C F Greeley1, A R Frost. 1. Department of Pathology, George Washington University Medical Center, Washington, D.C. 20037, USA.
Abstract
OBJECTIVE: To define cytologic criteria that reliably distinguish lobular (LC) and ductal (DC) carcinoma and classical LC from its variants-alveolar, solid and mixed infiltrative patterns. STUDY DESIGN: Fine needle aspiration (FNA) specimens from 29 cases each of histologically proven LC and DC were reviewed without knowledge of their histologic diagnoses. Malignant cells were seen and evaluated in 23 cases each of DC and LC (13 classic and 10 variant) for cellularity; cell size, pleomorphism and arrangement (group size, Indian filing, single cells); nuclear features (molding, nucleoli, intranuclear inclusions); and cytoplasmic features (vacuoles, signet-ring cells, granules). RESULTS: Features associated with LC included low cellularity (DC: 26%, LC: 70%), low pleomorphism (DC: 26%, LC: 61%), vacuoles (DC: 22%, LC: 61%), small nucleoli (DC: 43%, LC: 78%), and signet-ring cells (DC: 4%, LC: 48%). Features associated with DC included granular cytoplasm (DC: 39%, LC: 0%), high cellularity (DC: 57%, LC: 13%), high pleomorphism (DC: 48%, LC: 0%), medium to large nucleoli (DC: 35%, LC: 0%) and large sheets of cells (DC: 48%, LC: 17%). The range of cell size was larger in DC (1-20 times the size of RBCs) than in LC (1-8 times). Few cytologic differences were seen between the classic and variant patterns of LC. Variant patterns were slightly more likely to have intranuclear inclusions than classic LC; that was the only statistically significant difference seen. CONCLUSION: Most features studied were seen in at least a few examples of both DC and LC and in both classic and variant patterns of LC. However, some features were more characteristic of one entity than others.
OBJECTIVE: To define cytologic criteria that reliably distinguish lobular (LC) and ductal (DC) carcinoma and classical LC from its variants-alveolar, solid and mixed infiltrative patterns. STUDY DESIGN: Fine needle aspiration (FNA) specimens from 29 cases each of histologically proven LC and DC were reviewed without knowledge of their histologic diagnoses. Malignant cells were seen and evaluated in 23 cases each of DC and LC (13 classic and 10 variant) for cellularity; cell size, pleomorphism and arrangement (group size, Indian filing, single cells); nuclear features (molding, nucleoli, intranuclear inclusions); and cytoplasmic features (vacuoles, signet-ring cells, granules). RESULTS: Features associated with LC included low cellularity (DC: 26%, LC: 70%), low pleomorphism (DC: 26%, LC: 61%), vacuoles (DC: 22%, LC: 61%), small nucleoli (DC: 43%, LC: 78%), and signet-ring cells (DC: 4%, LC: 48%). Features associated with DC included granular cytoplasm (DC: 39%, LC: 0%), high cellularity (DC: 57%, LC: 13%), high pleomorphism (DC: 48%, LC: 0%), medium to large nucleoli (DC: 35%, LC: 0%) and large sheets of cells (DC: 48%, LC: 17%). The range of cell size was larger in DC (1-20 times the size of RBCs) than in LC (1-8 times). Few cytologic differences were seen between the classic and variant patterns of LC. Variant patterns were slightly more likely to have intranuclear inclusions than classic LC; that was the only statistically significant difference seen. CONCLUSION: Most features studied were seen in at least a few examples of both DC and LC and in both classic and variant patterns of LC. However, some features were more characteristic of one entity than others.