by Sarah Vogel at the Pump Handle

On Friday, August 15, the FDA released its draft assessment of the safety of bisphenol A (BPA). To the frustration and deep consternation of many, the regulatory agency upheld the current safety standard for human exposure to BPA in food. The agency based their decision on two large multigenerational studies funded by the American Plastics Council (part of the American Chemistry Council) and the Society of the Plastics Industry.[1] As for the large body of literature on low dose effects of BPA that originally raised concerns about the chemical’s ability to disrupt reproductive, neurological and metabolic development and function at levels of exposure within the range found in humans,[2] the FDA broadly found these studies to be inadequate or of limited utility in evaluating safety. By relying solely on the industry-funded studies, the agency reaffirmed the trade associations’ ability to control what is considered to be reliable, credible science. This protracted battle over BPA safety is about more than just this one chemical. It reveals a larger struggle to define whose science gets to count, how information shapes public policy decisions and how scientific knowledge informs chemical safety.

Whose definition of safety?
Safety, for the FDA, “means that there is reasonable certainty in the minds of credible scientists that the substance is not harmful under the intended conditions of use.” Further, “absolute harmlessness” can never be certain.[3] Based on this definition used to guide the FDA’s BPA assessment, one could draw the following conclusions: 1. the agency speaks from a position of credibility and scientific authority; 2. industry-funded studies represent credible research for evaluating safety; and 3. the body of peer-reviewed, published, low dose research on BPA that used “non-traditional” design methods and endpoints does not represent valid research for evaluating safety.

If we then accept these conclusions, what does this mean about the “minds” of scientists that are reasonably certain that BPA is not safe at current levels of human exposure? What about the group of 38 researchers, responsible for close to 100 published studies on low dose effects of BPA, who came to scientific consensus about the inadequacy of the current safety standard in 2007?[4] What does it say about the credibility of the National Toxicology Program’s staff scientist who stated they have “some concern for effects on the brain, behavior and prostate gland in fetuses, infants and children at current human exposures to bisphenol A”, and that although there are limitations to and controversy over the low dose effects of BPA, “the possibility that bisphenol A may alter human development cannot be dismissed”?[5]

By casting the large body of low dose research on BPA as unreliable and of questionable value for assessing human safety, the agency refused to address the pressing scientific and regulatory issue of today: how can low dose research better inform our understanding of chemical risks to public health? What is perhaps most unsettling is the similarity of the FDA’s approach to the low dose literature to “weight of the evidence” reviews of BPA conducted with funding from the American Plastics Council.[6]

Industry’s “weight of the evidence”
It should be noted here that the FDA report does not disclose the funding conflict with several “weight of the evidence” evaluations. Indeed, the FDA report reads as if these “weight of the evidence” evaluations were government reviews. It states, “Toxicity data have been summarized in numerous reviews and assessments prepared by government bodies.”[7] A single footnote, which appears after ‘reviews’, lists the evaluations funded by the American Plastics Council;[8] a paper written by employees of CANTOX Health Sciences International, a Canadian based “product defense” firm;[9] and a review by C.C. Willhite, an employee of the California EPA who conducted a review on his own time[10], along with the NIEHS’s Center for the Evaluation of Risks to Human Reproduction (CERHR) assessment and review of the NTP’s 1982 carcinogenesis bioassay written by a NIEHS scientist (who participated on the 1982 review panel at the NTP)[11]. Perhaps this is a careless oversight, but it raises the question as to why the agency so readily holds industry-sponsored research up as the gold standard (and apparently easily confused with government-sponsored reviews).

Similarly, in its assessment of developmental neurotoxicity, the FDA’s Office of Food Additive Safety conducted a review of the literature that relied heavily on an industry-funded document. The FDA assessment reviewed and updated a “weight of the evidence” evaluation conducted for the American Plastics Council by Exponent, Inc, a product defense firm. Once again the FDA findings concurred with the industry-funded assessments.[12]

Is the use of these industry-funded “weight of the evidence” assessments indicative of a conflict of interest? Not directly or overtly, but it certainly suggests a disturbing pattern of the FDA reliance on industry-funded research and assessment of the literature. It is clear from this report that the industry trade associations wield considerable influence over the FDA decision-making process and the agency’s understanding of what constitutes reliable, credible science.

