June 2000 Charge to the SAP
June 27 - 29, 2000
Review the draft document entitled Atrazine: Hazard and Dose-Response
Assessment and Characterization.
Over 10 years ago, atrazine was found to induce mammary tumors in female Sprague-Dawley (SD) rats (Mayhew, 1986). At that time EPA classified atrazine as a possible human carcinogen (i.e., a "Group C" classification category). In 1988, EPA asked the FIFRA Scientific Advisory Panel (SAP) to comment on the cancer classification.
During that meeting, the SAP agreed with the classification and
raised the issue that atrazine exposure may result in an endocrine
imbalance. EPA, therefore, encouraged the registrant of atrazine
to pursue studies on a potential endocrine mechanism. Since that
time, the registrant has completed numerous studies concerning atrazine's
mode of carcinogenic action to explain the mammary gland and pituitary
tumor responses in female Sprague Dawley rats. EPA's National Health
and Environmental Effects Laboratory has also generated information
on atrazine's neuroendocrine site of action, as well as its effects
on reproductive development. EPA has evaluated this information,
and now submits this preliminary assessment to the SAP for comment
on the Office of Pesticide Program's (OPP) evaluation of the chemical's
postulated mode of action and relevance to human hazard potential.
The draft document presents an integrative approach that uses a
common neuroendocrine mode of action to evaluate potential for both
cancer and noncancer health effects (especially reproductive and
Attachments: Document For Review With Respect to Charge ' EPA May 2000 Preliminary Draft Document on Atrazine: Hazard and Dose-Response Assessment and Characterization (enclosed).
Background Materials ' EPA's July 1999 Draft revisions to Guidelines for Carcinogen Risk Assessment (Chapter 2 of this document is enclosed which includes the framework for evaluating a postulated mode of action)
1999 Draft atrazine report of the Cornell University Program on Breast Cancer and Environmental Risk Factors in New York State (The final draft is available as a pdf file. Go to http://www.cfe.cornell.edu/bcerf/CriticalEval/Pesticide/CE.atrazine.pdf Scroll down to the"List of Critical Evaluations" and click on "Atrazine's Breast Cancer Risk". )
1999 International Agency for Research on Cancer (IARC) review of atrazine (enclosed)
Submitted manuscripts from EPA's National Health and Environmental Effects Laboratory on atrazine's reproductive/developmental effects;
Cummings, A.M., Rhodes, B.E., and Cooper, R.L. (Submitted). Effect of Atrazine on Implantation and Early Pregnancy in Four Strains of Rats.
Laws, S.C., Ferrell, J.M., Stoker, T.E., Schmid, J., and Cooper, R.L. (Submitted). The Effects of Atrazine on Female Wistar Rats: An Evaluation of the Protocol for Assessing Pubertal Development and Thyroid Function.
Narotsky, M.G., Best, D.S., Guidici, D.L., and Cooper, R.L. (Submitted). Strain Comparisons of Atrazine-induced Pregnancy Loss in the Rat.
Stoker, T.E., Laws, S.C., Guidici, D., and Cooper, R.L. (Submitted). The Effects of Atrazine on Puberty and Thyroid Function in the Male Wistar Rat: An Evaluation in the Male Pubertal Protocol.
Charge to the Panel: Provide comment and advice on the following questions.
Issue 1: Rodent Tumor Findings
The focus of atrazine induced tumor responses has been on mammary gland neoplasia in female Sprague Dawley rats. The 1999 December draft assessment document concludes, as in previous assessments, that treatment of female Sprague Dawley (SD) rats with atrazine, but not male SD rats or Fischer 344 rats or CD-1 mice of either sex, results in neoplastic responses expressed as an increased incidence and/or an early onset of mammary gland tumors. This preliminary hazard assessment further concludes that the mammary gland adenomas/carcinomas and the fibroadenomas should not be combined because they are of a different cell origin. The current draft assessment also points out that there is an early onset of pituitary adenomas. It is the preliminary conclusions regarding the rodent tumor findings that form the basis of the mode of action analysis.
Question 1.1: Please comment on the EPA preliminary findings that atrazine treatment induces an increased incidence and early onset of: (a) mammary gland adenomas/carcinomas, and (b) fibroadenomas. Please comment on the importance and significance of the data showing histomorphologic changes in mammary tissues (e.g., increased incidences or increased severity of alveolar development, acinar development, dilated ducts, increased secretory activity, and galactoceles) and their relevance to data showing an early onset and increased incidence of mammary fibroadenomas and pituitary adenomas in female SD rats following atrazine exposure. Also, are these histomorphologic effects valid indicators of increased or prolonged exposure to estrogen and prolactin.
Question 1.2: Please comment on the overall EPA preliminary conclusion that atrazine treatment does not lead to tumor formation in male SD rats, and F344 rats and CD-1 mice of either sex. Please comment specifically on OPP's evaluation of the Pinter et al., 1990 study, which reported neoplastic findings in the male F344 rat strain.
Issue 2: Mode of Action Analysis The draft OPP cancer assessment on atrazine includes evaluation of toxicological and mechanistic information to explain the rodent tumor responses and to identify what is believed to be the important elements of the carcinogenic process.
Question 2.1: Sufficiency of Evidence: Does the draft assessment adequately describe the data used to identify the key events in atrazine's mode of carcinogenic action?
