Cytologic differential diagnosis of papillary pattern in breast aspirates: Correlation with histology

Papillary neoplasms, fibroadenoma, fihrocystic change, low-grade ductal carcinoma, and apocrine carcinoma are among "gray zone" lesions in breast cytology. They often have cellular smears with a papillary or pseudopapillary pattern. To better define cytologic criteria useful in distinguishing these entities, we correlated them with histology. Papanicolaou and giemsa stained smears from 33 aspirates and their corresponding histology were reviewed. Of these, 28 had an initial cytologic diagnosis or suspicion of papillary neoplasm, while five cases were not diagnosed cytologically as papillary but the histologic diagnosis was a papillary neoplasm. Cytologic features evaluated included cellularity, architecture, apocrine/single/ columnar cells, nuclear atypia, intranuclear inclusions, calcifications, background, myoepithelial cells, and bipolar, naked nuclei. Discriminating cytologic features grouped by final histologic diagnosis were as follows: Papillary neoplasm (14 cases): Hemorrhagic/cystic background, 3-dimensional papillary clusters, columnar cells, and fibrovascular cores. Myoepithelial cells within clusters and background naked, bipolar nuclei were inconspicuous. Fibroadenoma (4 cases): Two-dimensional branching clusters often with folding, moderate myoepithelial cells in clusters, moderate to numerous background bipolar nuclei, often forming doublets in smear background, cellular stroma. Ductal carcinoma (11 cases): Papillary ductal carcinoma in situ in 5 of 11 cases, cribriform/tubular architecture in 6 of 11. Absence or paucity of myoepithelial within clusters and background bipolar nuclei was noted. Fibrocystic change (4 cases): Two-dimensional clusters, moderate myoepithelial cells within clusters, and moderate bipolar nuclei in the background. The presence and quantity of myoepithelial cells, bipolar naked nuclei in the background, and ductal cell architecture were the only consistently useful cytologic features in distinguishing breast lesions with a papillary pattern.