Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

DOI: 10.1148/radiol.2423060163
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yang, W. T.
Right arrow Articles by Krishnamurthy, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yang, W. T.
Right arrow Articles by Krishnamurthy, S.
(Radiology 2007;242:725-734.)
© RSNA, 2007

Breast Imaging

Mammary Angiosarcomas: Imaging Findings in 24 Patients1

Wei Tse Yang, MD, Bryan T. J. Hennessy, MD, Mark J. Dryden, MD, Vicente Valero, MD, Kelly K. Hunt, MD and Savitri Krishnamurthy, MD

1 From the Departments of Diagnostic Radiology (W.T.Y., M.J.D.), Medical Oncology (B.T.J.H.), Breast Medical Oncology (V.V.), Surgical Oncology (K.K.H.), and Pathology (S.K.), University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030. Received January 31, 2006; revision requested March 28; revision received April 12; accepted May 17; final version accepted July 5. Address correspondence to W.T.Y. (e-mail: wyang{at}di.mdacc.tmc.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Purpose: To retrospectively evaluate the clinical, imaging, and pathologic findings of mammary angiosarcomas in 24 patients.

Materials and Methods: The institutional review board approved this HIPAA-compliant study and waived informed consent. Twenty-four patients with records in the surgical pathology database who had a diagnosis of mammary angiosarcoma (n = 26) and who underwent preoperative imaging with mammography, ultrasonography (US), or magnetic resonance (MR) imaging were included. Mean and median ages at time of diagnosis were 40 and 38 years, respectively (range, 15–77 years). Images were reviewed in consensus by two radiologists with the American College of Radiology Breast Imaging Reporting and Data System lexicon and were compared with pathologic findings. Total length of follow-up (in months) was determined by the interval from the time of diagnosis to the time of last follow-up. Information on overall and disease-free survival was also obtained.

Results: Mean tumor size at time of diagnosis was 5.9 cm (range, 1–12 cm). Nineteen tumors manifested as a palpable mass, four manifested with progressive breast swelling, and three were asymptomatic. Mammograms of 16 tumors showed a noncalcified mass in eight, focal asymmetry in five, and no abnormality in three. All three mammographically occult tumors were visible at US and MR imaging. US images of 21 tumors showed a solid, frequently oval-shaped, and hyperechoic mass in 13 tumors and random mixed hyper- and hypoechogenicity with associated architectural distortion in eight tumors. Dynamic contrast material–enhanced MR imaging of nine tumors showed large, lobular, and intensely and heterogeneously enhancing masses with rapid enhancement and the washout characteristics of a malignant lesion.

Conclusion: A mass that shows homogeneous or heterogeneous hyperechogenicity at US (with associated architectural distortion) and has a hypervascular, hemorrhagic, and heterogeneous appearance and typical malignant enhancement characteristics at MR imaging should alert the radiologist to a possible diagnosis of angiosarcoma.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Angiosarcoma of the breast is rare, accounting for 0.04% of malignant breast neoplasms and 8% of breast sarcomas (1–3). The annual incidence of mammary angiosarcoma is 5.8 per 10 million women (4). Nevertheless, the breast is one of the most common anatomic sites where patients develop angiosarcomas. A comprehensive review of the imaging findings of mammary angiosarcomas was published in 1992 (5). To date, the largest published series on mammary angiosarcomas involved the review of original ultrasonographic (US) images in three cases (5). To the best of our knowledge, there are four single case reports in the literature in which the magnetic resonance (MR) imaging findings of mammary angiosarcoma are described (5–8). One of these reports included a case of bilateral tumors (6), and another described the findings in a patient who underwent an MR imaging examination of the chest that included the breast lesion (5). The purpose of our study was to retrospectively evaluate the clinical, imaging, and pathologic findings of mammary angiosarcomas in 24 patients.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Patients
Patients with records in the surgical pathology database of M. D. Anderson Cancer Center who received a diagnosis of angiosarcoma of the breast from June 1993 to December 2005 and who underwent preoperative imaging with mammography, US, or MR imaging were the subjects of this study. Of 44 patients referred to this institution with a diagnosis of mammary angiosarcoma, 24 had images available for review. A total of 26 mammary angiosarcomas were diagnosed in the 24 patients. Mean and median patient ages at the time of diagnosis were 40 and 38 years, respectively (range, 15–77 years). A waiver of informed consent was obtained, and the institutional review board approved this Health Insurance Portability and Accountability Act–compliant study (9). Information on clinical follow-up was collected by one senior medical oncology fellow (B.T.J.H.) through a review of clinical charts. The total length of follow-up (in months) was determined by the interval from the time of diagnosis to the time of last follow-up. Information on overall and disease-free survival was also obtained.

Imaging and Image Interpretation
Two breast radiologists (M.J.D., W.T.Y., with 5 and 9 years of experience in breast imaging, respectively) reviewed all available mammograms and US and MR images. These reviews were performed independently of the findings obtained with other imaging modalities. Agreement on the presence (or absence) of findings was by consensus.

Mammography.—Mammography was performed by using a Lorad M3 mammography unit (Hologic, Boston, Mass) or a DMR series unit (GE Medical Systems, Milwaukee, Wis). Standard three-view diagnostic mammography was performed, with additional views as deemed necessary. Breast parenchymal density was classified according to the American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) (10). Mammograms were reviewed for focal masses, calcifications, asymmetric density, architectural distortion, and associated features such as skin thickening and retraction, nipple retraction, and axillary lymphadenopathy and were also classified according to the BI-RADS lexicon (10). The final BI-RADS assessment category for mammograms was assigned retrospectively and independently of the results of other imaging modalities.

