Atypical lobular hyperplasia and lobular carcinoma in situ at core breast biopsy: Use of careful radiologic-pathologic correlation To recommend excision or observation

Purpose: To assess the utility of precise radiologic and pathologic correlation for establishing imaging-histologic concordance or discordance as a method to limit the number of patients requiring surgical excision when atypical lobular hyperplasia (ALH) or lobular carcinoma in situ (LCIS) is diagnosed at core biopsy. Materials and Methods: This study was approved by the institutional review board, and the requirement to obtain informed consent was waived. The pathology database was searched from 2000 to 2010 for core biopsies yielding ALH or LCIS devoid of any additional lesion that independently necessitated excision. All cases had to have either subsequent surgical excision or a minimum of 2 years of imaging follow-up. This yielded 50 cases from 49 women aged 40-73 years (mean age, 59 years). The authors performed detailed radiologic- pathologic analysis while blinded to subsequent follow-up information, comparing all biopsy-related images with the histologic findings at core biopsy and then designating each core biopsy finding as concordant or discordant. Then, results of subsequent surgery or extended follow-up for each case were unblinded and compared with original concordant or discordant designations. Outcomes and confidence intervals (CIs) were calculated. Results: Of the 43 benign concordant core biopsy findings, none were upgraded at surgery (n = 38) or extended follow-up (n = 5) (95% CI: 0%, 8%). Of the seven discordant biopsy findings, two were upgraded to ductal carcinoma in situ at surgery (n = 5); none of the cases were upgraded at follow-up (n = 2). Conclusion: When careful radiologic-pathologic correlation is performed and concordance is achieved, women with ALH or LCIS at core biopsy can be observed.