Intentional Human Dosing Studies for EPA Regulatory Purposes: Scientific and Ethical Issues (2004)

Intentional human dosing studies require not only strict substantive restrictions, but also careful procedural requirements to guarantee that the substantive restrictions are followed and ethical standards are met. This chapter delineates those procedures, which include the recommendation that all proposed experiments be reviewed by a properly constituted Institutional Review Board (IRB) and by a new body to be constituted at the Environmental Protection Agency (EPA).

As described in Chapter 2, in the United States for the last several decades most federally funded human research has been covered by the Common Rule, which requires advance approval of such research by local IRBs. The Common Rule applies only to human research conducted or funded by signatory federal agencies and to any research performed at an institution that has promised to have all research reviewed by an IRB as part of its assurance of compliance with the Common Rule. In addition, the Food and Drug Administration (FDA) requires review of protocols by an IRB and informed consent of participants for any investigations of new drugs conducted in the United States. On the other hand, private sponsors of intentional human dosing studies submitted to EPA are not specifically required to obtain IRB approval for studies, particularly if the

studies are conducted at institutions that do not require IRB review of all research. Although it appears that all of the pesticide experiments reviewed by the committee were approved in advance by IRBs or their foreign equivalents, the committee believes that this decision should not be left to the discretion of the sponsors.

EPA may wish to use FDA’s implementation of its equivalent of the Common Rule (21 CFR Part 50) as a guide for its adoption of such a requirement (see Chapter 2).

EPA itself has sponsored intentional human dosing studies. At least some of those experiments were approved by IRBs at the institutions that conducted the research. The committee was informed that EPA does not have an IRB, but instead has an Ethics Review Officer who typically ensures that EPA-sponsored or conducted studies undergo IRB review.

If all EPA-sponsored human research is conducted at nonfederal institutions and those institutions have appropriate IRBs operating in compliance with the Common Rule, the federal requirements might be satisfied. If EPA were to conduct human research in-house without prior IRB review, it would be violating the Common Rule.

The IRB that reviews EPA-sponsored research should contain the requisite expertise to review human dosing studies (as well as other human research in support of EPA’s mandate). In preparing this report, the committee spent considerable time assessing protocols that were submitted to EPA for registration purposes and that were made available to the committee and/or its staff. In reviewing these protocols, the committee was tasked with ascertaining the meaning and applicability, in the context of EPA research, of “procedures … which do not unnecessarily expose subjects to risk” (40 CFR 26.111(a) (1)) and “the importance of the knowledge that may reasonably be expected to result” (40 CFR 26.111(a) (2)). Despite the expertise of the toxicologists, clinicians, and biostatisticians on the committee, this was not an easy task, but it was one that led the committee to appreciate the difficulty of making these determinations. Thus, for EPA’s IRB to carry out its assigned duties in reviewing intentional human dosing studies, it must include members with the range of disciplines and perspectives and the array of skills needed for this task. This requirement is equally true for IRBs reviewing studies sponsored by or conducted by non-EPA institutions. All IRBs that undertake the review of intentional

human dosing studies should have appropriate scientific and ethical expertise, which may require the appointment of additional members or consultants with expertise in toxicology or biostatistics.

Despite their limitations, IRBs remain a crucial part of the system of protection for participants in research. They do, however, in special situations require substantive supplementation. For example, gene transfer protocols receive not only local IRB review, but are also subjected to close and public scrutiny by the National Institutes of Health’s (NIH’s) Recombinant DNA Advisory Committee (RAC). This additional review was instituted after the death of a healthy volunteer in a gene transfer study, which raised concerns that local IRBs may lack the expertise needed to review such studies. The RAC includes scientists from various disciplines pertinent to gene transfer, as well as members with expertise in the ethics of human research. As another example, Subpart D of 45 CFR 46 contains a provision for the Secretary of the Department of Health and Human Services to convene a panel of experts to review research with children that is not otherwise approvable, but that presents an opportunity “to understand, prevent, or alleviate a serious problem affecting the health or welfare of children” (45 CFR 46.407).

