Clinicopathologic analysis of invasive micropapillary differentiation in breast carcinoma

Abstract

Invasive micropapillary carcinoma (IMPCa) of breast is histologically characterized by growth of cohesive tumor cell clusters within prominent clear spaces resembling dilated angiolymphatic vessels. In this study, eighty three breast carcinomas with IMPCa differentiation were identified by review of the invasive carcinoma cases in our institution and correlated retrospectively with standard clinicopathologic parameters and survival status relative to a control series of cases (mean follow up 7 years). IMPCa growth pattern was present in 6% of all breast carcinomas; it was generally a focal component in otherwise typical invasive ductal carcinoma. It comprised more than 80% of the total neoplasm in only 10 cases (12%), 50-80% of the neoplasm in 7 cases (8%), 20-50% of the neoplasm in 22 cases (26%) and less than 20% in 44 cases (53%). The mean tumor size was 4 cm, 22% invaded skin, and 58% were poorly differentiated, but 71% were ER positive. Axillary node metastases were present in 77% of cases, were typically multiple (51% had three or more positive), and usually contained an IMPCa component (81% of the cases). There was no significant difference in node status, ER status, size, tumor grade, or peritumoral angiolymphatic invasion between tumors with predominant (more than 50%) v/s focal IMPCa components. In both groups 46% of the patients died from their disease (mean interval to death = 36m). Skin involvement and nodal status were the only parameters which predicted poor survival (P =.01). The outcome of patients with IMPCa did not differ significantly from infiltrating ductal carcinomas of similar node status. In conclusion, our results suggest that IMPCa growth pattern may be a manifestation of aggressive behavior, as shown by frequent skin invasion and extensive nodal involvement. However, clinicopathologic features and outcome of IMPCa are not strongly dependent on the relative amount of micropapillary component.