US.—Real-time gray-scale and color Doppler US was performed with a Siemens Elegra unit (Siemens Medical Solutions, Issaquah, Wash) and a 10–5-MHz linear array transducer or an ATL Ultramark 9 unit (Philips Medical Systems, Bothell, Wash) and a 10–5-MHz linear array transducer by one of 10 attending radiologists (including M.J.D. and W.T.Y.) with 1–10 years of experience in breast US. US images were assessed for masses (solid or cystic), and their shapes, margins, echo patterns, posterior acoustic features, and calcifications were classified according to the BI-RADS US lexicon (11). The presence or absence of tumoral vascularity was determined with color Doppler imaging and was described as within or at the periphery of the mass. Disease was also assessed as unifocal, multifocal, or multicentric in all patients who underwent US. Results of US assessment of the regional lymph node basins, including the axillary, infraclavicular, internal mammary, and supraclavicular regions, were documented according to previously published criteria (12). Whole-breast and nodal-basin US were routinely performed in our institution, because the majority of patients present for staging of known malignancy. The contralateral breast was evaluated if there was an important clinical or mammographic finding. Final BI-RADS assessment category for US images was assigned retrospectively and independently of the results of other imaging modalities.

MR imaging.—MR imaging was performed with a 1.5-T whole-body imaging system (Signa EXCITE; GE Healthcare, Milwaukee, Wis). A four-channel phased-array receive-only open breast array coil (In Vivo, Orlando, Fla) was used. Each patient was examined in the prone position with the breast suspended in the breast coil. MR imaging was performed in the transverse and sagittal planes with fat suppression. Precontrast transverse acquisitions were performed by using a T1-weighted fast spin-echo sequence (repetition time msec/echo time msec, 500/12; section thickness, 4 mm with no intersection gap; field of view, 160–220 mm; matrix size, 256 x 256; two signals acquired; imaging time, 3 minutes 20 seconds). Precontrast sagittal T2-weighted imaging was performed with the following parameters: 5000/85; section thickness, 4 mm with no intersection gap; field of view, 160–220 mm; matrix size, 256 x 256; four signals acquired; imaging time, 4 minutes 25 seconds.

Before and then after the patient was given a bolus intravenous injection of gadopentate dimeglumine (Magnevist; Schering, Berlin, Germany) (0.2 mmol per kilogram of body weight) at a rate of 3 mL/sec with a power injector (Medrad, Pittsburgh, Pa), sagittal pre- and then postcontrast dynamic imaging was performed by using a fast spoiled gradient-recalled-echo pulse sequence with a zoom mode (minimum echo time, 4.6 msec; section thickness, 2–3 mm with no intersection gap; field of view, 160–220 mm; matrix size, 256 x 256; one signal acquired; imaging time, 1 minute 30 seconds for each phase). Transverse postcontrast T1-weighted images were acquired by using the fast spoiled gradient-recalled-echo sequence in the same manner as it was used to acquire the precontrast images, without a change in the patient's position. Subtraction imaging was routinely performed.

The presence or absence of abnormal enhancement was classified according to the BI-RADS MR imaging lexicon (13). Abnormal enhancement was described as masslike or nonmasslike, and enhancement kinetic features, including initial and delayed phase patterns, were noted. Associated findings such as skin thickening, lymphadenopathy, and chest wall invasion were also noted. Final BI-RADS assessment category for MR imaging was assigned retrospectively and independently of the results of other imaging modalities, by using the most suspicious finding at either morphologic or kinetic evaluation.

Pathologic Examination
Information about histopathologic features was documented by one breast pathologist (S.K., with 7 years of experience), who reviewed the pathologic material. Hematoxylin-eosin–stained slides were reviewed in 20 tumors to confirm the morphologic diagnosis and histologic grading. The pathologic slides for six tumors were not available. Medical records indicated that a dedicated breast or sarcoma pathologist had previously confirmed the diagnosis of angiosarcoma in those six cases. Per the revised American Joint Committee on Cancer criteria (14), the staging system for sarcoma specifically states that it does not apply to angiosarcoma of the breast, and the breast cancer staging system does not include angiosarcoma of the breast (15). Histologic subtyping or grading, generally accepted as a guide to prognostic evaluation, was documented for each tumor (16). The tumor size at diagnosis was determined with US when available or with the next most accurate method in the following descending order: MR imaging, mammography, and clinical examination (17).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Clinical Data
Eighteen (75%) of 24 patients had primary mammary angiosarcoma of the breast at presentation. Two (11%) of these 18 patients developed recurrence 2 and 7 years after the time of initial diagnosis. Two (8%) of 24 patients had contralateral recurrence a year after the diagnosis of mammary angiosarcoma, and their initial primary tumors were not reviewed. Four (17%) of 24 patients had angiosarcoma after radiation therapy to the breast. Three of these four patients had had a primary breast carcinoma that was treated with breast-conserving surgery and radiation therapy 5, 6, and 6 years prior to diagnosis, and one patient had undergone radiation therapy for Hodgkin disease 12 years prior to the diagnosis of angiosarcoma.

Of the 26 tumors, 19 (73%) manifested as palpable masses and four (15%) manifested with progressive enlargement of the breast (Table 1). Three (12%) of 26 tumors were asymptomatic and were detected with routine mammography (n = 2) or routine CT chest surveillance for prior malignancy (n = 1). Four (21%) of 19 palpable masses were painful. Six (23%) of 26 tumors were associated with overlying skin discoloration. Thirteen tumors each were in the left and the right breast. The mean tumor size at diagnosis was 5.9 cm (range, 1–12 cm).


