Pathology quality control in the cooperative clinical cancer trial programs.

Pathology quality control is an essential part of clinical cancer research. However, pathology review for every study may not be practical; therefore, review in most cooperative groups will be performed only for selected protocols. The selection process should include those neoplasms with an expected diagnostic error rate which would affect the outcome of the study. Unfortunately, complete information about the diagnostic error rate for various neoplasms is not available at the present time; however, enough information is available from the pathology review experiences of the last decade to draw some general guidelines. All lymphomas should have pathology quality control review as should rare or difficult-to-diagnose neoplasms. Early cancer is likely to cause diagnostic disagreement, so protocols designed to evaluate various treatment options for such neoplasms should include pathology review. The diagnostic error rate for far-advanced carcinoma, most adenocarcinomas of the gastrointestinal tract and squamous cell carcinomas, is so low that routine pathologic quality control review is not cost-effective, in our opinion. However, pathology review may be pathology research, and in this circumstance, pathology review will be carried out to answer research questions. Whether or not pathology quality control review is needed should be determined at the time of development of each protocol.