The mammographic appearance of primary breast sarcomas varies with the histologic subtype. In general, they lack calcification and spiculation and may be mistaken for benign breast lesions such as fibroadenomas. For this reason, in addition to a pathologic diagnosis, historical features such as rapid growth and imaging modalities such as fluorodeoxyglucose-positronemission tomography scanning may be useful in distinguishing malignant breast sarcomas from benign breast lesions. Incisional, excisional, and core-needle biopsies are all useful in arriving at a definitive diagnosis. Although fine-needle aspiration can be diagnostic, it does not allow for determination of subtype or grade. The histologic types of primary breast sarcomas vary widely. Evaluation of subtype distribution is limited not only by the rarity of the disease, but also by differences in the classification of sarcomas. In one series of 90 patients with primary breast sarcomas (excluding cystosarcoma phyllodes) treated at the Institut Gustave- Roussy, the histologic types included malignant fibrous histiocytoma (69.5%), angiosarcoma (10%), liposarcoma (8%), malignant peripheral nerve sheath tumor (2.5%), spindlecell sarcoma (2.5%), extraskeletal osteosarcoma (2.5%), rhabdomyosarcoma (2.5%), and leiomyosarcoma (2.5%). Another single-institution series included subtype descriptions for 78 patients with breast sarcomas. Their reported histologic types included malignant cystosarcoma phyllodes (55%), stromal sarcomas (24%), angiosarcomas (14%), fibrosarcomas (12%), liposarcomas (7%), and others (21%). Given the differences in classification, an accurate determination of the distribution of primary breast sarcomas remains challenging. In general, malignant fibrous histiocytomas, fibrosarcomas, liposarcomas, and angiosarcomas comprise the major subtypes of breast sarcomas. Prognostic Factors
The prognosis for primary breast sarcomas is based on the characteristics of soft-tissue sarcomas elsewhere in the body and primarily on the histologic grade and size of the tumor.[ 3,10,16] Tumor grade is an important prognostic factor in most sarcomas, and therefore constitutes the foremost variable in the staging system of primary breast sarcomas (see Table 2). Pathologic features used to define grade generally include differentiation, the mitotic count, necrosis, cellularity, and pleomorphism. The French series of 90 patients with grade 1, 2, and 3 breast sarcomas had corresponding 10-year overall survival rates of 82%, 62%, and 36%, respectively (P = .00007). Size was also an important prognostic factor in this series. The 10-year overall survival rates for tumors < 5 cm, 5 to 10 cm, and > 10 cm on univariate analysis were 76%, 68%, and 28%, respectively (P = .002). However, multivariate analysis did not corroborate the association in this or other studies.[3,16] Other factors such as age, number of mitoses, menopausal status, history of benign breast disease, and initial therapy have not been shown to significantly affect outcome.[ 3,10,16] The only subtype that may confer a poor prognosis is angiosarcoma. Primary breast sarcomas tend to spread locally or hematogenously and are not typically associated with regional lymph node involvement. Even when palpable axillary lymphadenopathy is present, pathologic exami- nation often fails to reveal disease spread. However, metastatic disease to the lymph nodes has been reported. McGowan and colleagues reported that two of three patients with carcinosarcomas who underwent axillary lymph node dissections were found to have nodal metastases. Notably, only epithelial components were present in the lymph nodes. We believe that carcinosarcomas actually represent a poorly differentiated carcinoma, rather than a sarcoma. Liposarcomas have also been noted to have up to a 10% incidence of nodal involvement. In addition, lymph node involvement has been implicated in cases of widespread disease, which can be considered more akin to metastatic than regional spread. Treatment Recommendations