View this table:
[in this window]
[in a new window]

  		Table 1. Clinical and Pathologic Features of 26 Mammary Angiosarcomas in 24 Patients

 
Imaging
Mammography.—Mammograms were available for review for 16 tumors and were not obtained for 10 tumors (Table 2). Three tumors involved masses that were too large (>10 cm) and too painful for the patient to undergo mammography. A mass was visible in eight (50%) of 16 tumors (Fig 1), focal asymmetry was seen in five (31%) of 16 tumors (Fig 2), and no abnormality was noted in three (19%) of 16 tumors (Fig 3). All three mammographically occult tumors were visible at US, and two mammographically occult tumors were visible at MR imaging, while MR imaging was not performed for the third tumor. All masses were noncalcified. The predominant features of masses were oval shape (six of eight tumors, 75%), circumscribed margins (five of eight tumors, 62%) (Fig 1), and isodensity (five of eight tumors, 62%). One (12%) of eight masses demonstrated interspersed low fat-containing density (Fig 1). Associated overlying skin thickening was evident in two patients. The BI-RADS category was 0, 2, or 4 for the 16 tumors (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2. Findings and BI-RADS Category at Mammography in 16 Tumors

 

Figure 1A
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a: Images in 66-year-old woman with mammographically detected mass in right breast 6 years after breast-conserving and radiation therapy for right breast carcinoma. Pathologic examination showed 2.2-cm high-grade angiosarcoma. (a) Right craniocaudal mammogram shows oval, circumscribed fat-containing mass (arrow) in medial part of breast. Surgical metallic scar markers are seen in lateral breast. (b) Right transverse color Doppler US image obtained with 13–5-MHz linear array transducer shows oval, circumscribed, hypoechoic mass (arrow) with a hyperechoic center that does not demonstrate internal vascularity. At superficial analysis, this lesion suggests an intramammary lymph node, albeit in an unusual location. Correlation with mammographic results should also raise the possibility of a fat-containing mass or tumor.

 

Figure 1B
View larger version (159K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b: Images in 66-year-old woman with mammographically detected mass in right breast 6 years after breast-conserving and radiation therapy for right breast carcinoma. Pathologic examination showed 2.2-cm high-grade angiosarcoma. (a) Right craniocaudal mammogram shows oval, circumscribed fat-containing mass (arrow) in medial part of breast. Surgical metallic scar markers are seen in lateral breast. (b) Right transverse color Doppler US image obtained with 13–5-MHz linear array transducer shows oval, circumscribed, hypoechoic mass (arrow) with a hyperechoic center that does not demonstrate internal vascularity. At superficial analysis, this lesion suggests an intramammary lymph node, albeit in an unusual location. Correlation with mammographic results should also raise the possibility of a fat-containing mass or tumor.

 

Figure 2A
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a: Images in 36-year-old woman with painful, palpable mass in left breast. Pathologic examination showed 8-cm intermediate-grade angiosarcoma. (a) Left mediolateral oblique mammogram shows focal asymmetry (arrow) corresponding to palpable finding marked by an overlying Beekley marker. (b) Transverse T1-weighted spin-echo MR image (400/8) shows large, heterogeneously hypointense mass (thin arrows) with internal high signal intensity (thick arrow) that reflects hemorrhage. (c) Sagittal T2-weighted fast spin-echo MR image (500/8) shows heterogeneously hyperintense mass (arrow) that occupies much of superior breast. (d) Sagittal fast spoiled gradient-recalled-echo MR image (18.9/4.2) obtained after administration of gadolinium chelate shows intense heterogeneous enhancement of the mass (arrows).

 

Figure 2B
View larger version (142K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b: Images in 36-year-old woman with painful, palpable mass in left breast. Pathologic examination showed 8-cm intermediate-grade angiosarcoma. (a) Left mediolateral oblique mammogram shows focal asymmetry (arrow) corresponding to palpable finding marked by an overlying Beekley marker. (b) Transverse T1-weighted spin-echo MR image (400/8) shows large, heterogeneously hypointense mass (thin arrows) with internal high signal intensity (thick arrow) that reflects hemorrhage. (c) Sagittal T2-weighted fast spin-echo MR image (500/8) shows heterogeneously hyperintense mass (arrow) that occupies much of superior breast. (d) Sagittal fast spoiled gradient-recalled-echo MR image (18.9/4.2) obtained after administration of gadolinium chelate shows intense heterogeneous enhancement of the mass (arrows).

 

Figure 2C
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2c: Images in 36-year-old woman with painful, palpable mass in left breast. Pathologic examination showed 8-cm intermediate-grade angiosarcoma. (a) Left mediolateral oblique mammogram shows focal asymmetry (arrow) corresponding to palpable finding marked by an overlying Beekley marker. (b) Transverse T1-weighted spin-echo MR image (400/8) shows large, heterogeneously hypointense mass (thin arrows) with internal high signal intensity (thick arrow) that reflects hemorrhage. (c) Sagittal T2-weighted fast spin-echo MR image (500/8) shows heterogeneously hyperintense mass (arrow) that occupies much of superior breast. (d) Sagittal fast spoiled gradient-recalled-echo MR image (18.9/4.2) obtained after administration of gadolinium chelate shows intense heterogeneous enhancement of the mass (arrows).

 

Figure 2D
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2d: Images in 36-year-old woman with painful, palpable mass in left breast. Pathologic examination showed 8-cm intermediate-grade angiosarcoma. (a) Left mediolateral oblique mammogram shows focal asymmetry (arrow) corresponding to palpable finding marked by an overlying Beekley marker. (b) Transverse T1-weighted spin-echo MR image (400/8) shows large, heterogeneously hypointense mass (thin arrows) with internal high signal intensity (thick arrow) that reflects hemorrhage. (c) Sagittal T2-weighted fast spin-echo MR image (500/8) shows heterogeneously hyperintense mass (arrow) that occupies much of superior breast. (d) Sagittal fast spoiled gradient-recalled-echo MR image (18.9/4.2) obtained after administration of gadolinium chelate shows intense heterogeneous enhancement of the mass (arrows).

