Pathologic Upgrade Rates of High-Risk Breast Lesions on Digital Two-Dimensional vs Tomosynthesis Mammography

Abstract

Our aim was to compare the pathologic upgrade rates of high-risk breast lesions (HRLs) on digital two-dimensional mammography (DM) vs digital breast tomosynthesis (DBT). The study cohort was composed of patients with HRLs diagnosed by image-guided core needle biopsy from December 2007 to February 2011 (DM group, before DBT integration) and from January 2013 to March 2016 (DBT group, after complete DBT integration). Medical records were reviewed for surgical outcomes and follow-up imaging. Eight hundred and ninety-three patients with 900 biopsy-proven HRLs underwent surgical excision (97.0% [873 of 900]) or had at least 2 years of imaging follow-up (3.0% [27 of 900]). The most common HRL was atypical ductal hyperplasia in the DM and DBT groups (37.4% [337 of 900]). The overall upgrade rate of HRLs to malignancy was 11.3% (102 of 900). There were no statistically significant differences in overall upgrades rates of HRLs on DM vs DBT (11.4% [54 of 475] vs 11.3% [48 of 425]; p= 0.97) or in upgrade rates of HRL subtypes. However, HRLs that upgraded on DBT were more likely to be invasive rather than in situ carcinoma compared with HRLs that upgraded on DM (39.6% [19 of 48] vs 20.4% [11 of 54]; p= 0.03). Overall, the most common HRL to upgrade was atypical ductal hyperplasia (18.4% [62 of 337]). The least common HRLs to upgrade in the DBT group were biphasic neoplasms (0% [0 of 22]), flat epithelial atypia (1.6% [1 of 62]), and papillomas without atypia (4.5% [2of44]). There is no difference in the upgrade rates of HRLs on DM vs DBT, but the proportion of HRL upgrades that are invasive rather than in situ carcinoma is higher with DBT.