rather, the issue is one of assurance of regulatory compliance. Public confidence in the research enterprise may erode unless the research community can reasonably and demonstrably guarantee that human subjects protection requirements are being met.
An Issue of Institutional Competence
Bioethics commentators have generally lauded the OIG’s criticisms as right on target. They have predicted shock, dismay, and outrage as FDA’s deficiencies become known. Outrage, though, may not be helpful here.
This is by no means the first negative assessment of FDA’s BiMo program. Over the past 10 years, FDA’s clinical trials oversight program, assurance of IRB performance, procedures for disciplining clinical investigators charged with research misconduct, as well as the agency’s capacity to monitor financial conflicts of interest affecting clinical investigators have each drawn the scrutiny of watchdog entities.
The OIG’s latest revelations paint an especially grim picture as to how well we protect subjects in human experimentation. The OIG ignores the long-term implications, but there is strong evidence that the FDA may not be the best agency for the job of securing the protection of human subjects.
It is undeniable that FDA’s authoritative gatekeeping role in the U.S. prescription and medical device marketplaces endows the agency with considerable and coercive influence over pharmaceutical firms and clinical investigators.
FDA is empowered to address violations of human subjects’ protections by disqualifying IRBs, rejecting or terminating INDs and thereby clinical studies, refusing consideration of morally tainted study data, and sanctioning individual clinical investigators. These threats can give rise to compelling incentives to uphold ethical standards. Even contract research organizations (CROs) and academic medical centers may be influenced, albeit indirectly, by the risk of losing business or lucrative collaborative deals if misconduct jeopardizes their sponsors an FDA approval.
Threats must be carried out to be credible, violations uncovered to be punished. As the OIG has persuasively shown, the FDA may be unable or unwilling to exercise the enforcement powers at its disposal. Certainly, this is presently true. We believe it is likely to remain true for the foreseeable future as well.
According to the OIG, ethical violations were detected at trial sites 348 times from fiscal 2000 to 2005. Yet, the agency barred investigators from future trials just 26 times—a rate of discipline which, at best, invites deeper scrutiny. Looking ahead, FDA will continue to face a broad array of challenges that will compete for its attention and resources, both of which are and can be expected to remain in short supply.
First, FDA must continue to fulfill its core mission—including evaluating the safety and efficacy of new drugs, monitoring drug labeling and advertising, and undertaking postmarket safety surveillance—a gargantuan task in the best of times. Second, the agency must contend with the current turmoil over food and drug safety, which represents an ongoing distraction. Third, it must address politically its chronic underfunding and understaffing problems. Fourth, public confidence in FDA is waning. The agency must respond to growing perceptions that it is disorganized, inefficient, susceptible to political interference in scientific processes, and generally too cozy with pharmaceutical firms.
Finally, the number of new drugs winning approval and coming to market is slowing. Yet, expenditures on clinical research are at all-time highs. The agency must implement numerous pre-existing initiatives to improve the development pipeline and speed access to new medicines.
Collectively, it is clear that FDA has its hands full. Protecting human subjects enrolled in clinical research is undeniably important. Shouldering an agency already stretched thin with substantially increased investigatory expectations, however, is not likely to be successful in our view.
Centralized and Uniform Research Ethics Review
A more efficient alternative to FDA oversight is available: uniform scrutiny of all ongoing clinical trials by a single regulatory body. To be sure, the OIG’s recommendations: creation of a comprehensive clinical trials database and IRB registry to facilitate identification of ongoing studies and their associated IRBs; development of a mechanism to track all site inspections; and procurement of legal authority to cover all participants in the management and conduct of a trial, including CROs and associate research personnel should be implemented.
Responsibility for managing these data sources and exercising this expanded legal authority need not be vested in the FDA alone. It would be preferable to develop a central, unified ethics organization responsible for ensuring regulatory compliance in all clinical trials subject to FDA jurisdiction. At a high level of abstraction, this organization’s role would mirror the oversight function of the Office of Human Research Protections (OHRP) but involve a wider operational scope and require greater authority.
Either created anew or subsumed within OHRP, this entity would be exclusively responsible for ensuring that research ethics requirements have been satisfied. The organization’s first task would be to promulgate a uniform set of regulations applicable to all research with human subjects, regardless of whether funded privately or by the government. The source of funding, be it governmental, academic, or industrial, should be irrelevant to the question of ethical treatment of human subjects.
It might also be appropriate for this organization to accredit and periodically recertify select IRBs, both institutional and private, for industry-sponsored trials, in addition to generating a standardized set of guidelines encompassing data-safety monitoring procedures, adverse event reporting obligations, and detailed conflict-of-interest rules—all issues of special relevance to research within FDA’s purview.
It is appropriate for the FDA to retain its authority to sanction noncompliance with applicable federal regulations. This is a quasi-judicial function that requires no fact finding. As such, it should not impose the magnitude of administrative burden we fear might debilitate the agency further. Rather, the organization we propose could accept this burden, incorporate the BiMo inspection function, and take responsibility for investigating violations, compiling data on noncompliance, and reporting offenders to FDA for recommended disciplinary action. Of course, no federal rules require IRB representatives to visit trial sites or inspect research centers; the new organization could seek to fill this gap by codifying an inspection requirement.
Finally, a central ethics organization could assuage mounting concerns about the impartiality of private, for-profit IRBs and CROs. At present, a single privately run IRB performs the ethics review of more than half of all new drug applications to the FDA. This private IRB has been singled out for ethics violations on a number of occasions.
Similarly, CROs currently run a majority of the pharmaceutical industry’s drug trials. In 1991, about 80% of industry-sponsored drug trials were conducted by academic researchers at universities. By 2005, 75% of all clinical trials sponsored by pharmaceutical companies were set in private test centers or doctor’s offices.
Given these volumes, it is vital for maintaining public trust in research that federal overseers investigate reports that CROs take liberties with ethics guidelines to remain attractive business propositions, e.g., recruiting vulnerable study subjects without an adequate explanation of risks, employing unqualified or unlicensed clinicians to administer study drugs at private test centers, or simply maintaining incomplete or illegible records.
Our proposed ethics organization could do so by promoting transparency in CRO operations, enforcing accountability of CRO personnel for mishandling human subjects and clarifying how pharmaceutical firms employing CROs should be held responsible for violations at their test sites.
We have proposed one solution to the FDA’s apparent lack of institutional capacity to monitor clinical trial sites aggressively. There may be others. Admittedly, the role of our proposed ethics organization demands further specification. In addition, there are drawbacks to such an approach. But an alternative approach must be found because the evidence strongly implies that protecting human subjects may be outside the perimeter of FDA’s traditional, and likely future, competencies.
Jonathan E. Rackoff, J.D., is a guest researcher in the department of bioethics, NIH and an associate at Sidley Austin LLP. (email@example.com). Namrata Kotwani (firstname.lastname@example.org) is a predoctoral fellow,
department of bioethics at the NIH. The opinions expressed are the authors’ own. They do not reflect
any position or policy of Sidley Austin LLP or its clients, the NIH, U.S. Public Health Service, or Department of Health and Human Services.