header advert
Orthopaedic Proceedings Logo

Receive monthly Table of Contents alerts from Orthopaedic Proceedings

Comprehensive article alerts can be set up and managed through your account settings

View my account settings

Visit Orthopaedic Proceedings at:

Loading...

Loading...

Full Access

3.O.05 HIGH-GRADE ANGIOSARCOMA OF BONE: A CLINICOPATHOLOGICAL STUDY OF 64 CASES



Abstract

High-grade angiosarcomas (HGAS) of bone are rare and represent less than 1% of the primary malignant bone tumours. Because of their rareness little is known. Clinically, it is accepted that they are extremely aggressive. Due to the lack of uniform terminology and accepted histological criteria, terminology and classification of primary malignant vascular tumours of bone has been highly controversial. Today, angiosarcoma is the most accepted term for high-grade primary vascular tumour of bone, recognized by the 2002 WHO Classification. However, distinct histological hallmarks to define a HGAS of bone are not clear.

We collected 64 HGAS of bone diagnosed between 1964 and 2007 from the files of the departments of pathology, Leiden University Medical Center (Leiden), Rizzoli Institute (Bologna) and University Hospitals (Leuven). All clinical, radiological, and pathological data were reviewed and different histological criteria were scored. A tissue micro-array was constructed containing 57 HGAS of bone. To confirm the vascular origin of all lesions and to investigate the diagnostic value of commonly used markers, immunohistochemistry was performed for CD31, CD34, Factor VIII, and keratin AE1/AE3. Staining was evaluated positive or negative.

Among 64 patients with HGAS of bone, there are 41 males and 23 females. There is a wide age distribution, with a nearly equal distribution from the second to the sixth decade. The solitary cases are mostly located in the extremities (66%) followed by trunk (12.8%), axial/central location (10.6%) and pelvis (10.6%). 17 cases (73%) have multifocal bone lesions. HGAS of bone show variable histological patterns. Association with clinical outcome (chi-square test) reveals that there is a significant poor survival when the tumour has tree or more mitoses (p=0.001), a macronucleoli (p=0.011) or there is an absence of an eosinophilic infiltrate (p=0.023). The HGAS of bone are positive for CD31 in 53/55 (96%), CD34 in 33/57 (58%), Factor VIII in 47/55 (86%), and keratin in 40/57 (70%). Only 15 out of 40 (38%) keratin positive angiosarcomas, showed an epithelioid phenotype at classical morphology. All tumours with an epithelioid phenotype are keratin positive.

Although HGAS of bone in general have a poor outcome, histological criteria such as three or more mitoses, the presence of a macronucleolus and the absence of an eosinophilic infiltrate can be useful to predict a more aggressive course, consistent with the clinical behaviour of a high-grade angiosarcoma. CD31 and Factor VIII are the best diagnostic markers for HGAS of bone. It is striking that keratin positivity is seen in the majority of cases, and is independent of epithelioid morphology. Pathologists should be aware of this to avoid misinterpretation as metastatic carcinoma.

Correspondence should be addressed to Professor Stefan Bielack, Olgahospital, Klinikum Stuttgart, Bismarkstrasse 8, D-70176 Stuttgart, Germany. Email: s.bielack@klinikum_stuttgart.de