GLUT-1 expression in mesenchymal tumors: an immunohistochemical study of 247 soft tissue and bone neoplasms

Transcription

1 Human Pathology (2008) 39, Original contribution GLUT-1 expression in mesenchymal tumors: an immunohistochemical study of 247 soft tissue and bone neoplasms William A. Ahrens MD, Robert V. Ridenour III MD, Bolette L. Caron, Dylan V. Miller MD, Andrew L. Folpe MD Division of Anatomic Pathology, Mayo Clinic, Rochester, MN 55905, USA Received 25 January 2008; revised 6 March 2008; accepted 10 March 2008 Keywords: GLUT-1; Perineurioma; Hemangioma; Hypoxia; Sarcoma; HIF1-α Summary GLUT-1, an erythrocyte-type glucose transporter protein expressed in juvenile hemangiomas, has recently been shown to be a sensitive marker of perineurial cells and their tumors in a small number of cases. However, GLUT-1 expression has not been systematically examined in other mesenchymal neoplasms. Prompted by a recent report of GLUT-1 expression in epithelioid sarcoma, a tumor not generally felt to show perineurial differentiation, we examined GLUT-1 expression in a wide variety of mesenchymal tumors. Sections from 247 mesenchymal tumors of a variety of histologic subtypes were retrieved from our archives and immunostained for GLUT-1 using heat-induced epitope retrieval and the DAKO ADVANCE detection system (DAKO, Carpinteria, CA). Scoring was as follows: negative (b5% of cells), 1+ (5%-25% of cells), 2+ (25%-50% of cells), and 3+ (N50% of cells). All benign nerve sheath tumors showed a peripheral rim of positive normal perineurial cells, with 2 neurofibromas and 3 schwannomas showing more extensive staining. Three of 4 perineuriomas showed strong GLUT-1 expression. All juvenile hemangiomas were GLUT-1 positive. GLUT-1 expression was also seen in a wide variety of benign and malignant mesenchymal tumors. However, GLUT-1 expression was absent in nonjuvenile hemangioma endothelial tumors and in almost all low-grade lesions that enter the histologic differential diagnosis of perineurial tumors, including low-grade fibromyxoid sarcoma, dermatofibrosarcoma protuberans, and myxofibrosarcoma. We conclude that GLUT-1 expression in mesenchymal tumors is by no means specific for perineurial differentiation, but may instead represent upregulation of this protein within hypoxic zones, secondary to upstream activation of proteins such as hypoxia-inducible factor 1-α Elsevier Inc. All rights reserved. 1. Introduction GLUT-1 is an erythrocyte-type glucose transporter protein and a member of the facilitative cell-surface glucose transporter family which includes 5 other isoforms [1]. It was Corresponding author. address: folpe.andrew@mayo.edu (A. L. Folpe). originally purified from human erythrocyte membranes [2] and has been subsequently identified in the brain capillary endothelium, where it plays a critical role in the transport of glucose across the blood-brain barrier [3]. In addition to its role as a glucose transporter, GLUT-1 is also known to play an important role in the cellular response to hypoxia, as a downstream target of hypoxia-inducible factor 1-α (HIF1-α) [4]. Constitutive GLUT-1 expression has been documented /$ see front matter 2008 Elsevier Inc. All rights reserved. doi: /j.humpath

