Routine Excision of Core Biopsy Diagnosed Flat Epithelial Atypia Frequently Reveals ADH and Lobular Neoplasia: Potential Implications for Management
CS Ho, WM Yap. Tan Tock Seng Hospital, Singapore, Singapore
Background: Flat epithelial atypia (FEA) is one of the earliest recognizable, non-obligate precursors of low grade breast cancer, of uncertain natural history. The management of FEA found on core needle biopsy (CNB) remains debatable, as limited data depict malignancy in 0-30% of cases excised. We aimed to determine the significance of FEA on CNB by studying the frequency of upgrade to ADH, lobular neoplasia (LN) and malignancy upon excision.
Design: Pathology and imaging data were reviewed for all cases of CNB-detected FEA between May 2005 and December 2008. CNB cases of FEA associated with ADH, LN or malignancy were excluded. Excision of FEA on CNB was routinely recommended. Statistical analysis (Fisher's exact test / t-test) was performed; differences with p<0.05 were deemed significant.
Results: Among 2657 CNBs, 28 (1.1%) cases of FEA were identified in 26 women with a mean age of 51 years. A mean of 12 cores of tissue were obtained per case, using 11 or 14 gauge biopsy needles, with at least 3 H&E sections examined per core. Indications for CNB included calcifications (n=23), density (n=1) and masses (n=4). All calcifications were of indeterminate nature. 22/28 (78.6%) cases were excised, all within 3 months of diagnosis, where none showed DCIS or invasive carcinoma. However, 45.5% of excision biopsies revealed breast lesions at significant risk of subsequent cancer: ADH (n=8), ADH+ALH (n=1) and classical LCIS (n=1). There were no statistically significant differences in patient age, laterality, biopsy needle gauge (11 vs 14) or number of cores obtained, between the group with (n=10) and the group without (n=12) ADH/LN on excision. Radiologic-pathologic concordance was achieved in all but 2 cases (a radiographic mass and a retroareolar density), where excision disclosed fibrocystic change with ADH, and an intraductal papilloma with adjacent ADH, respectively. Six patients who declined surgery (only 1 was biopsied for a mass, where histology showed fibroadenoma with adjacent FEA) have not presented with malignancy, 10 to 37 months post CNB diagnosis.
Conclusions: Despite no malignancy, routine excision of CNB diagnosed FEA frequently revealed lesions at moderate-to-high risk of breast cancer (ADH/LN), a finding hitherto unaddressed. Excision biopsy of CNB-detected FEA may be of value in identifying patients with ADH/LN who would benefit from appropriate follow-up and counseling, and possible chemopreventive therapy.
Wednesday, March 24, 2010 9:30 AM
Poster Session V # 36, Wednesday Morning