Testing toxic pesticides in humans: health risks with no health benefits.

Sass and Needleman (2004a) described circumstances under which industry-sponsored studies on erythrocyte cholinesterase inhibition by the pesticides dichlorvos and aldicarb showed significant adverse effects that were dismissed by the industry-sponsored authors. Sass and Needleman (2004a) also cautioned about the very limited value of human studies that are based on an examination of short-term pesticide-induced effects in a small number of healthy adults. In the same issue of EHP, several rebuttal letters from industry were published (Charnley and Patterson 2004; Chart et al. 2004; McAllister 2004; Tobia et al. 2004) advocating the value and need for conducting studies of toxic agents in human subjects.

Why would the chemical and pesticide industries want to conduct studies of toxic chemicals and carcinogens in humans? The likely answer to this question lies in the way regulatory agencies, such as the U.S. Environmental Protection Agency (EPA), establish allowable human exposure levels to toxic chemicals. For most agents, human exposure standards are based on toxicology findings from studies in animals. As a public health protective measure, uncertainty factors (for example, 10-fold for differences in human versus animal sensitivity to that agent) are typically applied to the no observed effect level (NOEL) or the lowest observed effect level (LOEL) in animals to set maximum allowable human exposures to these toxic agents. However, if, in the opinion of the regulated industry, data from exposed humans suggest that this factor may be too high, then they pressure the U.S. EPA to use a lower factor, or even no adjustment, for the animal-to-human extrapolation. If the industry wins their argument, then the general public will encounter much higher levels of these toxic agents in the air we breathe, the water we drink, the food we eat, and the places we work. A fundamental question in the debate on how to estimate risk and apply uncertainty factors to assure human safety is, when dealing with uncertainty, is it more important to be public health protective by minimizing human exposures to known toxic agents or is it more important to protect the profits of those who release toxic materials into our environment?

McAllister (2004), representing the pesticide industry trade group, equated studies of pesticides in humans to clinical trials of potential pharmaceutical agents. However, there is a major difference in the objectives of these types of studies. After extensive animal experiments, clinical trials are conducted to determine the efficacy of experimental drugs to slow or reverse a disease or predisease condition without inducing significant adverse or life-threatening side effects. In contrast, the motive for testing toxic pesticides in humans is to influence regulations that limit our exposure to known toxic and carcinogenic agents. The Implementation Working Group, a coalition of pesticide manufacturers, farm group, and food processors, acknowledged that the strategy for human experiments is to "avoid the need for the 10-fold UF [uncertainty factor] for interspecies extrapolation" (Environmental Working Group 1998). Pesticide residues in foods have no known health benefits, and the studies of toxic pesticides in humans are not performed to reveal any health gains. Although we disagree with human testing, it is critically important that such studies, if they are to be conducted, must provide sufficient and reliable measures of human risk and that information is properly used in making policy decisions that affect the health of the general public. Reliable extrapolation factors are necessary to protect the public from unnecessary exposures to known hazardous chemicals.

A first consideration is that the biomarker measured in these studies is the most sensitive and reliable end point for estimating human risk. For example, dichlorvos, discussed by Sass and Needleman (2004a), is carcinogenic at multiple sites in rats and mice (Chan et al. 1991), and it induces gene mutations and chromosomal damage in mammalian cells. Carcinogenic risks of dichlorvos are not revealed by measurements of red blood cell acetylcholinesterase (ACHE) inhibition. Also, Sass and Needleman (2004b) noted that measurement of erythrocyte AChE activity in healthy adults is a poor surrogate of neurologic effects that can result from low-dose exposure to pesticides during critical states in fetal and neonatal develoment.

Second, the duration of exposure is important because effects that require long-term or chronic exposure will not be revealed in a single-dose or 3-week study. Human studies must capture the time- and dose-dependent responses that may occur in larger populations that are chronically exposed.

