By Jennifer Sass and Sarah Janssen

As described in earlier posts (here and here), the NIH’s National Institute of Environmental Health Sciences (NIEHS) has contracted the work of the Center for the Evaluation of Risks to Human Reproduction (CERHR) out to the consulting company Sciences International. This issue received public attention just as CERHR’s scheduled review of the chemical Bisphenol A (BPA) was taking place.

We submitted comments to CERHR detailing concerns about the content and the process of the BPA review. Our concerns include:

The NIEHS NTP Center for the Evaluation of Risks to Human Reproduction (CERHR) outsourced responsibility for developing the first draft scientific review, while failing to provide adequate oversight.
It is our understanding that the current process of reviewing chemicals for the  CERHR program relies on an outside contractor, Sciences International, to conduct a literature review and write up. This includes selecting and evaluating/ranking the literature to be included in the review, and is therefore very influential and vulnerable to abuse. The contractor then submits the review to the NTP. The NTP staff select an expert panel. This is also a very influential step. The expert panel develops a final expert panel report for the NTP. This final panel report is used by the NTP in the development of an “NTP-brief”, which is an official government opinion. The final NTP-CERHR monograph contains the NTP brief along with the expert panel report and other supplementary material (public comments, etc.). The Monograph is considered to be an authoritative body and, for example, can trigger listings under California Prop65.

NTP has failed to include some important studies in their draft report of Bisphenol A (BPA):
There are a few studies which were not included in the draft BPA report that should be evaluated by the expert committee. These include studies previously published that highlight the potential of BPA to mimic estrogen at extremely low doses. (For details, see Sarah Janssen’s comments (PDF).)

NTP should consider the choice of experimental animal when evaluating studies.
Several of the studies under consideration in the CERHR were done using inappropriate animal models and this information was left out of the evaluation process. It has been noted by other investigators that the Charles-River Sprague-Dawley (CD-SD) rat is relatively insensitive to exogenously administered estrogen, including potent estrogenic drugs.  Several of the studies under consideration by NTP’s CERHR used this animal strain in their experimental design and found either no effect of BPA or effects only at high doses.

It is critical that use of an appropriate animal model is determined when evaluating the strength of a study. There was no discussion by the committee of the relative insensitivity of the CD-SD strain to exogenous estrogen treatment in their reviews of these studies. 

NTP should consider whether an adequate positive control was included when evaluating the strength of studies.
Several of the studies under consideration in the CERHR did not have the expected results using positive controls, indicating problems in the study design. Establishment of a positive control is essential for determining the quality of a research study and its ability to predict differences in treatment outcomes.  In addition to the studies described previously [NTP references (#293 Tyl et al.); (#297 Kwon et al.); (#411 Tyl et al.); and (#438 Ashby et al.)] that did not have adequate positive controls, there were other studies determined to be adequate for analysis by the committee that did not have positive controls (#294 Cagen et al., #342 Cagen et al.).

In their evaluation, the CERHR review did not justify how studies lacking adequate positive controls are determined to be appropriate for evaluation by the CERHR.

NTP should consider the impact of bias in industry-funded research.
A recently published study demonstrated that the source of funding had significant impact on whether or not adverse effects were found with low-dose BPA treatment.2 Of a total 130 studies, 119 were done by government funded entities and 11 were done by chemical corporations.  Ninety-two percent (109/130) of the government funded studies found evidence of adverse effects after low dose BPA treatment while only 8% found no adverse outcome. Some of the government funded studies finding no effect used the previously described estrogen insensitive strain of CD-SD rats. In contrast, 100% (11/11) industry funded studies found no evidence of harm from low dose BPA treatment. 

The bias of industry-funded research has been well documented for many products such as tobacco, pharmaceuticals, and pesticides.  NTP should carefully consider the source of funding when evaluating the adequacy of studies.  A bias towards no-effect has already been demonstrated for low dose effects of BPA and several of those studies reviewed by the NTP CERHR committee have been determined to be adequate for evaluation.