 

Figure 3A
View larger version (169K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a: Images in 48-year-old woman with palpable mass in left breast associated with overlying red skin discoloration. This patient had been treated with breast-conserving surgery and radiation therapy for ductal carcinoma in left breast 5 years earlier. Pathologic examination showed 12-cm intermediate-grade angiosarcoma. (a) Transverse US image obtained with 10–5-MHz linear array transducer shows abnormal mixed hyper- and hypoechoic areas (arrows), with alteration of normal tissue planes and overlying skin thickening (*). (b) Photograph of left mastectomy specimen. Note reddish and purplish nodules and skin discoloration (arrows) in grossly enlarged breast. (c) Photograph of gross specimen shows ill-defined hemorrhagic and friable tumor in center of breast. Note diffuse areas of red discoloration (arrowheads) indicating diffuse spread of the tumor beyond the main mass (arrows).

 

Figure 3B
View larger version (142K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b: Images in 48-year-old woman with palpable mass in left breast associated with overlying red skin discoloration. This patient had been treated with breast-conserving surgery and radiation therapy for ductal carcinoma in left breast 5 years earlier. Pathologic examination showed 12-cm intermediate-grade angiosarcoma. (a) Transverse US image obtained with 10–5-MHz linear array transducer shows abnormal mixed hyper- and hypoechoic areas (arrows), with alteration of normal tissue planes and overlying skin thickening (*). (b) Photograph of left mastectomy specimen. Note reddish and purplish nodules and skin discoloration (arrows) in grossly enlarged breast. (c) Photograph of gross specimen shows ill-defined hemorrhagic and friable tumor in center of breast. Note diffuse areas of red discoloration (arrowheads) indicating diffuse spread of the tumor beyond the main mass (arrows).

 

Figure 3C
View larger version (122K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3c: Images in 48-year-old woman with palpable mass in left breast associated with overlying red skin discoloration. This patient had been treated with breast-conserving surgery and radiation therapy for ductal carcinoma in left breast 5 years earlier. Pathologic examination showed 12-cm intermediate-grade angiosarcoma. (a) Transverse US image obtained with 10–5-MHz linear array transducer shows abnormal mixed hyper- and hypoechoic areas (arrows), with alteration of normal tissue planes and overlying skin thickening (*). (b) Photograph of left mastectomy specimen. Note reddish and purplish nodules and skin discoloration (arrows) in grossly enlarged breast. (c) Photograph of gross specimen shows ill-defined hemorrhagic and friable tumor in center of breast. Note diffuse areas of red discoloration (arrowheads) indicating diffuse spread of the tumor beyond the main mass (arrows).

 
US.—US was performed for 21 tumors and showed an abnormality in all of them (Table 3). A focal mass was identified in 13 (62%) of 21 tumors (Figs 1, 2), and diffuse abnormal mixed hyper- and hypoechogenicity without a discrete mass was noted in eight (38%) of 21 tumors (Figs 3, 4). All masses were solid. The predominant features of masses were oval shape (10 of 13 tumors, 77%) and circumscribed margins (eight of 13 tumors, 62%). The remaining masses were lobular shaped (three of 13 tumors, 23%) and showed indistinct (three of 13 tumors, 23%) or microlobulated (two of 13 tumors, 15%) margins. Six (46%) of 13 masses were hypoechoic, and the rest were hyperechoic (three of 13 tumors, 23%) or mixed hyperechoic and hypoechoic (four of 13 tumors, 31%) (Figs 1, 2). Posterior acoustic phenomena were typically absent in masses (10 of 13 masses, 77%); posterior enhancement was seen in three of 13 masses (23%). No posterior acoustic shadowing was noted. The tumors that demonstrated abnormal heterogeneous echoes without a discrete mass were generally larger, with a mean size of 6 cm (range, 7–12 cm). Color Doppler imaging was performed in 10 tumors and revealed hypervascularity in all tumors (Fig 4). Associated skin or subcutaneous edema was noted in three (14%) of 21 tumors (Fig 4). Abnormal enlarged hypoechoic lymph nodes involving the axillary, infraclavicular, internal mammary, and supraclavicular nodal basins were noted in one patient. Metastatic disease was confirmed at axillary lymph node dissection in this patient. The BI-RADS category assigned to the 21 tumors was 4 or 5 (Table 3).


View this table:
[in this window]
[in a new window]

  		Table 3. Findings and BI-RADS Category at US in 21 Tumors

 

Figure 4A
View larger version (55K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a: Images in 29-year-old woman with large palpable mass in left breast. Disseminated metastasis was present at time of diagnosis. Pathologic examination showed 12-cm high-grade angiosarcoma. (a) Left extended-field-of-view US image obtained with 13–5-MHz linear array transducer shows heterogeneous echoes (arrow) replacing nearly the entire breast and marked overlying subcutaneous edema (arrowheads). (b) Color Doppler US image obtained with 13–5-MHz linear array transducer shows hypervascularity. (c) Transverse T1-weighted spin-echo MR image (400/8) shows large, heterogeneously hypointense mass (arrowheads). (d) Sagittal T2-weighted fast spin-echo MR image (500/8) shows heterogeneously hyperintense mass (arrowhead) that occupies the entire breast and markedly hyperintense central venous lakes (arrows). (e) Sagittal dynamic MR images obtained after gadolinium chelate administration (top left: first image; top right: second image; bottom left: third image; bottom right: fourth image) show rapid, patchy early enhancement of the mass.

