Ptosis

A 66-year-old woman presented with headaches, left eye proptosis, and loss of vision from her left eye for 2 years. Ophthalmologic examination showed a left proptotic eye with complete ptosis, blindness, no extraocular movements, and a 4+ afferent pupil defect. Vision in the right eye corrected to 20/20 and was otherwise unremarkable. Whole brain magnetic resonance imaging showed a 7.0 x 5.8 X 4.0-cm contrast-enhancing left orbital mass extending through the orbital apex along the optic nerve to the suprasellar cistern, middle cranial fossa, and posterior left maxillary sinus (Figure 1, A). She underwent left frontotemporal craniotomy and orbital decompression with removal of a highly vascular neoplasm.

Histologic sections stained with hematoxylin-eosin showed a hypercellular neoplasm predominantly composed of randomly oriented polygonal and rounded cells with ill-defined cytoplasm. Some cells had conspicuous nucleoli, and others had cytoplasmic clearing (Figure 1, B). Large and small blood vessels lined by a single layer of endothelium with a branching or "staghorn" appearance were readily apparent. There were foci of low cellularity with dense perivascular and intercellular collagen (Figure 1, C). Cells in mitosis were occasionally encountered. Necrosis, psammoma bodies, intranuclear pseudoinclusions, and whorl formations were not identified. A reticulin stain showed dense pericellular positivity surrounding every cell (Figure 1, D). The tumor cells showed focal immunoreactivity to vimentin, CD34, and factor XIIIa. The tumor cells were negative for glial fibrillary acid protein, S100 protein, epithelial membrane antigen (EMA), cytokeratin, neurofilament protein, desmin, muscle-specific actin, and smooth muscle actin. The MIB-1 (Ki67) labeling index was approximately 3%. Ultrastructural studies showed cells that generally lacked external or basal lamina material. No interdigitating cell processes, tight junctions, or desmosomes were identified.

What is your diagnosis?

Pathologic Diagnosis: Hemangiopericytoma of the Orbit

Hemangiopericytoma has evoked controversy since its description nearly 60 years ago in 1942 by Dr Arthur Purdy Stout.1 Although it is a diagnosis likely to be made during the career of most pathologists, hemangiopericytoma is certainly an entity with an identity crisis. Recent articles have questioned whether it is a "histopathologic pattern or clinicopathologic entity," a "diagnosis of exclusion," a "wastebasket" diagnosis, or a "dying breed."2-4 Stout even remarked, "It has been my general attitude in regard to hemangiopericytoma to reject it as a diagnosis if I can think of any other reasonable explanation for a tumor."3 The dubiousness attached to hemangiopericytoma stems from a number of features. First, the histologic, immunohistochemical, and ultrastructural features of hemangiopericytoma are not specific. The classic "staghorn" vascular pattern can be present in many neoplasms, including synovial sarcomas, mesenchymal chondrosarcomas, benign and malignant fibrous histiocytomas, malignant peripheral nerve sheath tumors, solitary fibrous tumors, malignant mesotheliomas, leiomyosarcomas, thymomas, sarcomatoid carcinomas, and endometrial stromal sarcomas, among others.2 Second, there is marked variation in the clinical course, and histologic grading does not necessarily correlate with behavior.5 According to the World Health Organization, hemangiopericytoma is graded as grade II or III, depending on mitotic rate (>5 mitoses per 10 high-power fields), high cellularity, nuclear pleomorphism, hemorrhage, and necrosis. However, the differences in time to recurrence, rate of metastasis, and survival appear marginal.6 Third, there is a lack of understanding about the nature and function of the pericyte. Zimmermann described pericytes as contractile cells with long processes that wrap around capillaries and serve to change the caliber of their lumens and suggested they are modified smooth muscle cells.1 Others have postulated that pericytes represent versatile precursors of other mesenchymal cells, such as fibroblasts. Stout recognized myoid differentiation in his tumors, and a recent study3 has suggested a unifying designation of "perivascular myoma" for adult myofibromatosis, glomangiopericytoma, and myopericytoma.

