The Poison Paradigm

BPA: Two Disciplines, Two Perspectives on Regulation

Franklynn Bartol

This week's documentary tells the story of endocrine distrupting chemicals, specifically BPA.

Over the course of the 1990s,  a theory of endocrine disruption began to emerge which argued “…that chemicals capable of interacting with the hormone system can elicit long-term and multigenerational effects at minute levels of exposure— levels detected in the ambient environment and inside human bodies.” (Vogel, 2009: 671)

But this went against our established understanding of toxicity and risk. How you may ask? Read on for a primer.

Although this is an oversimplification, there tend to be two perspectives on regulation—risk and hazard—which tend to map on to both disciplinary and geographic divides.

Toxicology: “Nearly 500 years ago, Swiss physician and chemist Paracelsus coined an adage that would become a basic principle of toxicology: ‘The dose makes the poison.’ In other words, the higher the concentration of a toxic chemical, the more toxic it is.” (EHN, 2019).  This perspective assumes a monotonic response (a linear relationship between the amount of a substance and its harm: more is worse). Toxicologists often follow a risk approach, especially in the U.S.

The assumption with a risk approach is that while there is a risk of endocrine disrupting compounds causing adverse effects within human and animal populations, this risk can be managed through the determination of “thresholds” and acceptable levels of exposure.

Endocrinology:  chemicals may be harmful to human health through interactions with the endocrine system. This means that chemicals can be most harmful at very low doses at which they best mimic hormones. These effects often follow a non-monotonic dose-response curve, which might take the shape of a "U" or a bell: harmful effects are seen at low and high doses, but not mid-range doses. T
here are several potential biological mechanisms for these non-monotonic dose-responses, including the opposing effects from multiple hormone receptors. Endocrinologists tend to follow a hazard approach.

The hazard approach is more common in the E.U., and is based on it not being possible to determine safe thresholds of exposure, in part because of the ways in which endocrine disrupting compounds interact with organisms hormonal systems (nonmonotonic). It also holds that testing the adverse effects of each EDC does not reveal how they interact with other compounds and the potential cumulative effects of multiple EDCs or their pathways of exposure. Therefore, thresholds cannot be determined.

Regulating BPA: The Toxicological Model

In her history of the politics of BPA, Vogel (2009) outlines how its regulation was, for a long time, firmly rooted in a toxicological paradigm: “… although BPA’s estrogen like properties (or estrogenicity) were never completely forgotten, its safety was defined by its commercial use in plastics and, accordingly, by its toxic rather than hormone like properties.”
When its use first became widespread in the 1950s, BPA’s safety was treated as a toxicological question: is this substance poisonous, and at what levels?
Inherent in this question was the assumption of a monotonic dose-response relationship; that is, increases in dose increase the adverse effects in a linear fashion. This framing was used to determine ‘acceptable’ levels of exposures: those that are low enough to not cause adverse effects.

The 1958 US Federal Food, Drug and Cosmetics Act enshrined the monotonic model of chemical safety in law: “Prior to this law, hazards were prohibited from the food supply as dangerous per se, regardless of dose” (Vogel, 2009).
The 1958 law applied to thousands of chemicals  (e.g. preservatives, pesticide residues, chemicals in packaging), which now had to have ‘safe’ thresholds of exposure approved by the FDA. Because BPA migrates from packaging and can linings into food, the FDA considered it an indirect food additive.
Early studies showed low general toxicity, fast metabolism in animals, and low levels of contaminants in food. This supported its approval by the FDA for use in food packaging. The FDA did not establish a regulatory standard for BPA until 1988, and work-place standards were never set.

A key assumption about BPA at that time was that it was non-carcinogenic. The 1958 law included a separate set of safety standards for carcinogens, called the Delaney Clause, in which any chemical found to be carcinogenic was considered unsafe at any level. Therefore, BPA was only regulated by dose because it was assumed to be non-carcinogenic.

In the late 1970s researchers at the National Cancer Institute began the first carcinogenesis studies of BPA. The studies operated on the assumption that if there were cancer-causing effects, these would be most evident at high-doses (e.g. a toxicological model). In other words, they did not test for the low-dose carcinogenic effects of estrogen-mimicking compounds that were being discovered in DES at the time.