Similarly centralized and elevated review would be useful for any proposed intentional human dosing studies conducted for EPA regulatory purposes, given their unique risk-benefit calculus, whether EPA sponsored or sponsor initiated. Previous chapters of this report have high-lighted the difficult and controversial ethical and scientific issues involved in some of these experiments, particularly those concerning pesticides. The committee concludes that EPA should not consider such experiments unless they meet high scientific and ethical standards (see Chapters 3, 4, and 5).

The committee also concludes that another level of review is required for these studies to establish a body of knowledge and expertise that can then be communicated and disseminated to the pertinent members of the research community. The committee understands that adding additional review burdens research with additional costs, which may not be trivial. The committee concludes that for this kind of research, however, the benefits of such review outweigh the costs, as such review may provide valuable advice to study sponsors regarding how to structure these experiments. In addition, it was not clear to the committee that local IRBs can be expected to conduct a thorough assessment of this kind of research, as evidenced by problems in IRB review of some of the air pollution studies. The envisioned additional review would provide greater specialized ex-

pertise than IRBs usually will have available in considering the special scientific and ethical problems raised by this kind of research, as well as in assessing the potential benefits of the research. Because this research is publicly sensitive, additional review will help build public trust that, when approved, the research is appropriate.

The proposed board’s basic function would be to help assure that EPA only uses intentional human dosing studies that meet the rigorous scientific and ethical standards specified in Chapters 3 through 5. This new board would undertake an integrated evaluation of the science and ethics of human research studies (IOM, 2003). In its review of study protocols, the proposed board would not function as a national IRB. Instead, it would provide advice to both the sponsors of the research and to the IRBs that would review it. The board would give the sponsors proposing such research (including EPA itself) advice on how to meet the high standards required. Its report would provide to any reviewing IRBs expert analysis that would help them consider such protocols. After the experiments are completed and the results are submitted to EPA, the Human Studies Review Board would advise EPA’s relevant program offices on whether, and to what extent, the results should be considered. The board also would collect and analyze information about these experiments that could allow it to suggest ways in which the research could be improved or to assess whether EPA should continue to consider the results of these experiments.

The committee recommends that the board be relatively small and that it should report directly to the Office of the EPA Administrator. The committee considered whether the board’s functions could be discharged within EPA’s existing structure. In light of the types of expertise that would be needed in both science and ethics, the committee concludes that no existing EPA office could perform the necessary tasks. Either the EPA Science Advisory Board (SAB) or the Federal Insecticide, Fungicide, and

Rodenticide Act Scientific Advisory Panel, with appropriately enhanced ethical and trial design expertise, might be able to perform those tasks; however, EPA would have to determine whether performing these enhanced functions would interfere with the current obligations of those bodies. Finally, and perhaps most importantly, creating a new board accountable directly to the Office of the EPA Administrator would highlight the importance of this new level of review.

With appropriate staff support, the board’s work could be performed by a relatively small but broadly knowledgeable group of experts, with core expertise in human toxicology, biostatistics, and research ethics. It should have the ability to make use of expertise on special subject matters as needed, either by expanding the board temporarily or by using other EPA experts as consultants. In addition, it should be encouraged to coordinate its efforts, as appropriate, with EPA’s SAB, SAP, its Ethics Review Officer, and EPA’s authorized IRB. Creation of the proposed board raises many other questions, such as conflict of interest limitations and compensation for service, which EPA should address.

The committee deliberated at some length over whether the board should be internal or external to EPA. Arguments can be made for either approach. However, on balance, the committee determined that a formal, permanent, internal structure would be best suited for the kind of integrated scientific and ethical review it envisioned. First, there seems no clear alternative to EPA for the location of such a body, and the creation of a body independent of EPA seems both logistically and politically complex. Furthermore, through the creation of an internal body EPA would take responsibility for the structure and own both the process and the results. The board over time would further specify the general ethical principles and conditions for justified research and develop a series of case judgments and commentaries, enhancing its ability to conduct the best possible reviews. Although the board would be an internal one, EPA could and should invite outside individuals to participate in order to ensure that the necessary expertise is included. Importantly, the board would not replace local IRB review. Rather it would supplement local review by looking at the toxicological and ethical aspects of protocols, and it would assist in improving and refining the science required as part of EPA’s regulatory mission (see Chapter 7 and its recommendations for EPA review and the use of scientific data).

