DOI: 10.1148/rg.27si075016
RadioGraphics 2007;27:S125-S130
© RSNA, 2007
Best Cases from the AFIP
Angiosarcoma of the Breast1
Robert F. Lim, MD and
Reginald Goei, MD
1 From the Department of Radiology, Atrium Medical Center, Henri Dunantstraat 5, 6419 PC Heerlen, the Netherlands. Received February 2, 2007; revision requested March 7 and received April 2; accepted April 18. Both authors have no financial relationships to disclose.
Address correspondence to R.F.L. (e-mail: robert.f.lim{at}gmail.com).
 |
History
|
|---|
An 81-year-old woman with a history of breast cancer presented with a palpable mass in the upper outer quadrant of her left breast. She did not complain of tenderness or pain.
At physical examination a mass was palpable, but not movable, from the surrounding tissue. There was discreet dimpling of the skin laterally, and the nipple was slightly retracted. There was no alteration of skin color.
The patient’s medical history revealed an adenocarcinoma in the left breast (pT1N0M0) 18 years ago, which was treated with breast-conserving surgery and axillary lymph node dissection. She also received adjuvant radiation therapy to the left breast with 5-MV photons to a total dose of 50.00 Gy in fractions of 2.00 Gy. This course was followed by a booster with 5-MV photons in fractions of 2.00 Gy, to a total dose of 64.00 Gy.
|
Imaging Findings
|
|---|
An opaque, ill-defined, and lobulated 4 x 3.5-cm lesion was shown at mammography at the site of the palpable mass. The lesion was best appreciated on the craniocaudal projection (Fig 1a). Oval, partly well-defined, heterogeneous lesions were situated dorsally of the nipple. In the mediolateral oblique projection, an irregular, 7 x 3.5-cm lesion without microcalcifications was seen in the cranial half of the fibroglandular disk (Fig 1b). Some coarse calcifications were also present; these findings most likely represented fatty necrosis due to the previous operation.
View larger version (76K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a. Craniocaudal (a) and mediolateral oblique (b) mammograms of the left breast show a lobulated lesion in the upper half of the breast (arrows), with a diffuse hazy extension toward the nipple. Note the round to oval structures in the retroareolar region (arrowheads).
|
|
View larger version (93K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b. Craniocaudal (a) and mediolateral oblique (b) mammograms of the left breast show a lobulated lesion in the upper half of the breast (arrows), with a diffuse hazy extension toward the nipple. Note the round to oval structures in the retroareolar region (arrowheads).
|
|
Ultrasonography (US) showed an extensive, ill-defined, hypoechoic region in the upper outer quadrant of the left breast with some acoustic shadowing (Fig 2a). The lesion extended into the overlying subcutaneous fatty tissue, where obvious acoustic shadowing was present (Fig 2b). In the vicinity of the lesion, there were round hypoechoic regions that showed posterior acoustic enhancement, indicating a more cystic nature (Fig 2c). These findings were all suggestive of an invasive, growing malignancy with satellite lesions. No lymphadenopathy was visualized in the left axilla.
View larger version (135K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. (a) US image of the left breast shows an extensive, diffuse, and heterogeneous region with a more oval lesion in the subcutaneous fat. (b) US image reveals extension of the more solid part of the lesion into the subcutaneous fat with obvious acoustic shadowing. (c) US image demonstrates one of the oval lesions in the retroareolar region and its acoustic enhancement, a finding that indicates a more cystic nature.
|
|
View larger version (131K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. (a) US image of the left breast shows an extensive, diffuse, and heterogeneous region with a more oval lesion in the subcutaneous fat. (b) US image reveals extension of the more solid part of the lesion into the subcutaneous fat with obvious acoustic shadowing. (c) US image demonstrates one of the oval lesions in the retroareolar region and its acoustic enhancement, a finding that indicates a more cystic nature.
|
|
View larger version (192K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2c. (a) US image of the left breast shows an extensive, diffuse, and heterogeneous region with a more oval lesion in the subcutaneous fat. (b) US image reveals extension of the more solid part of the lesion into the subcutaneous fat with obvious acoustic shadowing. (c) US image demonstrates one of the oval lesions in the retroareolar region and its acoustic enhancement, a finding that indicates a more cystic nature.
|
|
|
Pathologic Evaluation
|
|---|
Three 16-gauge, US-guided core-needle biopsy specimens were taken from the region of suspected infiltration. Pathologic examination of these samples showed fatty necrosis.
