Breast sarcomas are a diverse group of neoplasms arising from the nonepithelial and mesenchymal tissues of the breast. Their behaviors vary from the more indolent tumors like cystosarcoma phyllodes to the extremely aggressive angiosarcoma. They should be approached in a similar fashion to their soft tissue sarcoma counterparts in other locations and managed by multidisciplinary sarcoma specialists with attention to wide-margin surgical excision. The use of adjuvant chemotherapy is controversial and should be discussed on a case-by-case basis and preferably given in the context of a clinical trial.
Key points
- •
Breast sarcomas are a distinct tumor group and should be treated differently than breast carcinomas.
- •
Angiosarcomas as a subgroup are aggressive neoplasms and one of the more common subgroups identified in breast sarcomas.
- •
Cystosarcoma phyllodes, often not included in sarcoma reviews, are a unique tumor group and should be approached in a similar fashion to breast sarcomas.
Introduction
Breast sarcomas are a diverse group of neoplasms arising from the nonepithelial components of the breast. Because of the heterogeneity of the histologic subtypes, it is always difficult to lump them into one overarching consensus review. They comprise less than 1% of breast cancer and less than 5% of all sarcomas. Annual incidence has been estimated at approximately 45 new cases per 10 million women. There is mounting concern that this will increase over time given the common use of adjuvant radiation for breast carcinoma. Despite the diversity of subtypes, it is important to separate breast sarcomas out as a unique tumor group in an effort to highlight the divergent clinical course from breast carcinoma and encourage the importance of referring patients to a multidisciplinary sarcoma team when logistically feasible.
Introduction
Breast sarcomas are a diverse group of neoplasms arising from the nonepithelial components of the breast. Because of the heterogeneity of the histologic subtypes, it is always difficult to lump them into one overarching consensus review. They comprise less than 1% of breast cancer and less than 5% of all sarcomas. Annual incidence has been estimated at approximately 45 new cases per 10 million women. There is mounting concern that this will increase over time given the common use of adjuvant radiation for breast carcinoma. Despite the diversity of subtypes, it is important to separate breast sarcomas out as a unique tumor group in an effort to highlight the divergent clinical course from breast carcinoma and encourage the importance of referring patients to a multidisciplinary sarcoma team when logistically feasible.
Subtypes and risk factors
By and large, all histologic subtypes of sarcoma have the possibility of arising in the breast. Some of the more common subtypes include undifferentiated pleomorphic high-grade sarcoma, myxofibrosarcoma, angiosarcoma, and spindle cell sarcoma. Other subtypes (leiomyosarcoma, Ewing, rhabdomyosarcoma, synovial sarcoma, chondrosarcoma, and extraosseous osteosarcoma) have also been reported in smaller case series or case reports; however, these are exceedingly rare. In general, with the exception of cystosarcoma phyllodes and primary or secondary breast angiosarcoma, when encountering a soft tissue sarcoma in the breast, the first course of action should be to exclude the possibility of another primary source that has metastasized to the breast.
The two most commonly described risk factors for breast sarcoma include radiation, typically as a treatment for a previous malignancy (ie, mantle radiation for Hodgkin disease or breast carcinoma) and lymphedema (Steward-Treves syndrome). Cahan and colleagues in 1948 defined the criteria for postradiation sarcoma. The requirements are (1) evidence of an initial distinct malignant tumor different from the subsequent sarcoma, (2) development of the second malignant tumor in an irradiated field, (3) long interval between irradiation and development of sarcoma, and (4) histological confirmation of sarcoma. Occupational exposure to vinyl chlorides has been linked to hepatic angiosarcoma, and artificial breast implants have been implicated but are controversial. TP53 mutation can predispose to sarcomas in general, with some occurring in the breast. Despite the possible causes, most breast sarcomas present without any obvious predisposing factor.