2 1520 W. A. Ahrens et al. in a variety of normal cell types, including placental trophoblast and perineurial cells [5,6]. Upregulation of GLUT-1 expression, presumably due to enhanced glycolytic metabolism, has recently been shown to be a relatively common feature in various carcinomas [7]. In mesenchymal neoplasms, expression of GLUT-1 has been shown to be a constant feature of juvenile capillary hemangiomas, where its expression is useful in the discrimination of such tumors from various mimics, such as vascular malformations and kaposiform hemangioendothelioma [8-11]. GLUT-1 expression has also been noted in a subset of soft tissue perineuriomas, consistent with its expression in normal perineurium [12-14]. Most recently, Smith et al [15] have documented GLUT-1 expression in a small number of epithelioid sarcomas and have suggested that this observation supports the concept of perineurial differentiation in this otherwise enigmatic sarcoma. We undertook a large, retrospective study of GLUT-1 expression in a broad variety of mesenchymal tumors with the goal of better establishing the range of GLUT-1 expression in mesenchymal tumors. 2. Materials and methods Formalin-fixed, paraffin embedded blocks from 247 well-characterized mesenchymal tumors were retrieved from the archives of Mayo Clinic. These cases included alveolar soft part sarcoma (2 cases), angiomatoid (malignant) fibrous histiocytoma (1 case), angiosarcoma (2 cases), atypical fibroxanthoma (1 case), angiomyolipoma/perivascular epithelioid cell tumor (3 cases), atypical lipoma/ well-differentiated liposarcoma (1 cases), benign fibrous histiocytoma (4 cases), chondroblastoma (1 cases), chondrosarcoma (5 cases), chordoma (13 cases), clear cell sarcoma (2 cases), dermatofibrosarcoma protuberans (1 case), desmoplastic small round cell tumor (2 cases), dermatofibrosarcoma protuberans (9 cases), desmoid-type fibromatosis (7 cases), endometrial stromal sarcoma (3 cases), epithelioid hemangioendothelioma (1 case), epithelioid sarcoma (8 cases), extraskeletal myxoid chondrosarcoma (1 case), Ewing sarcoma (11 cases), fibroma (2 cases), fibrosarcoma (4 cases), gastrointestinal stromal tumor (14 cases), adult capillary hemangioma (1 case), juvenile capillary hemangioma (9 cases), inflammatory myofibroblastic tumor (1 case), leiomyosarcoma (10 cases), lipoma (9 cases), leiomyoma (1 case), liposarcoma (10 cases), low-grade fibromyxoid sarcoma (2 cases), malignant tenosynovial giant cell tumor (1 case), granular cell tumor (1 case), melanoma (6 cases), malignant peripheral nerve sheath tumor (3 cases), myoepithelioma (2 cases), myofibrosarcoma (1 case), myxofibrosarcoma (5 cases), neurofibroma (11 cases), nodular fasciitis (2 cases), perineurioma (4 cases), plexiform fibrohistiocytic tumor(1 case), rhabdomyosarcoma (4 cases), solitary fibrous tumor (11 cases), schwannoma (18 cases), synovial sarcoma (10 cases), osteosarcoma (2 cases), undifferentiated pleomorphic sarcoma (22 cases), and xanthoma/xanthogranuloma (2 cases). The diagnoses for all cases were confirmed on rereview of available histologic sections and immunostains by an experienced soft tissue and bone pathologist (ALF). For immunohistochemistry, GLUT-1 expression was evaluated using a rabbit polyclonal antihuman GLUT-1 antibody at a dilution of 1:200 (DAKO, Carpinteria, CA). Paraffin slides were deparaffinized and placed in a preheated solution (95 C-97 C) of 1 mmol/l EDTA, ph 8.0, for 30 minutes. This pretreatment was accomplished using a vegetable steamer. The slides were placed on a DAKO Autostainer. Slides were incubated with GLUT-1 antibody for 30 minutes. Detection was accomplished using a highly sensitive 2-step polymer system from DAKO, ADVANCE. The chromogen used was DAKO DAB+. All slides were counterstained with hematoxylin and coverslipped for microscopic examination. The antibody conditions had been optimized using normal perineurium and surrounding nonneural tissue as positive and negative controls, respectively. The tumors were scored as negative (b5% of cell positive), 1+ (5%-25% of cells positive), 2+ (25%-50% of cells positive), and 3+ (N50% of cells positive). Only a membranous pattern of positivity was considered to represent true GLUT-1 expression. 3. Results The immunohistochemical results are summarized in Table 1. When available, normal perineurial cells invariably served as a positive internal control. All benign nerve sheath tumors showed at least some GLUT-1 positive cells, presumably representing perineurial cells, with 4 neurofibromas and 4 schwannomas showing more extensive expression in the form of a peripheral rim of positive cells (Fig. 1). Three of 4 perineuriomas showed GLUT-1 expression (the negative case was an epithelial membrane antigen (EMA) positive low-grade perineurial sarcoma). GLUT-1 expression was seen in 5 of 8 epithelioid sarcomas (Fig. 2) and in all chordomas (Fig. 3). All juvenile capillary hemangiomas were uniformly GLUT-1 positive; all other endothelial neoplasms were negative. Among lesions which typically enter the histologic differential diagnosis of perineurioma, such as low-grade fibromyxoid sarcoma, dermatofibrosarcoma protuberans, and myxofibrosarcoma, GLUT-1 expression was confined to only focal positivity in a single myxofibrosarcoma. As detailed in Table 1, GLUT-1 expression was also seen in a wide variety of benign and malignant bone and soft tissue tumors (Fig. 4). GLUT-1 expression was frequently seen in association with foci of spontaneous or therapy-induced tumor cell necrosis, often chiefly in the neoplastic cells in closest proximity to the necrotic zones (Fig. 5). Overall, 87 (35%) of 247 cases showed at least 1+ GLUT-1 expression.

