Conspiracy, incompetence, a federal agency out of control. A recent Mother Jones story by Mariah Blake indicts the Food and Drug Administration (FDA) as a threat to science and public health over the way it’s conducting research into bisphenol A (BPA), the never-ending chemical scare story of the 21st century. Raise the alarm (again), stir the pot (again), marshal outrage (again).
And, if you have no other sources of information, the arguments that the FDA is trying to undermine a major research initiative by the National Institutes of Health, dubbed CLARITY, by introducing contaminated study data to claim that BPA was safe might seem persuasive and damaging, if somewhat perplexing and perhaps unbalanced. Why does the FDA keep insisting that BPA is not a threat to health? Why does the Environmental Protection Agency concur? And Europe’s Food Safety Authority and the World Health Organization? The Manichean tapestry woven by Mother Jones seems like such a complete narrative—the arc of a despicable covenant between industry and regulators—until you see the loose threads; and pull.
The unraveling begins with the kind of tedious procedural detail around which a Law and Order episode might build a counter-intuitive twist. Frederick vom Saal, the University of Missouri biologist who, like a cross between Tiresias and Helen of Troy, launched a thousand fearsome, yet seemingly incredible, studies on BPA, protests to Mother Jones about the perfidy of the Food and Drug Administration.
The FDA is charged with violating an agreement on how to conduct the next phase of research into BPA, the so called CLARITY program, by doing a rival, pre-emptive study and then propagating the results—BPA only does bad things at millions of times the amount we are exposed to—through the media. “Vom Saal and his colleagues are also livid that the FDA would publish a study based on older testing methods in the midst of their collaboration and treat it as the near-final word on BPA,” adds Mother Jones.
But a quick doxing of federal records shows that the FDA study, which was conducted at its National Center for Toxicological Research (NCTR) as part of an interagency agreement with the National Toxicology Program (NTP)-- a part of the National Institute of Environmental Health Sciences (NIEHS)--was presented as “approved” to the FDA at its Science Board meeting on August 17, 2009, which means the NCTR and NTP had been discussing its goals and methods since 2008.
The CLARITY program study, which is also funded by the NIEHS and is also being run by the NTP, wasn’t announced till 2010. In other words, the same agencies approved both studies, and the idea that somehow the FDA snuck in a study to undermine CLARITY without anyone noticing is chronologically impossible. Moreover, the lead author of the study, Barry Delclos, previewed the study results over a year ago at the American Association for the Advancement of Science. “Vom Saal and his colleagues” are spinning Mother Jones.
“Disengenous”
So what about the idea that the control group of rats in the “FDA” study were contaminated, thereby undermining the entire study—an opinion voiced by longtime vom Saal collaborator Gail S. Prins, a professor of physiology at the University of Illinois at Chicago. Prins tells Mother Jones she was “baffled that any journal would even publish” such a flawed study.
Moreover, “the FDA study glossed over this detail, which was buried near the end of the paper,” says Mother Jones. “Prins and her colleagues also complain that the paper omitted key information—including the fact that some of them had found dramatic effects in the same group of animals. ‘The way the FDA presented its findings is so disingenuous,’ says one scientist, who works closely with the agency. ‘It borders on scientific misconduct.’”
This sounds terrible.
But the study—Delclos et al.—which was published in what is arguably, the world’s top toxicology journal, Toxicological Sciences—mentions this issue in the discussion section, which is where you would expect to see this kind of thing; more importantly, the authors explicitly state that this issue is dealt with in detail in a second study, which was published simultaneously in the same journal: Churchwell et al. That Prins and the other scientists quoted attacking the FDA didn’t know about the Churchwell study strains credulity. It also suggests that no one at Mother Jones bothered to read the Delclos study, which you would think would be the decent thing to do, editorially, in a story accusing a federal agency of “scientific misconduct.”
So what do the Delclos and Churchwell studies say about contamination? Despite elaborate measures to control unintentional exposure to BPA in the rats’ diets, water, cage materials and bedding, BPA metabolites were found in the blood at levels similar to the lowest dose group, which was receiving a dose 70 times higher than the median daily American exposure: (2.5 ug/kg body weight/day versus 0.037 ug/kg bw/d). But as there were no adverse effects found in any of the dose groups except the ones given 100,000 and 300,000 ug/kg bw/d—or three to nine million times the median human exposure— it was not seen as a major issue.
I asked Gary W. Miller, Editor-in-Chief of Toxicological Sciences, to comment on the accusation that the Delclos study should not have been published. Miller who is the Asa Grigs Candler Professor in the Rollins School of Public Health at Emory University also directs a major, innovative toxicological program funded by the NIEHS—HERCULES (Health and Exposome Research Center: Understanding Lifetime Exposures). He responded by email:
“Toxicological Sciences is a premier journal in the field, and all submitted manuscripts undergo peer review that includes evaluation by the editor, an associate editor, and at least two reviewers who are subject matter experts. The point of contention in the Delclos 90-day study regards the presence of BPA in the control groups. The authors were very upfront with this element of the research and devoted significant text to the concern. The editorial staff weighed the input of the peer reviewers on this issue, and based upon their recommendations, the manuscript was accepted for publication.”