Significant issues were found with the laboratories testing BPA at the time. A government investigation found rampant fraudulent practices and poor quality-control measures in several of the private laboratories to which NCI had contracted the testing (Vogel, 2009). Despite these issues, in 1982 the newly-established National Toxicity Program, which had taken over management of the studies, released a report claiming there was “no convincing evidence” that BPA was a carcinogen. (According to Vogel: The only other category of evidence at the time was “convincing evidence.”)

As Vogel notes, “This study provided the basis for the first regulatory safety standard for BPA set by the Environmental Protection Agency (EPA) in 1988 and adopted by the FDA as a reference dose. Considering BPA to be a noncarcinogen, the EPA used the lowest dose from the carcinogenesis study as the ‘‘lowest observed adverse effect level’’ and divided this number by an uncertainty factor of 1000 to determine a reference dose of 50 lg/kg of body weight per day”. This has remained the FDA’s reference dose to this day.

Regulating BPA: Challenging the Toxicity Model

By the 1980s BPA production had reached almost a billion pounds per year in the US, making its way into ever more products, such as CD’s and baby bottles. Ironically, a common use of BPA—in plastic lab materials—would lead to its discovery as an actively endocrine-disrupting compound.
In 1993, researchers at Stanford University noted an estrogen-like compound had seeped from polycarbonate flasks (Krishnan et al., 1993). Later, in 1998, reproductive biologist Patricia Hunt discovered BPA had seeped from polycarbonate lab cages, causing genetic abnormalities in the reproductive process of her mice.
She reproduced this ‘accident’, confirming startling the effects: 40% of the eggs from BPA-exposed mice exhibited errors in meiosis (the process by which cells divide to create sex cells for reproduction).

These discoveries spurred on research amidst an emerging field testing the hormone-mimicking effects of synthetic compounds. A key researcher in these efforts was biologist Frederick vom Saal.  Departing from the toxicity model, his seminal studies in the 1990s tested low doses of BPA in pregnant mice—levels at which synthetic estrogens were known to have physiological impacts. vom Saal published his results in 1997, reporting higher prostate weights and higher-than-expected estrogen levels in the BPA-exposed mice . Concurrently, other studies on the effects of low doses of BPA on mammary glands and female reproductive systems echoed Van Saal’s results. These studies upset the view that BPA was safe at low-doses.

In 1996 the US Congress passed the Food Quality Protection Act, which required the EPA to introduce screening programs for endocrine disruptors, which meant agreeing on a definition of endocrine disruptors and adverse effects, as well as establishing testing guidelines and protocols.

Particularly controversial in the development of these guidelines was the debate over whether tests should be conducted at low-doses or if high-dose toxicity tests were sufficient.

In 2000, the EPA asked the National Toxicity Program to review studies of low-dose effects of endocrine disruptors, including BPA. In 2001 it concluded that there was credible evidence of adverse effects from low exposures of BPA, suggesting “…the current testing paradigm used for assessments of reproductive and developmental toxicity should be revisited to see if changes are needed regarding dose selection, animal model selection, age when animals are evaluated, and the end points being measured following exposure to endocrine-active agents.”

Industry responded to this by attacking the NTP’s study, claiming that they did not conduct a “weight of evidence assessment”. The American Plastics Council then funded the Harvard Centre for Risk Analysis to conduct their own review of studies on BPA. The review, published in 2004, cast doubt on BPA’s effects at low doses.  It claimed that only two longitudinal studies—both industry-funded—were reliable enough to consider.

Both studies would continue to be heavily relied upon in FDA reviews. In 2005, vom Saal and an original participant of the Harvard review panel published a critique of the Harvard review, arguing that industry-funded studies had distorted the role of BPA as an EDC at low doses and highlighting the discrepancy between the findings of government-funded and industry-funded studies.

Regulating BPA: Government Regulations and Debates

Amidst the rising pile of studies supporting adverse health effects at BPA exposures far below ‘safe’ levels, governments got more involved. In 2007, the US federal government invited 38 experts to Chapel Hill, NC to assess the scientific literature on BPA.

