Re: “A Love Letter to Canned Foods”

Last week, The New York Times published an article in their Food section highlighting meal ideas based on canned food. In response, Dr. Leonardo Trasande (NYU) and I wrote a letter to the editor with some of our concerns. This letter did not get published, so I’m posting here instead.

As avid cooks, we love reading columns from Melissa Clark. But as environmental health researchers, we were concerned that her recent piece, “A Love Letter to Canned Food,” fails to discuss potential health concerns associated with metal cans. Their linings contain bisphenols, such as bisphenol-A (BPA), or the wide array of “regrettable substitutes,” which can interfere with our body’s hormones and disrupt our developmental, reproductive, neurological, and immune systems. All of this is described in our American Academy of Pediatrics technical report and policy statement on “Food Additives and Child Health.” For canned food to continue to be a convenient, affordable and nutritional option for feeding our families, we need systemic policy changes that ensure that any additives are fully tested for safety prior to use in the marketplace. Our own work suggests that replacing BPA in cans with safer alternatives may produce economic benefits to society greater than the costs.  

Rachel M. Shaffer, MPH
PhD Candidate, Environmental Toxicology
Department of Environmental and Occupational Health Sciences
University of Washington School of Public Health

Leonardo Trasande, MD, MPP
Jim G. Hendrick, MD Professor and Vice Chair, Department of Pediatrics
Chief, Division of Environmental Pediatrics
Professor of Environmental Medicine & Population Health
NYU School of Medicine

High Impact Report on Low-Dose Toxicity

A condensed version of this post was originally published on The Conversation with the title “Can low doses of chemicals affect your health? A new report weighs the evidence.

Toxicology’s founding father, Paracelsus (1493-1541), is famous for his paraphrased proclamation: “the dose makes the poison.” This phrase represents a pillar of traditional toxicology: Essentially, chemicals are harmful only at high enough doses.

But, increasing evidence suggests that even low levels of “endocrine disrupting chemicals (EDCs)” can interfere with hormonal signals in the body in potentially harmful ways.

Standard toxicity tests don’t always detect the effects that chemicals can have at lower levels. There are many reasons for this shortcoming, including a focus on high dose animal testing as well as failure to include endpoints relevant to low dose disruption. And, even when the data do suggest such effects, scientists and policymakers may not act upon this information in a timely manner.

Recognizing these challenges, the U.S. Environmental Protection Agency (EPA) asked the National Academy of Sciences convene a committee to study the issue in detail. How can we better identify whether chemicals have effects at low doses? And how can we act on this information to protect public health?

After several years of work, the committee’s report was released in July. This landmark report provides the EPA with a strategy to identify and systematically analyze data about low-dose health effects, as well as two case study examples. It is an evidence-based call to action, and scientists and policymakers should take notice.

Delving into definitions

Before discussing the report, let’s review some definitions…

We know that animal experiments usually use high doses, but in comparison, what is a “low dose?”

This issue was a matter of considerable debate, but ultimately, the committee decided to proceed with a fairly general definition of low dose as “external or internal exposure that falls with the range estimated to occur in humans.” Therefore, any dose that we would encounter in our daily lives could be included, as well as doses that would be experienced in the workplace.

The committee also clarified the meaning of “adverse effects.” When a chemical produces a visible malformation, it is easy to conclude that it is adverse. But, when a chemical causes a small change in hormone levels, it is more difficult to conclusively state that the change is adverse. Are all hormone changes adverse? If not, what is the threshold of change that should be considered adverse?

In this context, an adverse effect was defined as “a biological change in an organism that results in an impairment of functional capacity, a decrease in the capacity to compensate for stress, or an increase in susceptibility to other influences.”

A strategy to identify low dose toxicity

With these semantics settled, the committee developed a 3-part strategy to help with timely identification, analysis, and action on low-dose toxicity information:

(1) Surveillance: Active monitoring of varied data sources and solicitation of stakeholder input can provide information on low dose effects of specific chemicals, especially since EPA’s standard regulatory testing framework may not always identify such effects. Human exposure and biomonitoring data should also be collected to help define relevant exposure levels of concern across the population.

(2) Investigation & Analysis: Systematic review and related evidence integration methods can be used to conduct targeted analysis of the human, animal, and in vitro studies identified in the surveillance step. Each of these approaches has different strengths and weaknesses, so examining the evidence together offers insight that a single approach could not provide.

(3) Actions: New evidence can be incorporated into risk assessments or utilized to improve toxicity testing. For example, protocols could be updated to include newly identified outcomes relevant to endocrine disruption.