When asked about the cross contamination, the FDA also said, in a statement by email, that it was “very low,” and “likely due to the proximity of the test groups.” Moreover, the agency said that it was able to characterize the contamination such that it could be taken into account when evaluating the study’s results. “The results do not invalidate the study.”
The rats from the Charles River.
The third major contention voiced by Mother Jones is that the FDA and NTP are using and relying on a rat strain—Charles River Sprague Dawley—that is insensitve to estrogenic compounds like BPA, and so these studies tell us nothing. It is worth noting some background here. Vom Saal, Prins, and other researchers added their names to a letter written by Tom Zoeller, a biologist at U. Mass. Amherst, to FDA Commissioner Margaret Hamburg dated September 21, 2009, which protested both the use of the Sprague Dawley rat strain and, indeed, the need to spend any further money on researching BPA given that there were so many studies identifying adverse effects.
“We strongly recommend that the agency immediately halt the studies with rats,” Zoeller et al. write. “At the very least, if they are to be carried out, they should be designed in ways that incorporate the scientific expertise of the National Institute of Environmental Health Sciences. Currently, not only are they scientifically flawed, they also represent a serious waste of time and public money.”
But fast forward to 2010, when the NIEHS announced that the CLARITY Program would disburse $32 million in grants for more BPA research—and magically those adamantine views about the pointlessness of more BPA research and the use of Sprague Dawley rats completely disappeared. The NIEHS—an agency the letter writers believe is far better equipped to do BPA research—would use Sprague Dawley rats exclusively. One can only imagine how difficult it must be for Prins, vom Saal and Zoeller to have to accept millions of dollars in taxpayer money for precisely the kind of research they’d said is pointless and should, as a matter of urgency, be halted at the FDA.
Elsewhere, Europe’s Food Safety Authority (EFSA) found that there is no evidence to think the Sprague Dawley rat is insensitive to BPA; and France—which executed a volte-face on the safety of BPA with the victory of Francois Hollande’s Socialist party—used findings from a 2008 Sprague Dawley rat study as the baseline for its ultra conservative risk assessment in 2013. And why would they all do that? Because there are numerous studies showing that the Sprague Dawley rat is sensitive not only to BPA but to other estrogens; indeed, one of the studies claiming low dose sensitivity to BPA is from... Gail Prins. It would seem that attacks on the Sprague Dawley rat are situational: when they show a result contrary to the anti-BPA crusade, they are bad; when they support the crusade and deliver precious grant money to universities, they are good.
Why all this spin?
The alarm over BPA has, since it began in 1998, been a perpetual spin machine. On PBS Frontline in 1998, vom Saal claimed his research on BPA would overturn everything we thought we knew about toxicology, and he invoked the paradigm shift theory of scientific progress as proposed by Thomas Kuhn in The Structure of Scientific Revolutions to explain the significance of his findings. What Frontline didn’t tell viewers was that this paradigm “inversion” was based on two studies, each with seven dosed mice and a total of 22 controls. Never in the history of science or statistics had 14 mice given their lives for so much rhetorical glory, while, at the same time, becoming the canaries in the coalmine of modern life. The idea that we were all being secretly estrogenized by the chemistry of mass consumption turned vom Saal into a mixture of scientific messiah and media celebrity. (Last year, I put a hypothetical “paradigm shift” based on the same statistics to several leading statisticians; you can read their perplexed responses here.)
The alchemical mixture of science and spin propelled the Federal Government into an interminable quest to prove whether vom Saal’s findings—and those of his collaborators—were true. For journalists, studies based on tiny numbers of rodents may bear the narrative weight of tablets from God; but for regulators, they bear no more than a hypothesis that must be stress tested with robust statistics and careful analytic procedures.
The pattern has been unvarying: The Environmental Protection Agency (which has funded and conducted some of the key studies on BPA) and the FDA find no adverse effects and nothing for the public to worry about on the basis of results from large scale, multi generational rodent studies and tiny human exposures; the vom Saal wing denounce these studies as incompetently done and wrong, and continues to get funded by the NIEHS to do more small scale studies, which find new alarming effects (the latest focus, apparently, is the risk of absorbing BPA from using toilet paper). The cost of this exercise since 1998? There is no formal, public account. But, based on grant amounts and estimates from government scientists on the cost of individual studies, 100 million dollars wouldn't be too far off.
But it has been a brutal and unnerving two years on the BPA research front if you have hedged your academic career on it being a threat to mankind. According to two government scientists who attended a public NIEHS meeting of CLARITY grantees, there was a frank admission that “the data was not going their way” on BPA. And CLARITY itself represents a serious challenge to the vom Saal-ites because they have agreed to do, what they have as yet not done: transparent large-scale testing of their hypotheses so as to produce statistically interpretable results.