Based on the existing data and comments made above, we strongly feel that BPA should be considered a hazard to human development and reproduction.  BPA has been demonstrated to have multiple developmental and reproductive toxicities at low and environmentally relevant doses in a number of well designed government funded studies.  In addition, significant new data indicates that in utero exposure to BPA causes aneuploidy in developing oocytes, which could contribute to miscarriage and birth defects.

The problem of outsourcing at NTP
To the extent that NTP farms out critical task, such as scientific reviews and assessments, in some cases to contractors that are also working for regulated industries, without adequate transparency, oversight, or accountability, it seriously compromises both the public trust and the NTPs ability to ensure that it is meeting its obligation to the American public.

The Center for the Evaluation of Risks to Human Reproduction (CERHR) contracts with Sciences International (SI), a private consulting firm, to provide scientific and administrative support. According to a recent news report, “the company also oversees the nomination of chemicals for review and of panel members who evaluate the chemicals”.   The same report also revealed that, “since the late 1990s, SI has helped the tobacco company RJ Reynolds fight stricter regulations on toxic pesticides.” The report also noted that, scientists who serve on CERHR panels to assess chemicals are required to sign conflict-of-interest forms, but that according to a letter to Environmental Working Group from CERHR director Michael Shelby, “no specific [conflict of interest] restrictions are placed on [SI].”  Public reports such as these represent a potential disastrous failure of public trust in the NTP program.

Congress has also raised its hackles on this issue. In a letter sent to Dr. David Schwartz, the Director of NIEHS, Senator Boxer (Chair, Senate Environment and Public Works Committee) and Representative Waxman (Chair, House Oversight and Government Reform Committee) expressed concern at potential conflicts of interest with Sciences International regarding the review of Bisphenol-A (PDF). Congressmen stated they were concerned that, “extensive use of contractors at CERHR may undermine” the ability of NIEHS to appropriately review the health impacts of toxic chemicals.

In fact, NTP is spending millions of dollars on contractors that also serve industries impacted by NTP reports and assessments.  These contractual arrangements are not subject to “sunshine” laws include the Federal Advisory Committee Act (FACA) and the Freedom of Information Act (FOIA),  that play a critical role in ensuring government accountability. When NTP outsources tasks without ensuring transparency, without adequate oversight, and without demanding public accountability, it has compromised public trust and the credibility of its products.  To erode public trust, it matters not whether the conflicts are perceived or actual, and whether the product is scientifically sound or flawed.

Process Recommendations 
 We recommend the following:
• If NIEHS intends to continue to contract out the task of developing a literature review and write up, then it is essential that:
* the review be conducted by experts in the field, and be as inclusive as possible of relevant literature
*the review be conducted by experts that have no financial conflicts
* the review be subjected to rigorous scrutiny for errors by NIEHS scientific experts, prior to review by the Expert Panel.
• Following internal review and approval by NIEHS experts, the scientific review should be available for public comment, and comments forwarded with the review to the Expert Panel.
• The Expert Panel must be comprised of independent scientific experts without financial conflicts. The experts must be familiar with the relevant literature, and with the relevant experts in the field. This is critical to provide rigorous scientific review and overall credibility to the document. Experts will be sufficiently familiar with the cutting edge research to identify misstatements, mis-interpretations, and missing literature in the review document. The use of experts will enable the Panel and NTP to identify where other experts or public commentors may have conflicts that may be undisclosed. These should be disclosed and, if appropriate, experts removed from the Panel.

The use of public funds to support poor-quality or poor-credibility science is of tremendous concern.  Congress adopted strong sunshine laws in part to prevent clandestine manipulation of the government process, and that objective is in serious jeopardy to the extent NTP is permitted to outsource critical responsibilities.  NIEHS must ensure that the public funding it is entrusted with is used in a manner that preserves the scientific integrity of the regulatory process and that any important science activities that the NTP funds are conducted with adequate transparency and direct lines of accountability.

Jennifer Sass, Ph.D is a senior scientist in the Health and Environment program at Natural Resources Defense Council.
Sarah Janssen, MD, PhD, MPH is a science fellow in the Health and Environment program at Natural Resources Defense Council.