- Surgery-Given the rarity of breast sarcomas, no prospective randomized trials have evaluated potential therapies. Because the behavior, histology, and prognosis of primary breast sarcomas are similar to that of other soft-tissue sarcomas, therapeutic recommendations are based on data from the latter setting. Results of retrospective analyses on outcomes related to therapy for breast sarcoma also aid in management. As with all soft-tissue sarcomas, a multidisciplinary approach at an experienced center involving the surgical oncologist, radiation oncologist, and medical oncologist is necessary. Tumors are then treated according to both histologic grade and size. Surgery remains the mainstay of therapy. The type and extent of surgery should be based on feasibility and the size of the tumor and breast. Retrospective studies have shown no statistically significant difference in cause-specific survival with breastconserving surgery vs mastectomy, if negative margins are achieved. Given the lack of multicentricity of the majority of primary breast sarcomas, wide local excision should be adequate.[ 16,12] An adequate resection margin is the single most important determinant of long-term survival. In cases of local recurrence, salvage mastectomy may be effective. Due to the natural history of breast sarcomas, axillary dissection is rarely necessary, but it may be considered in cases of palpable lymphadenopathy, carcinosarcoma, or liposarcoma.[2,10]
- Radiotherapy-Adjuvant radiotherapy is still controversial due to conflicting data from small retrospective studies.[10,12,21,22] Johnstone and colleagues reported a case series of 10 patients treated with mastectomy and adjuvant radiation who experienced no local or regional failures.[ 21] Other studies have suggested that radiation does not improve disease-free survival but that it may be useful in the treatment of high-grade but not low-grade lesions.[16,12] Given the retrospective nature of these reports and the attendant confounding factors, it is helpful to examine the larger pool of experience with sarcomas in general. In prospective studies of radiation plus surgery for lower-extremity sarcomas, adjuvant radiation is associated with a significant improvement in local control of both low- and high-grade tumors but without improvement in overall survival. Adjuvant radiation is also likely to be of some benefit, especially in the setting of breast-conserving therapy and high-grade or large tumors. Because radiotherapy of the breast may be associated with occasional quality-of-life and rare life-threatening consequences (see below), a thor- ough evaluation of the risks of adjuvant radiation therapy should be made depending on the individual patient and malignancy. The emphasis should be on treating tumors with questionable or positive margins, high-grade features, and those that are larger in size.
- Chemotherapy-Adjuvant chemotherapy does not have a clearly defined role in the treatment of primary breast or soft-tissue sarcomas. A meta-analysis by the Cochrane Collaborative evaluated 14 trials involving a total of 1,568 adults with resectable soft-tissue sarcoma randomized to adjuvant chemotherapy or no chemotherapy following local therapy. The hazard ratio for recurrencefree survival with chemotherapy was 0.75 (P = .0001). No significant benefit in overall survival was associated with adjuvant chemotherapy, but a possible trend for improvement was noted. Subgroup analysis revealed that the best overall survival was associated with extremity lesions (hazard ratio = 0.8, P = .029). A subsequent Italian study of intensified adjuvant chemotherapy, limited to adult soft-tissue sarcomas of the extremities and pelvis, revealed a significant benefit in median overall survival for treated vs untreated patients (75 vs 46 months, P = .03). Although primary breast sarcomas have not been studied exclusively, data from studies regarding sarcomas in general suggest there may be a role for adjuvant chemotherapy in patients with high-grade and large tumors, ideally in the setting of a clinical trial. Some active agents for sarcoma include doxorubicin(Drug information on doxorubicin), epirubicin (Ellence), and ifosfamide (Ifex) or combination chemotherapy regimens such as MAID (mesna [Mesnex], doxorubicin [Adriamycin], ifosfamide, dacarbazine(Drug information on dacarbazine) [DTIC-Dome]) or MAP (mitomycin [Mutamycin], doxorubicin, cisplatin(Drug information on cisplatin) [Platinol]).[25,27,28] Gemcitabine(Drug information on gemcitabine) (Gemzar) has been shown to have limited activity against soft-tissue sarcomas, especially in a salvage situation. The combination of gemcitabine and docetaxel(Drug information on docetaxel) (Taxotere) has shown early promise; however, further investigation is necessary.[30,31]