Leading by example: systematic review case studies

To put their strategy into practice, the committee conducted two systematic reviews of low dose EDC effects.

The first case study looked at phthalates, chemicals that increase the flexibility of plastic products such as shower curtains and food wrapping.

The committee found that diethylhexyl phthalate and other selected phthalates are associated with changes in male reproductive and hormonal health. Overall, the data were strong enough to classify diethylhexyl phthalate as a “presumed reproductive hazard” in humans.

The second case study focused on polybrominated diphenyl ethers (PBDEs), flame retardants used for over 30 years. Though they are now being phased out, these chemicals remain a concern for humans. They are still present in older products and can persist in the environment for many years.

Based on data showing the impact of these chemicals on learning and IQ, the panel concluded that developmental exposure is “presumed to pose a hazard to intelligence in humans.”

Questions and challenges for the future

During its review, the committee encountered a variety of barriers that could impede similar investigations into specific chemicals.

First, when reviewing evidence, it’s important to assess any systematic errors – also known as biases – that might have led to incorrect results. These errors can arise from study design flaws, such as failure to properly blind the researchers during analysis.

Some journals have strict guidelines for reporting details related to bias, but many do not. Better adherence to reporting guidelines would improve scientists’ ability to assess the quality of evidence.

Second, the committee noted a discrepancy between the concept of doses used in human and animal studies, which made it difficult to compare data from different sources.

For example, most toxicologists simply report the dose that they delivered to animals. But some of that administered dose might not actually be absorbed. The actual internal dose of chemical circulating in the body and causing harm may differ from the amount that was administered. By contrast, epidemiologists usually think about dose as the level of chemical they detect in the body, but they may not know how much of the chemical an individual was actually exposed to.

Biological modeling techniques can help scientists draw the connection between administered and internal doses and more closely compare results from animal and human studies.

Finally, many toxicology studies focus on only a single chemical. This is a valuable way to identify how one chemical affects the body. However, given that we are all exposed to chemical mixtures, these procedures may be of limited use in the real world.

The committee suggested that toxicologists incorporate real-world mixtures into their studies, to provide more relevant information about the risk to human health.

Leveraging toxicity testing for answers about low dose effects

This report demonstrates one of the challenges facing the field of toxicology and environmental health: How well can existing and emerging laboratory techniques predict adverse outcomes in humans? (If you’ve read some of my previous posts, you know that this issue is of particular interest to me.)

Traditional animal experiments usually use high doses, which don’t necessarily reflect the real world. These studies can be an important first step in identifying health hazards, but they cannot accurately predict how or at what levels the chemicals affect humans. The committee noted that more relevant doses and better modeling could help mitigate this problem.

Emerging high-throughput testing techniques use cell-based methods to detect how a chemical changes specific molecular or cellular activities. These newer methods are increasingly used in toxicology testing. They have the potential to quickly identify harmful chemicals, but have yet to be fully accepted or validated by the scientific community.

For these two case studies, the committee noted that high-throughput tests were not particularly helpful in drawing conclusions about health effects. Many of these studies are narrowly focused – looking at, for example, just a single signaling pathway, without indicating a chemical’s overall influence on an organism. Nevertheless, these methods could be used to prioritize chemicals for further in-depth testing, since activity in one pathway may predict a chemical’s capacity to cause harm.

Putting the report into action

Despite the imperfections of our testing methods, there’s already ample evidence about low-dose effects from many chemicals (including the two cases studies from the committee). The EPA should implement this new strategy to efficiently identify and act on problematic endocrine-disrupting chemicals. Only through such strong, science-based efforts can we prevent adverse effects from chemical exposures – and allow everyone to live the healthy lives that they deserve.

Lessons from A Toxicology Detective Story: Use Caution with Controls

This article was originally posted on the Massive Science Consortium website, with the title “An Unexplained Result Shows Why Studying the Effects of Chemicals is so Tricky

Toxicologists are no strangers to mysteries. In fact, understanding unexpected results caused by unintentional chemical contamination in the laboratory has a storied history in the environmental health field.

In the late 1980’s, researchers at Tufts University accidentally discovered that certain plastic components caused their estrogen-responsive cells to grow uncontrollably. Their findings spurred extensive work on endocrine disrupting chemicals (EDCs), which can interfere with normal hormonal activity in the body. Bisphenol-A (BPA), found in many plastic products, is a common example of an EDC.

