Regulatory Forum Opinion Piece*

Abstract

The transparency and documentation of the peer review process have been discussed recently. Our position is that transparency is best achieved when peer review is a collaborative process, in which both parties are open-minded but both also realize that the study pathologist retains complete control over the findings (raw data) and over the content of the pathology report. For these reasons, we believe that histopathology raw data should be defined as the observations made by the study pathologist (printed and/or electronic formats) rather than as the tissue slides recommended by the Organisation for Economic Co-operation and Development (OECD). Also, because the study pathologist retains control over the histopathology raw data, any notes or tabulations of findings by the study pathologist and peer review pathologist during the peer review are interim notes and should not be included as an appendix to the pathology report though they may be retained if desired, as currently recommended. Because the histopathology raw data have not been created until completion of the peer review, the performance of a peer review should be documented in the study report, as currently recommended, but that it not be a GLP-compliant process.

In the October 2013 issue of Toxicologic Pathology, opposing views on retaining peer review documents (worksheets) were offered by Engelhardt, Hardisty, and Mann (2013) and by Kerlin and Schaffer (2013). Both opinions addressed the same point, which is the need for transparency of the peer review process, particularly for regulatory reviewers. The need for transparency is derived from the appearance of possible undue sponsor influence during the peer review process. Our position is that transparency is best achieved when the study pathologist and the peer review pathologist each understand that peer review is a collaborative process, in which both parties are open-minded. But both also realize that the study pathologist retains complete control over the findings (raw data), and over the content of the pathology report. For these reasons, we believe that, as recently proposed, histopathology raw data should be defined as the observations made by the study pathologist (printed and/or electronic formats) rather than the tissue slides as recommended by the OECD or the pathology report as is the current position of some regulatory agencies (Leininger and Parker 2013; OECD, Advisory Document of the Working Group on Good Laboratory Practice [GLP], Guidance on the GLP Requirements for Peer Review of Histopathology in press). Also, because the study pathologist retains control over the histopathology raw data, any notes or tabulations of findings by the study pathologist and peer review pathologist during the peer review are interim notes and should not be included as an appendix to the pathology report though they may be retained if desired, as currently recommended (Society of Toxicologic Pathology [STP]; Morton et al. 2010). Thus, because the histopathology raw data have not been created until completion of the peer review, the performance of a peer review should be documented in the study report, as currently recommended, but that it need not be a GLP-compliant process.

In the process of generating histopathology raw data, the study pathologist enters findings into a data management program and then writes his or her pathology report. Because data management programs provide for password-protected control over the entered findings, only the study pathologist can make changes to the findings. If the database file is locked (audit trail activated) and printed prior to peer review, a clear record of the study pathologist’s original findings is created. Following peer review, the study pathologist changes diagnoses, as needed, to reflect agreement/disagreement with the peer review pathologist via the audit trail. The post–peer review database file is then printed. These printed findings represent the “official” histopathology raw data and can be compared to the study pathologist’s original entries (working raw data) if desired.

The pathology report should contain a section referencing the peer review process, in which it is clearly stated that the histopathology raw data reflect the opinion of the study pathologist following the peer review. The printed original entries (pathologist’s interim notes, or working raw data) and the printed histopathology official raw data (or the printed audit trail) can then be included as appendices to the pathology report as documentation of the process. The peer review statement can either indicate agreement by the peer review pathologist with all of the histopathology raw data or specify the data with which they disagree. In addition to clarifying the peer review process, this approach clearly shows that the histopathology raw data consist a set of data that the study pathologist entered and printed, making it more consistent with other raw data (necropsy observations, organ weights, and clinical pathology data). These steps do not reveal the thought processes that the study pathologist went through during peer review, but they do provide clear, written, documentation that the histopathology data represent his or her opinion.

For the above mentioned to be accepted, it is important to recognize that only the study pathologist can determine what the histopathology raw data are and what the content of the report narrative will be. It is also imperative that the peer review process be collaborative. These points must be clearly stated in the pathology report and in the peer review statement. If the study pathologist and peer review pathologist are not open-minded and willing to listen to each other’s comments and opinions (i.e., follow best practices for peer review), the point of peer review is lost anyway.

In over 50 years of toxicologic pathology experience between us, including experience as contract pathologists, sponsor pathologists, and peer review pathologists, only one of us has had instances (two) in which it was clear that a sponsor was applying undue pressure to have the pathology report reflect their opinion rather than the study pathologist’s opinion. In one case, the sponsor representative was a non-pathologist and in the second he or she was a pathologist. In both instances, even though the sponsor never said “change this whether you want to or not,” it was clear that was what was expected. If the steps described previously are followed, even in the very rare instances of sponsor pressure to make unwarranted report and/or data changes, the transparency of the peer review process will be clearly documented to anyone who reads the pathology report, its appendices, and the peer review statement.

If a non-pathologist reviewer chooses to consider histopathology data and information solely as a black box and not try to understand what the peer review process really is, one can never remove all doubt about the transparency of peer review. However, we believe that the previously outlined steps clearly demonstrate the transparency of peer review by (1) documenting the control of the process by the study pathologist and (2) the documentation of agreement or disagreement by the peer review pathologist with the study findings and interpretations.