Neurotoxicity of PBDEs: Dingemans et al. Respond

Banasik and Suchecka express their discontent with our recent review on the (in-)direct neurotoxic effects of parent and hydroxylated (OH-) polybrominated diphenyl ethers (PBDEs) on the (developing) nervous system (Dingemans et al. 2011). Their main discontent appears to be once more related to the experimental design in a number of cited behavioral studies. However, our aim was to identify and review the mechanisms underlying the observed adverse (behavioral) effects, not to evaluate the experimental design of behavioral studies within a regulatory setting. Nonetheless, approximately 10% of our review was dedicated to describing a number of behavioral studies [12 different studies from seven different research groups, including a 2008 EPA study (Gee and Moser 2008)] that all indicated the occurrence of neurobehavioral effects following developmental exposure to PBDEs. We used this information to create a starting point for the main part of our review of direct and indirect cellular and molecular mechanisms. For readability and space limitations, we were not able to include all studies, concerns, or critiques that have ever been raised. The absence of a citation to a particular study does not mean that we regard it as less credible. 
 
The view that (developmental) exposure to PBDEs induces adverse neurotoxic effects is widely supported by numerous in vivo, ex vivo, and in vitro studies reporting both structural and functional effects (Dingemans et al. 2011). For some time, a lively discussion has been taking place within the scientific community on the experimental design for behavioral developmental neurotoxicity studies for regulatory purposes, in particular considering the statistical unit (Alcock et al. 2011). In short, there is disagreement about whether direct dosing of pups precludes the need to control for litter effects (e.g., Eriksson 2008; Hardy and Stedeford 2008). However, we did not address this topic in our paper because we consider the potential occurrence of a litter effect to be irrelevant for the reviewed cellular and molecular in vitro studies, which all indicate that exposure to PBDEs induces neurotoxic effects. 
 
Critical remarks can be found throughout our review (Dingemans et al. 2011), but they are related to cellular and molecular findings, data gaps, or aspects that warrant further investigation. Our main conclusions are related to the specific (developmental) neurotoxic hazard of OH-PBDEs compared with that of their parent congeners via direct neurotoxicity and thyroid disruption. We also pointed out the need to further investigate the impact of active metabolites, concentrations of PBDEs and metabolites in the (developing) brain, and the potentially increased neurotoxic hazard following exposure to mixtures of different environmental contaminants. 
 
Nonetheless, Banasik and Suchecka raise an important issue: the existence of differences in experimental designs for in vivo investigation of (developmental) neurotoxicity. Differences exist in the selection of investigated end points and also in methodologies for the investigation of a specific end point, as reviewed for effects on motor activity by brominated flame retardants (Williams and DeSesso 2010). These differences in experimental design may underlie observed differences in sensitivity to detect neurotoxicity, possibly because of differences in biokinetics and exposure during sensitive windows of development. Fortunately, much effort is taking place in the scientific community to optimize experimental designs at different levels of biological complexity, including (developmental) neurobehavioral studies. Although a critical review on the impact of different experimental designs for in vivo (developmental) neurotoxicity studies would be very useful, it was beyond the scope of our review (Dingemans et al. 2011).

