Pediatric angiosarcoma of soft tissue: a rare clinicopathologic entity.
|Abstract:||Angiosarcomas are rare tumors that predominantly affect adults and elderly patients. Although angiosarcomas are well described in a variety of clinical settings, they have been poorly studied in children and little is known about their biology, natural history, or optimal treatment. Childhood angiosarcomas are exceedingly rare. The head and neck region and mediastinum seem to be the preferred locations. Most tumors are high-grade tumors. Vasoformative architecture is not always obvious on light microscopy requiring the benefit of immunohistochemical study. The differential diagnosis includes Kaposi sarcoma, epithelioid hemangioendothelioma, hemangiopericytoma, and spindle cell hemangioendothelioma whose prognosis is different. Complete resection is required for patients with localized tumors. Malignant vascular tumors are rare in children in the first 2 decades of life and when they do occur they seem to be more aggressive than in adults. Pathologic diagnosis is difficult particularly in poorly differentiated angiosarcomas requiring immunohistochemical study to confirm vascular differentiation.|
|Article Type:||Disease/Disorder overview|
Cancer in children
Cancer in children (Diagnosis)
Cancer in children (Care and treatment)
Cancer in children (Prognosis)
Sarcoma (Risk factors)
Sarcoma (Care and treatment)
|Publication:||Name: Archives of Pathology & Laboratory Medicine Publisher: College of American Pathologists Audience: Academic; Professional Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2010 College of American Pathologists ISSN: 1543-2165|
|Issue:||Date: March, 2010 Source Volume: 134 Source Issue: 3|
|Geographic:||Geographic Scope: Tunisia Geographic Code: 6TUNI Tunisia|
Angiosarcomas are rare tumors occurring predominately in adult
patients. They are extremely rare in children and adolescents. (1)
Although angiosarcomas are well described in adults, they have been
poorly studied in childhood. Pediatric malignant vascular tumors are
rare with less than 13 cases reported in the English literature. (1-3) A
review of 20 years of experience in the Italian and German Soft Tissue
Sarcoma Cooperative Group identified 12 children with angiosarcoma. (1)
The previous cases of angiosarcoma of childhood have been described in patients younger than 16 years. Age at diagnosis ranges from 3 months to 16 years. (3) Pediatric angiosarcomas arise in both sexes, but there is a male predilection (female to male ratio, 3:4). (1) They usually occur in deep soft tissue and liver and show a wide anatomic distribution with a predilection for axial soft tissue (trunk, pelvis, head and neck, mediastinum) and visceral organs (liver, heart). (1,3) The head and neck region and mediastinum seem to be the preferred locations. (2) Among the 12 cases studied by Ferrari et al, (1, 4) cases (30%) were located in the head and neck region and 3 cases (25%) in the pelvis. The classic presentation is an enlarging, painful mass of several weeks' duration and they are occasionally associated with acute hemorrhage, anemia, or a coagulopa thy. (4)
Radiologically, these lesions have a destructive pattern (Figure 1). Magnetic resonance imaging demonstrates the presence of a lobulated mass, signal voids, and hemodynamic flow characteristics. Magnetic resonance imaging also provides details about anatomic extent of the lesion, proximity to vital structures, and involvement of multiple tissue planes.
In the study conducted by Ferrari et al, (1) 25% of pediatric angiosarcomas were metastatic at diagnosis, whereas in a recent study of angiosarcomas arising predominantly in adult patients 31 (19%) had metastases. In contrast, Deyrup et al (5) reported that staging has played no role in prognosis of angiosarcomas and, consequently, angiosarcomas have been excluded from the American Joint Committee on Cancer staging system.