Following in this tradition, researchers at the University of Massachusetts (UMass) Amherst recently conducted some important toxicology detective work after they noticed that mammary glands of adult male mice raised in a commercial laboratory (that supplies mice for scientists) were larger and more developed than the mammary glands of the same type of mice raised in their own laboratory.

Scientists use rodent mammary glands as models of human breasts, which allow them to better understand growth and development as well as risk factors and treatments for diseases like breast cancer. Toxicologists have paid particular attention to how mammary glands change after exposure to EDCs, since the chemicals can interfere with hormonal activity and mammary glands are especially responsive to hormonal signals.

So, when the UMass Amherst researchers noticed a difference in the mammary glands between the two groups of theoretically similar mice, it set off some alarm bells. Could an unidentified EDC exposure in the commercial lab be the culprit? And if so, could this potential EDC exposure impact the ability of scientists and policy-makers to draw conclusions from toxicological experiments?

Comparing Gland Growth

To find out, the researchers carefully compared the mammary glands and blood hormones between the two groups of mice. One group was ordered directly from a commercial supplier. The other group was ordered from the same commercial supplier and then bred for two generations to produce offspring that were raised in their own lab under controlled conditions to minimize exposure to EDCs.

The findings supported their preliminary observations: the mammary glands in the male mice raised in the commercial lab were larger and more developed than those of male mice raised in their own lab. The researchers noted that the commercially-raised mouse mammary glands actually mirrored those of mice from different experiments that had been intentionally exposed to BPA during early development.

By contrast, female mice had smaller and less developed mammary glands at puberty. While this difference in response may at first seem counter-intuitive, endocrine disruptors are complicated. Because they interfere with hormones, which are gender-specific, EDCs have the potential to affect males and females in different ways.

Contemplating the Culprit

Although the researchers found differences between the two sets of mice, they didn’t immediately know why. A simple explanation would have been that different amounts of circulating hormones at the time they examined the mice might have influenced their body composition. Yet, they detected no significant differences in estrogen, the primary hormone that drives mammary gland development, between the two groups.

It is also unlikely that genetic differences could have contributed to the distinct mammary gland growth, given that the non-commercially raised animals were actually only two generations removed from the original commercial strain. Such drastic genetic changes do not usually occur over such short cycles.

Therefore, the researchers hypothesized that the difference was more likely due to the effects of EDC exposures during their early life in the commercial laboratory. (However, they were not able to confirm this theory through specific tests.) Exposures to EDCs, among other chemicals, during sensitive windows of development in early life have the potential to cause long-lasting changes, including in the mammary gland.

Considering the Consequences

These findings, which suggest that animals raised in commercial labs may be exposed to EDCs, could impact how we interpret toxicological studies.

One main reason is what researchers call the “two-hit” model, which suggests that an exposure in early life makes an individual more sensitive to the effects of a second exposure later in life. In this context, early exposures to EDCs might prime laboratory animals for more pronounced responses when treated with a test chemical in an experiment later in life. In other words, some laboratory experiments may be erroneously linking the test chemical to an outcome that is actually due to a combination of the test chemical and being exposed to EDCs earlier in life.

This study only evaluated animals from one commercial lab, and conditions may differ in other labs. The possibility that animals could be subject to different unintentional exposures may affect our ability to compare studies and pool data from diverse labs to make science-based policies. This issue may also partially explain why research and policy on EDCs has been so highly controversial, with distinct labs generating very different results about the same chemicals.

These differences also have implications for a controversial practice in the field of toxicology: the use of “historical controls.” Sometimes, scientists compare the changes in treated animals from one study to a database of untreated animals from a collection of previous studies. This practice can provide researchers with a better sense of whether the treated animals they are studying are truly different from normal. Yet, this study suggests that control animals raised in some laboratories may be exposed to EDCs, and therefore it would not be appropriate to compare them to treated animals raised in different environments. The presence of too many differing variables would make it difficult to make an accurate comparison.

Taken together, these findings suggest that both scientists and the public should be cautious when interpreting certain studies. While the researchers did not conclusively confirm the linkage to early life EDC exposure, this highly likely explanation illustrates that, at least in some cases, there may be factors behind the scenes that could be influencing the results of toxicology experiments.

How should the scientific community address the implications of this study? While researchers could try to mitigate the impact of these exposures on their subsequent toxicological experiments by screening animals prior to beginning their work, a better approach would be to improve handling of animals in commercial facilities through strict standards. Exposures to EDCs, among other chemicals, should be minimized. If such exposures continue, it may be hard to trust the results of these toxicological studies, which could impede the development of appropriate, evidence-based environmental health policies and protections for the population.