Elsevier

Human Pathology

Volume 45, Issue 5, May 2014, Pages 1039-1044
Human Pathology

Original contribution
Alveolar soft part sarcoma and granular cell tumor: an immunohistochemical comparison study

https://doi.org/10.1016/j.humpath.2013.12.021Get rights and content

Summary

Although the histologic features of alveolar soft part sarcoma and granular cell tumor are typically distinctive, occasional cases show a significant morphologic overlap. Differentiating these entities is crucial because granular cell tumor is almost always benign and alveolar soft part sarcoma is invariably malignant. We evaluated a panel of immunohistochemical stains (S-100 protein, inhibin, SOX10, nestin, calretinin, and TFE3) in 13 alveolar soft part sarcomas and 11 granular cell tumors. Tissue sections were also stained by the periodic acid–Schiff method after diastase digestion (PAS-D) and evaluated for coarse cytoplasmic granularity or crystalline cytoplasmic inclusions. S-100 protein, inhibin, SOX10, and nestin each distinguished granular cell tumor and alveolar soft part sarcoma with 100% sensitivity and specificity. PAS-D staining also distinguished cases with 100% accuracy, as granular cell tumor consistently demonstrated coarsely granular, PAS-D–positive cytoplasm and alveolar soft part sarcoma showed only focal intracytoplasmic crystalline inclusions. Although all granular cell tumors were calretinin positive, so were 46% of alveolar soft part sarcomas. TFE3 was positive in 91% of granular cell tumors and all alveolar soft part sarcomas. Together with PAS-D, immunohistochemical stains for S-100 protein, inhibin, SOX10, and nestin accurately identify alveolar soft part sarcoma and granular cell tumor. Although TFE3 has been reported as a relatively specific marker for alveolar soft part sarcoma, it should be recalled that it is also expressed in most granular cell tumors.

Introduction

In general, the characteristic histologic patterns of both alveolar soft part sarcoma and granular cell tumor are sufficiently distinctive to establish the diagnosis of either entity by routine hematoxylin and eosin (H&E) staining. However, occasional cases show histomorphologic overlap such that a definitive diagnosis requires the use of special and immunohistochemical (IHC) stains [1], [2], [3]. Whereas almost all granular cell tumors are benign and adequately treated with simple excision, alveolar soft part sarcoma is invariably malignant [4]. Although the clinical course of alveolar soft part sarcoma is often slow and indolent, these patients need to be followed up indefinitely for local recurrence and metastatic disease [3], [4], [5]. Although alveolar soft part sarcoma responds poorly to radiotherapy and conventional chemotherapy, molecularly targeted therapies hold promise for the systemic treatment of this tumor [6], [7], [8].

Granular cell tumor has a wide anatomical distribution and is composed of a uniform population of polygonal to spindle-shaped cells with abundant eosinophilic cytoplasm containing namesake lysosomal granules that impart a coarsely granular appearance to the cytoplasm upon staining with the periodic acid–Schiff reagent after diastase digestion (PAS-D). The nuclei are usually small and pyknotic. Some tumors may demonstrate stromal elastosis, and secondary epithelial hyperplasia may be seen in submucosal tumors. Alveolar soft part sarcoma also has a wide anatomical distribution and is characterized by well-defined nests of cells separated by thin fibrovascular septa. Like granular cell tumor, the constituent cells are large and polygonal, with finely granular, eosinophilic cytoplasm. However, they are generally discohesive and typically have large, vesicular nuclei with prominent nucleoli. Rod-shaped or globular crystalline inclusions are often noted PAS-D stains, which are morphologically distinct from the coarsely granular cytoplasmic staining seen in granular cell tumors [4], [9].

Given the dramatically different prognoses and recommended treatment of these 2 tumors, correct pathologic diagnosis is critical. We recently encountered a case of alveolar soft part sarcoma with small, pyknotic nuclei, focal spindling of tumor cells and a sheet-like, cohesive growth pattern that closely resembled granular cell tumor (Fig. 1), which prompted us to evaluate a panel of IHC stains and the PAS-D stain in resolving this differential diagnosis.

Section snippets

Materials and methods

Nine cases of alveolar soft part sarcoma and 11 cases of granular cell tumor were identified through a search of the surgical pathology archives. Four additional cases of alveolar soft part sarcoma were retrieved from the personal consult files of one of the authors (C. M. C.). After obtaining institutional review board approval, H&E-stained slides were reviewed to confirm the histologic diagnosis and select blocks for further study.

For IHC analysis, formalin-fixed, paraffin-embedded tissue

Results

Representative results for each marker are demonstrated in Fig. 2, Fig. 3, and the results are summarized in the Table. S-100 protein, inhibin, SOX10, and nestin were positive in each case of granular cell tumor and negative in each case of alveolar soft part sarcoma (100% accuracy). PAS-D staining demonstrated large, coarse cytoplasmic granules in all cases of granular cell tumor, whereas each alveolar soft part sarcoma showed at least focal intracytoplasmic crystalline inclusions. Although

Discussion

Occasionally, alveolar soft part sarcoma and granular cell tumor show significant overlap in their histologic features on H&E-stained sections, rendering definitive diagnosis particularly challenging. A correct diagnosis is crucial in this setting because the prognosis and management of these soft tissue tumors are significantly different. A number of previous studies have examined the IHC profiles of each of these tumors separately (results summarized in Fig. 4 and Supplementary Table S1), but

Supplementary data

The following are the supplementary data to this article.

Chamberlain Supplementary Table 1

References (25)

  • P. Masson

    Human tumors: histology, diagnosis, and technique

    (1970)
  • H. Jeffreys

    Theory of probability

    (1961)
  • Cited by (0)

    Disclosures: The authors disclose no conflicts of interest and source of funding.

    View full text