Quality assurance of pathology in clinical trials. The National Surgical Adjuvant Breast and Bowel Project experience.

Large cooperative clinical trials such as those conducted by the National Surgical Adjuvant Breast and Bowel Project (NSABP) require measures for assurance of quality and consistency of their pathologic findings. The NSABP experience indicates that these issues may in large part be resolved by a central or headquarters review of pathologic materials submitted by institutional pathologists. This cooperation is implicit with participation in NSABP protocols, and its importance is emphasized at annual NSABP meetings. Assessments of gross characteristics of tumors and total number of lymph nodes and number with metastases used in analyses are performed by institutional pathologists and may be confirmed if necessary from routine surgical pathology reports or special D-1 forms submitted to the central pathology headquarters. The microscopic assessments by headquarters pathologists are performed without knowledge of clinical events and require preliminary training. The requirement of preliminary training emphasizes use of established criteria for the many qualitative and quantitative evaluations required. In some studies there may be as many as 35 quantitative evaluations. Agreement of findings of at least 90% with other experienced members of the pathology center or on subsequent individual re-review is required before an individual's assessments are accepted for analyses. Studies performed with limited pathologic material require assurance that members of the test cohort exhibit similar pathologic and clinical characteristics with those excluded from participation. Those performed without microscopic sections require confirmation that the material used contains adequate viable tumor tissue. This algorithm has resulted in a consistent demonstration of pathologic characteristics that affect diagnosis, prognosis, and treatment of patients with breast cancer.