As We Were Saying…

Rather than addressing the substance of our Forum Article concerning Good Laboratory Practice (GLP) (Borgert et al., 2016), Tweedale takes the opportunity to advocate for the low-dose (non-monotonic dose) hypothesis, challenge the validity of test guidelines, and resurrect the ad hominem attack on industry scientists. While we do not wish to debate here issues that are outside the focus of our article, we will respond to a few points. Contrary to Tweedale’s central argument contending we claim that industry toxicity studies are “accurate,” even a casual reading of our Forum Article demonstrates we did not claim the “accuracy” of all industry studies. We clearly acknowledged errors can be made when conducting GLP studies and that uncontrolled variables can affect results of any study, including non-GLP, non-guideline, and academic studies. We went into some detail, however, as to how conducting a test with a standardized guideline in accordance with GLP reduces the likelihood of systematic errors, which is something we think all scientists should strive to achieve. Tweedale’s claim that non-guideline studies are not routinely considered also is wrong. We discussed how the Klimisch and other scoring systems explicitly include non-guideline studies and studies that are not GLP-compliant, especially in the regulatory context. Furthermore, the regulatory methods for evaluating data quality make it evident that non-guideline studies from the open literature are used routinely. (See our discussion in Borgert et al., 2016 and accompanying cites.) While acknowledging chemicals are extensively tested prior to marketing, Tweedale’s primary claim is that test guidelines are deficient because they do not require the use of sufficiently low doses (as suggested by the “low-dose hypothesis”). This, of course, is not a deficiency of test guidelines per se or of GLP. Test guidelines are developed using a rigorous, transparent scientific process designed to establish the reliability and relevance of a particular test for a specific purpose (i.e., to test a specific hypothesis). To the extent test guidelines do not included certain doses, those doses are not within the hypothesis being tested. And to be clear, the requirement for the use of GLP and OECD Test Guidelines was not established by industry, nor are the guidelines developed, evaluated, and approved by industry. OECD Test Guidelines are developed, evaluated, and approved under the OECD program utilizing the expertise of some of the most knowledgeable regulatory toxicology experts from almost 40 countries. As to Tweedale’s concern then, the issue appears to be that the low-dose hypothesis has not been accepted by the broader scientific and regulatory communities—at least to the extent it would alter existing testing protocols. The low-dose hypothesis has been debated for almost two decades and we do not feel it appropriate to continue that debate here. We must mention, however, that contrary to Tweedale’s claim, there have been a number of large, well-conducted studies that do not support the low-dose hypothesis. (For a broader discussion of the literature see Goodman et al., 2009; Lamb et al., 2014; Rhomberg and Goodman, 2014.) To date, regulatory agencies have found that existing evidence does not support requiring routine testing of substances for low-dose effects (NTP, 2001; Melnick et al., 2002; EPA, 2002). Recently, EPA concluded “[f]or estrogen, androgen or thyroid MoA that provide adequate information to make an assessment, our evaluation shows that there is not sufficient evidence of NMDRs [non-monotonic dose responses] for adverse effects below the NOAELS or BMD derived from the current testing strategies” and current testing approaches are “highly unlikely to mischaracterize a chemical that has the potential to adversely perturb the endocrine system because of an NMDR” (EPA, 2013). We expect that if the low-dose hypothesis was validated and generally accepted, test guidelines would be revised as appropriate and use of GLP would be instrumental in ensuring the proper conduct of those assays. Finally, as is the case with Tweesdale’s response to Borgert et al. (2016), it has become a common practice in some circles to dismiss differing views concerning scientific evidence as mere industry conflict of interest. To state the obvious, one would be hard pressed not to find some conflict of interest and bias in any group, including academia and advocacy groups. Indeed, ideology alone can be a significant source of bias such as what has been termed "white hat bias"—bias leading to distortion of information in the service of what may be perceived to be righteous ends (Cope and Allison, 2010). Anyone remotely aware of the academic world knows that tenure and promotion is commonly based upon publishable research and obtaining funding, and that positive as opposed to negative findings are more readily published in journals. Furthermore, scientists who have established a reputation and funding record in support of a particular hypothesis might be less eager to present data that counter their hypothesis. The conflict and bias issue has been discussed in greater detail (Allison, 2009; Borgert, 2007a,b; McCarty et al., 2012) and we welcome a more focused consideration of this issue in an appropriate place. While ad hominem attacks may be a useful advocacy technique, they have no value in earnest scientific discourse. We suggest focusing on the quality of science and finding ways to manage potential conflict and bias in all camps. While we would like to believe that concern for one’s professional reputation would ameliorate any issues concerning conflict and bias, we believe it is wise to require transparency and to use standardized methods for generating and reporting scientific evidence. The use of GLP is certainly one method to manage conflict and bias in this regard. Supplementary data are available online at http://toxsci.oxfordjournals.org/. The authors declare no competing interests in the subject matter of this submission. In the course of their employment, the authors may conduct, direct, oversee, or interpret studies that are required by statute to comply with GLP, however, their employment status, compensation, advancement, and other benefits do not depend in any way upon the requirement for compliance with or use of GLP. CJB and TFQ received compensation from the Endocrine Policy Forum for the draft letter. Available as a Supplementary File.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.