Credible, relevant and reliable science, according to the FDA
Credible, relevant and reliable research on the safety of BPA, according to the FDA, consists of large, multigenerational studies that use multiple doses, expose animals orally and continuously, and — here’s an important point — apply Good Laboratory Practices (GLP). And lo and behold…the trade associations funded just such work. Let’s start with the last condition because the conflation of GLP as the gold standard for scientific research is just blatantly misleading. Good Laboratory Practices were established in response to abysmal conditions and deliberate mismanagement of data found in private laboratories in the 1970s.[13] In effect, GLP outlines the very basic standards of animal care, study design, data recording, and monitoring. This is not to argue that GLP standards are not absolutely necessary for industry-funded research; however, to suggest these guidelines for conducting science confer a greater degree of validity or relevance is entirely misleading. A study that has followed GLP standards is by no means more rigorous, reliable, relevant or valid than a peer-reviewed paper. Peer-reviewed research, funded by the federal government, is subject to extensive review; first, after the submission of a proposal for research funding and second, when individual studies are submitted for publication.

The battle over the safety of BPA isn’t simply about the quality of the industry-funded GLP studies, it is also about their relevance to public health. In assessing these studies, it is imperative to ask what questions these studies were designed to answer and whether this research is relevant to human exposure and disease experience (i.e. environmentally relevant doses and diseases such as reproductive cancers and abnormalities, metabolic disorders, and neurobehavioral abnormalities). These industry-funded studies were specifically designed to assess reproductive toxicity, not the endocrine effects that can disrupt development millions of times below the toxic threshold and within the range reported in humans.[14] As such, the multigenerational studies measured traditional toxic endpoints. But these are not the endpoints of concern associated with low-dose exposure to BPA. For example, these toxicity studies did not examine any effects on the brain, which is of the highest concern for the National Toxicology Program—effects which were just recently replicated in non-primate humans.[15] Similarly, they did not examine the mammary or prostate gland for possible precancerous lesions.

In its report the FDA acknowledges that these multigenerational studies do not address the “non-traditional” endpoints measured in much of the low dose literature.[16] The question that must be asked then is whether the use of higher doses and the examination of toxic endpoints provide relevant evidence for assessing the safety of the public’s exposure experience. The answer can only be yes if you adhere to the notion that effects of endocrine disrupting chemicals at low doses can be predicted from high dose studies. This assumption fails to fit the reality of the effects of low-doses of endocrine disrupting chemicals, a position that was confirmed by the 1999 National Research Council panel on Hormonally-Active Agents in the Environment.[17]

As for those studies that did examine the effects of extremely low doses (levels relevant to the human experience) the FDA’s assessment discusses a number of issues that the agency contends renders such work unreliable. For example, the FDA notes that because of the limited space available in published studies, they could not often verify the integrity of the published data.[18] (Those familiar with the Data Quality Act will note a familiar argument here.) Such an argument effectively disqualifies most government funded, published research as unreliable for assessing safety.

The FDA report also argues that because many of the low dose studies exposed animals only during a short critical window of development, these studies are considered less relevant for assessing human safety. This is because, they argue, such “’windows’ of exposure to BPA are not expected to occur in the human population as exposure is continuous.” [19] This is a ridiculous argument. Humans may be exposed continuously (with levels potentially increasing)[20], but it is exposure during early development when cells are proliferating and differentiating and neurological, reproductive and sexual systems are being laid down that is of the greatest concern. This understanding draws on strong epidemiological and laboratory work on the fetal origins of disease.[21] Additionally, exposure during early development is of further concern because of the reduced ability for human fetuses and infant to metabolize BPA.[22]

The relevance of low dose research in assessing safety
The Centers for Disease Control and Prevention’s biomonitoring program now tests for and has detected over a hundred synthetic compounds in Americans, including BPA.[23] Environmental health researchers, many of whom are funded by the National Institutes of Health, are studying the effects of these low levels of BPA present in human bodies on multiple health endpoints of concern in the U.S. population—metabolic disorders, like obesity, diabetes and heart disease; neurobehavioral abnormalities; reproductive abnormalities like polycystic ovarian disease and increased susceptibility to cancer.[24] The FDA referred to studies that examined developmental effects of extremely low doses of BPA on the mammary and prostate gland development as mechanistic research. This ‘mechanistic’ work, they concluded, made these data of “questionable usefulness.” [22]

There is no doubt that the research questions posed by the low dose research on BPA are incredibly complex. BPA appears to disrupt many endocrine signaling systems with different effects in different tissues and in different species. As such, no single study will confirm all that is known or not known about BPA. Isolating the mechanisms of action or examining the effects of exposure during critical windows of development on specific developmental processes contributes to a complex web of knowledge about BPA. This body of knowledge, which has expanded rapidly since the late 1990s, cannot be simply undermined by two studies that asked very different questions about the chemical’s toxic effects. The entire body of literature or the entire picture must be considered as was done by the NTP and by the Chapel Hill group.

The FDA’s failure to integrate low dose research in the assessment of safety reveals a serious lack of understanding about the direction of environmental health research today. One of the most pressing issues facing chemical policy today is how to integrate the mounting biomonitoring data with the growing body of literature examining the health effects of these low doses of exposure to improve the process by which chemical safety is evaluated. The agency offers some important recommendations for expanding research on BPA, such as the need to gain a better handle on the internal dose of free BPA (estrogenic form) and the pharmcokinetics in non-human primates. But demanding more and more certainty about low dose effects while upholding the safety of the compound based on toxicity testing reveals a flawed logic.

Studies like the industry-sponsored developmental and reproductive toxicity tests, which tested doses in the toxic range, only provide evidence of high dose effects. As such, these studies do not provide evidence of effects at low doses. This statement is fundamentally different than stating that these studies provide evidence of no effects at low (untested) doses.

What is desperately needed is an open and participatory discussion about the impact of low dose research and biomonitoring data on our understanding of what constitutes a safe chemical. This is an issue that will not go away. Safety can no longer be defined by the absence of overt toxic effects; instead, researchers can and should examine developmental effects and those effects that better reflect diseases of public health concern, like susceptibility to cancer, as well as cancerous tumors, metabolic disorders, reproductive and neurological disorders and disease.

The FDA’s draft assessment has made no effort in addressing the major issue of low dose exposure confronting environmental health today. Indeed, the agency’s reasoning for excluding the body of low dose research is myopic and at times ridiculous. But the most unsettling aspect of this draft assessment is its unwavering support not only for research funded by the industry trade associations, but also for the industry’s definition of what constitutes relevant and reliable science. In its refusal to acknowledge the validity of peer-reviewed, government funded low dose work, the agency impedes progress in public health prevention.

Sarah Vogel received her PhD from Columbia University in the Department of Sociomedical Sciences’ Center for the History and Ethics of Public Health and Medicine; her dissertation was entitled “Politics of Plastic: the economic, political and scientific history of bisphenol A.” She holds master’s degrees in public health and environmental management from Yale University. She authored the case study “Battles Over Bisphenol A” at DefendingScience.org.