Because there was a high suspicion of an invasively growing malignancy at physical examination and on radiologic images, a few days later stereotactic equipment was used to take five 14-gauge core-needle biopsy samples from the largest mass. A malignancy was suspected at pathologic examination of the second series of biopsy specimens. Because the samples were fragmented, further specification was not provided by the pathologist. US images of the liver and a chest radiograph showed no evidence of metastatic disease.
Because the patient had a history of malignancy and because the radiologic and pathologic findings were highly suggestive of malignancy, it was decided that mastectomy of the left breast should be performed, although an exact diagnosis had not yet been reached. The patient agreed to this procedure.
Mastectomy of the left breast was performed. The fascia of the major pectoral muscle and part of the anterior serratus muscle were resected because of adhesions that formed after the first operation 18 years ago.
Radiography of the gross specimen, sectioned in thick slices, showed that the lesion was completely resected (Fig 3a). No axillary fat was present. In two slices, two lesions with a 2-cm diameter were noted. These lesions displayed hemorrhage (Fig 3b), which was represented by the oval lesions seen at both preoperative mammography and specimen radiography. It is likely that these regions of hemorrhage were represented by the more cystic-appearing lesions seen at US, based on their shape and location.
View larger version (106K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Specimen radiograph (a) shows the extensive lesion (arrows), with the round to oval structures in the retroareolar region (arrowheads). These findings correlate with the regions of hemorrhage seen in the gross specimen (arrowheads) in b.
|
|
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Specimen radiograph (a) shows the extensive lesion (arrows), with the round to oval structures in the retroareolar region (arrowheads). These findings correlate with the regions of hemorrhage seen in the gross specimen (arrowheads) in b.
|
|
Specimen radiography also revealed a more solid-appearing region of breast tissue next to the lumpectomy scar that was excised and examined. This solid region represented the more focal lesion seen on mammograms and the ill-defined lesion seen on US images.
Microscopic examination revealed tumorous tissue lining the regions of hemorrhage and necrosis, which explains the diffuse and heterogeneous appearance of the lesion at US. The tumor cells had large polymorphic nuclei and were partly spindlelike. In one specimen, there was a small region in which the tumor cells seemed to form the borders of the lumina of the blood vessels (Fig 4a). The blood vessels themselves had an irregular border and a crevicelike appearance. In some regions, they appeared as a large lumen filled with blood cells. Exact dimensions were difficult to give, although the tumorous region was at least several centimeters in diameter. In some ducts, intraductal epithelial proliferation was seen, a characteristic that suggested a diagnosis of ductal carcinoma in situ grade 1. Additional immunohistochemical staining showed a strong positive reaction of the defective endothelium to CD31 (Fig 4b). The cut surface was free of tumor cells. Based on these findings, the pathologic diagnosis was angiosarcoma.
View larger version (178K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. (a) Photomicrograph (original magnification, x200; hematoxylin-eosin stain) shows clusters of tumor cells with polymorphic nuclei (arrows) and crevicelike lumina filled with blood cells and lined by the defective cells (a smaller one is outlined by arrowheads). (b) Photomicrograph (original magnification, x200; CD31) shows a strong positive reaction to CD31 (brownish cells), with obvious polymorphic nuclei (arrows).
|
|
View larger version (156K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. (a) Photomicrograph (original magnification, x200; hematoxylin-eosin stain) shows clusters of tumor cells with polymorphic nuclei (arrows) and crevicelike lumina filled with blood cells and lined by the defective cells (a smaller one is outlined by arrowheads). (b) Photomicrograph (original magnification, x200; CD31) shows a strong positive reaction to CD31 (brownish cells), with obvious polymorphic nuclei (arrows).
|
|
|
Discussion
|
|---|
Angiosarcoma is a malignancy of endovascular origin. A primary and a secondary type are known. Primary angiosarcoma can arise anywhere in the body. When it occurs in the breast, it affects women in their 3rd and 4th decade and accounts for one in 1700–2300 cases of primary breast cancer (1,2). There are no known risk factors for developing primary angiosarcoma. Because of high rates of local recurrence and early development of metastases after treatment, the prognosis is dismal. Angiosarcoma of the breast tends to metastasize hematogenously, rather than lymphogenously (3). Metastases to the lungs, skin, subcutaneous tissue, bone, liver, spleen, omentum, adrenal gland, psoas muscles, brain, and ovaries have been described. Primary angiosarcoma can occur bilaterally and is often associated with pregnancy. Bilateral lesions can also be due to metastatic spread (4).