Are the available data sufficient to describe these events? Have the uncertainties and limitations of these data been adequately and clearly characterized? Which event(s) is viewed as critical to the carcinogenic process? Are the preliminary conclusions as to atrazine's mode of action supported by the analyses presented in the draft EPA document, and consistent with the mode of action framework analysis described in EPA's July 1999 Draft Guidelines for Carcinogen Risk Assessment?
Question 2.2: Have alternative modes of carcinogenic action been sufficiently discussed and ruled out?
Issue 3: Dose-Response The attenuation of the LH surge is viewed by EPA as the necessary step that eventually leads to neoplasia in rodents. EPA concluded that this critical event is consistent with a nonlinear phenomenon, and thus it proposed that dose-response assessment proceed by a margin of exposure analysis.
Question 3.1: Because it is the attenuation of the preovulatory LH surge that results in disruption of the estrous cycle and is a necessary step in the mode of carcinogenic action, the Draft EPA document proposes that the cancer dose-response assessment should proceed by a nonlinear dose-response extrapolation. The document further proposes that the NOAEL for the attenuation of the LH surge be used as the point of departure for a margin of exposure analysis. We would like the panel to comment on this proposed dose-response approach and the consistency of other effects going along with the LH suppression effect.
Question 3.2: Is there a common endpoint in the mode of action that could be used as a point of departure that would be protective of both cancer and certain reproductive developmental effects (e.g., delayed puberty, prostatitis)? The EPA draft document proposes the LH data as such.
Issue 4: Human Relevance It is EPA science policy that animal tumor responses are presumed to be indicative of human cancer potential unless there is substantive information to the contrary. When there is information on an agent's mode of action in laboratory animals, it is important to address whether or not it would be anticipated to be operative in humans. For atrazine, two questions need answering to determine the relevance of the animal model to humans. One deals with the possibility that atrazine produces neuroendocrine disruption in humans as it does in rats; the second is whether potential adverse human consequences including carcinogenicity may ensue if these neuroendocrine effects develop from atrazine exposure.
Question 4.1: The EPA draft document concludes that there is suggestive evidence of a possible association of triazine exposure and cancer occurrence for three hormone-responsive cancers in humans - ovary, breast and prostate cancer. However, these associations should not be considered as conclusive evidence of an association of triazine exposure with these tumor types, and triazine exposure should not be interpreted as a causal factor in these tumor types. Please comment on EPA's evaluation of the epidemiologic studies, and how much weight it should be given in the hazard and mode of action assessment.
Question 4.2: Does the document provide a thorough and adequate discussion of the similarities and differences in Sprague-Dawley (and Long Evans) rats compared to humans with respect to this hypothalamic-pituitary-ovarian axis perturbation? Given the steps outlined in Figure 2-1 contained in Part A, Chapter 2, what events are conserved and show homology between rats and humans, and when does the process diverge? Currently the document indicates that CNS-GnRH control of pituitary function is conserved and similar.
Question 4.3: Given the neuroendocrine associated effects (i.e.,decreased secretion of hypothalamic catecholamine levels and gonadotropin releasing hormone, attenuation of the LH pituitary surge, prolonged estrogen and prolactin exposure) found with atrazine treatment in rats, please comment on the conclusions drawn in the Draft EPA document regarding relevance and the possible ramifications in humans. What are commonalities between humans and rats in the endocrine effects found that would raise a concern for human health consequences including carcinogenicity? The document considers human conditions of anovulation as a means to judge the potential human health risks. Please comment on the appropriateness of these models in evaluating atrazine.
Question 4.4: Given the toxicological and mechanistic information available on atrazine, OPP has proposed that atrazine be classified as a likely human carcinogen (see EPA's July 1999 draft revisions to the guidelines for carcinogen risk assessment). Please comment on this proposal.
Issue 5: Children's Health Concerns Risks to infants and children from environmental exposure of chemicals may differ qualitatively or quantitatively from those of adults due to biological, physiological and metabolic differences. As stated in the July 1999 Draft revisions to the EPA's cancer risk assessment guidelines, when information is developed to show a mode of carcinogenic action that is expected to be relevant to adults, an evaluation also needs to be made as to whether this mode of action is relevant to children. Because of the absence of direct relevant information on atrazine, the draft document develops a "cogent biological rationale" on whether the postulated mode of carcinogenic action is applicable to children by considering what is understood about the observed endocrine effects in children versus adults and by considering animal data on developmental and reproductive effects. The document also emphasizes the concern for developmental effects.
Charge to the FIFRA Scientific Advisory Panel
Question 5.1: Please comment on the applicability of the neuroendocrine mode of carcinogenic action in adult rats to the fetus, infants, and peripubertal children. Also, please comment on the adequacy of the EPA document in addressing children's cancer concern.
Question 5.2: Does the panel agree that the reproductive developmental findings in rats (e.g., delayed puberty, prostatitis in young animals) are a result of atrazine primary action on the hypothalamic-pituitary-gonadal axis. And if so, would the panel also provide their view of the commonality between the cancer mode of action and the underlying basis leading to adverse reproductive/developmental outcomes, and how well the EPA document addresses this commonality.
Question 5.3: Do the rodent studies showing delayed puberty in both female and male rats raise a concern for children from a clinical perspective? What does it mean to have a delay in puberty in females and males caused by a compound that does not bind to an estrogen receptor?
Question 5.4: Given that atrazine treatment of dams during lactation results in a decrease in prolactin, which may lead to altered TIDA neuron development and eventually prostatitis in young male offspring, what kind of concern does this finding mean to humans, especially as an early life stage susceptibility?