Testing: Advance Notice of Public Rulemaking
Information and Records Integrity Branch
Re: Docket #: OPP-2003-0132
Dear Sir or Madam:
These comments address the Advanced Notice of Public Rulemaking on concerning the standards to be applied for accepting intentional dosing studies of human subjects. They are submitted on behalf of Farmworker Justice Fund Inc.
The Farmworker Justice Fund Inc. (FJF) is a nonprofit organization, based in Washington D.C., which is dedicated to improving the living and working conditions of migrant and seasonal farmworkers throughout the nation. Throughout its 21-year history, FJF has advocated for policies which would increase protections for farmworkers from exposure to toxic pesticides.
The U.S. Environmental Protection Agency (EPA) has asked for public comment on a multiplicity of questions to assist it in determining whether and under what circumstances and subject to what standards, the EPA should accept, consider and rely on studies involving the intentional dosing of human subjects with pesticides or other toxic environmental agents.
We urge the EPA
(1) to categorically
reject any study using the intentional, non-therapeutic dosing of human
subjects with toxic pesticides for the primary purpose of establishing
a No Observeable Adverse Effect Level (NOAEL) or a No Observeaable Effect
Level (NOEL) on the grounds that any such study is unethical and inconsistent
with accepted national and international legal standards;
Prior to the publication of the ANPR, the EPA had convened a Joint Subcommittee of its Science Advisory Board (SAB) and the FIFRA Scientific Advisory Panel (SAP). In a Report entitled Comments on the Use of Data from the Testing of Human Subjects (SAB/SAP Report), the Joint SAB/SAP Subcommittee found, nter alia, that the intentional testing of pesticides on human subjects should be subject to limitations "ranging from 'rigorous' to 'severe,' that such studies should not be conducted if data is available from other sources (e.g., animal studies, models or unintentional human exposures) or if the resulting data will lack adequate statistical power and that such studies should never involve children, pregnant women or other vulnerable populations.. SAB/SAP Report at 3. Further, the SAB/SAP Report stated that the "Subcommittee, in general, would not support human experimentation primarily to determine a No Observeable Adverse Effect Level." (SAB/SAP Report at 11). While we do not endorse all of the Majority Report's findings, these conclusions of the Joint SAB/SAP Committee are sound and should be adopted by EPA..
The ethical concerns raised by the intentionalnontherapeutic testing of toxic pesticides on human subjects should be informed by the international and national standards established to regulate human experimentation of these kinds. The applicable standards include: 1) The Nuremberg Code, which was adopted by the United States after revelations about the heinous Nazi experiments on humans, conducted during World War II, came to light. It establishes minimum ethical standards for human testing, which include the requirement that the study be expected to "yield fruitful results for the good of society, unprocurable by other methods or means of study." 2) The Helsinki Declaration, which was adopted by the World Medical Association to govern medical testing of humans. It requires, inter alia¸ that there be a "reasonable likelihood" that the test subjects would benefit from the study. 3) FIFRA, which prohibits the "use [of] any pesticide in tests on human beings unless such human beings (i) are fully informed of the nature and purposes of the tests and of any physical and mental health consequences which are reasonably foreseeable therefrom, and (ii) freely volunteer to participate in the test." 4) The Common Rule, adopted by EPA at 40 CFR part 26, which in addition to requiring oversight by an Institutional Review Board and the informed consent of participants, establishes the principle that the "risks to subjects be reasonable in relation to anticipated benefits, if any, to subjects, and the importance of the knowledge that may reasonably be expected." 40 C.F.R. Part 26. Applying these standards, it is clear that the EPA should establish a policy of refusing to accept human studies for purposes of setting NOELs/NOAELs and that it should refuse to accept or rely on the human studies which registrants previously submitted to it for this purpose. In addition, the intentional dosing of human subjects with pesticides should be allowed in very limited circumstance to establish adequate protections for workers and handlers who are occupationally exposed to pesticides.
I. THE TESTING OF HUMANS TO ESTABLISH A NOEL/NOAEL IS ETHICALLY AND LEGALLY IMPROPER
Pesticides are poisons. Many, like the neurotoxic organophosphate or n-methyl carbamate insecticides, act on humans in much the same manner as they do on the insects they are designed to kill. (Moses 1998) These compounds cause acute symptoms ranging from nausea, headache, and dizziness, to convulsions, coma and death. Some overexposure incidents with these chemicals lead to delayed neuropathy and neuropsychological effects, including a loss of intellectual functioning. (Savage 1988). Even at low exposure levels, some pesticide ingredients may cause chemical sensitization or hormone disruption. Some pesticides have also been associated with cancer, infertility, birth defects and neurological disorders including Parkinson's Disease. (Moses 1998) These data establish that exposure to pesticides is highly risky to human health. In such circumstances, only the most compelling reasons could justify the use of human test subjects.
The testing of human subjects to establish a NOEL/NOAEL cannot meet this standard.
