The problems and promise of central pathology review: development of a standardized procedure for the Children's Oncology Group.

The practice of central histopathologic review by expert pathologists has accompanied entry of patients onto multi-institutional therapeutic research protocols for the past half-century, and it is still a key component in the majority of therapeutic protocols conducted by the Children's Oncology Group (COG). Policies regarding pathology review have historically varied but have recently been standardized by the COG. Pretreatment central pathology review has been used to improve the accuracy of pathology data that is critical to the conduct of the study and to ensure that ineligible patients are not enrolled on a study. Pretreatment central review is unnecessary when there are established criteria for diagnosis, when institutions can uniformly apply these criteria, and when diagnoses are reliably reported. Pretreatment central review is appropriate when there are established diagnostic criteria and expert reviewers can consistently apply them but when institutional diagnoses show significant variability. It is inappropriate when there are no standard criteria or when variability exists among experts. Retrospective reviews (for example, those performed after protocol entry and treatment have occurred) historically have been used to verify institutional diagnoses and to record other features of tumors, prior to analysis of research questions. However, delayed reporting of discrepant diagnoses from retrospective reviews introduced obvious concerns regarding the appropriateness of treatment and exposed institutional pathologists and clinicians to significant professional risks. This article examines the historical usage and current status of the central pathology review process and the rationale and indications for its performance.