Performance validation in anatomic pathology: successful integration of a new classification system into the practice...

The new, international, multidisciplinary classification of lung adenocarcinoma, from the International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society, presents a paradigm shift for diagnostic pathologists.

To validate our ability to apply the recommendations in reporting on non-small cell lung cancer cases.

A test based on the new non-small cell lung cancer classification was administered to 16 pathology faculty members, senior residents, and fellows before and after major educational interventions, which included circulation of articles, electronic presentations, and live presentations by a well-known lung pathologist. Surgical and cytologic (including cell-block material) reports of lung malignancies for representative periods before and after the educational interventions were reviewed for compliance with the new guidelines. Cases were scored on a 3-point scale, with 1 indicating incorrect terminology and/or highly inappropriate stain use, 2 indicating correct diagnostic terminology with suboptimal stain use, and 3 indicating appropriate diagnosis and stain use. The actual error type was also evaluated.

The average score on initial testing was 55%, increasing to 88% following the educational interventions (60% improvement). Of the 54 reports evaluated before intervention, participants scored 3 out of 3 points on 15 cases (28%), 2 of 3 on 31 cases (57%), and 1 of 3 on 8 cases (15%). Incorrect use of stains was noted in 23 of 54 cases (43%), incorrect terminology in 15 of 54 cases (28%), and inappropriate use of tissue, precluding possible molecular testing, in 4 out of 54 cases (7%). Of the 55 cases after intervention, participants scored 3 out of 3 points on 46 cases (84%), 2 of 3 on 8 cases (15%), and 1 of 3 on 1 case (2%). Incorrect use of stains was identified in 9 of 55 cases (16% of total reports), and inappropriate use of tissue, precluding possible molecular testing, was found in 1 of the 55 cases (2%).

The study results demonstrated marked improvement in the pathologists' understanding and application of the new non-small cell lung cancer classification recommendations, which was sufficient to validate our use of the system in routine practice. The results also affirm the value of intensive education on, and validation of, pathologists' use of a classification or diagnostic algorithm.