2.1 Histology and immunohistochemistry

Tumors were identified using a retrospective chart review of the pathology tissue archives of the Department of Pathology and Laboratory Medicine at the University of California, Los Angeles (UCLA), using the search terms “glomus tumor,” “malignant glomus tumor,” “myopericytoma,” and “angioleiomyoma.” Slides were reviewed by 2 independent pathologists to ensure accuracy of diagnosis (SMD and AWJ). Diagnostic criteria for malignancy in glomus tumors were used as described by Folpe and colleagues,^15,16 including deep-seated tumors greater than 2cm, tumors with atypical mitotic figures, or tumors with moderate to high nuclear grade and >5 mitotic figures in 50 HPF. Patient information was obtained, including age, sex, tumor location, and tumor size. Formalin-fixed paraffin embedded (FFPE) tumor tissue from patients was acquired from the tissue archives, under IRB approval # 13-000918.

Immunohistochemistry for pericyte markers was performed using the ABC method (Vectastain Elite ABC; Vector Laboratories, Burlingame, CA) using diaminobenzidine (DAB) as the chromogen (ImmPACT DAB, Vector Laboratories). The following primary antibodies were used: goat anti-angiopoietin-1 IgG (1:50, R&D Systems, Minneapolis, MN, USA) and mouse anti-angiopoietin-2 IgG (1:50, Santa Cruz Biotechnology, Santa Cruz, CA, USA). The following secondary antibodies were used: polyclonal rabbit biotinylated anti-goat IgG (1:500, Sigma, St Louis, MO) and polyclonal horse anti-mouse IgG (1:500 [H+L], Vector Laboratories).

Heat-mediated antigen retrieval was performed for all immunohistochemical stains in 1mmol/L Tris–EDTA, 0.01% Tween-20 (Sigma), pH 8. Nonspecific antibody binding was blocked (IHC-TEK Antibody Diluent, pH 7.4; IHC World, LLC, Woodstock, MD). Endogenous peroxidase and alkaline phosphatase blocking solution was used (BLOXALL endogenous peroxidase and alkaline phosphatase blocking solution, Vector Laboratories). Mayer's hematoxylin was used as a nuclear counterstain (1:5, ABCAM), and slides were mounted using aqueous media (VectaMount AQ, Vector Laboratories).

2.2 Immunohistochemical semi-quantitation

Semiquantitative grading of immunohistochemical stains was performed by 3 blinded independent observers, as previously performed.⁴ Intensity of staining was graded on a 3-point scale (0–3+), defined as follows: 0, absent stain; 1+, weak, focal cytoplasmic staining; 2+, moderate, focal to diffuse cytoplasmic staining; 3+, strong, diffuse cytoplasmic staining. In cases of disagreement between observers, tumor staining was re-evaluated by the same observers and the majority opinion was selected. In addition, the percentage of tumor cells stained was also evaluated, using a 5% incremental scale. Means and standard deviations were calculated for all semi-quantitative measurements.

3.1 Histologic appearance of perivascular soft tissue tumors

A collection of perivascular soft tissue tumors was interrogated for angiopoietin protein expression. Tumors included glomus tumor (n=9), malignant glomus tumor (n=4), myopericytoma (n=3), and angioleiomyoma (n=9). Each tumor type showed typical histologic features (Fig. 1). Clinical and demographic information for each patient sample are shown in Supplemental Table 1. Benign glomus tumors were all located on the digits, and showed either solid or glomuvenous architectural patterns (Fig. 1A, B, solid glomus tumor shown). Malignant glomus tumors were deep-seated (including deep soft tissue or visceral locations). Increased mitotic rate was seen in three of four tumors (9–25 mitoses per 10 HPF). Moderate to high nuclear grade was seen in a single case (Fig. 1C, D). Myopericytoma and angioleiomyoma specimens were found in the superficial soft tissue or dermis, most often of the lower extremity. Typical histologic features of myopericytoma included a perivascular proliferation of monomorphic ovoid to spindled cells with brightly eosinophilic cytoplasm (Fig. 1E, F). Angioleiomyoma tumors showed characteristics of both solid and venous growth patterns, being composed of eosinophilic tumor cells with indistinct cell borders, consistent with smooth muscle differentiation (Fig. 1G, H). Where appropriate, immunohistochemical stains were performed and supported each diagnosis (summarized in 4).

A collection of perivascular soft tissue tumors was interrogated for angiopoietin protein expression. Tumors included glomus tumor (n=9), malignant glomus tumor (n=4), myopericytoma (n=3), and angioleiomyoma (n=9). Each tumor type showed typical histologic features (Fig. 1). Clinical and demographic information for each patient sample are shown in Supplemental Table 1. Benign glomus tumors were all located on the digits, and showed either solid or glomuvenous architectural patterns (Fig. 1A, B, solid glomus tumor shown). Malignant glomus tumors were deep-seated (including deep soft tissue or visceral locations). Increased mitotic rate was seen in three of four tumors (9–25 mitoses per 10 HPF). Moderate to high nuclear grade was seen in a single case (Fig. 1C, D). Myopericytoma and angioleiomyoma specimens were found in the superficial soft tissue or dermis, most often of the lower extremity. Typical histologic features of myopericytoma included a perivascular proliferation of monomorphic ovoid to spindled cells with brightly eosinophilic cytoplasm (Fig. 1E, F). Angioleiomyoma tumors showed characteristics of both solid and venous growth patterns, being composed of eosinophilic tumor cells with indistinct cell borders, consistent with smooth muscle differentiation (Fig. 1G, H). Where appropriate, immunohistochemical stains were performed and supported each diagnosis (summarized in 4).