The committee also strongly recommends that the board report directly to the Office of the EPA Administrator, rather than be located in any one EPA operational unit, for two reasons. First, it should review experiments sponsored by or relevant to many different EPA functions. Much of the testimony to this committee focused on pesticides, but the committee is making broad recommendations encompassing all inten-

tional human dosing studies sponsored by or submitted to EPA, whether they involve pesticides, air pollutants, water pollutants, or any other relevant EPA jurisdiction. Second, placement of the board within an operational unit of EPA would raise the possibility that the unit’s interests might conflict with the board’s free consideration of scientific and ethical issues.

The committee recommends that the Human Studies Review Board should review in advance proposed intentional human dosing studies sponsored or conducted by EPA and all such studies whose sponsors intend to submit them to EPA. The committee concludes that it would be optimal if this review were mandatory, but, because of legal and logistical concerns, it recommends only that EPA consider making it mandatory. Any conclusions reached by the board should be advisory and not binding on the sponsoring companies or reviewing IRBs. The board’s process should take place in advance of local IRB review (or EPA internal IRB review for EPA-conducted studies). It would not replace IRB review. The committee also recommends that the results of this review be made public, taking into account the sponsors’ need to protect trade secrets.

As detailed earlier in this report, intentional human dosing studies raise many difficult issues about their scientific worth and ethical propriety. The committee concludes that if such experiments promise scientifically valid and important information, they can be conducted ethically. Instituting a process of advance screening and advice should improve the scientific and ethical quality of protocols. It could lead sponsors to abandon some ill-conceived experiments while, in other cases, it would provide advice that will improve both the scientific value and ethical acceptability of the experiments.

The committee envisions a process similar to the one through which FDA often provides informal advice to firms that are conducting clinical trials on a new drug or biologic. Anyone seeking to experiment with drugs or biologics on humans is required to file with FDA an Investigational New Drug application (IND) and to amend the IND with any new protocols. FDA can put any study it considers unsafe on hold (not allowing it to proceed), or it can engage in discussions with the sponsor about improving the proposed trials. FDA also can put studies on hold if they are not likely to provide useful information—certainly FDA has an interest in identifying studies of no scientific value. In addition, FDA is available to discuss approaches more generally, even before protocols are written. These discussions often lead to clinical trials that are more scientifically valuable, safer, and thus more ethically appropriate than they would have been without the FDA-sponsor interaction. However, FDA has no central

board, but rather staff and numerous advisory boards that review protocols in various drug and/or device categories.

The committee discussed at length whether pre-experiment review by the proposed board should be mandatory or voluntary. The main argument for mandatory review was the importance of this review process. The committee wants to prevent inappropriate intentional human dosing studies whenever possible, and requiring review of proposed experiments in advance would lead to fewer inappropriate studies. In addition, making pre-experiment review mandatory should build public confidence that problematic experiments are being minimized and would guarantee that EPA knew of all relevant industry-sponsored experiments, making it impossible for sponsors to keep EPA from learning of experiments that yielded negative results.

Some committee members, however, argued for a voluntary system for experiments not sponsored by EPA. (Everyone agreed that all proposed EPA-sponsored intentional human dosing studies should be reviewed in advance by the board.) They pointed out that few, if any, sponsors would refuse an opportunity to obtain early advice from the board, particularly when it also would review the completed experiment. They further noted that a voluntary system could be easily implemented, while a mandatory system would appear to require, at a minimum, changes in EPA regulations, and possibly new legislation. A voluntary system also would avoid an implementation problem inherent in a mandatory system—the need to distinguish between studies intended for submission to EPA, for which the pre-experiment review would be mandatory, and studies independent of a commercial sponsor that later turned out to be relevant to an EPA decision. Of course, if experience were to reveal that many protocols were not submitted for advance review, EPA could take steps to require such submission.

Ultimately, the committee concludes that pre-experiment review of studies intended for submission to EPA should be mandatory, if legally and logistically feasible. If not, EPA should strongly encourage study sponsors to seek such review.

The committee strongly urges that any research sponsored, funded, or conducted by EPA that intentionally exposes research participants to toxic substances also should be submitted to the Human Studies Review Board, to ensure consistency in EPA evaluations of such studies and educate EPA program offices about the issues involved and the board about such research within EPA. As discussed previously, under the Common Rule research conducted, sponsored, or funded by EPA also must have IRB approval.