3 GLUT-1 expression in mesenchymal tumors 1521 Table 1 Immunohistochemical results Diagnosis n Negative Total positive (%) Alveolar soft part sarcoma (0) Angiomatoid (malignant) fibrous histiocytoma (0) Angiomyolipoma/other PEComa (0) Angiosarcoma (0) Atypical fibroxanthoma (0) Atypical lipomatous tumor (100) Benign fibrous histiocytoma (25) Chondroblastoma (0) Chondrosarcoma (0) Chordoma (100) Clear cell sarcoma (50) Dermatofibrosarcoma protuberans (0) Desmoplastic small round cell tumor (100) Desmoid-type fibromatosis (29) Endometrial stromal sarcoma (0) Epithelioid hemangioendothelioma (0) Epithelioid sarcoma (63) Extraskeletal myxoid chondrosarcoma (100) Ewing sarcoma (27) Fibrosarcoma (0) Fibroma (0) Adult capillary hemangioma (0) Juvenile capillary hemangioma (100) Granular cell tumor (100) Gastrointestinal stromal tumor (14) Inflammatory myofibroblastic tumor (0) Leiomyosarcoma (40) Liposarcoma (0) Low-grade fibromyxoid sarcoma (0) Lipoma (0) Leiomyoma (0) Melanoma (67) Malignant peripheral nerve sheath tumor (33) Myoepithelioma (50) Myofibrosarcoma (0) Myxofibrosarcoma (20) Neurofibroma (36) Nodular fasciitis (50) Perineurioma (75) Plexiform fibrohistiocytic tumor (100) Rhabdomyosarcoma (75) Schwannoma (22) Solitary fibrous tumor (9) Synovial sarcoma (30) Malignant tenosynovial giant cell tumor (100) Osteosarcoma (0) Undifferentiated pleomorphic sarcoma (60) Xanthoma/xanthogranuloma (50) Total (36) 3.1. Conclusions We conclude that GLUT-1 expression in mesenchymal neoplasms is far more widespread than has been previously recognized and is by no means confined to tumors with perineurial differentiation and juvenile hemangiomas. Our results suggest that GLUT-1 expression may be seen in tumors of a wide variety of mesenchymal lineages, including

4 1522 W. A. Ahrens et al. Fig. 1 Plexiform neurofibroma (A), with strong expression of GLUT-1 in peripherally located perineurial cells (B). Fig. 2 Epithelioid sarcoma (A) with variable GLUT-1 expression (B).

5 GLUT-1 expression in mesenchymal tumors Fig Chordoma (A) with intense, diffuse GLUT-1 expression (B). Fig. 4 Metastatic endometrial stromal sarcoma (A) with GLUT-1 expression in neoplastic cells distant from, but not immediately adjacent to, a small blood vessel ( hypoxic pattern ) (B).

6 1524 W. A. Ahrens et al. Fig. 5 Undifferentiated pleomorphic sarcoma (A). GLUT-1-positive cells are present at the interface between viable and necrotic tumor (B). (but not limited to) chordomas, epithelioid sarcomas, gastrointestinal stromal tumors, adipocytic tumors, myogenous tumors, and undifferentiated pleomorphic sarcomas. In the area of mesenchymal neoplasia, GLUT-1 expression was first described, by North et al [8], as a useful marker in the discrimination of juvenile hemangiomas (GLUT-1 positive) from other tumors in its differential diagnosis, including vascular malformations, pyogenic granulomas, kaposiform hemangioendotheliomas, and epithelioid hemangioendotheliomas (all GLUT-1 negative). A subsequent larger series, by Lyons et al [16], also observed absent GLUT-1 expression in kaposiform hemangioendothelioma, reinforcing the essential biologic differences between these histologically somewhat similar-appearing lesions. Our findings of GLUT-1 expression in all studied juvenile capillary hemangiomas and in no other vascular tumors are in agreement with these previous results. Somewhat more recently, GLUT-1 has been promoted as an alternative marker to EMA and claudin-1 for the diagnosis of perineurial tumors [12]. Using immunohistochemistry, Hirose et al [12] demonstrated GLUT-1 expression in perineurial cells forming the capsule of all schwannomas, in isolated perineurial cells in 90% of neurofibromas, in 100% of benign perineuriomas, in 2 of 6 conventional-type malignant peripheral nerve sheath tumor (in perineurial-appearing cells forming whorls and perineurium-like structures), and in 4 of 4 perineurialtype malignant peripheral nerve sheath tumor. GLUT-1 expression has also been noted in 5 of 5 sclerosing perineuriomas [13] and in 2 of 2 reticular perineuriomas [14]. Our findings of GLUT-1 expression in the majority of studied perineuriomas, in the perineurial cells of neurofibromas, and in the normal perineurial cells in essentially all studied nerve sheath tumors are in agreement with these previous studies. With the exception of the present study, we are aware of only 2 studies which have evaluated GLUT-1 in other types of mesenchymal tumors. Rao et al [17], in a study of uterine and extra-uterine smooth muscle tumors, noted GLUT-1 expression in 25% to 50% of uterine and extra-uterine leiomyosarcomas, but not in leiomyomas. Most recently, Smith et al [15,18] have shown GLUT-1 expression and a perineurial-like pattern of cadherin expression in epithelioid sarcomas, findings which suggested to these authors the possibility of perineurial differentiation in this distinctive sarcoma. Although the present study confirms that GLUT-1 expression is relatively common in epithelioid sarcoma, we believe it is premature to ascribe perineurial differentiation to such tumors on the basis of this finding. Indeed, as discussed below, we suspect that GLUT-1 expression in epithelioid sarcoma is simply a reflection of intratumoral hypoxia, a concept consistent with the wellknown tendency of epithelioid sarcoma to undergo spontaneous necrosis. In addition to its role as a glucose transporter in tissues with presumed barrier functions, GLUT-1 is known to be

