Letters to the Editor
To the Editors:
Hornick et al1 have reported that the 69% of the hybrid peripheral nerve sheath tumors (HPNSTs) are situated primarily in the subcutis. These tumors are biphasic, showing abrupt transition between the 2 components or an intimate admixture of the 2 cell types.1 The distinction between 2 different cellular proliferations may not be appreciated on hematoxylin-eosin (HE)-stained slides. The perineural component may be inconspicuous because it may be obscured by Schwann cells and is not evident on HE-stained slides. The identification of schwannomatous areas may be difficult because the histological hallmarks of schwannoma (eg, perivascular hyalinization and Antoni A and B zonation) are lacking in the hybrid tumors.1 The cutaneous neurofibroma consists of an admixture of various cell types including Schwann cells, fibroblasts, perineural cells, and scattered axons. This cell lineage may be identifiable with difficult in the hybrid tumors because the usual cutaneous neurofibromas are less cellular than the hybrid tumors and lack their architectural features.1 Immunohistochemistry only demonstrates the dual cell lineages. The perineural cells are epithelial membrane antigen (EMA) positive,2 and Schwann cells are S-100 protein positive. In the histological practice, the pathologist does not usually search with accuracy for the presence of 2 components in the benign nerve sheath tumors especially when one lesional component is predominant. The present case demonstrates that the routinary use of immunohistochemistry may increase the frequency of hybrid tumors. Recently, we have observed a case of hybrid neurofibroma/perineurioma in a 45-year-old woman, presenting a 1-year history of 1 × 0.8 cm palpable nodule on the back. The lesion was well circumscribed and unencapsulated. The cut surface was homogeneous and gray-white. The patient had no signs of neurofibromatosis. The histological specimen had been sent for consultation to Prof Christopher D. M. Fletcher, director of surgical pathology, Brigham and Women's Hospital. According to the histological report by Fletcher, the lesion consisted of a poorly marginated intradermal spindle cell lesion with neural-like cytomorphology (Fig. 1). Atypia, pleomorphism, and mitosis were not evident. Two cellular populations might be observed. Spindle cells with dark, elongated, variably wrinkled nuclei were intermingled with cells characterized by vesicular or stellated nuclei (Fig. 2). Biphasic pattern was not discernable on HE-stained slides, but it was evident on immunohistochemical slides. Immunostains showed positivity for S-100 protein in approximately 50% of the cells, whereas the other 50% showed a very striking positivity for EMA, suggesting an almost even admixture of the 2 cell types (Figs. 3, 4). There were scattered intralesional axons highlighted by positivity for neurofilament protein (Fig. 1, inset). This lesion showed a mixed schwannian and perineural differentiation, but it lacked the sharp circumscription and the more whorled appearance of a usual hybrid schwannoma/perineurioma. The histological hallmarks of schwannoma-like areas (eg, perivascular hyalinization and Antoni A and B zonation) were not found. Taking into account the overall architectural features of the lesion, the appearances were those of a benign nerve sheath tumor with hybrid features of neurofibroma and perineurioma. Only 2 cases of hybrid neurofibroma-perineurioma have been reported.3 In the reported HPNSTs, the biphasic proliferative pattern was generally evident in HE-stained slides and was confirmed by immunohistochemistry. A biphasic pattern was rarely not evident on HE-stained slides as in case 2 of hybrid neurofibroma-perineurioma series described in the study by Kazakov et al.3 The contribution of dual cell lineages is evident by immunohistochemistry, as confirmed by our case. In the hybrid tumors, a double staining for S100 protein and EMA clearly illustrates that these tumors are indeed composed of 2 separate cell types, arranged in parallel arrays, rather than a single population of cells expressing both antigens.1 EMA-positive cells are rarely found only in pericapsular and perivascular areas of some neurofibromas, and they are considered as remnants of the nerve sheath of the affected nerve. Typical cutaneous schwannoma are surrounded by an EMA-positive perineural capsule, whereas the hybrid tumors contain perineural cells mixed with S-100 protein-positive cells. Moreover, hybrid tumors differ from pure schwannomas because the former are unencapsulated. We believe that the frequency of HPNSTs is underestimated. The use of immunohistochemistry may facilitate the correct diagnosis of HPNSTs and increase the frequency of HPNSTs. Positivity for both EMA and S100 protein assists in revealing the biphasic nature of the neoplasm, when this proliferative feature is not evident on HE-stained sections.
Teresa Pusiol, MD
Maria Grazia Zorzi, MD
Doriana Morichetti, MD
Francesco Piscioli, MD
Institute of Pathology, Rovereto Hospital, Rovereto (TN), Italy
The authors are grateful to Prof Christopher D. M. Fletcher, professor of pathology, Harvard Medical School, director of surgical pathology, Brigham and Women's Hospital, Boston, MA, for consultation with this case.
1. Hornick JL, Bundock EA, Fletcher CD. Hybrid schwannoma/perineurioma: clinicopathologic analysis of 42 distinctive benign nerve sheath tumors. Am J Surg Pathol
2. Ariza A, Bilbao JM, Rosai J. Immunohistochemical detection of epithelial membrane antigen in normal perineurial cells and perineurioma. Am J Surg Pathol
© 2011 Lippincott Williams & Wilkins, Inc.
3. Kazakov DV, Pitha J, Sima R, et al. Hybrid peripheral nerve sheath tumors: schwannoma-perineurioma and neurofibroma-perineurioma. A report of 3 cases in extradigital locations. Ann Diagn Pathol