Neurotoxicity of PBDEs on the Developing Nervous System doi:10.1289/ehp.1103907 Dingemans et al. (2011) published a review article on polybrominated diphenyl ethers (PBDEs) and the developing nervous system. However, the authors summarized but failed to criti cally evaluate the articles cited in their review. They also did not discuss or cite literature that contradicted the studies on which they based their conclusions. For example, the U.S. Environmental Protection Agency (EPA) cosponsored an expert panel on neurodevelop mental end points, which concluded that an experimental design used in nine of the studies cited by Dingemans et al. (2011) failed to control for litter effects (Holson et al. 2008).
Although some investigators have set forth the argument that direct dosing of pups precludes the need to control for litter effects, a U.S. EPA cosponsored expert panel (Moser et al. 2005) evaluated this issue and concluded otherwise.
Regardless of whether Dingemans et al. (2011) view the studies by Holson et al. (2008) and Moser et al. (2005) as credible, the authors should have discussed them to some degree. It is understandable that because of space limitations not all studies can be included in a review. However, it was unacceptable to exclude studies that carry the weight of U.S. EPA cosponsored expert panels or other reviews that critically evaluated many of the studies cited by Dingemans et al. (2011) (e.g., Goodman 2009Hardy et al. 2009;Williams and DeSesso 2010) and came to opposite conclusions.
Although the article by Dingemans et al. (2011) was peer-reviewed, it presents information in a selective, noncritical manner, which is best reserved for public relation pieces communicated in the non-peer-reviewed media. Banasik and Suchecka express their discontent with our recent review on the (in-)direct neurotoxic effects of parent and hydrox ylated (OH-) polybrominated diphenyl ethers (PBDEs) on the (developing) nervous system (Dingemans et al. 2011). Their main discontent appears to be once more related to the experi mental design in a number of cited behavioral studies. However, our aim was to identify and review the mecha nisms under lying the observed adverse (behavioral) effects, not to evaluate the experimental design of behavioral studies within a regulatory setting. Nonetheless, approximately 10% of our review was dedicated to describing a number of behavioral studies [12 different studies from seven different research groups, including a 2008 EPA study (Gee and Moser 2008)] that all indicated the occurrence of neuro behavioral effects following developmental exposure to PBDEs. We used this information to create a starting point for the main part of our review of direct and indirect cellular and molecular mechanisms. For readability and space limitations, we were not able to include all studies, concerns, or critiques that have ever been raised. The absence of a citation to a particular study does not mean that we regard it as less credible.
The view that (developmental) exposure to PBDEs induces adverse neuro toxic effects is widely supported by numerous in vivo, ex vivo, and in vitro studies reporting both structural and functional effects (Dingemans et al. 2011). For some time, a lively discussion has been taking place within the scientific community on the experimental design for behavioral developmental neuro toxicity studies for regulatory purposes, in particular considering the statistical unit (Alcock et al. 2011). In short, there is disagreement about whether direct dosing of pups precludes the need to control for litter effects (e.g., Eriksson 2008; Hardy and Stedeford 2008). However, we did not address this topic in our paper because we consider the potential occurrence of a litter effect to be irrele vant for the reviewed cellular and molecular in vitro studies, which all indicate that exposure to PBDEs induces neurotoxic effects.
Critical remarks can be found throughout our review (Dingemans et al. 2011), but they are related to cellu lar and molecular findings, data gaps, or aspects that warrant further investigation. Our main conclusions are related to the specific (developmental) neuro toxic hazard of OH-PBDEs compared with that of their parent congeners via direct neuro toxicity and thyroid disruption. We also pointed out the need to further investigate the impact of active metabolites, concentrations of PBDEs and metabolites in the (developing) brain, and the potentially increased neurotoxic hazard following exposure to mixtures of different environmental contaminants.
Nonetheless, Banasik and Suchecka raise an important issue: the existence of differences in experimental designs for in vivo investigation of (developmental) neuro toxicity. Differences exist in the selection of investigated end points and also in methodologies for the investigation of a specific end point, as reviewed for effects on motor activity by brominated flame retardants (Williams and DeSesso 2010). These differences in experi-A 332 volume 119 | number 8 | August 2011 • Environmental Health Perspectives mental design may under lie observed differences in sensitivity to detect neuro toxicity, possibly because of differences in bio kinetics and exposure during sensitive windows of development. Fortunately, much effort is taking place in the scientific community to optimize experi mental designs at different levels of biological complexity, including (developmental) neuro behavioral studies. Although a critical review on the impact of different experi mental designs for in vivo (developmental) neuro toxicity studies would be very useful, it was beyond the scope of our review (Dingemans et al. 2011 (Jacob et al. 1999) with two methodo logical limitations: use of grouped doses and poor control of confounding. With these new findings, evidence from ecological, case-control, and cohort studies are consistent; thus, an interesting question is whether there was an ecological fallacy. Although ecological studies are important to epidemiology (especially in environmental and social epidemiology), public health practitioners seem afraid of ecological studies. It is a common practice to assume the presence of ecological fallacy (Robinson 1950) and lowlevel validity when analyzing an ecological study. Most epidemiologists prefer an exclusive individualistic approach, although the importance of a multi level causal approach is widely recognized (Diez-Roux 2002). In this sense, some authors suggest that it is as important to recognize the presence of ecological fallacy as to recognize psychologistic or individualistic fallacy (Subramanian et al. 2009) (Figure 1).
Thus, it is necessary to have clear guidelines on when there is or not an ecological fallacy. In this sense, I propose three criteria for the identification of ecological fallacy; all three of these should be present to confirm its existence: • Results must be obtained with ecological (population) data. • Data must be inferred to individuals.
One use of ecological studies is to explore individual-level association when individual data are not available. When the focus of the study was contextual or based on popula tion effects and there is no inference to individuals, ecological fallacy is not possible. When only the first two criteria are present-which is insufficient to affirm ecological fallacy-it is appropriate to acknowledge that there is a possible relation ship and that further study is required. • Results obtained with individual data are contradictory.
Only when empirical data are available is it possible to confirm that an ecological fallacy is present.
The author declares that he has no competing financial interests. doi:10.1289/ehp.1103444 In "Research Recommendations for Selected IARC-Classified Agents," Ward et al. (2010) identified research gaps for 20 occupational agents "based on evidence of widespread human exposures and potential carcinogenicity in animals or humans." (Ward et al. 2010) For carbon black, the authors suggested that

Carbon Black
Research needs include updating epidemiology cohorts with data on work histories and exposures in relation to particle size and surface area, and recruitment of additional carbon black facilities. The relationship between occupational exposure to carbon black and validated bio markers of oxidative stress should be examined and exposureresponse relationships in humans and rodents quantified, including the role of particle size. Ward et al. (2010) referred to a study of British carbon black workers in which carbon black was suggested as a possible "late stage carcinogen" (Sorahan and Harrington 2007). In that study, Sorahan and Harrington (2007) called for similar analyses of other carbon black cohorts (i.e., evaluating the possibility of carbon black acting as a late stage carcinogen via the concept of "lugging," which considers only recent exposures and not historical exposures). In response to suggestions made by Sorahan and Harrington, we conducted such analyses on a large German carbon black cohort (Morfeld andMcCunney 2007, 2009). We were unable to reproduce the results of the British analysis, despite the elevation noted in lung cancer among German cohort workers, thus providing no support for the late stage-lugging hypothesis. Results of a detailed analysis of the German cohort using Bayesian methodology showed smoking and exposure to occupational carcinogens prior to work at the carbon black plant as confounders probably responsible for the lung cancer excess