The histologic differential diagnosis of orbital hemangiopericytoma includes meningioma, solitary fibrous tumor (SFT), and fibrous histiocytoma (FH). As noted herein, a host of other tumors may show a hemangiopericytomatous pattern. It has become apparent that intracranial hemangiopericytomas are not meningiomas, but rather are indistinguishable from hemangiopericytomas that occur in soft tissues. Unlike hemangiopericytomas, meningiomas are immunoreactive to EMA, ultrastructural studies show complex interdigitating processes and desmosomes, and cytogenetics show deletions of chromosome 22. Thus, the concept of "angioblastic meningioma" has been abolished. Increasingly, SFTs are being recognized in the orbit and central nervous system. These typically meningeal-- based neoplasms histologically resemble pleural-based SFTs and are characterized immunohistochemically by strong and diffuse CD34 staining. In addition, ultrastructural studies show fibroblastic differentiation. In contrast, CD34 reactivity in meningiomas and hemangiopericytomas is usually either focal or not as strong. Benign FH, the most common orbital mesenchymal tumor, may mimic hemangiopericytoma histologically.7,8 Although the typical storiform architecture of benign FH can be focally present in hemangiopericytoma, it usually is not the predominant pattern. Conversely, FH may display a significant vascular pattern. Foam cells and giant cells can usually be found in FH. The reticulin stain in FH is almost always moderately heavy, but haphazard or erratic,8 unlike the dense pericellular reticulin staining of hemangiopericytoma. However, FH may rarely demonstrate a pericellular reticulin pattern.4 Both FH and hemangiopericytoma may exhibit factor XIIIa immunoreactivity.9 Hemangiopericytomas, unlike FHs, tend to be encapsulated. Despite these differences, it may be difficult to confidently distinguish these tumors in some cases.

The ability to detect the products of chimeric gene fusions by reverse transcriptase polymerase chain reaction has been instrumental in the ability to diagnose a subset of soft tissue tumors. Little is known about the molecular genetics of hemangiopericytoma. A subgroup of both intracranial and extracranial hemangiopericytomas is characterized by a reciprocal translocation t(12;19)(q13;q13.3), but the gene fusion products have not yet been detected.10 Interestingly, 12q13 (TLS [FUS] gene) is the site of aberration in myxoid or round cell liposarcoma, malignant melanoma of soft parts (clear cell sarcoma) (ATF1 gene), and congenital fibrosarcoma (NTRK3 gene).

When taken together, the staghorn appearance, perivascular collagen, dense pericellular reticulin, absence of diffuse CD34 and EMA staining, and lack of interdigitating cell processes support the diagnosis of hemangiopericytoma in this case.

References

1. Stout AP, Murray MR. Hemangiopericytoma: a vascular tumor featuring Zimmermann's pericytes. Ann Surg. 1942;116:26-33.

2. Nappi 0, Ritter JH, Pettinato G, Wick MR. Hemangiopericytoma: histopathologic pattern or clinicopathologic entity? Semin Diagn Pathol. 1995;12:221232.

3. Granter SR, Badizadegan K, Fletcher CDM. Myofibromatosis in adults, glomangiopericytoma, and myopericytoma: a spectrum of tumors showing perivascular myoid differentiation. Am] Surg PathoL 1998;22:513-525.

4. Fletcher CDM. Hemangiopericytoma-a dying breed? reappraisal of an "entity" and its variants: a hypothesis. Curr Diagn Pathol. 1994;1:19-23.

5. Stout AP. Hemangiopericytoma: a study of twenty five new cases. Cancer. 1949;2:1027-1035.

6. jaaskelainen J, Louis DN, Paulus W, Haltia MJ. Hemangiopericytoma. In: Klei hues P, Cavenee WK, eds. World Health Organization Classification of Tumors, Pathology and Genetics, Tumors of the Nervous System. Lyon, France: IARC Press; 2000:190-193.

7. Croxatto JO, Font RL. Hemangiopericytoma of the orbit: a clinicopathologic study of 30 cases. Hum Pathol. 1982;13:210-218.

8. Jakobiec FA, Howard GM, Jones IS, Wolff M. Hemangiopericytoma of the orbit. Am] Ophthalmol. 1974;78:816-834.

9. Nemas Z. Differentiation markers in hemangiopericytoma. Cancer. 1992; 69:133-140.

10. Henn W, Wullich B, Thonnes M, Steudel WI, Feiden W, Zang KD. Recurrent t(12;19)(q13;q13.3) in intracranial and extracranial hemangiopericytoma. Cancer Genet Cytogenet. 1993;71:151-154.

Kelly J. Butnor, MD; Thomas J. Cummings, MD

Accepted for publication December 6, 2001.

From the Department of Pathology, Duke University Medical Center, Durham, NC.

Reprints not available from the author.

Copyright College of American Pathologists Dec 2002
Provided by ProQuest Information and Learning Company. All rights Reserved