In terms of the sequence of submissions to the board and the local IRB, the committee believes it would be beneficial generally to have each

proposed third-party study submitted to the board in advance of the local IRB review, because the board would probably have greater scientific expertise in evaluating these experiments than most IRBs. The board would then offer its views first, which the sponsor would be required to forward to the IRB, thus assisting the IRB’s evaluation of such studies. The sponsors should have to submit the IRB report to the board, which would serve to provide feedback to the board from the IRB’s perspective—the model RAC uses at NIH. Of course, such review must be done in a timely manner.

The board’s pre-experiment review would be advisory. It would make nonbinding recommendations to researchers or sponsors of research about the scientific and ethical aspects of protocols. However, it is likely that the advice of the board usually would be accepted, given the role of the same board in reviewing research results later. The committee does not believe, however, that the board would require veto power over this kind of research, in the way that the IND process gives veto power to FDA. Although a sponsor could proceed with an experiment in the face of a negative board conclusion, the committee believes that few, if any, sponsors would do so without compelling arguments to support their position.

Another issue debated by the committee with regard to pre-experiment review was whether results should be made public. On the one hand, the general availability of the reviews would guide other applicants and help to reassure the public that only scientifically valuable and ethically appropriate studies were being conducted. The NIH RAC has adopted this model, and its deliberations, as well as its conclusions, are public. On the other hand, sponsors may well have legitimate concerns about disclosure of trade secrets or other confidential business information. FDA’s discussions with IND applicants are not public, either in substance or results. The RAC also does not make trade secrets public. Although RAC members do see details of vector and gene construct, they promise to keep this information confidential. This is a valuable model for allowing reviewers to see all pertinent information while respecting the confidentiality of trade secrets.

The committee decided to seek the best of both systems and recommended that the board’s deliberations should not be public, but that reports on its deliberations and conclusions should be publicly available, except to the extent that they might reveal protected trade secrets or confidential business information. Alternatively, the board could hold public sessions and convene in closed session to review confidential materials. For many of the pesticides undergoing registration after decades of use, the committee expects that few, if any, legitimate claims of trade secrets or confidential business information will be of issue. The board should, how-

ever, monitor the effectiveness of the recommended compromise solution and may well, based on its experience, choose to make changes.

When the results of an intentional human dosing study are submitted to EPA for its regulatory consideration, including studies conducted, sponsored, or funded by EPA, those results should be submitted first to the board for its review. The review should be based on all information collected as part of the study and reported with completeness comparable to that required by FDA for clinical trial submissions. In Chapter 7 the committee recommends a process for internal EPA review of scientific data submitted for regulatory decisions. The results of that staff review should be communicated to the Human Studies Review Board for a second level of review of the scientific value and ethical propriety of the experiments. This model of dual review is used at NIH. The Human Studies Review Board would then provide recommendations to EPA on the scientific and ethical acceptability of such studies. The results of the board’s review should be made public, subject to legitimate claims of trade secrets or confidential business information. (The board also may need to consider some delays in the publication of some parts of its report to allow the private investigators to publish their results.) The board also should review studies submitted to EPA that were completed before the effective date of the changes proposed by this report and other studies submitted for EPA’s consideration, including those submitted as part of the peer-reviewed literature, for the purposes set out in Chapter 5.

The post-experiment review function of the board is distinct from the kind of review that EPA undertakes for the purpose of incorporating results from particular experiments into the regulatory process. It would not replace or modify the structures and procedures for the latter kind of review. Instead, it would offer nonbinding advice to the relevant EPA units about the scientific and ethical acceptability of the submitted and completed research, not about whether the research should alter the standards for human exposure to toxic substances.

The committee considered whether the pre-experiment and post-experiment reviews should be conducted by the same body. It determined that consistency in judgments and the need for the companies conducting or sponsoring research and submitting data to EPA to be able to rely on the advice given point to a single body discharging both functions. A board that receives and reviews both pre-experiment protocols and their ultimate results would be better informed and more capable of undertaking either review.