Secondary angiosarcoma is distinguished from primary angiosarcoma in that it is apparently associated with a certain number of risk factors. A correlation has been shown between breast-conserving therapy and the development of secondary angiosarcoma, especially after breast-conserving therapy accompanied by radiation therapy. Cahan et al (5) listed criteria for radiation-associated neoplasms, which are also often applied to angiosarcoma (6,7): (a) the site of origin must be within the field of previous irradiation, (b) patients should have received a significant amount of radiation therapy (between 25 and 80 Gy), (c) an interval of at least 3–5 years must elapse between the time of irradiation and the development of the sarcoma, and (d) the secondary sarcoma should be histologically different from the primary neoplasm. These criteria were later modified to include tissues adjacent to the irradiated field and a shorter latency period of 3–4 years (8). Exposure to radiation is not the only cause of secondary angiosarcoma, however, as is illustrated in the case of Stewart-Treves syndrome. Stewart-Treves syndrome is characterized by angiosarcoma that arises in regions of chronic lymphedema, especially in the breast after breast-conserving therapy, regardless of radiation exposure (9,10).
The exact etiology is still not completely understood, but the presence of lymphedema (11,12), radiogenic DNA mutations (12), and a genetic predisposition (2) are suggested as causative factors for developing secondary angiosarcoma, as well as congenital, idiopathic, and traumatic factors (1). Likely, these factors have a synergistic effect (11). Monroe et al (1) postulated, however, that the distinction between primary and secondary angiosarcoma may not be important because both the clinical behavior and outcome of primary and secondary angiosarcoma are very similar.
The mean age at diagnosis of secondary angiosarcoma is about 65 years (13). Estimated prevalence of angiosarcoma after breast-conserving therapy ranges from 0.9 to 1.59 cases per 1000 (13) with an estimated relative risk of 15.9 of developing angiosarcoma after breast-conserving therapy and radiation therapy, compared with breast-conserving therapy alone (11). Our case can be categorized as secondary angiosarcoma.
The clinical manifestations of secondary angiosarcoma vary, which could explain why there is often a delay in the diagnosis of this disease (1). Secondary angiosarcoma can manifest as either a palpable mass or as just an alteration in skin color (1,6,7). The size and location of palpable masses vary from small cutaneous nodules (the cutaneous subtype) (13) to large lumps that constitute the entire breast (the intraparenchymal subtype). The breast can be firm on palpation, and sometimes dimpling is present. Unlike our case, the majority of cases in the literature exhibit skin changes. Changes in color range from hyperpigmentation to blue, red, and purple discoloration. Often, these changes are accompanied by an ecchymotic, telangiectatic, erythematous, or eczematous component. The skin can show different morphologies such as macules, papules, and vesicles. Advanced disease is typically marked by edema and ulcerative lesions in the breast (7). Angiosarcoma usually manifests with a mixture of the above-mentioned features, either in a single region or multifocally.
Radiologic findings are often nonspecific, and angiosarcoma may easily be overlooked. Mammograms may appear completely normal in 33% of cases of primary angiosarcoma (14), but this has also been reported in cases of secondary angiosarcoma. Sometimes only skin thickening is noted (13). Visible masses may be ill-defined or circumscribed, and may be round to oval or more lobulated. Associated coarse calcifications may be present (15). Luini et al (16) reported mammographic findings in almost all patients in their 16-patient series with mixed primary and secondary angiosarcoma. However, they correctly remarked that most authors report normal mammographic findings in cases of primary angiosarcoma. US tends to be nonspecific and only seems to have equivocal additional value if a mass is visible at mammography. If a lesion is detected with US, it may appear as a hypoechoic, a hyperechoic, or a heterogeneous region, with or without acoustic shadowing (3,14) in the case of primary angiosarcoma. These characteristics also seem applicable to secondary angiosarcoma, as in our case.
Although we did not perform magnetic resonance (MR) imaging in our case, a few characteristics should be mentioned, as MR imaging is being used more and more as a promising imaging modality for breast diseases. Kikawa et al (3) presented a case of primary angiosarcoma that exhibited low signal intensity on T1-weighted images and markedly high signal intensity on T2-weighted images. On dynamic MR images the tumor demonstrated prolonged enhancement (a characteristic not seen with ductal carcinoma) that was suspected to be due to the blood-filled vascular spaces and channels seen in histologic images. These findings were similar to those described earlier by Liberman et al (14) and by Marchant et al (4). Sanders et al (17) presented two cases of secondary cutaneous angiosarcoma, of which one showed a rapid initial enhancement followed by a steady state, and the other showed rapid wash-in followed by a rapid decrease in enhancement. Hence, they suggested that detection of a small focus of enhancement in irradiated thickened skin on dynamic breast MR images is more helpful than enhancement kinetics are.
Results from fine-needle aspiration and punch biopsies are generally not diagnostic and can be misleading because they may resemble those of other entities, such as radiation-related morphea or metastatic breast carcinoma (6,7,13,18). Findings from full-thickness incisional biopsies or excisional biopsy specimens should be conclusive, and these procedures are indicated if angiosarcoma is suspected (4,7,13,18).
Angiosarcoma is divided into low and high histologic grades, and ranges from grade 1 to grade 3. The cells of low-grade angiosarcoma resemble endothelium both morphologically and functionally. The cells are typically spindle-shaped and have single, large, oval nuclei with vesicular chromatin. They form irregular, anastomosing vascular channels and sometimes seem to connect to dermal capillaries. High-grade angiosarcoma—the majority of diagnosed angiosarcomas—is characterized by larger cells, pleomorphic nuclei, prominent nucleoli, and a high degree of mitotic activity. Hemorrhage into the surrounding stroma, known as "blood lakes," is also common and could account for the sometimes heterogeneous appearance of angiosarcomas on US images and the presence of acoustic shadowing (1,7,14,19,20). At the margins of a high-grade angiosarcoma, a low-grade appearance can be found (16), which should be taken into consideration when an incisional biopsy is performed.
Immunohistochemical staining does not usually lead to the diagnosis of angiosarcoma. It does, however, allow tumors of vascular origin to be differentiated from other types of neoplasms. Vascular-associated immunohistochemical markers used in the literature are factor VIII-related antigen, CD34, ulex-lectin antigen, desmin, and vimentin (1,7,19). CD31 seems to be the most sensitive and specific for endothelial differentiation (1).
The prognosis of angiosarcoma is thought to depend on the histologic grade. Patients with a higher-grade lesion are more prone to develop a recurrence (14) and have a lower survival rate than are those with a lower-grade lesion (1). Tumor size is also thought to be linked to the patient’s prognosis. Furthermore, low-grade angiosarcoma may become high-grade angiosarcoma (1). The reported median survival period ranges from 14.5 to 37 months, with a 5-year survival rate of 15% (1,13). To date, 24 months after mastectomy, our patient shows no signs of recurrence or of metastases.
Almost all authors advocate aggressive surgical resection as the treatment of choice for angiosarcoma. Although irradiation may seem to be a contradiction for a possibly radiation-induced tumor, surgery may be accompanied by hyperfractioned radiation therapy in patients with high-grade sarcoma in order to prevent recurrence (1). The effect of systemic therapy has not yet been established.
|
Acknowledgments
|
|---|
The authors thank Stephanie Meens-Koreman for drawing attention to this particular case and for being such a helpful colleague, Ineke Schreutelkamp for being an excellent mammography instructor, and Marius Nap for providing insight into the histologic properties of breast disease.
|
Footnotes
|
|---|
Editor’s Note.—Everyone who has taken the course in radiologic pathology at the Armed Forces Institute of Pathology (AFIP) remembers bringing beautifully illustrated cases for accession to the Institute. In recent years, the staff of the Department of Radiologic Pathology has judged the "best cases" by organ system, and recognition is given to the winners on the last day of the class. With each issue of RadioGraphics, one or more of these cases are published, written by the winning resident. Radiologic-pathologic correlation is emphasized, and the causes of the imaging signs of various diseases are illustrated.
|
References
|
|---|
- Monroe AT, Feigenberg SJ, Price Mendenhall N. Angiosarcoma after breast-conserving therapy. Cancer 2003;97(8):1832–1840.[CrossRef][Medline]
- Shet T, Malaviya A, Nadkarni M, et al. Primary angiosarcoma of the breast: observations in Asian Indian women. J Surg Oncol 2006;94(5):368–374.[CrossRef][Medline]
- Kikawa Y, Konishi Y, Nakamoto Y, et al. Angiosarcoma of the breast: specific findings of MRI. Breast Cancer 2006;13(4):369–373.[CrossRef][Medline]
- Marchant LK, Orel SG, Perez-Jaffe LA, Reynolds C, Schnall MD. Bilateral angiosarcoma of the breast on MR imaging. AJR Am J Roentgenol 1997;169(4):1009–1010.[Free Full Text]
- Cahan WG, Woodard HW, Higinbotham NL, Stewart FW, Coley BL. Sarcoma arising in irradiated bone: report of 11 cases. Cancer 1948;1:3–29.[CrossRef]
- Cafiero F, Gipponi M, Peressini A, et al. Radiation-associated angiosarcoma: diagnostic and therapeutic implications—two case reports and a review of the literature. Cancer 1996;77(12): 2496–2502.[CrossRef][Medline]
- Rao J, Dekoven JG, Beatty JD, Jones G. Cutaneous angiosarcoma as a delayed complication of radiation therapy for carcinoma of the breast. J Am Acad Dermatol 2003;49(3):532–538.[CrossRef][Medline]
- Tahir M, Hendry P, Baird L, Qureshi N, Ritchie D, Whitford P. Radiation induced angiosarcoma a sequela of radiotherapy for breast cancer following conservative surgery. Int Semin Surg Oncol 2006; 3:26.[CrossRef][Medline]
- Stewart FW, Treves N. Lymphangiosarcoma in postmastectomy lymphedema: a report of six cases in elephantiasis chirurgica. Cancer 1948;1:64–81.[CrossRef]
- Stewart NJ, Pritchard DJ, Nascimento AG, Kang YK. Lymphangiosarcoma following mastectomy. Clin Orthop Relat Res 1995;320:135–141.[Medline]
- Huang J, Mackillop WJ. Increased risk of soft tissue sarcoma after radiotherapy in women with breast carcinoma. Cancer 2001;92(1):172–180.[CrossRef][Medline]
- Cozen W, Bernstein L, Wang F, Press MF, Mack TM. The risk of angiosarcoma following primary breast cancer. Br J Cancer 1999;81(3):532–536.[CrossRef][Medline]
- Strobbe LJ, Peterse HL, van Tinteren H, Wijnmaalen A, Rutgers EJ. Angiosarcoma of the breast after conservation therapy for invasive cancer, the incidence and outcome: an unforeseen sequela. Breast Cancer Res Treat 1998;47(2): 101–109.[CrossRef][Medline]
- Liberman L, Dershaw DD, Kaufman RJ, Rosen PP. Angiosarcoma of the breast. Radiology 1992; 183(3):649–654.[Abstract/Free Full Text]
- Berg WA, Birdwell RL, Gombos EC, et al. Diagnostic imaging breast. Amirsys, 2006.
- Luini A, Gatti G, Diaz J, et al. Angiosarcoma of the breast: the experience of the European Institute of Oncology and a review of the literature. Breast Cancer Res Treat 2006 Nov 18; [Epub ahead of print]
- Sanders LM, Groves AC, Schaefer S. Cutaneous angiosarcoma of the breast on MRI. AJR Am J Roentgenol 2006;187(2):W143–W146.[Free Full Text]
- Lamblin G, Oteifa M, Zinzindohoue C, Isaac S, Termine L, Bobin JY. Angiosarcoma after conservative treatment and radiation therapy for adenocarcinoma of the breast. Eur J Surg Oncol 2001; 27(2):146–151.[CrossRef][Medline]
- Stokkel MP, Peterse HL. Angiosarcoma of the breast after lumpectomy and radiation therapy for adenocarcinoma. Cancer 1992;69(12):2965–2968.[CrossRef][Medline]
- Antman KH, Corson J, Greenberger J, Wilson R. Multimodality therapy in the management of angiosarcoma of the breast. Cancer 1982;50(10): 2000–2003.[CrossRef][Medline]
Top
History
Imaging Findings
