Introduction
•At the time of this publication, columnar cell lesions were increasingly being found in breast biopsies performed for mammographic microcalcifications. The lack of standardized terminology and limited information on their clinical significance created difficulties in both pathologic diagnosis of CCLs and management of patients with these lesions.
•The term “flat epithelial atypia” (FEA) was introduced by the WHO Working Group on the Pathology and Genetics of Tumours of the Breast to encompass CCLs with low-grade/monomorphic-type cytologic atypia that lacked the architectural features of atypical ductal hyperplasia or low-grade DCIS
•Studieshave documented an association between FEA and synchronous ADH, DCIS, invasive carcinoma (esp tubular carcinoma), and lobular neoplasia, thus showing distinguishing FEA from non-atypical CCLs to be of practical importance as FEA may serve as a ‘red flag,’ particularly in core biopsies for the possible presence of these other lesions
•This study served to show if pathologists could reproducibly separate FEA from CCLs without atypia
Findings
•A reference pathologist prepared a PowerPoint tutorial that included written criteria for the diagnosis of non-atypical CCLs (columnar cell change and columnar cell hyperplasia) and for FEA (see Table 1) , as well as digital images of each
•Following review of the training material, seven study pathologists (with an interest in breast pathology) were given a test set of 30 CCLs. They were instructed to categorize each case as either FEA or not atypical
•The diagnoses of the reference pathologist and those of the other study pathologists were independently submitted directly to the study statistician
•The overall agreement among the eight pathologists for the 30 cases was 91.8% (95% CI, 84.0–96.9%) and the multi-rater kappa value was 0.83 (95% CI, 0.67–0.94), which is within the ‘excellent agreement’ range.
•Agreement was slightly better for determining the absence of FEA (92.8%: 95% CI, 84.1–97.4%), than for determining the presence of FEA (90.4%: 95% CI, 79.9–96.7%).
Conclusions
•The ability of pathologists to reproducibly diagnose FEA and to distinguish it from CCLs without atypia had not been previously evaluated
•Observer agreement is fostered by the use of standardized histologic criteria by the study participants, and in keeping with this, the results of this study demonstrate for the first time that FEA, as defined by the WHO Working Group, can be distinguished from non-atypical CCLs with a high degree of consistency using available diagnostic criteria.