Downloaded from http://journals.lww.com/pathologycasereviews by BhDMf5ePHKbH4TTImqenVNDRplvZecTJCFZ+46Saueq1g+u0YQCUgJMQH/p3xRpUletb3xEPd+o= on 07/04/2018 Malignant Spindle Cell Tumors of the Breast: A Guide for the Practicing Pathologist Olaoluwa Bode-Omoleye, MD, MPH, Komal Arora, MD, and I-Tien Yeh, MD Abstract: Spindle cell tumors of the breast are problematic when encoun- tered by the practicing pathologist because they are relatively infrequent, especially those that are not accompanied by an epithelial component. This review offers an approach to the diagnosis of malignant spindle cell tumors with an emphasis on key features and the differential diagnosis. Key Words: breast sarcoma, spindle cell tumor, monophasic, pleomorphic, metaplastic carcinoma, sarcoma (AJSP: Reviews & Reports 2016;21: 16–23) M alignant spindle cell tumors of the breast are relatively uncommon in comparison with epithelial tumors of the breast and therefore represent a diagnostic problem for the practic- ing pathologist. Most commonly, spindle cell tumors of the breast occur as part of a biphasic proliferation, such as metaplastic carcinomas or phyllodes tumors. Others represent a variety of mesenchymal tumor types, most of which are morphologically identical to the tumors that arise in extramammary tissues. Careful attention to clinical history as well as gross and microscopic details will aid in making the correct diagnosis. Immunohistochemistry and molecular studies can help define the specific type of spindle cell tumor. Clinically, the important differential diagnoses are among benign spindle cell lesions, spindle cell carcinoma (metaplastic carcinoma or carcinosarcoma), sarcoma, and metastasis. The par- ticular type of sarcoma is less important than defining the tumor as purely sarcomatous versus carcinosarcoma or metastasis. This review focuses on the approach to making the correct diagnosis of spindle cell malignancies of the breast, followed by key features of specific malignant spindle cell tumors and their differential diagnosis. THE APPROACH Clinical History Check to see if there was a previous tumor (breast or nonbreast), and if so, what kind of therapy did the patient have. A history of breast radiation for any tumor type raises the possib- ility of postradiation sarcoma, especially angiosarcoma. Other postradiation sarcomas include osteosarcoma, undifferentiated pleomorphic sarcoma (AKA malignant fibrous histiocytoma), and fibrosarcoma. The latent period is typically at least 4 years af- ter radiation. Patients with a history of Hodgkin lymphoma may have had breast tissue included in the radiation field. Family history of sarcoma is also important, especially in patients with Li-Fraumeni syndrome because both breast cancer and sarcomas are part of the syndrome. Gross Examination What is the tumor size? A larger tumor is more likely malig- nant, although size is never the sole criterion for malignancy. Where is the tumor location, that is, is it dermal/subcutaneous/ within the breast/deep to the breast? A superficial tumor may be dermatofibrosarcoma protuberans, whereas a tumor deep to the breast may be a soft tissue sarcoma of the chest wall. Is the tumor circumscribed or infiltrative? An infiltrative tumor is more likely malignant, although again, it is never the sole criterion because le- sions such as fibromatosis are typically infiltrative but not consid- ered malignant and many malignancies will look grossly well circumscribed. Margin status is especially important in infiltrative tumors to predict recurrence, even if the lesion is benign. Is there hemorrhage or necrosis? Significant hemorrhage and necrosis are more likely to be associated with malignancy. Microscopic Examination At low power, the margins of the tumor should be noted, as well as the location of the tumor. As in the discussion of gross characteristics earlier, dermal/subcutaneous and chest wall tumors will have a differential diagnosis different from that of tumors that arise within the breast tissue. At medium-to-high power, a careful search for epithelial ele- ments should be made. Oftentimes, clear-cut glands are not pres- ent, but the cohesiveness of tumor cells may hint at the epithelial elements. Some pure spindle cell tumors are metaplastic carcino- mas without obvious epithelial component. Their epithelial nature can be demonstrated by immunohistochemistry for cytokeratins (see section on immunohistochemistry). The presence of differ- entiated elements such as chondroid or osteoid tissue is more com- mon in metaplastic carcinomas than finding pure chondrosarcoma or osteosarcoma in the breast. Malignancy must be established by noting cellularity, prolif- erative index (mitotic count and/or Ki-67 staining), necrosis, and cytologic atypia. Some benign lesions will have increased mitotic activity and be infiltrative, such as fibromatosis, but will not have pleomorphism. Most pleomorphic lesions with necrosis will in fact be malignant. In addition, the site of origin should be established because some spindle cell tumors in the breast may be metastatic from an- other site. Specific differentiation can be by findings of differenti- ated tissue, such as vessels or osteoid, but in the absence of such features, history and ancillary studies are paramount. Once the determination has been made that the tumor is a pure spindle cell proliferation within the breast and is malignant, primarily in the breast, further characterization of the tumor may be performed by noting growth patterns, considering the differen- tial diagnosis and using ancillary testing. From the Department of Pathology, University of Texas Health Science Center at San Antonio, San Antonio, TX. Reprints: I-Tien Yeh, MD, Department of Pathology, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Dr, San Antonio, TX 78229. E-mail: yehi@uthscsa.edu. The authors have no funding or conflicts to declare. Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 2381-5949 DOI: 10.1097/PCR.0000000000000124 REVIEW 16 www.pathologycasereviews.com AJSP: Reviews & Reports • Volume 21, Number 1, January/February 2016 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.