The product of this review, the Chapel Hill Consensus Statement stated, “BPA at concentrations found in the human body is associated with organizational changes in the prostate, breast, testis, mammary glands, body size, brain structure and chemistry, and behavior of laboratory animals.”

It also argued that while BPA is not persistent in the environment or people, exposure is continuous. Which is problematic because most studies used to determine safe levels were based on acute exposure, not chronic, continuous low-level exposure.

In 2008 Lang et al published some of the first evidence linking BPA to heart disease, liver toxicity, and diabetes in humans. Alongside these health concerns, more evidence started to emerge that BPA had become a regular part of our biology; for example, the urine of 91% of Canadians surveyed contained BPA, with children aged 6 to 11 years having the highest concentrations (Haines et al, 2010).

Following the Chapel Hill Consensus Statement, the Centre for the Evaluation of Risks to Human Reproduction, a part of the NTP, published a review that found “some concern for effects on the brain, behaviour and prostate gland in foetuses, infants and children at current human exposures to BPA.”

In August, 2008, the FDA released a draft assessment of BPAs safety in which they maintained that there was “no observed adverse effect level” reported in major studies. Their review had relied heavily on industry-funded studies, rejecting others as having insufficient methodological rigor. The FDA Science Board Subcommittee on Bisphenol A, charged with reviewing the FDA report, disagreed with the conclusion and criticized the report’s exclusion criteria as not scientifically-grounded.

Health Canada also departed from the FDA. It had published its own findings on BPA in April 2008, concluding that “early development is sensitive to BPA”. In the wake of the report Health Canada announced they would “…ban polycarbonate baby bottles and set limits on how much BPA can migrate from infant formula cans”. A number of states in the US began to seek bans on BPA in children’s products.

In 2009, the Endocrine Society published an unprecedented scientific statement declaring endocrine disruptors a “significant concern to public health”. They called for stricter regulations, acknowledging EDC’s non-monotonic responses and that “even infinitesimally low levels of exposure — indeed, any level of exposure at all — may cause endocrine or reproductive abnormalities (Fagin, 2012).

That same year, Chuck Schumer introduced the BPA-free kids act, which would ban the use of BPA in any product designed for a child. The bill died in committee. However, with growing public concern, retailers began pulling BPA-containing baby products from their shelves.

Reviewing and accepting the report of the Science Board Subcommittee, in summer of 2009 the FDA  announced plans to conduct extensive toxicity tests of BPA at the NCT. The EPA soon followed suite, announcing their own plans for toxicity testing. The NTP’s review concluded there was some concern that BPA could have adverse effects on the brain, prostate gland in foetuses, infants and children at levels then deemed safe by the EPA.

In 2010, the FDA, which in 2008 declared BPA safe, stated that it concurred with the NTP; yet, the agency took no action at this time. That same year, Canada declared BPA a toxic substance.

At the same the European Food Safety Authority and a WHO panel concluded that there was insufficient evidence that BPA was a risk to human health. Indeed, the WHO expert panel recommended no new regulations limiting or banning BPA, stating “initiation of public health measures would be premature.”

In 2012, the FDA concluded an assessment of scientific research on the effects of BPA and stated in the March 2012 Consumer Update that "the scientific evidence at this time does not suggest that the very low levels of human exposure to BPA through the diet are unsafe" although recognizing "potential uncertainties in the overall interpretation of these studies including route of exposure used in the studies and the relevance of animal models to human health.

The FDA is continuing to pursue additional research to resolve these uncertainties." That same year the FDA banned BPA from baby bottles and children’s cups. As Resnik and Elliott note, the ban was not based on a conclusion of BPA’s negative health effects, but was introduced at the request of manufacturers.

That same year, WHO finally reversed its long-held, risk model position that there was insufficient evidence to classify it as a toxic chemical at low exposure levels. After a global review of the evidence, the WHO’s landmark report recognized BPA’s dose-dependent impacts at different developmental stages.

The report concluded that, through alterations of the endocrine system, BPA is harmful to humans and wildlife and associated with health outcomes such as prostate cancer, breast cancer, testicular cancer, thyroid cancer, ADHD, asthma, obesity, and diabetes (WHO/UNEP 2012)

As you can see, the does doesn't always make the poison.