Columnar Cell Lesions and Flat Epithelial Atypia in Breast Biopsies Performed Due to the Presence of Mammographic Suspicious Abnormalities
M De Brot, CB Nunes, H Gobbi. Federal University of Minas Gerais - UFMG, Belo Horizonte, Minas Gerais, Brazil
Background: Columnar cell lesions (CCL) are characterized by the presence of columnar epithelial cells lining enlarged TDLUs of the breast. There is renewed interest in it because these lesions are been encountered with increasing frequency in excisional biopsies recommended by the mammographic finding of microcalcifications. The manner in which CCL are classified has varied among different authors. The aim of our study was to investigate frequency and types of CLL in breast biopsies performed due to the presence of mammographic suspicious abnormalities.
Design: We selected 249 breast biopsies required by suspicious mammographic findings, previously categorized as BI-RADS 3 or 4. Original H&E stained sections were reviewed and CCL were classified according to Schnitt S criteria as columnar cell change (CCC), columnar cell hyperplasia (CCH), or flat epithelial atypia (FEA).
Results: CCL were detected in 91/249 (36.5%) biopsies, being categorized as: CCC-81/91 (89.0%); CCH-4/91 (4.3%); FEA-6/91 (6.5%). CCL coexisted with epithelial hyperplasias in 70/91 (76.9%) cases (p value<0.05), 70% comprising usual ductal hyperplasia, 9% being atypical ductal hyperplasia, and 7% diagnosed as atypical lobular hyperplasia. We also detected benign neoplasias in 24/91 (26.3%) CCL cases and malignant neoplasias in 29/91 (31.8%) CCL cases (p value<0.05). Microcalcifications were identified in 124/249 (50.0%) specimens and were present within the CCL in 72/124 (58.0%) biopsies, being this association statistically significant (p value<0.05). Microcalcifications were exclusively present within malignant lesions in 30/124 (24.1%) biopsies. Besides CCL, we observed other benign lesions in 69/249 (27.7%) specimens, including ductal ectasia/cysts, apocrine metaplasia, sclerosing adenosis, and radial scar. Benign neoplasias were the sole diagnosis in only 2 biopsies, while malignant neoplasias were noted in 95 cases.
Conclusions: Our results showed that the majority of CCL are represented by lesions without cytologic atypia or architectural complexity which frequently coexisted with other lesions. The coexistence of CCL and invasive carcinomas was statistically significant (p value<0.05). We also found that CCL were the main site of microcalcifications either mammographically or histologically detected (p value<0.05). Most importantly, microcalcifications associated with these lesions allowed the incidental diagnosis of clinically silent carcinomas present elsewhere within the specimens.
Monday, March 22, 2010 1:00 PM
Poster Session II # 34, Monday Afternoon