This is one reason why the NCTR Delclos study has everyone aflutter in Mother Jones: it is the most extensive study yet that fails to replicate their claims that very low doses of BPA cause effects which cannot be seen at higher doses. The CLARITANS have all committed to using the same rats and their tissues and to be “blinded” as to the animal exposure groups (so as to guarantee reliable interpretation once unblinded by separate researchers at the NTP).
After Delcos, the long odds that the CLARITY study group will find low dose effects have lengthened even further. As the FDA explains (via email):
“One main difference between the recent 90 day study and the previous comprehensive multigenerational toxicity studies conducted in rats and mice was the use of direct oral dosing of the neonates rather than relying on transfer of BPA to the pups through the mother’s milk. While the broad dose range used in the previous multigenerational studies certainly resulted in significant exposure to the pups, the level of exposure in the present study was more readily quantifiable. In fact, measurements of internal doses achieved were made at several ages, including careful evaluation of levels in control animals... Another difference in the present study was the inclusion of endpoints not typically covered in guideline studies but that address effects of BPA that have been reported in the literature, e.g., fat pad weights and fasting serum glucose and insulin. No significant effects were observed for these additional endpoints in the dose range of interest.”
Another reason the Delclos study caused such dyspepsia is that it was reported by National Public Radio (NPR). The anti BPA crusaders have long enjoyed wide immunity in the media from any kind of skepticism about the quality of their science, but that is increasingly unconvincing in the face of the constant failure to replicate their findings. FDA NCTR research on the exposures to BPA in young non human primates found that they were much lower than in young rats, and thus—being closer to human infants—eased concern about postnatal exposure. An EPA-funded study conducted by the Pacific Northwest National Laboratory found that the overwhelming majority of those hundreds of “low dose” studies that had generated media headlines were not low at all, and that there was nowhere close to being enough active BPA in human exposures to bind with the receptors in quantities to trigger estrogenic effects. Massive European exposure monitoring blew away the idea that people were somehow absorbing high amounts of BPA from the environment—even from cash register receipts.
And then there were moments that should give even the most ardent Mother Jones reader pause to wonder whether the science on BPA could be nearly as black and white as claimed. One of the top environmental scientists at the Silent Spring Institute, Ruthann Rudel, publicly praised the work of the FDA at the NCTR at the American Association for the Advancement of Science in 2013. “I actually just want to thank you for the work that you’ve done, because I’ve found your studies to be some of the most clarifying and helpful pieces of information in making my way through the bisphenol A woods,” Rudel said after a presentation by the FDA’s Dan Doerge, one of the world’s leading experts in BPA pharmacokinetics—the way the chemical diffuses through and interacts with the body’s organs. There is little ambiguity in this research: it is unlikely that BPA is present in sufficient amounts to affect tissues given the tiny exposures.
And then Antonia Calafat, the Centers for Disease Control expert on human biomonitoring, weighed in on one of the issues that was continuing to fuel claims about the risk of BPA from consumer products: studies, which claimed levels of the chemical are much higher in blood than the exposure monitoring and pharmacokinetic data indicate. The reason, said Calafat, in a letter to the journal Breast Cancer Research, was since it is very difficult, if not impossible, to control for BPA in all the lab equipment (think lots of plastics and tubing), you’d end up with contaminated blood samples. The only reliable way of assessing exposure to BPA was through urine metabolites. What’s striking about the letter is not the content—the contamination issue has been repeatedly noted by the EFSA, EPA and FDA—but the signatories on the letter—a who’s who of environmental health scientists.
On top of all this, the NIEHS in-house and tax-funded journal, Environmental Health Perspectives had to retract a recent study claim that BPA caused breast cancer rats after it was pointed out that the statistics blatantly failed to support such a claim. The study was conducted by CLARITY grantee Ana Soto of Tufts University... using Sprague Dawley rats.
Endgame
Perhaps Mother Jones is okay with supporting team vom Saal, and taking one for the cause. But it might think what is at stake if the attempts to ban BPA do not outrun the regulators. If you’ve talked, as I have done, for years with scientists employed by the government, one consistent cause for worry has been the endgame on BPA—what happens when all the research is in and everyone realizes that it was, as predicted years ago, an almost complete waste of time, effort and money? Who will be blamed? Will it be vom Saal and Prins and all the others who’ve gotten millions of dollars in research grants? Undoubtedly—which might explain their pre-emptive strikes against the FDA in Mother Jones (indeed, one wonders whether the article is laying the ground for their withdrawal from CLARITY).
Will it be the NIEHS official who funded so much of this research over the past decade, Jerry Heindel? Highly likely, not least because it seems the NIEHS doesn’t think it’s a conflict of interest for its principal grant officer on endocrine disruption to collaborate on scientific papers with the scientists he is, in the course of his job, funding with taxpayer dollars.
But what worries everyone is not just that the NIEHS will be dragged through a Republican-dominated congressional mill that will grind the institute small; what worries government scientists and officials is that they are all going to feel the wrath of Congress over BPA and it won’t matter who was right and who was wrong on the science.