The risk of developing soft-tissue sarcomas after breast cancer is highest when treatment involves radiotherapy. Ionizing radiation is a known carcinogen that can induce sarcoma.[ 6,13] Axillary radiation-induced lymph node sclerosis and resultant lymphedema is another possible mechanism for radiation-related tumors (see Mammary Angiosarcomas). A correlation may also exist between the integral dose of radiation and the development of a sarcoma in the radiation field. The possible effects of radiation therapy are reported in numerous population-based studies of patients treated for breast carcinomas. In one large series of 274,572 breast cancer patients identified in SEER registries, the incidence of new cases of sarcoma at 15 years posttreatment among those receiving radiation therapy was 3.2 per 1,000, compared to 2.3 per 1,000 among those not receiving radiation (P = .001). Another retrospective study reported the relative risk of breast sarcoma as 2.2 (95% confidence interval [CI] = 1.4-3.3) for breast cancer patients who received radiotherapy compared to those who did not. In contrast, Obedian and colleagues did not find an excess of subsequent malignancies in 1,029 breast cancer patients who underwent lumpectomy and radiation therapy, compared to 1,387 patients treated with mastectomy.[ 34] However, given the rarity of sarcomas (only seven were diagnosed in this trial), it may be difficult to detect a difference in this particular subgroup. Histopathology and Treatment Protocols
Patients with therapy-related breast sarcomas are usually women in their 6th or 7th decades who have been treated for breast carcinomas approximately 11 years (range: 3-44 years) earlier.[5-7] Mean tumor size at diagnosis is 4.2 cm. The tumors can present similarly to primary breast sarcomas. However, they may be more difficult to detect on physical examination or mammogram, as abnormalities of the breast or mastectomy site are often attributed to postirradiation changes. The histopathology of these tumors includes a wide range of soft-tissue sarcomas, with proportions that vary significantly depending on the series. The most frequently reported subtypes include leiomyosarcomas, malignant fibrous histiocytomas, liposarcomas, fibrosarcomas, and angiosarco mas. Compared with other tumors, angiosarcomas appear to occur more frequently in patients who have had prior therapy.[4,5,8,13] Similar to primary breast sarcomas, prognosis correlates with tumor grade. The overall prognosis in breast cancer patients after diagnosis of sarcoma is generally poor, with a reported median survival of 2.3 years and 5-year survival rates of 27% to 35%.[5,8,13] The cornerstone of therapy continues to be surgery, although outcomes remain poor due to local aggressiveness. Operative procedures are often difficult, as the sarcomas arise in irradiated areas. In a report of 15 patients who underwent radical resection of radiation-induced chest wall sarcomas, 3 patients suffered complications of surgery, 7 had local recurrences, and 4 died of metastatic disease. Radiotherapy has also been used with little success, as these tumors tend to be radiation-resistant. There are also limits to the amount of additional therapy that may be given in previously irradiated areas, and chemotherapeutic options are limited as well. The indisputable benefit of radiation therapy in the treatment of breast carcinoma would still appear to outweigh the small risk of sarcoma after therapy. Moreover, available data regarding the relationship of radiation and sarcomas are gathered from retrospective studies that span back many decades. Significant technical advances have since been made in radiation oncology and may serve to lower the reported incidence rate even further. Some retrospective studies include sarcomas developing outside the radiation field for breast cancer therapy, which casts further doubt on the causality of radiotherapy. Nevertheless, one must be cognizant of the increased risk of this rare tumor for purposes of early detection, patient counseling, and treatment. Mammary Angiosarcomas Angiosarcomas are aggressive, heterogeneous malignancies that arise from endothelial cells. These tumors are sometimes referred to separately as lymphangiosarcomas or hemangiosarcomas derived from lymphatic or capillary endothelium, respectively. As this pathologic distinction can be quite difficult and is not always clearly defined, we will refer to both of these subtypes collectively as angiosarcomas. The annual incidence of mammary angiosarcomas has been reported as 5.8 per 10 million women, and although angiosarcomas in general are extremely rare, the breast is one of the more common sites of occurrence. These malignancies are notable as sequelae of therapy for breast carcinoma. Of all sarcomas arising after breast cancer therapy, many studies report angiosarcomas as the most common histologic subtype.[8,13] In a population- based study in Los Angeles County, the adjusted relative risk of developing an angiosarcoma among women with a prior diagnosis of breast cancer compared to women without a prior diagnosis was 59.3 (95% CI = 21.9-152.8). However, this relationship has not been borne out in all studies. Pathogenesis