First, the human subjects who bear the risk will not personally benefit from the results of the study. Under both the Common Rule and the Helsinki Declaration, lack of personal benefit would render the study unethical. Nor can the paltry payment of $500 to $1,000 be viewed as an adequate benefit. While people in desperate poverty (especially in the Third World) may be willing to risk their health for such small sums, our society has already determined that it is unacceptable to compromise health for mere financial gain. For this reason, the sale of human organs is unlawful. Most testing of pharmaceuticals is done on individuals who themselves suffer from the condition to be alleviated by the drug under review. This is the type of personal benefit, which is comencerate with the risk. Here there is no benefit to the test subjects (and the EPA has alternate means of obtaining adequate information on which to base its judgment). The only purpose of these tests is to enable the pesticide manufacturers (i.e., the registrants) to make it easier for themselves to register their products by reducing applicable margins of safety. This is not the kind of benefit that could justify the health risks involved. See SAB/SAP Report.
Second, the information to be obtained is not of crucial importance to society and adequate information can be secured by other less harmful means. See Nuremberg Code; Common Rule, SAB/SAP Report. Far from being necessary, the EPA has for decades established NOELs/NOAELs based primarily on animal studies without the need for human studies. Moreover, supplemental data is often available from case studies of unintentional human exposures and, where appropriate computer models. In addition, the limited range of test subjects - usually healthy young adults - will not provide useful information for evaluating health effects on the highly diverse US population, especially the most vulnerable subgroups of fetuses, infants, pregnant women, people with compromised immune systems, the elderly etc. Thus, these studies cannot be said to "yield fruitful results for the good of society, unprocurable by other methods or means of study." Nuremberg Code.
Third, in conducting these studies the companies cannot "fully inform" the test subjects of the "reasonably foreseeable" health consequences. See FIFRA. Studies to establish a NOEL/NOAEL are generally conducted at the beginning of the toxicological test battery before the full range of short- and long- term health effects have been established. Even now, however, many pesticides which have been on the market for decades have not been fully tested for chronic health effects. In addition, even though dozens of pesticides have been identified as possible endocrine disruptors, scant testing of this potential adverse effect has as yet been conducted. Moreover, individual genetic make-up and metabolism, as well as a host of health conditions, will make some individuals more susceptible to the toxic effects of the pesticide - but the companies lack the information to identify which individuals would be at risk. Similarly, the companies could not fully inform potential test subjects of the potential adverse effects the pesticide might cause in combination with medicines the person is taking or other pesticides orchemicals to which the individual has been exposed at home or work - because pesticide manufacturers rarely, if ever, undertake experiments to determine the effects of their product in combination with others. In these circumstances, it is not possible for the registrant to provide full disclosure or obtain fully informed consent. As such, these tests would not meet the requirements of FIFRA or the Common Rule.
Finally, it should be noted that the companies are not seeking to use human subjects in the pursuit of sound science. The questions that these tests would answer are not the most important unknowns. As noted above, it would be far more useful for the companies to use their resources to identify endocrine disrupting effects at extremely low exposure levels, determine the effects of pesticides in combination with each other, other chemicals or drugs, establish clinical tests to determine when overexposure occurs, or conduct epidemiological studies to determine the full extent of long term health effects which are now experienced by people who have been overexposed to their pesticide products. But these studies are not on the companies agenda.
II. THE STUDIES PREVIOUSLY CONDUCTED ARE SCIENTIFICALLY, ETHICALLY AND LEGALLY FLAWED
The studies previously conducted are ethically, legally and scientifically flawed and should not be considered by the EPA.
One of the guiding principles established by FIFRA and the Common Rule is that human test subjects must be fully informed of the health consequences and give informed consent to participate. Numerous examples exist to show that the registrants did not conduct these studies under conditions of informed consent. For example, in a study conducted on azinphosmethyl, the substance to be tested was sometimes referred to as a pesticide but other times referred to as a drug. One study on chlorpyrifos recruited test subjects by claiming that they would be advancing medical research. Nor did the consent forms describe the potential for these organophosphate (OP) insecticides to have long term effects, such as such as chemical sensitization, or neurobehavioral effects or cancer - even though studies have associated OPs with each of these conditions. In the absence of fully informed consent, the study violates both applicable legal and ethical standards.
The studies that have been conducted lack statistical power. Most concern dose groups of between seven and 50 participants. By contrast, Professor Herbert Needleman of the University of Pittsburgh, who was a member of the Joint SAB/SAP Subcommittee noted, that thse studies should have had been 1,000 and 5,000 subjects. Indeed, drug companies like Bayer normally use about 3,000 subjects when they are doing a comparable clinical trial of a drug. Where the studies lack scientific validity, they cannot pass ethical or legal muster. See SAB/SAP Report, the Common Rule.
The human subject studies done on organophosphates and n-methyl carbamates were unnecessary because this information could be obtained in other ways. In addition to the animal studies - which show the more significant effect of brain cholinesterase inhibition - there are numerous studies of unintentional exposure to these products which demonstrate human reactions at varying dose levels.
In an article describing a test of chlorpyrifos, conducted by MDS Harris Laboratories for then Dow AgroSciences, a Dow spokesman was quoted as saying that previous research guided the doses selected, so that they "would not cause any harm to the volunteers." (Thompson 1999). But in the study itself, volunteers suffered a variety of symptoms including nausea, shortness of breath, and lack of sensation, which the researchers acknowledged were probably or possibly due to the pesticide exposure. The companies'claim that they try to avoid adverse health effects, is in essence, a tacit admission that allowing people to suffer adverse health consequences - solely to generate greater profits for chemical companies or growers - is not ethically justifiable.
Finally, the EPA has asked whether it may consider studies, which were conducted in the past that do not meet current ethical and legal standards. The answer must be a resounding no. When the companies conducted those studies they took a risk that the studies would not be accepted. At least on two previous occasions, during the tenure of EPA Administrator Lee Thomas and EPA Administrator Carol Browner, the EPA has not accepted intentional nontherapeutic dosing studies of human subjects. In addition, the standards established by the Nuremberg Code predate any of the tests conducted so that the companies should have known that their efforts did not meet internationally recognized ethical requirements. Moreover, it has long been an established tenet of American law that information obtained in violation of accepted standards - like a confession obtained without Miranda warnings - cannot be used because to do so would be to condone and encourage improper conduct. Similarly, here, EPA use of studies conducted in violation of current ethical and legal standards must be rejected.
III. Studies Conducted with Pesticide Handlers or Field Workers to Determine the Extent of Needed Protections Can Be Justified in Limited Circumstances
Studies conducted on pesticide handlers or field workers to determine the extent of personal protective equipment needed or the appropriate duration of a restricted entry interval (REI) do not raise the same ethical and legal objections as those conducted to set NOAEL/NOELs. Nonetheless, because they involve the intentional exposure of people to toxins they should only be undertaken in limited circumstances where they will provide essential information which could not be adequately obtained by other means.
Studies conducted on workers to determine the extent of protections needed may be permitted for the following reaons: First, unlike the NOAEL/NOEL studies, the subjects themselves are likely to benefit because they may be exposed to the pesticide in the course of their work, and would benefit from having adequate protections derived from the sudy results. Second, these studies are conducted after the registrant has already conducted a full battery of tests on animals to determine the full range of acute and chronic effects. This information could be used to ensure that workers were not exposed to too great a dose of the chemical and to provided volunteer subjects with information on the reasonably foreseeable consequences of exposure to the product. Third, the EPA itself has established guidelines for the conduct of studies, so that the methodology has not been set solely by the company that stands to benefit economically from a favorable result.
Nonetheless, there are risks to the human subjects, because the companies will probably not know which individuals are genetically or otherwise more susceptible to the toxic effects of a pestide exposure. Nor will the product have been tested in combination with other chemicals to which the volunteers are routinely expoed in their homes or work. For that reason, when adequate information can be obtained without human studies, human studies should not be conducted.
IV. Additional Observations
In response to other questions posed by EPA we suggest the following additional guiding principles:
In no event should intentional dosing of human subjects experiments be conducted on populations that are not in a position to fully and voluntarily consent. This would preclude the involvement of infants, children prisons, the mentally incapacitated and those who work for the manufacturer or the testing organization.
When third party tests are submitted the EPA should carefully review all aspects of the study, including an evaluation of the participation of the Indepent Review Board (IRV). From studies previously submitted it is apparent that such IRBs may be "independent" in name only. Because IRBs, too, may stand to benefit financially by approving a particular test protocol, the appropriateness of the IRB's determinations, like every other aspect of the study design, should be carefully reviewed to insure compliance with the Common Rule and other applicable legal and ethical standards.
The EPA should not accept the results of studies involving the intentional dosing of human subjects unless it can determine that a study meets the current requirements of the of Common Rule and other applicable legal and ethical standards. If studies available in the open literature were not conducted in compliance with these requirements, such studies should not be considered.
In applying legal and ethical standards, the toxicity and range of adverse health effects of a substance must be taken into account when evaluating the possible benefit to the test subject and the appropriateness of the study. For example, it would never be appropriate to intentionally dose a pregnant woman with a product that might cause teratogenic effects. Nor would it be ethical, given our current limited state of knowledge, to expose anyone to a likely human carcinogen.
In no case should developing humans (i.e., the fetus, infant, young children, or adolescents) be exposed to toxic chemicals. There are currently too many unknown dangers to justify such studies.
Finally, we have attached several newspaper editorials condemning the human subject experimentation for the purpose of setting pesticide NOAELs/NOELs.
Carrier G & Brunet RC, A Toxicokinetic Model to Assess the Risk of Azinphosmethyl Exposure in Humans Through Measures of Urinary Elimination of Alkylphosphates, Toxicological Sciences 47: 23-32 (1999)
Common Rule, 45 C.F.R. Part 46; 40 C.F.R. Part 26
Federal Insecticide, Fungicide & Rodenticide Act, 7 USC 136 et seq
Moses M: Pesticides. In Wallace, RB (ed): Public Health and Preventive Medicine, Stamford CT: Appleton & Lang 1998, pp: 593-605
Savage, E.P., et al., (1988) "Chronic Neurological Sequelae of Organophosphate Pesticide Poisoning," Archives of Environmental Health, 43:38-45.
Thompson J: Lincoln Lab is Only one in US to Test Pesticides in Humans, Omaha World Herald, December 19, 1999