Radiation-associated sarcomas of the breast can develop after a wide range of time intervals, with a cumulative incidence of 0.3% at 15 years after radiation treatment reported in a case series of patients diagnosed between 1973 and 1997. Whether this will decrease over time with improvement in radiation techniques and with dose reductions, or conversely, increase, as radiation is used more frequently and operations become more conservative, remains to be seen. Various subtypes have been reported in postradiation breast sarcomas, but angiosarcoma seems to be one of the most common. The median time to occurrence after radiation ranges from 5 to 8 years with a wide range of 1 to 16 years reported. The risk increases with higher dose of radiation, exposure during childhood, concurrent dose of chemotherapy, and preexisting genetic conditions, such as BRCA-1 mutation.
Diagnosis and staging
Diagnosis of breast sarcomas is made histologically by a percutaneous biopsy. Core biopsy is generally preferred to fine-needle aspiration, which will typically yield insufficient material. Punch biopsies may be sufficient for radiation-associated angiosarcoma, which is typically a cutaneous sarcoma. For local staging, breast mammography, ultrasound, and breast MRI are most commonly used. In mammography, microcalcifications are not seen as frequently as in breast carcinomas; therefore, MRI may be a better modality.
The 10-year overall survival of breast sarcomas was 62% in one series ; the 5- and 10-year relapse-free survival was 47% and 42% in another series whereby patients did not have distant metastases at presentation. It is difficult to lump such a heterogeneous tumor group together and derive any meaningful data with regard to overall survival. But in various retrospective case series and reports, characteristics like histology, margin status, size, depth of tumor, and grade seem to be the overall driving prognostic factors.
Treatment
The treatment paradigm for breast sarcoma has mirrored that of its soft tissue counterparts with the main objective being surgical resection with wide margins. This treatment may require a mastectomy depending on the size and infiltration of the tumor; extent of surgery is histology dependent. Unlike breast carcinomas, lymph node metastasis is not common, with few notable exceptions, including epithelioid sarcomas and angiosarcomas. In a retrospective case series including 34 lymph node dissections for breast sarcomas, none were positive. Conversely, in a series of 28 patients with breast angiosarcoma who underwent axillary lymph node dissection as part of their primary surgery, 2 (7%) had positive lymph nodes.
Chemotherapy for breast sarcomas, similar to their extremity and retroperitoneal counterparts, has remained controversial. In locally advanced cases, neoadjuvant chemotherapy and/or radiation may be warranted to decrease tumor size and subsequently allow for a margin negative resection. Caution should be advised, however, when using chemotherapy to try to reduce size to perform lumpectomy versus mastectomy. As responses to chemotherapy are highly variable, the risk of rendering a resectable patient unresectable is not negligible. This risk is particularly troublesome given the close proximity to the chest wall. Invasion into the underlying muscle and bone may necessitate a more morbid surgery.
The data supporting or refuting adjuvant chemotherapy in breast sarcoma are exceedingly controversial. Advocates supporting the use for adjuvant chemotherapy in soft tissue sarcoma (STS) have often turned towards the sarcoma meta-analyses, which compared adjuvant chemotherapy versus observation in resected localized STS. The more recent meta-analysis reviewing 18 phase III trials was an update on an earlier meta-analysis analyzing 14 trials. At the time of the initial publication, there did not seem to be an overall survival advantage. However, there was a significant improvement in time to local and distant recurrence and overall recurrence-free survival, with a trend toward improved overall survival. Critics of the article centered on the lack of adequate dosing and the utilization of single-agent doxorubicin. The updated article incorporated 4 more eligible studies using ifosfamide, and the results this time suggested a statistically significant survival advantage. Whether this was due to a larger sample size, narrowing the confidence intervals, or dose intensification with use of multiagent regimens is unknown. The evidence was most significant in extremity sarcomas. However, caution is always advised when interpreting subgroup analyses in a meta-analysis. A more recent, retrospective, single-institution study evaluating adjuvant chemotherapy over a 10-year period also found a benefit to adjuvant chemotherapy with a median disease-free survival of 29.6 months compared with 7.8 months and a median overall survival of 67.0 months versus 33.7 months in treated versus untreated patients. Interpreting these data is always troublesome given the retrospective pooled data and the heterogeneity of tumor types involved. Retrospective data, heterogeneity of histologic sub-types, coupled with the dosing regimens used, makes relying on these data as support to use chemotherapy challenging.
