Hormones and Endocrine Disrupting Chemicals: 

Low-­-Dose Effects and Nonmonotonic Dose Responses

This FAQ by Pete Myers

1.   What are the main findings of this paper? 
For years, scientists have looked at single chemicals and asked whether they have effects at low doses, in the range of what humans typically experience. That single-­-chemical 
approach has produced a huge amount of data, but little analysis has been done to make conclusions on a wider scale. We approached this huge body of literature and asked whether there was sufficient evidence to conclude that many or most hormone-­-mimicking chemicals have effects at low doses. We compiled examples of 30 endocrine-­-disrupting chemicals  (EDCs) that have been shown to cause low dose effects—effects very different than those at high doses.   
 

We wanted use take this broad perspective to address whether current regulatory practices for determining chemical safety were sufficient to protect human health. We found overwhelming evidence that low doses of chemicals are not safe for humans or wildlife. The failure of traditional testing methods to detect low dose responses has large implications for chemical regulation. It means that many chemical safety standards will have to be strengthened substantially to protect public health.

2.   What are “low dose effects?

Usually, “low dose” refers to doses that are similar to what humans experience in their regular lives, meaning a dose that is environmentally relevant. When scientists look at whether exposure to a chemical is “safe” or not, what we are really asking is whether exposure to that chemical has an effect in the environmentally relevant range. “Low dose” can also refer to the amount of a hormonally active or endocrine disrupting chemical (EDC) that has to be given to an animal to produce the same amount of chemical in blood levels that 
is similar to human blood levels of that chemical. 
 
But sometimes we have no idea how much of a chemical humans are exposed to, or how much EDC must be given to produce human blood levels. In this case, we consider “low doses” to be any dose below the doses that regulatory agencies have tested. 
 
Low dose “effects” – or the impact/consequence of exposure to an EDC at a “low dose” -­–­- include a huge range of biological changes including increased cancer incidence, altered development of the reproductive tract, abnormal brain development and behaviors, impaired metabolic responses related to diabetes and obesity, and many others.

3.   The study concludes that high dose tests can’t be relied upon to reveal low dose effects. Why?

Hormones, and chemicals that behave like hormones, can cause very different effects at low doses compared to high doses. We show many examples of this in the review. One clear example involves the cancer-­-fighting synthetic hormone tamoxifen, commonly used to treat breast cancer. At high doses it stops the growth of breast tumors. But at low doses it actually causes breast tumors to grow. Traditional testing methods used to test for chemical safety would completely miss this low dose effect.

The paper presents many examples where low dose effects cannot be predicted from high dose testing, and it reviews extensive information now available about the biological mechanisms that can cause this pattern. The paper concludes that this should be expected for any chemical that behaves like a hormone. The failure of traditional testing methods to detect low dose responses has large implications for chemical regulation. It means that many chemical safety standards will have to be strengthened substantially to protect public health.

4.   Why are these effects not predicted in the toxicology studies performed to determine the safety of chemicals on the market?

Under current FDA testing procedures, when toxicology studies are performed, very high doses of a chemical are given to animals. The animals are then examined to see if they die or have other, very obvious and serious health effects. These tests do not focus on the kinds of developmental effects that you would see if, for example, you gave an animal birth control pills or another chemical that disrupts hormones. Hormones don’t kill animals, unless they are given at extremely high doses. But they do a lot of other things to animals, and those are the endpoints that are often ignored in toxicology testing.

The traditional way of thinking about chemicals is that ‘the dose makes the poison’, or that more of a chemical produces more damage. Toxicology tests use very high doses, and then predict that a much lower dose of the chemical would be safe. But the lower dose – the “safe” dose – is never actually tested for safety. So when humans are exposed to low doses of a chemical, regulatory agencies almost never have any data to predict what types of biological effects can be expected. They don’t know, because no tests have been performed that would tell them. But we do see the effects of low dose exposures in prevalent human health problems like learning disabilities, infertility, and a variety of cancers.

5.   What evidence do we have that low dose exposure to EDCs adversely affects human health?

The first line of evidence we have comes from animal testing, which is the standard for determining the safety of any chemical exposure. Critics often say: “But animals aren’t people!” While this is true, the endocrine systems of animals and humans are remarkably similar. This means that the effects of EDCs, especially chemicals that act like hormone mimics, are expected to be very similar in animals and humans. And in fact, there are several examples in which animal studies predicted exactly the kinds of effects from EDCs that were later seen in exposed humans. A study of DES, the drug infamously given to pregnant women to prevent complications prior to 1970, predicted the harmful effects of this drug on the mammary gland 25 years before the daughters of those who took it were old enough to develop breast cancer.

We compiled examples of 30 EDCs that have been shown to cause low dose effects in animals. That means that low doses of these chemicals affect a range of endpoints including development of sex differences in the brain, behavioral changes, development of the male and female reproductive tracts, hormone levels in the body, bone health, obesity, and fertility, among other impacts. The second line of evidence comes from human epidemiology studies. In these studies, people that are exposed to a chemical can be compared to people that are not exposed to a chemical, and these studies indicate that EDC exposures affect human health. Additional epidemiology studies compare the amount that people are exposed to an EDC, and find that the level of exposure matters, too. In human studies, the kinds of effects associated with EDC exposures include fertility, cardiovascular disease, obesity, arthritis, bone health, endometriosis, high blood pressure, and metabolic syndrome, including type 2 diabetes.