 

Figure 4B
View larger version (61K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b: Images in 29-year-old woman with large palpable mass in left breast. Disseminated metastasis was present at time of diagnosis. Pathologic examination showed 12-cm high-grade angiosarcoma. (a) Left extended-field-of-view US image obtained with 13–5-MHz linear array transducer shows heterogeneous echoes (arrow) replacing nearly the entire breast and marked overlying subcutaneous edema (arrowheads). (b) Color Doppler US image obtained with 13–5-MHz linear array transducer shows hypervascularity. (c) Transverse T1-weighted spin-echo MR image (400/8) shows large, heterogeneously hypointense mass (arrowheads). (d) Sagittal T2-weighted fast spin-echo MR image (500/8) shows heterogeneously hyperintense mass (arrowhead) that occupies the entire breast and markedly hyperintense central venous lakes (arrows). (e) Sagittal dynamic MR images obtained after gadolinium chelate administration (top left: first image; top right: second image; bottom left: third image; bottom right: fourth image) show rapid, patchy early enhancement of the mass.

 

Figure 4C
View larger version (101K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4c: Images in 29-year-old woman with large palpable mass in left breast. Disseminated metastasis was present at time of diagnosis. Pathologic examination showed 12-cm high-grade angiosarcoma. (a) Left extended-field-of-view US image obtained with 13–5-MHz linear array transducer shows heterogeneous echoes (arrow) replacing nearly the entire breast and marked overlying subcutaneous edema (arrowheads). (b) Color Doppler US image obtained with 13–5-MHz linear array transducer shows hypervascularity. (c) Transverse T1-weighted spin-echo MR image (400/8) shows large, heterogeneously hypointense mass (arrowheads). (d) Sagittal T2-weighted fast spin-echo MR image (500/8) shows heterogeneously hyperintense mass (arrowhead) that occupies the entire breast and markedly hyperintense central venous lakes (arrows). (e) Sagittal dynamic MR images obtained after gadolinium chelate administration (top left: first image; top right: second image; bottom left: third image; bottom right: fourth image) show rapid, patchy early enhancement of the mass.

 

Figure 4D
View larger version (103K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4d: Images in 29-year-old woman with large palpable mass in left breast. Disseminated metastasis was present at time of diagnosis. Pathologic examination showed 12-cm high-grade angiosarcoma. (a) Left extended-field-of-view US image obtained with 13–5-MHz linear array transducer shows heterogeneous echoes (arrow) replacing nearly the entire breast and marked overlying subcutaneous edema (arrowheads). (b) Color Doppler US image obtained with 13–5-MHz linear array transducer shows hypervascularity. (c) Transverse T1-weighted spin-echo MR image (400/8) shows large, heterogeneously hypointense mass (arrowheads). (d) Sagittal T2-weighted fast spin-echo MR image (500/8) shows heterogeneously hyperintense mass (arrowhead) that occupies the entire breast and markedly hyperintense central venous lakes (arrows). (e) Sagittal dynamic MR images obtained after gadolinium chelate administration (top left: first image; top right: second image; bottom left: third image; bottom right: fourth image) show rapid, patchy early enhancement of the mass.

 

Figure 4E
View larger version (155K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4e: Images in 29-year-old woman with large palpable mass in left breast. Disseminated metastasis was present at time of diagnosis. Pathologic examination showed 12-cm high-grade angiosarcoma. (a) Left extended-field-of-view US image obtained with 13–5-MHz linear array transducer shows heterogeneous echoes (arrow) replacing nearly the entire breast and marked overlying subcutaneous edema (arrowheads). (b) Color Doppler US image obtained with 13–5-MHz linear array transducer shows hypervascularity. (c) Transverse T1-weighted spin-echo MR image (400/8) shows large, heterogeneously hypointense mass (arrowheads). (d) Sagittal T2-weighted fast spin-echo MR image (500/8) shows heterogeneously hyperintense mass (arrowhead) that occupies the entire breast and markedly hyperintense central venous lakes (arrows). (e) Sagittal dynamic MR images obtained after gadolinium chelate administration (top left: first image; top right: second image; bottom left: third image; bottom right: fourth image) show rapid, patchy early enhancement of the mass.

 
MR imaging.—MR imaging was performed for nine tumors (Table 4). Preoperative images were available for six tumors (67%), and three of the nine tumors (33%) were imaged after excisional biopsies were performed at outside facilities. The preoperative MR images for these three patients were not available for review and were not included in the study analysis. The mean size of these nine tumors was 10 cm (range, 5–12 cm). All tumors consisted of large, lobular masses that were heterogeneously hypointense on T1-weighted images and heterogeneously hyperintense on T2-weighted images (Figs 2, 4). These masses contained irregular areas of high signal intensity on T1-weighted images (Fig 2) that reflected the hemorrhagic nature of the tumors and their markedly heterogeneous architecture. Cystic cavities representing venous lakes were common, and a large venous cavity was noted in one tumor (Fig 4). Some masses were partially irregular and showed diffuse infiltration adjacent to the dominant mass. Skin thickening was noted in one patient, and extension to the pectoralis fascia without direct involvement of the pectoralis muscle was seen in two tumors (Figs 2, 4). All tumors demonstrated intense and heterogeneous enhancement, with a typical malignant pattern on dynamic contrast material–enhanced images that was characterized by rapid initial enhancement followed by washout kinetic features (defined as an abrupt decline in signal intensity after the initial enhancement) in the delayed phase. In three of these nine tumors, postbiopsy MR imaging revealed residual disease and helped in the planning of surgical treatment. The BI-RADS category assigned to the nine tumors was 5 or 6 (Table 4).


View this table:
[in this window]
[in a new window]

  		Table 4. Findings and BI-RADS Category at MR Imaging in Nine Tumors

 
Treatment and Pathologic Findings
US-guided core biopsies were performed in five patients without complications and yielded satisfactory material for diagnosis in four. Grading of tumors in these four patients was deferred to final surgery. Histopathologic evaluation showed that 11 tumors were high-grade angiosarcomas; eight were intermediate-grade angiosarcomas; six were low-grade angiosarcomas; and one was a mixed low-, intermediate-, and high-grade angiosarcoma. The patient with the mixed-grade angiosarcoma was given a diagnosis on the basis of incisional biopsy results and did not undergo definitive surgery because disseminated metastasis was diagnosed at the time of presentation and she was treated palliatively.

For 23 of the 26 tumors, the patients underwent surgery: Mastectomy was performed in 20 patients (one patient also underwent axillary lymph node dissection), and wide local excision was performed in three patients. One patient underwent repeated arterial embolization, one patient had disseminated metastases at the time of diagnosis and was not considered a surgical candidate, and one patient refused therapy. Eleven patients received adjuvant chemotherapy, and five patients received radiation therapy.

Survival
As of December 30, 2005, eight women were dead (five had high-grade and three had intermediate-grade angiosarcomas), and 16 were alive. The mean and median follow-up duration was 37 and 36 months, respectively (range, 1–126 months). The mean and median survival for the eight women who died was 32 and 36 months, respectively (range, 14–36 months). All eight women died of angiosarcoma; three had disseminated metastasis at the time of death, two had bone metastasis, two had bone metastasis and local recurrence, and one died of liver metastasis with hemorrhage. Of the 16 women still alive, 12 were disease free, one had pulmonary metastasis that was in remission, one had had multiple local recurrences, one had pulmonary and pelvic metastasis, and one had disseminated metastasis involving the lungs, liver, pelvis (including adnexae), and bone.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 References
 
Mammary angiosarcoma has been described as rarely clinically occult, and a palpable mass was present in each of the 29 patients (100%) in the study by Liberman et al (5). In our study, however, 19 (73%) of 26 angiosarcomas were palpable, four manifested with breast enlargement (15%), and three were detected with routine surveillance in asymptomatic women (12%). This difference in clinical manifestation may reflect differences in the study patients. Four of the tumors in this study occurred in women who had received prior radiation therapy; two of these women were asymptomatic, and two presented with large masses. Skin discoloration was present in six (23%) of 26 tumors in our study, compared with 17% of the patients in the series of Liberman et al (5) and 34% of the patients in the review by Chen et al (18).

Unlike breast carcinoma, which most often occurs in women between the ages of 55 and 69 years (19), mammary angiosarcoma tends to occur in younger women. The mean and median age of patients in our study was 40 and 38 years, respectively; these data compare well with the previously reported mean age (42 years) of patients with angiosarcoma (5,18). This young age at diagnosis is, however, lower than the average age for carcinomas and falls below the recommended age for mammographic screening, possibly explaining why most patients were symptomatic.

Despite the large mean tumor size (5.9 cm) at diagnosis in our study, three (19%) of 16 angiosarcomas were not visible mammographically but were visible at US. This lack of visibility may be attributed to the vague, isodense, noncalcified nature of the masses; the absence of architectural distortion and calcifications; and the dense breast parenchyma characteristic of young women. Masses imaged with mammography in our study were typically oval, circumscribed, and noncalcified. This compares with a previous description of mammary angiosarcomas as noncalcified, ill-defined masses lacking spiculation, with 33% (seven of 21) being mammographically occult (5). A mammographic feature not previously described, to our knowledge, is focal asymmetry, which was noted in 31% (five of 16) of the tumors in this study. We postulate that this mammographic feature reflects the subtle infiltrating nature of this vascular tumor, which tends to be deeply embedded in the breast tissue. The single lesion that demonstrated low (fat-containing) density suggests the possibility of trapped fat. The presence of fat is uncommon but has been described in angiosarcomas (20,21) and raises the differential diagnosis of hemangioma and angiolipoma (8,20,21).

In our study, high-spatial-resolution gray-scale US showed that angiosarcomas were primarily solitary, circumscribed, hypervascular masses with variable echo patterns and without posterior acoustic phenomena. Angular margins and posterior acoustic shadowing, both characteristically associated with carcinoma, were not a feature. Fifty-four percent (seven of 13) of the masses in our series were hyperechoic or mixed hyper- and hypoechoic, which is unusual: Breast carcinomas are reportedly very rarely hyperechoic (22). This may reflect the vascular nature of angiosarcoma and the multiple interfaces of the vascular channels and may alert the radiologist to a possible diagnosis of angiosarcoma when this feature is encountered. A US feature not previously described, to our knowledge, is abnormal mixed hyper- and hypoechogenicity of the breast without a discrete mass. This finding was more frequently associated with large tumors and occurred in 37% (seven of 19) of tumors in our study. This may reflect the subtle infiltrating nature of mammary angiosarcoma, which tends to ramify within the breast tissue. This feature is therefore particularly important when imaging large tumors. Using a small-footprint transducer when imaging a patient who presents with a diffuse or vague clinical abnormality in the breast may result in a false-negative US study if the subtle finding of architectural distortion and diffuse abnormal echoes without a discrete mass is overlooked by the imager who is searching for a mass. Color Doppler US findings, which have not been extensively reported in the literature (8), showed hypervascularity in all eight tumors in this study.