More recently, Le Cesne and colleagues published a pooled analysis of 2 STBSG (Soft Tissue and Bone Sarcoma Group)-European Organization for Research and Treatment of Cancer (EORTC) phase III clinical trials specifically evaluating the use of adjuvant chemotherapy in soft tissue sarcoma. The data were pooled with the intent of prospectively identifying whether there were any small patient subgroups deriving benefit from adjuvant chemotherapy, given the disparities in previous findings. The investigators concluded that chemotherapy was not associated with a better overall survival in younger patients or in any histologic subgroup. The 2 consecutive EORTC studies are the largest adjuvant studies published to date. Both studies failed to demonstrate any advantage of adjuvant chemotherapy on overall survival. However, as has been the case with other adjuvant studies, there does seem to be an advantage to recurrence free survival (RFS).
The take-home message, once again, is that there seems to be no reliable take-home message. The EORTC data suggest, given no overall survival advantage, adjuvant chemotherapy should not be considered as standard of care. Counseling patients at high risk for recurrence regarding the variation of opinions and using adjuvant chemotherapy in the context of a clinical trial is advised.
Tale of 2 breast sarcomas
Cystosarcoma Phyllodes
Phyllodes tumor is a rare neoplasm that accounts for 1% of all breast cancer in women. Phyllodes tumors are often excluded from discussion of other breast sarcomas given the epithelial component that is present. However, given the similar clinical course and survival characteristics between the two entities, cystosarcoma phyllodes are included in this discussion.
The median age at presentation is typically the mid-40s. The tumors are often large and can occasionally reach sizes of up to 40 cm in diameter. Rates of recurrence vary significantly depending on the series, with distant metastases occurring in a minority of patients, and generally occurring in those with borderline or malignant grading. Metastases to the lymph nodes in one institutional case series were observed in fewer than 5% of patients. Johannes Müller initially described phyllodes tumors as cystosarcoma phyllodes in 1838. The current World Health Organization (WHO) classification was established in 2003. They are composed of biphasic stromal and glandular elements. The WHO classification distinguishes 3 histologic subtypes of phyllodes tumors: benign, borderline, and malignant, which account for 58.4% to74.6%, 15.0% to 16.1%, and 9.3% to 31.0% of all phyllodes tumors, respectively. The grading system is based on degree of stromal hypercellularity, mitoses, cytologic atypia, and degree of stromal overgrowth.
The behavior of most benign phyllodes tumors is fairly indolent. Local recurrences can occur in approximately 10% to 17% of patients, with metastases exceedingly rare. However, the more borderline and malignant variants carry a significant chance of metastases. Surgical resection remains the most important treatment, with tumor histology dictating whether excision, wide local excision, or mastectomy is favored. All patients should be followed prospectively following primary treatment. Although benign phyllodes tumors have low metastasis potential, it is still possible; therefore, patients should be followed for local recurrence as well as distant surveillance. Malignant phyllodes should be followed similar to a high-grade sarcoma with chest imaging every 3 to 4 months for 2 years, every 6 months out to 5 years, and annually thereafter. Bone scan should be considered as case reports of bone-only metastases exist.
In one large case series out of China, the researchers found that younger patients were more likely to develop local recurrence, with a median age of 33 years. Whether this was biology or extent of surgery was not commented on. Despite the difference in local recurrence, there was not an age-related difference in distant metastasis and overall survival. Their findings suggest that in younger patients, discussions about wider local excisions or mastectomy should be performed. In other larger series, this correlation did not hold true. Similar to other extended case series that found local recurrence rates ranging from 12.2% to 32.0%, 10% of patients developed distant metastases. However, of those that developed distant metastases, approximately 25% were classified as having malignant phyllodes at diagnosis.
The most frequent site of metastases is soft tissue, followed by lungs, thoracic cavity, bones, and pleura. The thoracic cavity and soft tissue lesions likely arose from local recurrences into adjacent tissues. Margin status cannot be overemphasized. The National Comprehensive Cancer Network’s “Breast Cancer” version 1.2016 ( http://www.nccn.org ) delineates a surgical margin of 1 cm as adequate.
