It was a bold plan, and even the person who conceived of it gave it slim odds of getting off the ground: government and academic scientists, sharing samples and data from the same cohorts of rats dosed with bisphenol A, would try to reach an overall conclusion on whether the ubiquitous food-packaging chemical has deleterious effects on health.
“The way I thought about it initially was, why would anybody want to do this?” says Jerry Heindel, who was then a health sciences administrator at the National Institute of Environmental Health Sciences (NIEHS) and has since retired. He thought a regulatory agency would be unlikely to take the risk that their previous conclusions—namely, bisphenol A (BPA) is safe for human at typical exposures—would be proven wrong by such a project. But as he saw it, there was a need for partnership. The US Food and Drug Administration (FDA) was declining to consider troves of academic data on BPA’s effects on the grounds that their study designs didn’t meet its criteria, while the agency’s own studies focused on endpoints with dubious relevance to human health, according to Heindel.
The FDA agreed to the idea of the partnered study, dubbed CLARITY-BPA, and it commenced under the direction of the agency, NIEHS, and the National Toxicology Program in 2012. At the time, Heindel “thought we were doing something good,” he recalls. Now, as the results roll out and FDA and the CLARITY-BPA grantees outside the agency prepare to produce their final report, he’s not so sure.
Some critics, including those funded by the program, have gone public with their concerns about the study design and interpretation of the data. Earlier this week (October 23), the Endocrine Society held a webinar for media in which Heather Patisaul, an endocrinologist at North Carolina State University who participated in the study, said the agency’s early statements about its results indicate that “FDA has put a stake in the ground, they’ve said that BPA is safe, and they’re very reluctant to back away from this stake.”
These are toxicologists, they are not endocrinologists, and they don’t understand that effects at a low dose but no effects at high dose are very biologically plausible.—Gail Prins, University of Illinois at Chicago
The webinar and other criticisms lobbed at FDA by scientists who are intimately familiar with CLARITY-BPA indicate the likely failure of a program undertaken at great expense and effort to close a gulf between the FDA and many academics regarding BPA’s health effects. As those involved in CLARITY prepare a final, integrated report for release next year, deep divisions remain about whether it’s safe for humans to consume low levels of the hormone-like chemical.
The goal of the project was to study a wide range of potential health effects from BPA exposure, and generate data to be used in making regulatory decisions. The NIEHS put up $30 million for the seven-year program.
The agreed-upon protocol for the study was that the FDA’s National Center for Toxicological Research (NCTR) would start the experiment. Scientists there gave Sprague-Dawley rats a daily dose of BPA, either prenatally through puberty or prenatally and for their entire year or two of life. Some control rats received no BPA or hormones, while other, positive controls were dosed with ethinyl estradiol, a synthetic hormone. Fourteen academic scientists who had successfully applied to participate in the study were sent tissue samples from the rats. Grantees used the samples to test BPA’s effects on the brain, reproductive organs, the heart, and other tissues and functions, while the NCTR conducted its own tests, called the core study.
All of the raw data from the core study and grantees are now available on the National Toxicology Program’s website, and the FDA released its report on the core study in late September, finding that lower-dose effects of BPA “were not dose responsive, sometimes occurring in only one low or intermediate dose group, and did not demonstrate a clear pattern of consistent responses within or across organs within the stop- and continuous-dose arms and sacrifice times.”
Some grantees have published peer-reviewed reports of their findings from the study. Patisaul, who focused on the brain, found that BPA’s effects included altering gene expression in the amygdala, hypothalamus, and hippocampus of newborn rats. The University of Illinois at Chicago’s Gail Prins, whose paper is pending publication, says her findings were similar to those of previous studies she’s conducted, namely, that low exposure to BPA during development increases the risk of prostate cancer for rats later in life.
Two of the grantees did not find clear effects from exposures to low-dose BPA, points out Steve Hentges of the American Chemistry Council, an industry group unaffiliated with CLARITY, in an email to The Scientist. Those studies, led by Norbert Kaminski of Michigan State University and Kim Boekelheide of Brown University, were on immune function and testes and sperm, respectively. Hentges agrees with FDA’s assessment that low-dose effects seen in the core study were not biologically relevant, and says that the lack of dose-responsiveness was just one criterion used in making that judgment. “Although the CLARITY Core Study is complex, its conclusion is quite straightforward. . . . The results clearly show that BPA has very little potential to cause health effects, even when people are exposed to it throughout their lives,” he says.
Other experts disagree. “As a scientist, I look at their data and I think some of those low-dose effects are very biologically plausible,” Prins tells The Scientist. “These are toxicologists, they are not endocrinologists, and they don’t understand that effects at a low dose but no effects at high dose are very biologically plausible.” While toxicologists are taught that the dose makes the poison, she says, endocrinologists commonly observe effects of hormones at lower levels that disappear at higher levels because the molecules can trigger different receptors depending on their concentration.
What I was shocked to find was that these studies were not replicable. . . . That, to me, suggests that there’s a much deeper story here.—Laura Vandenberg, University of Massachusetts Amherst
Last month, in a webinar hosted by Carnegie Mellon University’s Institute for Green Science, Prins, Heindel, and two other researchers argued the results so far indicate BPA does affect health at low doses—that is, at the levels most Americans are exposed to through products such as food containers and receipts.
The following day, September 13, the FDA hosted its own webinar previewing the results of the core study; the description reads, in part, “Results of the NCTR toxicity studies indicated that BPA produced adverse effects at high doses, but not at the low end of the dose range tested, consistent with its activity as a weak estrogen.” The webinar’s presenter, NCTR research pharmacologist K. Barry Delclos, stated early on that “there are no conclusions based on the results.” But he then suggested that statistically significant effects seen at lower doses in the study were not “biologically relevant” for several reasons—they occurred only in the puberty arm of the study, not the one that went on for the rats’ lifetime, for example, and did not increase in severity along with the dose of BPA. (FDA declined to make Delclos or another spokesperson available for an interview about the study.)
At the October 23 webinar, Patisaul cited “concern from the Endocrine Society and others that FDA is jumping the gun a little bit in its conclusion that BPA in food packaging is safe,” as its core study data have not yet been integrated with that of the grantees to produce the final report.
Critics of FDA’s handling of CLARITY-BPA also cite concerns with the study’s design. In the Endocrine Society webinar, the University of Massachusetts Amherst’s Laura Vandenberg, who was not involved in the project, said there were discrepancies between the core study’s data on the effect of the positive control, ethinyl estradiol, and those of previous FDA experiments. “What I was shocked to find was that these studies were not replicable. . . . That, to me, suggests that there’s a much deeper story here” about the reliability of regulatory agencies’ toxicity studies, she said.
Prins tells The Scientist there were aspects of the study design that also gave her pause. For example, “they stressed the hell out of the rats by putting a tube down their throats every single day to administer the BPA,” even as pups, and that stress may have been a confounding factor. The scrutiny of CLARITY-BPA by its grantees reveals, Heindel says, that the guideline studies regulatory agencies perform to gather data for use in policymaking are “not as great a study as everybody thinks they are.”
Pat Hunt, who studies mammalian germ cells at Washington State University and who was, in 2003, one of the first researchers to report the effects of BPA on mice, says the difficulty in making a regulatory decision about the chemical illustrates the need for a fundamental change in how such substances are evaluated. BPA “is a good demonstration of how wonderfully facile chemistry is now, because you can create replacement bisphenols and go from one chemical like BPA to a family of 50 or more,” she says, making it “impossible to even think about really regulating these chemicals effectively, or even testing them.”