Grossly, most cases are described as a firm, hemorrhagic, and multinodular mass with ill-defined margins. The size of tumors ranges from 3.5 to 13 cm (mean, 8.1 cm). (3) The tumor is more than 5 cm in 50% of the cases1; this large volume implies that tumor tissue shows dissimilar growth ability in dissimilar sites. Generally, angiosarcomas of soft tissue have both epithelioid and spindle areas with emphasis on the former, which is composed of epithelioid cells that are arranged in nests, clusters, papillae, and gaping vascular channels. Tumor cells have irregular, hyperchromatic, or vesicular nuclei and prominent nucleoli. Cytoplasm is eosinophilic with ill-defined borders. Hyaline globules may be seen and are most likely digested erythrocytes (Figure 2, A and B). The tumor typically shows features of angiosarcoma including vasoformative architecture with hobnailed appearance (Figure 3) and it has infiltrative margins. Extensive hemorrhage is a characteristic feature, but it tends to obscure the diagnosis in several cases. Most pediatric soft tissue angiosarcomas are high-grade tumors (1,4,6,7): The tumor is graded as 2 or 3 using the criteria of the Pediatric Oncology Group. (8) The role of histologic grading in angiosarcoma is unclear. A recent study of 161 angiosarcomas arising in adult patients found that higher histologic grade was a significant prognostic factor for overall survival. (9) Although the number of pediatric cases in the literature is insufficient for statistical analysis, it is possible that the high mortality rate is attributable to the fact that all cases of pediatric angiosarcomas are classified as either intermediate or high grade. (4) In general, grade can be used to predict outcome of nonrhabdomyosarcomatous soft tissue sarcomas in childhood. (8) The diagnosis is suspected on light microscopy by identifying cells forming rudimentary vascular channels. At one extreme, angiosarcoma may resemble carcinoma or other undifferentiated sarcoma and it may be difficult to diagnosis. Vasoformative architecture is not always obvious on light microscopy requiring the benefit of immunohistochemical study. Sastre-Garau et al (10) reported a pediatric case of poorly differentiated angiosarcoma lacking vasoformative pattern. Recently, Deyrup et al (3) published a series of 15 angiosarcomas arising in viscera and soft tissue of patients 21 years and younger. Histologic features including growth pattern (vasoformative versus solid) were assessed. Solid areas composed of spindled or epithelioid cells were seen in more than 50% of tumors. Epithelioid morphology is present at least focally in more than 50%. Hobnail morphology was rarely seen. (8) In adult patients, epithelioid angiosarcoma is also the most frequently observed pattern (4,11); 70% of cases have epithelioid cells, (4) but vasoformative pattern and hobnail morphology is not uncommon in adult soft tissue angiosarcoma. (11)
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
IMMUNOHISTOCHEMICAL AND ULTRASTRUCTURAL STUDIES
Immunohistochemical study is an important adjunctive procedure in the diagnosis of angiosarcomas particularly for poorly differentiated forms. Tumor cells stain for the usual vascular and endothelial markers including CD31, CD34, CD144 (VE-cadherin), and at least focally for factor VIII-related antigen. The new marker FLI-1, a nuclear transcription factor, appears to be expressed in almost 100% of different vascular tumors, including angiosarcoma. (11) Nearly all cells stain strongly for vimentin, which accentuates the endothelial cells and vessel lumen formation. BNH9 is detected in 72%, and cytokeratins in 35% of tumors. Epithelial membrane antigen, S100 protein, human herpesvirus 8, and HMB-45 are consistently negative. Fifty-five percent of the tumors have intracytoplasmic aggregates of laminin. Immunostains for [alpha]-smooth muscle actin demonstrate a prominent pericytic component in several tumors (24%). Ki-67 immunostains with MIB1 indicate high proliferative activity ([greater than or equal to] 10%) in 72% of cases (Figure 4, A through D). p53 immunoreactivity (>20% nuclear staining) is observed in 20% of cases. (4)
Ultrastructural studies performed on poorly differentiated areas show groups of cells, which are frequently epithelioid, surrounded by basal lamina and closely associated with pericytes, along with intercellular and intracellular lumina with or without red blood cells. Whorls of abundant intermediate filaments, occasional tonofilament-like structures, and pinocytotic vesicles are also noted. (4)
[FIGURE 4 OMITTED]
Soft tissue angiosarcoma in children must be differentiated from Kaposi sarcoma, epithelioid hemangioendothelioma, hemangiopericytoma, and spindle cell hemangioendothelioma whose prognosis is different. In Kaposi sarcoma, there is usually a more prominent spindle cell proliferation around preexisting blood vessels containing red cells. The spindle cells are typically factor VIII negative. (11) Angiosarcoma may occasionally duplicate the picture of epithelioid hemangioendothelioma. Histologically, this lesion is composed of epithelioid endothelial cells arranged in a typical strand or cordlike pattern set in a distinctive myxohyaline stroma. On immunohistochemistry, in addition to vascular antigens (CD31, CD34, FLI-1), it may also express epithelial membrane antigen, which is negative in angiosarcoma. (11) Spindle cell hemangioendothelioma may also be confused with angiosarcoma. The presence of cavernous vessels, which are not seen in angiosarcoma, is the most reliable feature for distinguishing the 2 tumors. (12) Composite hemangioendothelioma is a complex low-grade lesion mimicking angiosarcoma. Features that distinguish the 2 lesions include the presence of benign and intermediate vascular components. Factor VIII and Ulex europeus agglutinin I immunostaining are observed in both angiosarcoma and hemangioendotheloma. (7) The distinction of angiosarcoma from hemangiopericytoma may cause difficulties, but, in general, hemangiopericytoma can be recognized by its uniform cellularity and vascular pattern with typical ''antler'' or "staghorn" configuration. (12) Tumor cells do not stain for Ulex europeus agglutinin I and immunostaining for factor VIII is weak or absent. (10) In cases of angiosarcomas lacking vascular pattern, the principal significance is the close mimicry it provides with other sarcomas such as rhabdomyosarcomas, epithelioid osteosarcomas, malignant peripheral nerve sheath tumors, synovial sarcomas, and melanomas.
