Diagnosis of Papillary Breast Lesions on Core Needle Biopsy: Upgrade Rates and Interobserver Variability.

The histologic distinction between papillary breast lesions remains challenging, especially with core biopsy (CB) specimens. A retrospective review of the clinical, imaging, and histologic findings was performed for patients with papillary breast lesions on CB from 2013 to 2017. The interpretation accuracy was expressed as upgrade rate relative to the excision diagnosis. Diagnostic reproducibility with and without immunohistochemistry was analyzed as interobserver variability among 3 board-certified pathologists. Among 57 papillary lesions with biopsies and excisions available for review, the upgrade rates were 0% for benign papilloma, 30% for papilloma with atypical ductal hyperplasia, and 25% for papilloma with ductal carcinoma in situ, resulting in an overall upgrade rate of 11.1%. There were no statistical differences between patients in an upgrade group and others, when comparing the patient age, clinical presentation, BI-RADS (Breast Imaging Reporting and Database System) category, location, and histologic grade. The overall interobserver variability of the 60 consecutive core biopsies of papillary breast lesions by morphology alone was in the "substantial" agreement range (κ = 0.79, 86% agreement), with an excellent κ score of 0.88 for papilloma (92% agreement). "Substantial" and "fair" κ values were seen for papilloma with atypical ductal hyperplasia/ductal carcinoma in situ (0.74, 84% agreement) and invasive carcinoma (0.40, 60% agreement). Use of immunohistochemical stains improved the κ values into "excellent" range (0.92, 94% agreement). Our study favors a conservative approach in the management of benign papillomas, at least in cases of good radiologic-pathologic concordance. Papillary breast lesions with atypia/malignancy show lower diagnostic reproducibility on CB, and utility of immunohistochemistry is recommended in challenging cases.