6.   What kind of changes do we need to make in chemical testing to ensure the protection of human health?

Toxicology testing needs to be done so that rather than only testing high doses and guessing about which doses are safe, actual doses in the range of real life human exposures are tested for safety.

                                                        
1 Soto AM, Vandenberg LN, Maffini MV, Sonnenschein C.2008. Does breast cancer start in the womb? 
Basic Clin Pharmacol Toxicol 102 (2): 125-­-33.

Furthermore, toxicology testing needs to change its focus from asking only whether chemicals kill people or cause birth defects, to also addressing whether sensitive endpoints relevant to endocrine health, like development of the brain and sex-­-specific behaviors, or control of insulin and blood sugar levels, are affected. There are many, many academic labs doing these kinds of tests; however, regulatory agencies often ignore these data.

7.   If these chemicals are so pervasive, is it feasible to try to eliminate all human exposures?

EDCs are routinely used on crops (and therefore found on/in our foods), to package foods, on our lawns, in our houses, and on our bodies (soap) and clothes (detergents). They are everywhere. We are not suggesting that every single one of these chemicals needs to disappear today, but we are suggesting that they should not be considered ‘safe’. Can they be phased out and replaced with chemicals that do not have hormonal activity? Can we change the way new chemicals are tested to avoid this problem in the future? These are all questions that need to be answered.

8.   What are the cumulative effects of these exposures?

“Cumulative” can be interpreted in two ways. If by cumulative you mean the effects over time of exposures, the answer is complex. At different times in life, we are differently sensitive to 
chemicals including hormones. Early developmental periods (fetal, neonatal and infant periods) are incredibly sensitive to EDCs and hormones. Exposures to hormone mimics (or hormone blockers) in the womb will have permanent effects because hormone-­-induced changes during this sensitive period will alter development of organs in ways that can never be reversed. Puberty is also a sensitive period. We often think of adult EDC exposures as less harmful, but that might not be true. More and more human studies show links between adult exposures and diseases or dysfunctions such as infertility, cardiovascular disease, type 2 diabetes and obesity. There may also be sensitive periods late in life, i.e. elderly adults may have a heightened sensitivity to hormones. And EDCs may interfere with treatment for some cancers. If by cumulative you mean the effects of multiple chemicals, this is a very important issue that needs a lot more attention. Some studies show that when low doses of chemicals are mixed together, they have effects that could not be predicted from the effects of single chemicals alone. Humans are exposed to mixtures of chemicals – dozens if not hundreds every day. Knowing how these chemicals interact with each other, as well as their interactions with natural hormones, is an important research need. We do not yet have the answer to this question, but it’s one we should not ignore.

9.   In those cases where low doses are associated with worse effects than high doses, doesn’t a greater amount of exposure effectively eliminate the threat?

This is a common misunderstanding. It is not that these chemicals do not cause harm at high doses – they cause very different effects at high doses. High dose toxicology studies show that these chemicals can be deadly, or block pregnancies, or induce birth defects at high doses. These incredibly visible effects are not seen at low doses, which is why some think they are safe at low doses. But at low doses, they affect a wide range of other endpoints. The other important thing to keep in mind is that our bodies are not a single cell type or a single organ. Let’s say that you could identify a dose of an EDC that does not induce mammary cancer. That doesn’t tell you anything about whether that dose is ‘safe’ – because a dose that doesn’t affect the mammary gland could affect the brain, or fertility, or metabolism. 

10. If we are being harmed by low-­-dose exposure to EDCs, what can we do to protect ourselves?

Because this isn’t a single chemical issue, it is not easy to make suggestions to consumers. If this were a question of how to avoid a chemical applied to your lawn to control grubs, the answer would be “don’t use it.” But so many of these chemicals are contaminants in our foods, are in the air and water, and are in consumer products that aren’t labeled. Instead, this study suggests that we need to change how we think about chemical safety, and start to deal with the hundreds (or thousands) of chemicals on the market that are EDCs.
The real point of this paper is to show that this isn’t a single chemical issue. Low dose effects are observed for dozens of widely-­-used chemicals. We propose that scientists and regulators should expect EDCs to have effects at low doses and to test for them explicitly. That is not done today and it means many safety standards for chemical exposure are too weak.

Just published is a  huge review of low dose and non-monotonic effects in Endocrine Reviews, with Laura Vandenberg as first author. Nothing like it has ever been done.

Review:

Birnbaum’s editorial:

Cone’s story:

Vandenberg’s op-ed:

Don’t miss, in the review, an evaluation of how ‘weight of the evidence’ risk assessments must be done for EDCs using endocrinological principles, and what that means for bisphenol A.
Bottom line of the review: Non-monotonicity is common and should be the default expectation for endocrine disrupting chemicals. This means a large number of current safety standards are too weak.