PROGNOSIS AND TREATMENT
Previous reported cases showed that pediatric angiosarcomas were high-grade lesions with a capacity to invade into peripheral tissue and to metastasize. (1-3,13) Tumor size is a key prognostic variable for soft tissue sarcomas, and a tumor diameter of 5 cm is generally used as a cutoff for risk-grouping purposes. (14) Poorly differentiated neoplasms and mitotic count are also found to be independent prognostic factors. (2) Espat et al (6) demonstrated that the factor predictive of tumor-related mortality was presentation status. Treatment consists of local complete excision. (1,13) Nevertheless, treatment modalities are variable, depending on resectability and disease extent. In treatment planning of pediatric angiosarcomas, the role of chemotherapy is still unknown. In the review of the Italian and German Soft Tissue Sarcoma Cooperative Group, objective response to chemotherapy was observed in only 1 of 12 patients, and the overall behavior of this tumor is found to be identical to angiosarcoma in adults. (1) Drugs that target angiogenesis should be of particular interest in malignant vascular tumors. Because the origin of angiosarcoma is the endothelial cell, it is conceivable that the antiangiogenic activity of paclitaxel is a mechanism that contributes to its efficacy in angiosarcoma. Indeed, this drug could have a double action, targeting tumoral vascularization and tumor cells directly. In adult patients, the report from Fata et al (15) from the Memorial Sloan-Kettering Cancer Center seems particularly relevant, in which paclitaxel was shown to be effective as a single agent against cutaneous angiosarcoma of the scalp and face, even in previously treated patients. Probably thanks to the antiangiogenic and apoptotic activity already reported in Kaposi sarcoma, paclitaxel achieved 8 objective responses in 9 adult patients. (15) In the study conducted by Geller et al (16) on 15 children with relapsed or refractory solid tumors including 7 sarcomas, a dose-escalation phase I study of paclitaxel was administered in combination with standard dose ifosfamide. Paclitaxel hypersensitivity reactions were dose limiting when the drug was administered as a 6-hour infusion. The maximum tolerated dose and recommended phase II dose of paclitaxel administered as a 6-hour continuous intravenous infusion followed by standard dose intravenous ifosfamide is 425 mg/[m.sup.2] paclitaxel. But there is no ongoing antiangiogenic therapy for the pediatric angiosarcomas. Further investigations are needed as soon as possible to confirm these promising results on pediatric sarcomas. Radiation therapy may play a role, but the greater risks posed to children by treatment should be considered. Only early therapy seems to influence the outcome at all. There are some established guidelines for primary and palliative therapy. Interesting new options of biomodulatory and molecularly targeted therapy can be envisioned. The treatment of children with metastatic soft tissue angiosarcomas consists of combined treatment with chemotherapy, radiation therapy, and surgical resection of pulmonary metastases. (1-8) Angiosarcomas in children seem to be more aggressive than in adults. Bien et al (17) studied 10 patients with pediatric angiosarcoma and concluded that these neoplasms form a distinctive subset of pediatric soft tissue sarcomas that display an aggressive clinical behavior, typically with local recurrence and metastases; among them, 5 patients presented with local, 3 with systemic, and 2 with regional disease. Primary complete resection was performed in 7 patients, not resulting in local control in any of them. Radiotherapy was administered in 5 patients, chemotherapy in all. Seven patients relapsed, and 3 never entered clinical remission. Nine of 15 patients died of disease progression. Deyrup et al (3) studied 15 patients with high-grade angiosarcoma arising in children and young adults; on follow-up 10 (67%) died of disease (range, 27 months to 11 years; mean, 6 years 8 months) and 4 patients were alive without disease (range, 27 to 132 months; mean, 79 months). Three of 9 cases reported by Kauffmann and Stout (2) died within a delay ranging from a few months to 2 years.