7 GLUT-1 expression in mesenchymal tumors involved in the cellular response to hypoxia, as a downstream target of HIF1-α [4]. GLUT-1 upregulation and subsequent overexpression of GLUT-1 receptors on the plasma membrane of various tumor cells are thought to allow escape from the apoptosis-inducing effects of a hypoxic environment [19]. Indeed, in epithelial malignancies, GLUT- 1 expression has been noted in 25% to 100% of carcinomas of urothelial, breast, colonic, pancreaticobiliary, esophageal, renal, pulmonary, and ovarian surface epithelial origin, among others [7]. Previous studies have also noted an apparent gradient of GLUT-1 expression in epithelial malignancies between presumably hypoxic (ie, adjacent to necrosis) and normoxic zones, similar to our findings in mesenchymal neoplasms [20-26]. Our finding of frequent GLUT-1 expression in sarcomas of various types, frequently in association with spontaneous or therapy-related necrosis, suggests that GLUT-1 expression in mesenchymal tumors other than typical perineuriomas and juvenile hemangiomas may be simply a reflection of intratumoral hypoxia, rather than a lineage-restricted phenomenon. It is thus likely that the diagnostic uses of GLUT-1 in the evaluation of mesenchymal neoplasms are quite limited. Certainly, GLUT-1 immunostaining should continue to play a valuable role in the differential diagnosis of juvenile hemangioma from other vascular tumors with which it may be confused, such as kaposiform hemangioendothelioma. There may be some use for GLUT-1 in the evaluation of suspected perineurial lesions, particularly when EMA expression is very limited or absent. In this respect, it is worth noting that those tumors that are most likely to simulate perineurioma, such as low-grade fibromyxoid sarcoma and dermatofibrosarcoma protuberans, seldom, if ever, show GLUT-1 expression. It is unlikely that GLUT-1 immunostaining should play a role in the diagnosis of epithelioid sarcoma, despite the high frequency of GLUT-1 expression in this tumor. We base this view in part on our anecdotal experience that GLUT-1 expression may also be seen in essentially all of the superficial tumors that enter the differential diagnosis of epithelioid sarcoma, including squamous cell carcinoma, granuloma annulare, and rheumatoid nodule (data not shown), and in part on the widespread availability of much more specific reagents, such as antibodies to cytokeratins, CD34, and INI-1 protein. Finally, there may be some role for GLUT-1 in the distinction of chordoma (invariably GLUT-1 positive) and chondrosarcoma (GLUT-1 negative), particularly in the setting of skull base tumors, where material may be very limited. Compensatory GLUT-1 expression in the setting of hypoxia may also in part underlie the known radioresistance of chordoma. References [1] Mueckler M. Facilitative glucose transporters. Eur J Biochem 1994;219: [2] Kasahara M, Hinkle PC. Reconstitution and purification of the D-glucose transporter from human erythrocytes. J Biol Chem 1977;252: [3] Pardridge WM, Boado RJ, Farrell CR. Brain-type glucose transporter (GLUT-1) is selectively localized to the blood-brain barrier. Studies with quantitative western blotting and in situ hybridization. J Biol Chem 1990;265: [4] Mayer A, Hockel M, Vaupel P. Endogenous hypoxia markers in locally advanced cancers of the uterine cervix: reality or wishful thinking? Strahlenther Onkol 2006;182: [5] Takata K, Hirano H, Kasahara M. Transport of glucose across the blood-tissue barriers. Int Rev Cytol 1997;172:1-53. [6] Harik SI, Kalaria RN, Andersson L, et al. Immunocytochemical localization of the erythroid glucose transporter: abundance in tissues with barrier functions. J Neurosci 1990;10: [7] Younes M, Lechago LV, Somoano JR, et al. Wide expression of the human erythrocyte glucose transporter Glut1 in human cancers. Cancer Res 1996;56: [8] North PE, Waner M, Mizeracki A, et al. GLUT1: a newly discovered immunohistochemical marker for juvenile hemangiomas. HUM PATHOL 2000;31: [9] Leon-Villapalos J, Wolfe K, Kangesu L. GLUT-1: an extra diagnostic tool to differentiate between haemangiomas and vascular malformations. Br J Plast Surg 2005;58: [10] Mo JQ, Dimashkieh HH, Bove KE. GLUT1 endothelial reactivity distinguishes hepatic infantile hemangioma from congenital hepatic vascular malformation with associated capillary proliferation. HUM PATHOL 2004;35: [11] Hernandez F, Navarro M, Encinas JL, et al. The role of GLUT1 immunostaining in the diagnosis and classification of liver vascular tumors in children. J Pediatr Surg 2005;40: [12] Hirose T, Tani T, Shimada T, et al. Immunohistochemical demonstration of EMA/Glut1-positive perineurial cells and CD34-positive fibroblastic cells in peripheral nerve sheath tumors. Mod Pathol 2003;16: [13] Yamaguchi U, Hasegawa T, Hirose T, et al. Sclerosing perineurioma: a clinicopathological study of five cases and diagnostic utility of immunohistochemical staining for GLUT1. Virchows Arch 2003;443: [14] Mentzel T, Kutzner H. Reticular and plexiform perineurioma: clinicopathological and immunohistochemical analysis of two cases and review of perineurial neoplasms of skin and soft tissues. Virchows Arch 2005;447: [15] Smith ME, Awasthi R, O'Shaughnessy S, et al. Evaluation of perineurial differentiation in epithelioid sarcoma. Histopathology 2005;47: [16] Lyons LL, North PE, Mac-Moune Lai F, et al. Kaposiform hemangioendothelioma: a study of 33 cases emphasizing its pathologic, immunophenotypic, and biologic uniqueness from juvenile hemangioma. Am J Surg Pathol 2004;28: [17] Rao UN, Finkelstein SD, Jones MW. Comparative immunohistochemical and molecular analysis of uterine and extrauterine leiomyosarcomas. Mod Pathol 1999;12: [18] Smith ME, Brown JI, Fisher C. Epithelioid sarcoma: presence of vascular-endothelial cadherin and lack of epithelial cadherin. Histopathology 1998;33: [19] Lin Z, Weinberg JM, Malhotra R, et al. GLUT-1 reduces hypoxiainduced apoptosis and JNK pathway activation. Am J Physiol Endocrinol Metab 2000;278:E [20] Kalir T, Rahaman J, Hagopian G, et al. Immunohistochemical detection of glucose transporter GLUT1 in benign and malignant fallopian tube epithelia, with comparison to ovarian carcinomas. Arch Pathol Lab Med 2005;129: [21] Page T, Hodgkinson AD, Ollerenshaw M, et al. Glucose transporter polymorphisms are associated with clear-cell renal carcinoma. Cancer Genet Cytogenet 2005;163: [22] Alo PL, Visca P, Botti C, et al. Immunohistochemical expression of human erythrocyte glucose transporter and fatty acid synthase in

8 1526 W. A. Ahrens et al. infiltrating breast carcinomas and adjacent typical/atypical hyperplastic or normal breast tissue. Am J Clin Pathol 2001;116: [23] Younes M, Brown RW, Stephenson M, et al. Overexpression of Glut1 and Glut3 in stage I nonsmall cell lung carcinoma is associated with poor survival. Cancer 1997;80: [24] Urano N, Fujiwara Y, Doki Y, et al. Overexpression of hypoxiainducible factor-1 alpha in gastric adenocarcinoma. Gastric Cancer 2006;9:44-9. [25] Younes M, Juarez D, Lechago LV, et al. Glut 1 expression in transitional cell carcinoma of the urinary bladder is associated with poor patient survival. Anticancer Res 2001;21: [26] Burstein DE, Nagi C, Kohtz DS, et al. Immunodetection of GLUT1, p63 and phospho-histone H1 in invasive head and neck squamous carcinoma: correlation of immunohistochemical staining patterns with keratinization. Histopathology 2006;48: