The risk of developing breast sarcoma is largely unknown. Some authors showed a significant correlation between external beam radiation of the breast or chest wall and sarcoma.28Another study demonstrated a larger percentage of angiosarcomas as a histopathological subtype post-irradiation therapy sarcoma when compared with other subtypes of sarcoma. The latent time interval between radiotherapy and breast sarcoma was 12 years and 8 months. For breast angiosarcoma this period of time was even shorter with only 8 years and 8 months.7 Other studies had comparable results.7 Another proposed risk factor, especially for angiosarcomas, is a chronic lymphedema of the breast and the arm.38 In the future, the number of sarcomas may grow because the increasing numbers of patients undergoing radiotherapy. This UP-PGH study had no patients with prior exposure to radiation or chronic lymphedema.

Breast sarcoma is mostly diagnosed in patients who are in their 5^th or 6^th decade of life, and the UP-PGH data is comparable to the largest review done by Al-Benna42. In that review, the weighted mean age upon diagnosis was 50.0 years, with 98.5% occurring in women and 1.5% in men. The mean tumor size was between 4.8- 6.6 cm (range: 0.8 cm-40cm) in most studies. McGregor26 reported median tumor size of 3 cm and Adem43 had a mean tumor size of 10 cm for angiosarcomas. In contrast to the UP-PGH study the mean age of diagnosis is 46 years, with 87% occurring in women and 13% among men and the median tumor size is 7.4 cm. It has been postulated that the menopausal status influences breast sarcoma development, but there is little evidence to support this. The typical patient with a breast sarcoma has a unilateral, well-defined, mobile and painless breast lump which often grows more rapidly than epithelial breast carcinoma

Primary breast sarcomas are as histologically heterogenous as non-breast tissue sarcomas.39 It needs a thorough histologic review by a qualified and experienced soft tissue pathologist.40A subtype evaluation in the breast is limited due to its rarity and at times, the heterogeneity that these tumors present with may lead to misclassification.28 There is debate on the exclusion of cystosarcoma phyllodes from PBS studies. Some authors excluded the former because these tumors have a benign epithelial component in addition to a malignant mesenchymal component. However, some classify malignant phyllodes as a subtype of PBS due to its similar clinical course. PBS on the other hand, arises from the mesenchymal tissue of the mammary gland, and the pathologic definition of this entity is clear.11

Also, PBS should be distinguished from metaplastic carcinomas (carcinosarcomas) of the breast, which represent as various combinations of poorly differentiated ductal carcinoma with mesenchymal and squamous components. There are clinicians in literature who classify and manage them as a variant of ductal adenocarcinomas of the breast. Therefore, when facing a spindle cell neoplasm in an epithelial organ such as the breast, one should be careful in rendering the diagnosis of sarcoma. In this setting, immunohistochemistry using the right antibodies is a major contribution. Berg defined stromal sarcomas of the breast in 1962 as a group of mesenchymal malignant tumors with fibrous, myxoid and adipose components, excluding malignant cystosarcoma phyllodes and lymphomas.12

However, series in the literature have included many different entities under the rubric of sarcomas such as cystosarcoma phyllodes andlymphosarcoma. 13141516171819202122232425262728For this study, we chose to categorize primary breast sarcomas in histogenic terms, similar to other soft-tissue sarcomas, thus including angiosarcomas, and excluding malignant cystosarcomas phyllodes and carcinosarcoma (Table 4).12312293031

In this UP-PGH data, the most common histologies were spindle cell sarcoma, fibrosarcoma, and liposarcoma constituting 50% of the whole population. This was comparable with the study done by Berg12 and by North44.

The current 2010 American Joint Committee on Cancer TNM system45 incorporates four major changes compared with the 2002 system, two of the changes pertain to histology. First, it excludes four histologic types: gastrointestinal stromal tumor (GIST), desmoid tumor, Kaposi sarcoma, and infantile fibrosarcoma. Second, it adds the following histologic types: angiosarcoma, extra skeletal Ewing sarcoma, and dermatofibrosarcomaprotuberans. For the purposes of this study, re-classification of histologic subtypes before the changes were made was not done because of no data. The report of the pathologist was taken as it was and everything written in the report was recorded in the data collection sheet as it was.

The results from this UP-PGH study showed that the lung was the most common site of distant metastases (75%), followed by the liver (25%).

Metastases from breast sarcoma spread hematogenously, and typically to the lungs, bones, and liver. Lymph node metastases are rare. In most cases of lymph node metastasis, the histopathologic subtype carcinosarcoma should be ruled out. Breast skin and the nipple areola complex are rarely involved by breast sarcomas, but angiosarcomas may be associated with a bluish discoloration of the skin overlying the lesion.^7,8,15

There is still no definitive consensus regarding the treatment of PBS, even if simple mastectomy without axillary dissection is still widely regarded as the gold standard,11the only potentially curative modality for breast sarcoma. There are major variations in the extent of local therapy, ranging from wide local excision3032to radical mastectomy.3334 Recourse to chemotherapy and radiotherapy has also differed from one center to another 293132A key article related to soft tissue sarcomas was the collaborative meta-analysis published in 1997, which showed that adjuvant chemotherapy with doxorubicin-based regimens yielded a statistically significant improvement in disease-free survival but without a considerable impact on overall survival.35 Moreover, a recent randomized trial comparing intensive adjuvant epirubicin plus ifosfamide versus no treatment showed, despite its small size, a significant improvement in both disease-free and overall survival.36 Adjuvant radiotherapy may improve locoregional control but its impact on survival is still unclear.

According to the clinicopathologic series done at the Mayo Clinic, lymphatic spread is uncommon as shown by the absence of axillary lymph node metastasis in their cases, and therefore axillary node dissection is not necessary. When lymph node metastasis is present, the diagnosis of a metaplastic carcinoma should be considered even in the presence of a pure spindle cell neoplasm, or that the sarcoma is far advanced.5

Tumor size seems to be the most frequently reliable prognostic factor in many of these series, as in breast carcinomas and soft tissue sarcomas. 1303133 Other reported prognostic factors are the histopathological diagnosis,31infiltrative features,2histopathologic grading,2,30,31presence of positive margins,12,31and extent of surgery for local recurrence.12,31 Some authors also found age to be of prognostic importance.19Angiosarcoma was associated with poor outcomes in some series.1,2,30,31Margin status is a major risk factor for recurrence as it occurs in any neoplastic entity, and some authors advised adjuvant radiotherapy for cases with positive margins,3 or less than 2 cm of clear margins.28In this UP-PGH study, tumor size >10 cm, positive margins, tumor grades 2-3, and a histologic diagnosis of angiosarcoma are well correlated with lower DFS and OS, comparable with similar studies.

Breast sarcomas are primarily non-epithelial malignancies that arise from the connective tissue within the breast. They can be primary or secondary, as a result of treatment (usually RT) from previous breast malignancies. They are histologically heterogeneous, thus definitive histologic diagnosis maybe difficult if not seen by an experienced pathologist. They are extremely aggressive and once not properly diagnosed or managed early, have poor prognosis. Due to its rarity, it is difficult to conduct a prospective study to establish a standard of care for patients with primary breast sarcoma.

This UP-PGH series is similar to other series done in US and Europe, with findings as follows: (1) the larger the size, the poorer the prognosis; (2) the higher the grade, the poorer the prognosis; (3) angiosarcoma as a histologic subtype is associated with poor outcomes; and (4) the lung and liver are the most common sites of metastases. These are not different from the natural behavior of soft tissue sarcomas from other parts of the body. Therefore, primary breast sarcomas should be managed as such.

In the diagnosis of primary breast sarcoma, core needle biopsy is the procedure of choice for definitive diagnosis. Appropriate diagnostics should be ordered for accurate staging. According to the NCCN guidelines, chest CT scan should be ordered for all and abdominopelvic CT scans should be considered for angiosarcoma. For angiosarcomas, given the possible catastrophic outcomes, they should be offered early on maximal treatment, and that is, surgery, chemotherapy, and radiotherapy.

Due to the rarity of this disease, a multidisciplinary approach is recommended to properly manage this particular subset of sarcomas. Histologic review by an experienced soft tissue pathologistis critical in making the diagnosis and determining the histologic subtype. If diagnosis is still doubtful, appropriate immunostaining should be applied.