Although extremely rare, angiosarcomas do affect children and young adults and this diagnosis should be considered. Pediatric angiosarcoma seems to be more aggressive than angiosarcoma in adults because all cases of pediatric angiosarcomas are classified as either intermediate or high grade. Pathologic diagnosis is difficult in poorly differentiated angiosarcomas, requiring immunohistochemical study to confirm vascular differentiation. Enrollment in investigational trials for high-risk children with recurrent or metastatic disease is recommended to potentially improve survival.
We gratefully acknowledge the excellent assistance of Professor Jean Michel Coindre for performing immunohistochemical study and confirming the diagnosis.
(1.) Ferrari A, Casanova M, Bisogno G, et al. Malignant vascular tumors in children and adolescents: a report from the Italian and German Soft Tissue Sarcoma Cooperative Group. Med Pediatr Oncol. 2002;39(2):109-114.
(2.) Kauffman SL, Stout AP. Hemangiopericytoma in children. Cancer. 1960;13: 695-710.
(3.) Deyrup AT, Miettinen M, North PE, et al. Angiosarcomas arising in the viscera and soft tissue of children and young adults: a clinicopathologic study of 1 5 cases. Am J Surg Pathol. 2009;33(2):264-269.
(4.) Meis-Kindblom JM, Kindblom LG. Angiosarcoma of soft tissue: a study of 80 cases. Am J Surg Pathol. 1998;22(6):683-697.
(5.) Deyrup AT, McKenney JK, Tighiouart M, Folpe AL, Weiss SW. Sporadic cutaneous angiosarcomas: a proposal for risk stratification based on 69 cases. Am J Surg Pathol. 2008;32(1):72-77.
(6.) Espat NJ, Lewis JJ, Woodruff M, et al. Angiosarcoma: prognostic factors and outcome in 50 prospectively followed patients. Sarcoma. 2000;4:173-177.
(7.) Lezama-del Valle P, Gerald WL, Tsai J, Meyers P, La Quaglia MP. Malignant vascular tumors in young patients. Cancer. 1998;83(8):1634-1639.
(8.) Parham DM, Webber BL, Jenkins JJ III, et al. Nonrhabdomyosarcomatous soft tissue sarcomas of childhood: formulation of a simplified system for grading. Mod Pathol. 2008;7:705-710.
(9.) Fayette J, Martin E, Piperno-Neumann S, et al. Angiosarcomas, a heterogeneous group of sarcomas with specific behavior depending on primary site: a retrospective study of 161 cases. Ann Oncol. 2007;18(12):2030-2036.
(10.) Sastre-Garau X, Thiery JP, Ovtracht L, Gallet C, Quintana E, Laurent M. Soft tissue angiosarcoma in a child: immunohistochemical and ultrastructural features. Ann Pathol. 1992;12(1):34-40.
(11.) Fletcher CDM, Unni KK, Mertens F. Pathology and Genetics of Tumours of Soft Tissue and Bone. Lyon, France: IARC Press; 2002. World Health Organization Classification of Tumours.
(12.) Weiss SW, Goldblum JR. Enzinger and Weiss's Soft Tissue Tumors. 5th ed. St Louis, MO: The CV Mosby Co; 2007.
(13.) Harish S, Hosalkar JP, Dormans MD. Surgical management of pelvic sarcoma in children. J Am Acad Orthop Surg. 2007;15(7):408-424.
(14.) Ferrari A, Miceli R, Meazza C, et al. Soft tissue sarcomas of childhood and adolescence: the prognostic role of tumor size in relation to patient body size. J Clin Oncol. 2009;27(3):371-376.
(15.) Fata F, O'Reilly E, Ilson D, et al. Paclitaxel in the treatment of patients with angiosarcoma of the scalp or face. Cancer. 1999;86(10):2034-2037.
(16.) Geller JI, Wall D, Perentesis J, Blaney SM, Bernstein M. Phase I study of paclitaxel with standard dose ifosfamide in children with refractory solid tumors: a Pediatric Oncology Group study (POG 9376). Pediatr Blood Cancer. 2009; 52(3):346-350.
(17.) Bien E, Godzinski J, Balcerska A, et al. Malignant vascular tumours in children: report from the Polish Pediatric Rare Tumors Study. Med Wieku Rozwoj. 2004;8(2):145-158.
Lobna Ayadi, MD; Abdelmajid Khabir, PhD
Accepted for publication May 21, 2009.
From the Department of Pathology, Habib Bourguiba University Hospital, Tunisia.
The authors have no relevant financial interest in the products or companies described in this article.
Reprints: Lobna Ayadi, MD, Department of Pathology, University Hospital Habib Bourguiba, El Am Road Km 0,5, Sfax 3029, Tunisia (e-mail: email@example.com).
|Gale Copyright:||Copyright 2010 Gale, Cengage Learning. All rights reserved.|