Close

Fig. 1 Typical histologic appearance of each perivascular soft tissue tumor. Hematoxylin–eosin (H&E) staining of (A, B) benign glomus tumor, (C, D) malignant glomus tumor, (E, F) myopericytoma, and (G, H) angioleiomyoma. Scale bar: 100μm.

Export to PPT

3.2 Ang-1 and -2 expression in non-lesional blood vessels

The normal distribution of Ang-1 and Ang-2 was first interrogated by immunohistochemical staining among small and large blood vessels not associated with tumors (Fig. 2). In small caliber vessels, Ang-1 was observed in both endothelial and perivascular cell types (Fig. 2A, B), while in larger vessels Ang-1 predominantly localized to the endothelium (Fig. 2C, D). Ang-2 expression was found in both endothelial and perivascular cells in both small and larger vessels (Fig. 2E–H). In aggregate, these data confirm previous observation that Ang-1 and -2 are present in both endothelial and perivascular cell types. Moreover, Ang-1 and -2 expression was not observed in other mesenchymal tissues, including no immunostaining in adipocytes, skeletal muscle, or stromal fibroblasts.

Close

Fig. 2 Ang-1 and -2 expression in non-tumor associated blood vessels. (A, B) Ang-1 in small caliber vessels. (C, D) Ang-1 in larger caliber vessels. (E, F) Ang-2 in small caliber vessels. (G, H) Ang-2 in larger caliber vessels. Black arrowheads indicate endothelial staining. Red arrowheads indicate perivascular staining. Scale bar: 100μm.

Export to PPT

3.3 Ang-1 expression in perivascular soft tissue tumors

Next, Ang-1 expression was examined across perivascular soft tissue tumor samples. Results showed a near complete absence of Ang-1 expression across all tumor types (Fig. 3, Table 1). Ang-1 immunoreactivity was limited to intralesional blood vessels in most cases. The exception to this was malignant glomus tumor, which showed weak to moderate Ang-1 immunoreactivity (1–2+) in 50% of cases (2/4). Ang-1 expression in malignant glomus tumor was focal in all cases (≤10%). See also Supplemental Tables 2–5 for a more detailed analysis of individual tumors.

Next, Ang-1 expression was examined across perivascular soft tissue tumor samples. Results showed a near complete absence of Ang-1 expression across all tumor types (Fig. 3, Table 1). Ang-1 immunoreactivity was limited to intralesional blood vessels in most cases. The exception to this was malignant glomus tumor, which showed weak to moderate Ang-1 immunoreactivity (1–2+) in 50% of cases (2/4). Ang-1 expression in malignant glomus tumor was focal in all cases (≤10%). See also Supplemental Tables 2–5 for a more detailed analysis of individual tumors.

Close

Fig. 3 Ang-1 immunohistochemical staining among perivascular soft tissue tumors. (A, B) Benign glomus tumor, (C, D) malignant glomus tumor, (E, F) myopericytoma, and (G, H) angioleiomyoma. Across the majority of tumor samples, Ang-1 was essentially limited to intralesional blood vessels. Malignant glomus tumor samples showed focal staining in 2/4 of cases. Scale bar: 100μm.

Export to PPT

3.4 Ang-2 expression is perivascular soft tissue tumors

Next, Ang-2 immunohistochemical staining was performed across all perivascular soft tissue tumor samples. Results showed that Ang-2 expression was present throughout all tumor types and in all tumor samples (Fig. 4, Table 1). Ang-2 expression was higher in glomus tumor, malignant glomus tumor, and myopericytoma, with predominantly moderate staining intensity (2+) in the majority of tumor cells (66.1–82.5% of tumor cells showed positive stain on average across all three diagnoses). In contrast, Ang-2 immunoreactivity was usually weak in angioleiomyoma. Similarly, the distribution of Ang-2 immunoreactivity was more limited in angioleiomyoma, and was most often noted in the minority of tumor cells within a given sample (30% of cells showed Ang-2 staining within angioleiomyoma). Among angioleiomyoma, no discernable difference in staining was observed with different architectural patterns (i.e. sheet-like versus venous growth).

Close

Fig. 4 Ang-2 expression among perivascular soft tissue tumors. (A, B) Benign glomus tumor, (C, D) malignant glomus tumor, (E, F) myopericytoma, and (G, H) angioleiomyoma. Scale bar: 100μm.

Export to PPT