Third, inhibition of AChE by organophosphorous pesticides can occur at very low concentrations. The ability to detect a significant change in exposed humans is a function of the doses used, timing of evaluation, group size, and interindividual variability in factors affecting the pharmacokinetics and pharmacodynamics of the pesticide. Thus, group size must be sufficiently large to detect low-dose effects and to distinguish a LOEL from a true NOEL. Because testing of toxic pesticides in humans has been limited to healthy adults, these data do not inform us of variability due to exposures during different life stages or individual differences in health status, genetics, or coexposure to other agents that act on the same target or affect different steps in the multi-step processes leading to disease. For this reason, we recommend that any decisions made with these data be based on a predicted NOEL for 99% of the healthy adult population and not on calculated mean values. Relevant to the methods by which the U.S. EPA and other regulatory agencies identify an adverse NOEL concerns how these agencies address interindividual differences in susceptibility due to extrinsic and intrinsic factors, such as those listed above. Tests conducted in healthy adults are not predictive of effects that can occur in the fetuses, children, adolescents, pregnant women, the elderly, and the frail. In a recent study Whyatt et al. (2004) have shown that prenatal exposures to organophosphate pesticides (chlorpyrifos and diazinon) from residential use resulted in measurable levels of these insecticides in umbilical cord plasma that were associated with impaired fetal growth. Because a safety factor of 10-fold does not adequately account for the wide range of susceptibility that exists in human populations, there is a need to reexamine the adequacy of this uncertainty factor to provide adequate health protection.

If studies in humans do not adequately inform of the likelihood of time-dependent, dose-related effects in humans, then it is truly unethical to intentionally expose human volunteers to such poisons and carcinogens. Further, if pesticide-exposure studies in humans are deemed essential and necessary, then it is important that volunteers in these studies be duly informed that short-term and particularly long-term health risks from participation in these studies are not known.

Sass and Needleman (2004a) raised valid concerns of potentially misleading conclusions being drawn from studies of toxic pesticides in small numbers of healthy adults. For some reason, and counter to the common practice of EHP, the journal published rebuttal letters promoting the industry perspective in the same issue. These letters claimed that the industry-sponsored pesticide studies in humans were conducted in accordance with ethical standards at the time and principles of good laboratory practice. However, good laboratory practice does not compensate for an inadequate experimental design. We agree with Sass and Needleman (2004a) and with the U.S. EPA Science Advisory Board and the FIFRA (Federal Insecticide, Fungicide, and Rodenticide Act) Scientific Advisory Panel (U.S. EPA 2000), who concluded that justification for the human subjects in pesticide testing "cannot be to facilitate the interests of industry or of agriculture, but only to better safeguard the public health." It is wrong to intentionally dose people with toxic pesticides for the purpose of lobbying the U.S. EPA to lower interspecies extrapolation factors.

The authors declare they have no competing financial interests.

REFERENCES

Chan PC, Huff J, Haseman JK, Alison R, Prejean JD. 1991. Carcinogenesis studies of dichlorvos in Fischer rats and B6C3F1 mice. Jpn J Cancer Res 02:157-164.

Charnley G, Patterson J. 2004. Ethical standards of studies involving human subjects [Letter]. Environ Health Perspect 112:A152-A153.

Chart IS, Manley A, Youngren SH. 2004. Study criticisms unjustified [Letter]. Environ Health Perspect 112:A151-A152.

Environmental Working Group. 1998. The English Patients. Human Experiments and Pesticide Policy. Washington, DC:Environmental Working Group.

McAllister RS. 2004. Statement of CropLife America on pesticide testing involving human subjects [Letter]. Environ Health Perspect 112:A154-A155.

Sass JB, Needleman HL. 2004a. Industry testing of toxic pesticides on human subjects concluded "no effect," despite the evidence [Letter]. Environ Health Perspect 112:A150-A151.

Sass JB, Needlemen HL. 2004b. Human testing: Sass and Needleman respond to industry [Letter], Environ Health Perspect 112:A340-A341.

Tobia A, Ayers A, Blacker A, Hodges L, Carmichael N. 2004. Aldicarb study misrepresented in human testing debate [Letter]. Environ Health Perspect 112:A155-A156.

U.S. EPA. 2000. Comments on the Use of Data from the Testing of Human Subjects. EPA-SAB-EC-00-017. Washington, DC:U.S. Environmental Protection Agency. Available: http://www.epa.gov/sab/pdf/ec0017.pdf [accessed 30 April 2004],

Whyatt RM, Rauh V, Barr DB, Camann DE, Andrews HF, Garfinkel R, et al. In press, Prenatal insecticide exposures, birth weight and length among an urban minority cohort. Environ Health Perspect doi:10.1289/ehp.6641 [Online 22 March 2004].

Ronald L. Melnick

James Huff

National Institute of Environmental

Health Sciences

National Institutes of Health

Department of Health and Human Services 

Research Triangle Park, North Carolina

E-mail: [e-mail]