Therapeutic radiation has been closely implicated in the pathogenesis of angiosarcomas (see Therapy- Related Breast Tumors). In one series of over 274,000 patients, the mean incidence at 15 years was 0.9 per 1,000 cases receiving radiation and 0.1 per 1,000 (P = .0001) cases not receiving radiation. Another series of over 194,000 women with breast carcinoma demonstrated a relative risk of developing angiosarcoma of 15.9 (95% CI = 6.6-38.1) for patients treated with radiation therapy compared to those treated with other modalities. A second association in the pathogenesis of angiosarcomas is lymphedema. Angiosarcomas can occur in chronically lymphedematous extremities from various etiologies. Stewart and Treves originally described angiosarcomas (or lymphangiosarcomas) of the upper extremity, breast, and axilla arising in women with chronic lymphedema who had been previously treated for breast cancer.[ 14] This occurrence has since been designated as Stewart-Treves syndrome. Lymphedema of the upper extremity and breast may occur as a result of surgery (particularly axillary dissections) or radiation therapy (via axillary lymph node sclerosis). Thus, it is difficult to distinguish whether an angiosarcoma is secondary to lymphedema, primary radiotherapy, or a combination thereof. Clinical Presentation
Mammary angiosarcomas tend to occur in either young women with no prior cancer or in older women following therapy for breast carcinoma.[ 11,36] The clinical presentation is that of a palpable growing breast mass that may have an overlying bluish tint to the skin. The age at diagnosis ranges from 17 to 70 years with a median age of 38 years, which is younger than for typical breast carcinomas.[ 37] Mammographically, angiosarcomas can appear ill-defined without spiculae or calcifications or may not be apparent at all, even in patients with large palpable masses.[2,39] Tumor size ranges from 1 to 15 cm with a median of 5 cm. Lymphedema-associated angiosarcomas can appear as single or multiple purplish macular or papular lesions in the edematous upper extremity or as a deeper ecchymoses or cellulitis. The average age of occurrence is 64 years (range: 44-84 years). In the Los Angeles study, the mean time from development of the original breast carcinoma to subsequent Stewart-Treves syndrome was 9.7 and 4.4 years for upper extremity and chest or breast tumors, respectively. Prognosis
The single most important indicator of prognosis for patients with angiosarcomas is histologic grade. A series of 32 patients with angiosarcoma reported 10-year recurrence-free survival rates of 76%, 70%, and 15% for patients with grade 1, 2, and 3 tumors.[ 40] The natural history of angiosarcomas is also highly dependent on tumor grade. Low-grade tumors follow an indolent course, whereas high-grade tumors are extremely aggressive.[ 42] These angiosarcomas metastasize early, most often to the lung and liver. However, they also tend to metastasize to bones, skin, and the contralateral breast, which are unusual locations for other soft-tissue sarcomas. Overall survival in angiosarcoma patients is generally quite poor, with a median lifespan after diagnosis of approximately 15 months. With more aggressive management strategies, survival rates appear to be improving in recent years, but further improvement is required given the poor prognosis. Treatment Options
Therapy for angiosarcoma is based primarily on local control. Surgical management, ranging from wide local excision to chest wall resection, should be chosen to optimize chances of complete resection and preservation of form. Despite aggressive surgical management, local recurrences are common, and repeat surgery is often necessary. Radiation therapy may play a useful role in the neoadjuvant or adjuvant setting, but few data exist to support this hypothesis at present. Similarly, little is known about the utility of chemotherapeutic agents, but agents used to treat other soft-tissue sarcomas may be employed (see Primary Breast Sarcomas). In summary, angiosarcomas are locally aggressive soft-tissue sarcomas that arise primarily in the breast and are associated with prior breast cancer therapy (with or without radiation) or in the setting of chronic lymphedema. Prognosis is based on histologic grade, and therapy is centered on surgical resection with the possible addition of radiation. Because angiosarcomas are composed of vascular tissue, future therapy may include vascular-targeting agents or antiangiogenesis drugs.