[1] R. W. Tyl et al., “Three-Generation Reproductive Toxicity Study of Dietary Bisphenol a in Cd Sprague-Dawley Rats,” Toxicol Sci 68, no. 1 (2002), ———, “Two-Generation Reproductive Toxicity Study of Dietary Bisphenol a in Cd-1 (Swiss) Mice,” Toxicol Sci 104, no. 2 (2008).
[2] L. N. Vandenberg et al., “Exposure to Environmentally Relevant Doses of the Xenoestrogen Bisphenol-a Alters Development of the Fetal Mouse Mammary Gland,” Endocrinology 148, no. 1 (2007).
[3] FDA, “Draft Assessment of Bisphenol a for Use in Food Contact Applications,” (2008), 3.
[4] F. S. Vom Saal et al., “Chapel Hill bisphenol A expert panel consensus statement: Integration of mechanisms, effects in animals and potential to impact human health at current levels of exposure,” Reprod Toxicol 24, no. 2 (2007).
[5] National Toxicology Program, U.S. Department of Health and Human Services, Center for the Evaluation of Risks to Human Reproduction. NTP-CERHR Monograph on the Potential Human Reproductive and Developmental Effects of Bisphenol A. September 2003, NIH Publication No. 08-5994, 7.
[6] J. E. Goodman et al., “An Updated Weight of the Evidence Evaluation of Reproductive and Developmental Effects of Low Doses of Bisphenol A,” Crit Rev Toxicol 36, no. 5 (2006), George M. Gray et al., “Weight of the Evidence Evaluation of Low-Dose Reproductive and Developmental Effects of Bisphenol A,” Human and Ecological Risk Assessment 10 (2004).
[7] FDA, “Draft Assessment of Bisphenol a for Use in Food Contact Applications,” 14.
[8] Ibid.
[9] L. A. Haighton et al., “An Evaluation of the Possible Carcinogenicity of Bisphenol a to Humans,” Regul Toxicol Pharmacol 35, no. 2 Pt 1 (2002).
[10] Wilhite C. C. Willhite, G. L. Ball, and C. J. McLellan, “Derivation of a Bisphenol a Oral Reference Dose (Rfd) and Drinking-Water Equivalent Concentration,” J Toxicol Environ Health B Crit Rev 11, no. 2 (2008).
[11] Center for the Evaluation of Risks to Human Reproduction CERHR, “Ntp-Cerhr Expert Panel Report on the Reproductive and Developmental Toxicity of Bisphenol A,” (2007), J. Huff, “Carcinogenicity of Bisphenol-a in Fischer Rats and B6c3f1 Mice,” Odontology 89, no. 1 (2001).
[13] Gerald E. Markowitz and David Rosner, Deceit and Denial : The Deadly Politics of Industrial Pollution (Berkeley, CA: University of California Press, 2002), 215.
[14] L. N. Vandenberg et al., “Human Exposure to Bisphenol a (Bpa),” Reprod Toxicol 24, no. 2 (2007).
[15] Csaba Leranth et al., “Bisphenol a Prevents the Synaptogenic Response to Estradiol in Hippocampus and Prefrontal Cortex of Ovariectomized Nonhuman Primates,” Proceedings of the National Academy of Sciences 105, no. 37 (2008).
[16] FDA, “Draft Assessment of Bisphenol a for Use in Food Contact Applications,” 29.
[17] National Research Council (U.S.). Committee on Hormonally Active Agents in the Environment., Hormonally Active Agents in the Environment (Washington, D.C.: National Academy Press, 1999), 109-18.
[18] FDA, “Draft Assessment of Bisphenol a for Use in Food Contact Applications,” 16.
[19] Ibid., 33.
[20] National Toxicology Program, U.S. Department of Health and Human Services, Center for the Evaluation of Risks to Human Reproduction. NTP-CERHR Monograph on the Potential Human Reproductive and Developmental Effects of Bisphenol A. September 2003, NIH Publication No. 08-5994, 5.
[21] J. J. Heindel, “The Fetal Basis of Adult Disease: Role of Environmental Exposures–Introduction,” Birth Defects Res A Clin Mol Teratol 73, no. 3 (2005), L. S. Birnbaum and S. E. Fenton, “Cancer and Developmental Exposure to Endocrine Disruptors,” Environ Health Perspect 111, no. 4 (2003).
[22] National Toxicology Program, U.S. Department of Health and Human Services, Center for the Evaluation of Risks to Human Reproduction. NTP-CERHR Monograph on the Potential Human Reproductive and Developmental Effects of Bisphenol A. September 2003, NIH Publication No. 08-5994, 6.
[23] A. M. Calafat et al., “Exposure of the U.S. Population to Bisphenol a and 4-Tertiary-Octylphenol: 2003-2004,” Environ Health Perspect 116, no. 1 (2008).
[24] R. R. Newbold, W. N. Jefferson, and E. Padilla-Banks, “Long-Term Adverse Effects of Neonatal Exposure to Bisphenol a on the Murine Female Reproductive Tract,” Reprod Toxicol 24, no. 2 (2007), T. J. Murray et al., “Induction of Mammary Gland Ductal Hyperplasias and Carcinoma in Situ Following Fetal Bisphenol a Exposure,” Reprod Toxicol (2006), Eric R. Hugo et al., “Bisphenol a at Environmentally Relevant Doses Inhibits Adiponectin Release from Human Adipose Tissue Explants and Adipocytes,” Environ Health Perspect in press (2008).