Finally, the committee strongly recommends that the results of the

board’s post-experiment review be made public. The same arguments for public disclosure apply as in the case of pre-experiment review. The arguments for confidentiality are more limited, however, as the experiments have now been completed and are being voluntarily submitted to EPA for its regulatory use. It is possible that there still may be some legitimate claims of trade secrets or confidential business information that would not be publicly disclosed as an inevitable consequence of the submission to EPA. Public disclosure of the board’s review should be limited as necessary to protect the sponsors’ legitimate claims for protection.

Most of the procedures set forth in this chapter are prospective only, applying to experiments completed or proposals for experiments made after the recommendations of this report are implemented. Earlier experiments or later experiments not sponsored by EPA or those submitting the results to EPA also can raise scientific and ethical issues. The board should review those studies for their scientific and ethical propriety under the standards set out in Chapters 3 through 5. The board also should review those studies to offer its advice on their scientific value to the relevant EPA unit.

The committee reached two other conclusions about the board and made one overall recommendation for review of the board itself. First, the committee believes that, over time, the board should do more than review the proposed experiments and experimental results put before it. It will have an excellent opportunity to study intentional human dosing studies and make general recommendations concerning them. On the one hand, the board might conclude, based on its experience, that such experiments, in fact, have little value or suffer from major unresolved ethical problems. On the other hand, it may be able to make specific recommendations for improving either the scientific value or the ethical propriety of these studies.

In particular, the committee strongly urges that the board should consider the issue of payments made to research participants (as discussed in Chapter 5). This is not a new issue or one unique to these studies. The ethical problems involved in paying research participants have been long recognized and debated, but they have not been resolved. The committee finds these issues particularly complicated in intentional dosing studies, in part because the altruistic motives that may inspire volunteers in other research, such as drug studies, seem more complex in the context of those involving pesticides. As the board gains experience with the payment arrangements made in various experiments and with the nature of the vol-

unteers enrolled, it may be able to reach some valuable, empirically grounded conclusions regarding these issues.

Second, the committee urges that the board institute “ethical audits” of experiments that involve the intentional administration of toxic substances to research participants. These audits would examine a sample of these experiments to determine whether those conducting them are following, or have followed, the protocols set out in their submissions to the board. Such an audit function has been called for in general reports on protecting research participants (IOM, 2003); the strong ethical concerns and public controversy regarding human toxicant experiments make them a particularly good subject for this action.

Finally, the committee recommends a structure for review of these experiments that should be both rigorous and practical, but it recognizes its limits in foreseeing how well the structure might work over time and whether it will continue to be needed. A timely review, preferably conducted by a body including individuals from outside EPA, should help ensure that the board plays the important role this committee envisions for it.

This chapter provides a procedural framework for EPA’s review of intentional human dosing studies, which should be used to implement the substantive recommendations offered in previous chapters. In this chapter, the committee recommends that EPA require all human research conducted for regulatory purposes be approved in advance by an appropriately constituted IRB or an acceptable foreign equivalent. The recommendation includes EPA-conducted research; thus, EPA should ensure that research involving humans that it sponsors or conducts undergoes appropriate IRB review. EPA may want to establish its own IRB or specifically authorize another IRB to fulfill that role. Any such IRB may need special expertise to review these types of studies.

Furthermore, to assure that intentional human dosing studies conducted for EPA regulatory decisions meet the highest scientific and ethical standards, the committee recommends that EPA establish a Human Studies Review Board to address in an integrated way the scientific and ethical issues raised by intentional human dosing studies. The board should prospectively review all EPA-sponsored or EPA-conducted stud-

ies. The committee recommends that, if legally and logistically feasible, private entities that anticipate submitting the results of intentional dosing studies to EPA for regulatory purposes also should be required to submit protocols to the board before beginning a study. The submission should include the proposed protocol and sufficient background information to establish the scientific value of the experiment and provide assurance of participant safety. The proposed board supplements but does not replace review by an IRB.

The post-experiment review function of the board is distinct from the kind of review that EPA undertakes for the purpose of incorporating results from particular experiments into the regulatory process. It would not replace or modify the structures and procedures for existing EPA review. Instead, it would offer nonbinding advice to the relevant EPA units about the scientific and ethical acceptability of the submitted and completed research. The proposed review board, its functions, and its record should be assessed after five years.

Institute of Medicine (IOM). 2003. Responsible Research: A Systems Approach to Protecting Research Participants. Washington, D.C.: The National Academies Press.