As is the case with many other soft tissue sarcomas, there are varying opinions on whether adjuvant chemotherapy adds any benefit to breast sarcomas and in general is not recommended outside of a clinical trial.
Unfortunately, large randomized prospective studies of radiation therapy in phyllodes tumors are lacking; therefore, there are no definitive guidelines. In general, in cases whereby adequate margins are difficult to obtain, and local recurrence would result in significant morbidity, radiation may be considered following the same guidelines applied to soft tissue sarcoma.
Angiosarcoma
Angiosarcomas affecting the breast are typically divided into 2 broad categories: primary (arising de novo in the breast) and secondary (typically the after effect of prior breast irradiation). The so-called Stewart-Treves syndrome or angiosarcoma arising from prior lymphedema would occur usually in the upper arm as a result of lymph node dissections that were performed during radical mastectomies. With more conservative management of breast carcinoma, including breast conservation surgery and smaller radiation fields, one does not encounter Stewart-Treves syndrome as often.
Angiosarcoma arising in the setting of prior radiation therapy mainly involves the skin, with or without local invasion of subjacent breast tissue. As more patients undergo breast-conserving surgeries with the addition of adjuvant radiation, the incidence may be increasing. The prognosis of secondary breast angiosarcoma is dismal and seems to be independent of tumor grade.
In a single-institution retrospective case series published by the Mayo clinic, the investigators found that, although secondary angiosarcoma and primary angiosarcoma occur in 2 distinct patient populations, their outcomes are similar. Most series report an overall survival of approximately 18 to 36 months. However, the Mayo patients had an overall median survival of 3 years. Although secondary angiosarcoma was more likely to be high grade in their review, the median tumor size and median survival were similar.
In several case series of breast angiosarcomas, histologic grade seemed to play an important role. However, Nascimento and colleagues noted that the tumors in their retrospective case series were evenly distributed among the 3 different histologic grades. Most were treated by mastectomy with only a minority of the patients receiving any form of adjuvant treatment. Although the rate of local recurrence was moderate, the rate of distant metastases was quite high, with approximately 60% of the patients developing metastatic disease. In contrast to older studies, there was no histologic grade correlation with patient outcomes.
D’Angelo and colleagues, in a retrospective series evaluating clinical characteristics, prognostic factors, and treatment outcomes with surgery and chemotherapy in patients with radiation-associated angiosarcoma of the breast, found a median disease-specific survival (DSS) of 2.97 years. Their 1-, 2-, and 5-year DSS was 84%, 66%, and 47%, respectively, which was worse than other previously reported series. Not surprisingly, their study suggested age and depth were independent predictors of DSS, with tumors that invade the underlying muscularis layers behaving more aggressively. However, somewhat surprisingly, and contrary to an earlier retrospective case series by the MD Anderson group, they did not identify margin status as a predictor of local recurrence; instead only depth was associated. Regardless of the type of resection, the local recurrence rates tend to be high in all the series. The 1-year local RFS was 55% in the D’Angelo study. However, in a smaller study reported by Monroe and colleagues, the 1-year local recurrence rate was 84%, though most of those patients did not undergo mastectomy.
In conclusion, the behavior of breast sarcomas can be as divergent as that of the more benign phyllodes tumors on one end of the spectrum and the more aggressive angiosarcoma on the other. Given the heterogeneity and rarity of the tumors, patients should be referred to centers with sarcoma multidisciplinary teams. Along with the clinical expertise found at such centers, it also allows for more standardization of data and the possibility of patients to be enrolled in clinical trials.
Disclosures: Novartis Speaker’s Bureau, and Advisory role to Jannsen Pharmaceuticals.
Related posts:
Contemporary Management and Controversies of Sarcoma
Refinements in Sarcoma Classification in the Current 2013 World Health Organization Classification of Tumours of Soft Tissue and Bone
Sarcomas 2016
Management of Sarcoma Metastases to the Lung
Combined Therapy of Gastrointestinal Stromal Tumors
Myxofibrosarcoma
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