MR imaging findings of mammary angiosarcomas in the literature are scant and have involved solitary cases (5–8). Very high signal intensity on heavily T2-weighted images in the periphery of the mass that likely represents the presence of vascular channels containing slow-flowing blood has been reported (5–8). MR imaging of mammary angiosarcomas in our study showed hemorrhagic, heterogeneous, and hypervascular masses. Similar findings have been noted for angiosarcomas in the liver, spleen, and retroperitoneum (23–25). All mammary angiosarcomas were large, lobular-shaped masses with indistinct borders, and some demonstrated infiltrative margins. Hemorrhagic areas and heterogeneous architecture were typical. All masses were hypervascular and enhanced intensely and heterogeneously, with rapid initial enhancement and washout kinetic features at dynamic imaging. Cystic cavities representing venous lakes were a feature, and a large venous cavity was noted in one lesion. In our study, we found that MR imaging was useful in defining disease extent to guide surgical planning and in evaluating residual disease after excisional biopsy.

Therapeutic radiation has been implicated in the pathogenesis of angiosarcoma (26–29), which is the most common subtype of all sarcomas arising after radiation therapy to the breast. Postradiation mammary angiosarcomas have become more frequently recognized with the increasing practice of breast-conserving surgery and irradiation for breast cancer therapy. Mammary angiosarcomas that occur after radiation therapy can arise from the overlying dermis (cutaneous angiosarcoma) (30–32) or from the parenchymal vascular endothelial cells (true mammary angiosarcoma) (33,34).

Despite ours being the largest series to date on the imaging (including MR imaging) findings of mammary angiosarcomas to our knowledge, the numbers reported are limited by being relatively small, necessitating a retrospective study design. This, however, reflects the rareness of an aggressive and frequently lethal disease. The small number of angiosarcomas in our series is secondary to our work in a large referral practice, which inherently introduces case-selection bias. It is possible that smaller, low-grade tumors were treated without referral; this fact can potentially introduce bias into the imaging and outcome results in our study. The relatively short median follow-up was another limitation that in part reflects the relatively short survival of patients with this disease. A further limitation of our study was that not every patient and every lesion underwent imaging with all three modalities. This, once again, reflects true clinical practice, in which many lesions in the breast are excised without appropriate imaging work-up or an attempt at percutaneous biopsy.

In conclusion, hyperechoic masses at US and hypervascular, hemorrhagic, and heterogeneous masses demonstrating malignant enhancement characteristics at MR imaging should alert the radiologist to a possible diagnosis of angiosarcoma.


   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
References
 


   FOOTNOTES
 

Abbreviations: BI-RADS = Breast Imaging Reporting and Data System

Authors stated no financial relationship to disclose.

Author contributions: Guarantor of integrity of entire study, W.T.Y.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, W.T.Y., B.T.J.H., S.K.; clinical studies, W.T.Y., B.T.J.H., M.J.D., K.K.H., S.K.; and manuscript editing, all authors


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 

  1. Agarwal PK, Mehrotra R. Haemangiosarcoma of the breast. Indian J Cancer 1977;14:182–185.[Medline]
  2. Myerowitz RL, Pietruszka M, Barnes EL. Primary angiosarcoma of the breast. JAMA 1978;239:403.[CrossRef][Medline]
  3. Stewart FW, Treves N. Lymphangiosarcoma in postmastectomy lymphedema: a report of six cases in elephantiasis chirurgica. Cancer 1948;1:64–81.[CrossRef]
  4. May DS, Stroup NE. The incidence of sarcomas of the breast among women in the United States, 1973–1986. Plast Reconstr Surg 1991;87:193–194.[Medline]
  5. Liberman L, Dershaw DD, Kaufman RJ, Rosen PP. Angiosarcoma of the breast. Radiology 1992;183:649–654.[Abstract/Free Full Text]
  6. Marchant LK, Orel SG, Perez-Jaffe LA, Reynolds C, Schnall MD. Bilateral angiosarcoma of the breast on MR imaging. AJR Am J Roentgenol 1997;169:1009–1010.[Free Full Text]
  7. Murakam S, Nagano H, Okubo K, et al. Angiosarcoma of the breast: report of a case and its findings of MRI. Breast Cancer 2001;8:254–258.[Medline]
  8. Glazebrook KN, Morton MJ, Reynolds C. Vascular tumors of the breast: mammographic, sonographic, and MRI appearances. AJR Am J Roentgenol 2005;184:331–338.[Free Full Text]
  9. Johnson MS, Gonzales MN, Bizila S. Responsible conduct of radiology research. V. The Health Insurance Portability and Accountability Act and research. Radiology 2005;237:757–764.
  10. American College of Radiology. ACR BI-RADS: mammography. In: ACR Breast Imaging Reporting and Data System, Breast Imaging Atlas. 4th ed. Reston, Va: American College of Radiology, 2003.
  11. American College of Radiology. ACR BI-RADS: ultrasound. In: ACR Breast Imaging Reporting and Data System, Breast Imaging Atlas. 4th ed. Reston, Va: American College of Radiology, 2003.
  12. Vlastos G, Fornage BD, Mirza NQ, et al. The correlation of axillary ultrasonography with histologic breast cancer downstaging after induction chemotherapy. Am J Surg 2000;179:446–452.[CrossRef][Medline]
  13. American College of Radiology. ACR BI-RADS: magnetic resonance imaging. In: ACR Breast Imaging Reporting and Data System, Breast Imaging Atlas. 4th ed. Reston, Va: American College of Radiology, 2003.
  14. Greene FL, Page DL, Fleming ID, et al. American Joint Cancer Committee cancer staging manual. 6th ed. New York, NY: Springer-Verlag, 2002; 221–240.
  15. Singletary SE, Allred C, Ashley P, et al. Staging system for breast cancer: revisions for the 6th edition of the AJCC Cancer Staging Manual. Surg Clin North Am 2003;83:803–819.[CrossRef][Medline]
  16. Rosen PP, Kimmel M, Ernsberger D. Mammary angiosarcoma: the prognostic significance of tumor differentiation. Cancer 1988;62:2145–2151.[CrossRef][Medline]
  17. Yang WT, Lam WW, Cheung H, Suen M, King W, Metreweli C. Sonographic, magnetic resonance imaging and mammographic assessments of preoperative size of breast cancer. J Ultrasound Med 1997;16:791–797.[Abstract]
  18. Chen KT, Kirkegaard DD, Bocian JJ. Angiosarcoma of the breast. Cancer 1980;46:368–371.[CrossRef][Medline]
  19. National Institutes of Health. SEER data 1984–1988. National Cancer Institute Statistics Review 1975–1988. NIH publication no. 91–2789. Bethesda, Md: National Institutes of Health, 1991.
  20. Rosen PP. Sarcoma. In: Rosen PP, ed. Rosen's breast pathology. 2nd ed. New York, NY: Lippincott Williams & Wilkins, 2001; 813–861.
  21. Tsuda N, Chowdhury PR, Hayashi T, et al. Primary renal angiosarcoma: a case report and review of the literature. Pathol Int 1997;47:778–783.[Medline]
  22. Stavros AT, Thickman D, Rapp CL, Dennis MA, Parker SH, Sisney GA. Solid breast nodules: use of sonography to distinguish between benign and malignant lesions. Radiology 1995;196:123–134.[Abstract/Free Full Text]
  23. Thompson WM, Levy AD, Aguilera NS, Gorospe L, Abbott RM. Angiosarcoma of the spleen: imaging characteristics in 12 patients. Radiology 2005;235:106–115.[Abstract/Free Full Text]
  24. Koyama T, Fletcher JG, Johnson CD, Kuo MS, Notohara K, Burgart LJ. Primary hepatic angiosarcoma: findings at CT and MR imaging. Radiology 2002;222:667–673.[Abstract/Free Full Text]
  25. Scialpi M, Galasso C, Di Maggio A, et al. Primary retroperitoneal angiosarcoma: MR imaging features. Eur Radiol 2001;11:791–795.[CrossRef][Medline]
  26. Yap J, Chuba PJ, Thomas R, et al. Sarcoma as a second malignancy after treatment for breast cancer. Int J Radiat Oncol Biol Phys 2002;52:1231–1237.[CrossRef][Medline]
  27. Lagrange JL, Ramaioli A, Chateau MC, et al. Sarcoma after radiation therapy: retrospective multiinstitutional study of 80 histologically confirmed cases. Radiation Therapist and Pathologist Groups of the Federation Nationale des Centres de Lutte Contre le Cancer. Radiology 2000;216:197–205.
  28. Monroe AT, Feigenberg SJ, Mendenhall NP. Angiosarcoma after breast-conserving therapy. Cancer 2003;97:1832–1840.[CrossRef][Medline]
  29. Cozen W, Bernstein L, Wang F, Press MF, Mack TM. The risk of angiosarcoma following primary breast cancer. Br J Cancer 1999;81:532–536.[CrossRef][Medline]
  30. Fant J, Grant M, May S, Talaasen L, Watkins D. Angiosarcoma of the breast: mammographic, clinical, and pathologic correlation. Breast J 2003;9:252–253.[CrossRef][Medline]
  31. Billings SD, McKenney JK, Folpe AL, Hardacre MC, Weiss SW. Cutaneous angiosarcoma following breast-conserving surgery and radiation. Am J Surg Pathol 2004;28:781–788.[CrossRef][Medline]
  32. Mills TD, Vinnicombe SJ, Wells CA, Carpenter R. Angiosarcoma of the breast after wide local excision and radiotherapy for breast carcinoma. Clin Radiol 2002;57:63–66. [Published correction appears in Clin Radiol 2002;57:541.][CrossRef][Medline]
  33. Mark RJ, Poen J, Luu MT, Yao SF, Selch MT, Parker RG. Postirradiation sarcomas: a single-institution study and review of the literature. Cancer 1994;73:2653–2662.[CrossRef][Medline]
  34. Kuten A, Sapir D, Cohen Y, Haim N, Borovik R, Robinson E. Post irradiation soft tissue sarcoma occurring in breast cancer patients: report of seven cases and results of combination chemotherapy. J Surg Oncol 1985;28:168–171.[Medline]



This article has been cited by other articles:


Home page
 RadioGraphicsHome page
E. M. Chung, R. Cube, G. J. Hall, C. Gonzalez, J. T. Stocker, and L. M. Glassman
From the Archives of the AFIP: Breast Masses in Children and Adolescents: Radiologic-Pathologic Correlation
RadioGraphics, May 1, 2009; 29(3): 907 - 931.
[Abstract] [Full Text] [PDF]

Home page
 J Ultrasound MedHome page
J. H. Youk, E.-K. Kim, G. I. Kim, and K. H. Ko
Unusual Sonographic Finding of Metastatic Invasive Lobular Carcinoma to the Contralateral Breast
J. Ultrasound Med., December 1, 2008; 27(12): 1771 - 1775.
[Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
K. N. Glazebrook, M. J. Magut, and C. Reynolds
Angiosarcoma of the Breast
Am. J. Roentgenol., February 1, 2008; 190(2): 533 - 538.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yang, W. T.
Right arrow Articles by Krishnamurthy, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yang, W. T.
Right arrow Articles by Krishnamurthy, S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE