Virchows Arch (2007) 450:463–470 DOI 10.1007/s00428-007-0378-7
ORIGINAL ARTICLE
Perivascular epithelioid cell neoplasms (PEComas): four malignant cases expanding the histopathological spectrum and a description of a unique finding Ilan Weinreb & David Howarth & Eleanor Latta & Danny Ghazarian & Runjan Chetty
Received: 18 December 2006 / Revised: 24 January 2007 / Accepted: 26 January 2007 / Published online: 15 February 2007 # Springer-Verlag 2007
Abstract Four cases of malignant PEComas were stained with smooth muscle actin, muscle specific actin, desmin, human melanoma black-45, melan-A, microphthalmia transcription factor, S100 and cyclin D1. One case was studied by electron microscopy (EM). Tumour locations were the thigh, elbow, retroperitoneum and bladder in association with a urachal cyst. There were two men and two women; the average age was 51.3 years, and the size ranged from 5.0–23.0 cm. In three cases, 50–95% of the tumour was composed of pleomorphic sarcomatous areas. All cases had at least focal clear-cell areas. One case showed a continuous single layer of perivascular clear cells remote from the tumour, transitioning to invasive nests and to PEComa. EM demonstrated these cells in apposition to and in direct contact with the abluminal surface of the basal lamina of the capillaries. We suggest the term “pecosis” for these areas. All cases were positive for two or more melanocytic markers and for at least one actin. S100 and I. Weinreb (*) : D. Ghazarian : R. Chetty Department of Pathology, University Health Network, 200 Elizabeth Street, 11th Floor, Toronto, ON M5G 2C4, Canada e-mail:
[email protected] D. Howarth Department of Pathology and Laboratory Medicine, Mount Sinai Hospital, Toronto, ON, Canada E. Latta Department of Pathology, St. Michael’s Hospital, Toronto, ON, Canada I. Weinreb : D. Howarth : E. Latta : D. Ghazarian : R. Chetty Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON, Canada
desmin were focally positive in one case. Cyclin D1 was positive in 3:4 cases. Four cases of malignant PEComa are described with the existence of a unique lesion (pecosis) in one case. These tumours may manifest largely as sarcomas appearing to be undifferentiated and should be considered in their differential diagnosis. Keywords Pecoma . Myomelanocytic tumour . Perivascular epithelioid cell . Pecosis
Introduction The perivascular epithelioid cell tumour (PEComa) family is a well-accepted and distinctive group, initially described by Bonetti et al. [5] who also first proposed the theoretical cell of origin [16, 23]. This terminology applies to angiomyolipomas of the kidney and liver, clear cell sugar tumours, lymphangioleiomyomatosis of the lung and tumours of the soft tissue and viscera with either spindled or epithelioid morphology and dual myomelanocytic differentiation [14–16]. The growing list of sites of occurrence include gastrointestinal, gynecological, genitourinary, extremities and the skin, as well as single reports in the oral cavity, skull base, orbit and prostate [16, 23, 24, 27, 28, 32]. The putative cell of origin (perivascular epithelioid cell) has not been identified to date in normal tissues, nor has a precursor lesion been described. Malignant PEComas are being increasingly reported as well [4, 7, 8, 10, 11, 13, 16, 21, 22, 28, 32, 33]. Folpe et al. [16] suggested criteria for malignancy including a size of >8.0 cm, mitotic count of >1 per 50 high power fields (HPFs) and necrosis, and these help to stratify PEComas into benign, uncertain malignant potential and malignant. Occasional cases with overt sarcomatous areas have been described and mim-
464
Virchows Arch (2007) 450:463–470
icked other malignancies, and these may, therefore, escape recognition [1]. To study these issues further, we examined four malignant PEComas with overt sarcomatous areas that mimicked other pleomorphic sarcomas. We also report the first example of a PEComa arising in a urachal remnant. This case had a unique lesion remote from the tumour, for which we suggest the term “pecosis” and discuss its possible implications. We also examine the role of cyclin D1 over-expression by immunohistochemistry, which has been reported previously in a single example of a malignant PEComa of the colon [43].
Materials and methods A search of the computer records of our archives for malignant cases belonging to the “PEComa” family was undertaken. All cases listed as malignant examples of PEComa, angiomyolipoma, clear-cell sugar tumour and myomelanocytic tumour were collected. Only cases that were histologically malignant (according to the criteria established by Folpe et al. [16]) were included. All four surgical specimens were fixed in neutral buffered formalin. Hematoxylin and eosin stains were performed on 3- to 4-μ thick sections of formalin-fixed paraffin-embedded tissue. Immunohistochemical stains were performed in sections from one or more representative paraffin blocks from each case, using the ultra-streptavidin–horseradish peroxidase (HRP) detection system (ID Labs Biotechnology, London, Ontario, Canada). Colour development was performed using the NovaRed substrate kit (Vector Labs, Burlingame, California, USA). The following antibodies were employed: AE1/AE3 (Dako, 1:800), EMA (Ventana, prediluted), smooth muscle actin (SMA; Dako, 1:200), muscle specific actin (MSA; Dako, 1:100), desmin (Dako, 1:200), S100 (Dako, 1:200), melan-A (Ventana, prediluted), human melanoma black-45 (HMB-45; Ventana, prediluted), microphthalmia transcription factor (MiTF; Neomarkers, 1:50) and cyclin D1 (Neomarkers, 1:80). Immunopositivity was assessed as follows: +++, >75% of tumour cells labelling; ++, 50–75% of tumour cells labelling; +, 25– 50% of tumour cells labelling; focal +, 0–25% of tumour cells labelling.
Electron microscopy was performed on one case with the areas of “pecosis” on the tissue extracted from a paraffin block (urachal case). Molecular analysis on one case was performed from the paraffin-embedded tissue to exclude the presence of the t(12;22) translocation present in clear-cell sarcoma of the tendons and aponeuroses.
Results Clinical findings The clinical findings are summarised in Table 1. There were four patients in total, two men and two women with an average age of 51.3 years and an age range of 39–69 years. They were found in the elbow, retroperitoneum, a urachal remnant and the left thigh. The tumours ranged in size from 5.0 to 23.0 cm with an average size of 10.7 cm. Follow-up information was available on three of the cases, with one patient lost to follow-up. Of these three cases, two are alive and well with no evidence of disease 8 and 11 months after resection. One patient with the retroperitoneal mass is alive with brain and lung metastases 4 years after surgery. No patient had the tuberous sclerosis complex or a current or subsequent history of melanoma. Histopathologic findings All four cases showed sheets of cells arranged in fascicles, nests or both and had infiltrative borders with invasion of the surrounding soft tissues. Focal areas of perivascular accentuation were seen in all cases and, in some cases, was a prominent feature. The cells ranged from small epithelioid and spindle cells to large pleomorphic malignant giant cells. The cytoplasm varied from clear to eosinophilic and granular. Melanin pigment was seen in one case by H&E alone and by melanin special staining (Fontana–Masson) in a second. Three cases showed dense stromal hyalinization. Case 1 (retroperitoneal) had >95% epithelioid and giant cell morphology, reminiscent of a myogenic sarcoma or undifferentiated malignant tumour and was noted grossly to be wrapping around the ureter, but not invading it. The tumour showed nested architecture with a fine fibrovascular compartmentalization. There were focal areas of perivas-
Table 1 Clinicopathological features Case
Age
Gender
Site
1 2 3 4
49 68 49 39
F M F M
Retroperitoneal Left thigh Elbow Urachal
Size (cm) 7.0 7.8 5.0 23.0
AWD Alive with disease, NED no evidence of disease recurrence
Histologic pattern
Follow-up
Myogenic sarcoma-like Dedifferentiation Fibrosarcoma-like Myomelanocytic
AWD (brain and lung metastases) NED×11 months Lost to follow-up NED×8 months
Virchows Arch (2007) 450:463–470
cular epithelioid clear cells (Fig. 1a). Marked pleomorphism was noted; the mitotic count was 28 mitotic figures (MFs) per 10 HPFs including atypical forms, and there were large areas of geographic necrosis. A large thickwalled vein identified as the inferior vena cava was noted to have a large tumour embolus within it. Case 2 (left thigh), which grossly involved and surrounded a large neurovascular bundle, showed a predomiFig. 1 a A pleomorphic sarcomatous PEComa mimicking a myogenic sarcoma, containing focal perivascular clear cell areas (inset), which is an important clue to the diagnosis (case 1; 200×). b Geographic necrosis in the case arising in the thigh (case 2; 50×). c The same case as in b (case 2) demonstrating an abrupt transition of clear cell areas to a more undifferentiated, sarcoma pattern (100×). Inset, higher power of clear cell areas. d Case 3, arising in the elbow with a fibrosarcoma-like pattern representing 50% of the tumour (200×). e The same case (case 3) with clear cell areas within hyalinized fibrous tissue representing the other 50% of the tumour arising in the elbow (200×). f Longitudinal section of a capillary involved by “pecosis” in case 4. Arrows point to the subtle clear cells lining the vascular space that correspondingly stain for HMB-45. Note the cytoplasmic melanin pigment at the top right corner of image (400×). g The same case demonstrating an additional example of “pecosis” within the urachal cyst wall (630×). h The urachal PEComa (case 4) showing infiltration of the bladder wall by clear cells in the top half of the picture. Note the uninvolved urachal remnant at the bottom half of the picture (100×). Inset, high power of classic PEComa
465
nant spindled eosinophilic and clear cell pattern. There were large areas of necrosis with the preservation of cells nearest the blood vessels (Fig. 1b). Little-to-no-cellular pleomorphism was evident in most of the tumour, but there was a high mitotic count of 42 MFs per 10 HPFs. In addition, there was an abrupt transition from the entirely clearspindled areas to an undifferentiated patternless sarcoma with prominent intra-tumoural chronic inflammation
466
Virchows Arch (2007) 450:463–470
(Fig. 1c). The post radiotherapy resection specimen showed mainly the undifferentiated sarcoma pattern with involvement and occlusion of the superficial femoral vein. Case 3 (elbow) showed a tumour-infiltrating skeletal muscle and fibroadipose tissue in a fascicular and herringbone, “fibrosarcoma-like” pattern (Fig. 1d). In addition, there were areas of dense hyalinization and a single file arrangement of epithelioid clear cells entrapped within the fibrous tissue in approximately 50% of the tumour (Fig. 1e). The mitotic count was 20 MFs per 10 HPFs including atypical forms, and there was focal single-cell necrosis. Multi-nucleated malignant giant cells were noted within the tumour, and there was focal vascular invasion by tumour cells. In case 4 (urachal cyst), the umbilicus and the dome of the bladder were resected in continuity with a 23.0-cm cystic lesion. Within the cyst was a 5.0-cm mural mass. Several centimeters away from the main tumour mass, at the umbilical end of the specimen, there were isolated capillaries surrounded by a single layer of clear cells (pecosis; Fig. 1f,g). These clear cells showed fine dusty melanin pigment in their cytoplasm. They were noted to have little to no nuclear atypia and no nucleoli. The involved capillaries merged to form complex interconnecting vascular spaces that were also surrounded by a single continuous layer of clear cells. There was no invasion between these individual vascular channels, but the lesion gradually merged with focal invasive nests within the stroma of the cyst wall. The invasive nests converged to form the mural tumour mass with the classical clear, spindled morphology of a “myomelanocytic tumour” (Fig. 1h). At the bladder end of the cyst, the tumour showed focal evidence of dedifferentiation: discohesive, pleomorphic eosinophilic cells containing larger nuclei and prominent nucleoli, focal single cell necrosis and isolated mitotic figures. No vascular invasion was noted.
(Fig. 2a). All but one of the cases were S100 negative; the positive case occurred in the urachal remnant and was patchy (20% of cells labelling). This case was t(12;22) negative (see later). All cases showed actin positivity with three of the cases showing multiple myoid marker positivity as well (Fig. 2b). The most frequent and strongest staining myoid marker present was SMA. A single case was focally positive for desmin. The case with the “pecosis” showed strong HMB-45 and MiTF positivity of the clear cells in apposition to the thin endothelial lined spaces (Fig. 2c). Only focal keratin or EMA staining was seen (one case each). Cyclin D1 staining was present in three of the four cases, with strong nuclear positivity in the tumour cells, suggesting its over-expression. In case 2 (thigh) where there was an abrupt transition to an undifferentiated sarcoma pattern, the cyclin D1 staining was absent in the clear-cell portion of the tumour and present in the area of dedifferentiation. All other positive tumours showed diffuse staining. The retroperitoneal tumour was cyclin D1 negative. The urachal tumour showed cyclin D1 staining in the main tumour mass but not in the peri-capillary “pecosis”. The degree of cyclin D1 staining increased from the early invasive nests to the main tumour mass. Ultra-structural analysis Electron microscopy was performed on the “pecosis” and mural mass of case 4 from the paraffin-embedded tissue. It showed single perivascular epithelioid cells surrounding endothelial-lined vascular channels in a single continuous layer (Fig. 3a). These cells were found to be in direct apposition to the abluminal surface of the basement membrane of these capillaries with no stroma in-between (Fig. 3b). No invasive nests were seen in-between these capillaries. The invasive areas in the mural mass showed primitive cellular junctions, but no melanosomes or premelanosomes were seen.
Immunohistochemical findings Molecular analysis The immunohistochemical profile of the four cases is summarised in Table 2. All four cases showed strong and diffuse staining with at least two melanocytic markers
Case 4 (focally S100 positive) was also tested by RT-PCR for the recurrent t(12;22) translocation (EWS/ATF-1 fusion
Table 2 Immunohistochemical results Case
SMA
MSA
Desmin
Melan-A
HMB-45
S-100
MiTF
Cyclin D1
1 2 3 4
Focal + +++ ++ +++
− ++ Focal + ++
− − − Focal +
+++ +++ ++ Focal +
++ − ++ +++
− − − Focal +
Not done +++ +++ +++
− +++a +++ +++b
+++, >75% of tumour cells labelling; ++, 50–75% of tumour cells labelling; +, 25–50% of tumour cells labelling; focal +, 0–25% of tumour cells labelling a Cyclin D1 staining in undifferentiated areas only b Cyclin D1 staining in tumour but not in “pecosis”
Virchows Arch (2007) 450:463–470
467
Fig. 3 a Electron microscopy demonstrating “pecosis” within the urachal cyst wall. Note the PEC cells surrounding the capillary in a single layer. b Higher power image of the same capillary showing the cells in direct apposition to the basal lamina
transcript seen in clear cell sarcoma/melanoma of soft parts) and was negative.
Discussion
Fig. 2 a Melan-A staining of the retroperitoneal tumour (case 1) with an increase in staining intensity noted in a perivascular locale (400×). b Strong SMA staining of the elbow PEComa (case 3; 400×). c HMB45 staining numerous capillaries within the cyst wall containing foci of “pecosis” in case 4. Note the perivascular staining and the lack of invasive tumour nests in-between capillaries (100×)
It appears clear that the PEComa family of tumours represent a unique and distinct group arising from the perivascular epithelioid cell [16, 23, 31]. The first suggestion of this possible origin is credited to Bonetti et al. [5, 6, 35, 36], based on a series of observations in clear-cell sugar tumours, angiomyolipomas and lymphangioleiomyomatosis of the lung. The perivascular accentuation of tumour and the intimate association with vessel walls is a prominent feature and a useful clue to the diagnosis. The differential diagnosis includes leiomyosarcoma, malignant melanoma, clear-cell sarcoma of tendons and aponeuroses and any other tumour with focal or prominent clear-cell change.
468
PEComas have now been reported in almost every body site including visceral and somatic soft tissues [23]. The majority has been reported in women, and only a few are associated with tuberous sclerosis [12, 16, 23, 31, 41]. Malignant PEComas are being increasingly reported as well [4, 7, 8, 10, 11, 13, 16, 21, 22, 28, 32, 33]. Criteria for malignancy have been suggested recently and include size greater than 8.0 cm, mitotic count >1 per 50 HPFs and necrosis, with benign, uncertain malignant potential and malignant categories based on zero, one or two of these criteria, respectively [16]. Infiltrative edges, cellularity and nuclear pleomorphism may be secondary features suggesting aggressive behaviour [16, 23]. An occasional case may even have no features of malignancy and still show clinical aggressiveness [16]. No case described in the skin has behaved aggressively to date, suggesting that site may also determine behaviour in PEComas [9, 23, 30]. An occasional case may have frankly sarcomatous areas and mimic other sarcomas [1]. All four malignant cases in this series contained at least focal clear-cell areas, and three had a large sarcomatous component. These latter cases were initially suspected to be other sarcomas, but based on the clear cell areas, malignant PEComas were considered. Myomelanocytic marker expression was found to be prominent, including in the sarcomatous areas. Controversy exists regarding the minimum criteria for the diagnosis of malignant PEComa [38]. In a study of uterine tumours demonstrating epithelioid morphology, clear cell areas and HMB-45 positivity, Silva et al. [38] argued that this group represented smooth muscle tumours with focal melanocytic differentiation and not PEComas. This was based largely on the fact that the tumours looked like typical leiomyosarcomas with spindled and epithelioid areas and stained with desmin. It is also noteworthy that only focal HMB-45 staining was seen, unlike the diffuse expression in our cases. More than 30% of PEComas showed at least focal desmin staining in the study by Folpe et al. [16], and this does not exclude the diagnosis of PEComa. Other authors have reported uterine, cervical and broad ligament tumours of the kind Silva et al. described as PEComas, and at least two cases were associated with tuberous sclerosis [2, 8, 10, 12, 13, 17– 21, 25, 26, 41]. We suggest that tumours with diffuse myomelanocytic differentiation should be regarded as being related to the PEComa family irrespective of site of origin or desmin positivity. It is the characteristic immunophenotype that warrants separating these tumours from classic leiomyosarcoma that exhibit only muscle differentiation. In addition to leiomyosarcoma, other malignant neoplasms have also to be considered. The most important of these (melanoma and clear cell sarcoma) can be differentiated based on S100 positivity; however, up to 18% of
Virchows Arch (2007) 450:463–470
PEComas express S100 as well [16]. Important clues to the diagnosis of PEComa in this context include perivascular accentuation of tumour cells, actin positivity, negative history for melanoma, visceral location of tumour and absence of the t(12:22) translocation. Although S100 negative melanomas have rarely been described, we feel tumours with diffuse melanocytic marker positivity, S100 negativity and actin immunoreactivity are PEComas based on morphology and immunophenotype. Despite this, an occasional case will still defy classification. The single case in our series with patchy S100 (20% cells staining) expression showed typical histology for a PEComa in most of the tumour, was diffusely actin positive, lacked a history of melanoma, had a visceral location and lacked the t (12;22) translocation. Malignant PEComas that are entirely undifferentiated may also exist and, in the absence of the aforementioned features, will escape recognition. Such lesions warrant a search for conspicuous clear cell areas and perivascular growth, as well as being investigated with a panel of myomelanocytic markers if the histological appearance is appropriate. However, purely undifferentiated sarcomas without clear cell areas do not warrant additional investigation. Pitfalls in this diagnosis include aberrant staining of cells with melanocytic markers. In our experience, diffuse and multiple melanocytic marker expression is highly reliable for melanocytic differentiation. Focal or weak positivity can be disregarded, however, and do not warrant the diagnosis of PEComa. All of our cases were diffusely and strongly positive with at least two melanocytic markers (excluding S100), which is not a finding in other sarcomas in our experience. Other modalities such as cytogenetics may help to refine these diagnostic categories in the future. Cytogenetic alterations have been described in PEComas, including losses on chromosome 16p, the location of the tuberous sclerosis complex 2 gene (TSC2) [31]. There is no known precursor lesion for PEComas, and the so-called PEC has not been identified in normal tissue. The presumed perivascular location and actin positivity of the cells may argue for a pericyte origin, but pericytic differentiation has never been suggested, to our knowledge, in ultra-structural studies. We report the first description of one case with a perivascular lesion, for which we propose the term “pecosis”, arising in a urachal cyst. In addition, we have encountered a benign PEComa located in the skin that also exhibited this phenomenon. The “pecosis” showed a gradual transition to an invasive PEComa with the usual features and, focally, to a poorly differentiated malignant neoplasm. This may imply that “pecosis” was a precursor to the PEComa in this case. Bladder PEComas have been reported previously [31, 34], but this is the first case specifically proven to arise within a urachal cyst and
Virchows Arch (2007) 450:463–470
display “pecosis” remote to the main tumour. An alternative explanation for “pecosis” may be malignant tumour extension along capillaries away from the tumour rather than a precursor. Although this is a possibility, the lack of atypia in these cells (Fig. 1f ) compared to the main tumour, lack of cyclin D1 expression (positive in tumour) and lack of any invasive nests in these areas argue against this. More cases showing “pecosis” need to be identified to resolve this issue. If “pecosis” is found to be a precursor lesion, it is still possible that most PEComas arise de novo and not from a precursor, and that this is a rare phenomenon. The term “pecosis” should not be confused with “pecomatosis”, a term first applied to a single case in the literature that demonstrated numerous aggregates of PEC cells in the lamina propria of small bowel, myometrium and ovarian hilum in a tuberous sclerosis patient with a cervical PEComa [12]. These aggregates were felt not to be metastases of the cervical tumour, but rather a “field effect” in the abdomen. These pecomatosis aggregates were not confined to the walls of capillaries in a single continuous layer and are best regarded as microscopic benign PEComas. Other than the tuberous sclerosis connection that does not appear to be important in the pathway to most PEComas, little is known of genes that may be involved in the pathogenesis of these tumours. A single malignant case in the colon was shown to over-express cyclin D1 by immunohistochemistry [43]. Cyclin D1 is a cell cycle regulator necessary for the transition through the G1→S checkpoint [3, 37, 42]. It is normally expressed transiently and may be constitutively expressed in many neoplasms including melanoma [40]. Soucek et al. [39] have shown cyclin D1 abnormalities associated with the loss of the TSC2 gene. Two of the three cyclin D1 positive cases in our series showed a diffuse expression in both classical and malignant areas, whereas one was positive only in the sarcomatous areas. If cyclin D1 has a role in the genesis of PEComa, it is not well understood and it warrants further study. In conclusion, we have described four additional cases of malignant PEComa, all of which had pleomorphic areas and three of which showed significant sarcomatous areas. Malignant PEComa should be considered in the light microscopic differential diagnosis for all pleomorphic sarcomas. The recent report of focal HMB-45 positivity in undifferentiated soft tissue sarcomas should not be regarded as malignant PEComas in our opinion, and reflex melanocytic marker staining of these tumours is unwarranted [29]. At least, focal light microscopic suspicion of PEComa and diffuse myomelanocytic marker positivity is required to support this diagnosis. We also describe the first PEComa arising in a urachal remnant. One case of “pecosis”, a lesion of uncertain etiology is described. Finally, cyclin D1 was
469
strongly expressed in three of the four cases and may have a role in the genesis of these tumours.
References 1. Agaimy A, Wunsch PH (2006) Perivascular epithelioid cell sarcoma (malignant PEComa) of the ileum. Pathol Res Pract 202:37–41 2. Azad NS, Aziz AB, Pervez S, Kayani N (2006) Uterine perivascular epithelioid cell tumour presenting as a cervical mass. J Pak Med Assoc 56:83–84 3. Beijersbergen RL, Bernards R (1996) Cell cycle regulation by the retinoblastoma family of growth inhibitory proteins. Biochim Biophys Acta 1287:103–120 4. Bonetti F, Martignoni G, Colato C, Manfrin E, Gambacorta M, Faleri M, Bacchi C, Sin VC, Wong NL, Coady M, Chan JK (2001) Abdominopelvic sarcoma of perivascular epithelioid cells. Report of four cases in young women, one with tuberous sclerosis. Mod Pathol 14:563–568 5. Bonetti F, Pea M, Martignoni G, Zamboni G (1992) PEC and sugar. Am J Surg Pathol 16:307–308 6. Bonetti F, Pea M, Martignoni G, Zamboni G, Iuzzolino P (1991) Cellular heterogeneity in lymphangiomyomatosis of the lung. Hum Pathol 22:727–728 7. Chen IY, Yang SF, Chen FM, Chai CY (2005) Abdominopelvic perivascular epithelioid cell tumor with overt malignancy: a case report. Kaohsiung J Med Sci 21:277–281 8. D’Andrea V, Lippolis G, Biancari F, Ruco LP, Marzullo A, Wedard BM, Di Matteo FM, Sarmiento R, Dibra A, De Antoni E (1999) A uterine pecoma: a case report. G Chir 20:163–164 9. de Saint Aubain Somerhausen N, Gomez Galdon M, Bouffioux B, Courtin C, Theunis A, Vogeleer MN, Myant N (2005) Clear cell ‘sugar’ tumor (PEComa) of the skin: a case report. J Cutan Pathol 32:441–444 10. Dimmler A, Seitz G, Hohenberger W, Kirchner T, Faller G (2003) Late pulmonary metastasis in uterine PEComa. J Clin Pathol 56:627–628 11. Evert M, Wardelmann E, Nestler G, Schulz HU, Roessner A, Rocken C (2005) Abdominopelvic perivascular epithelioid cell sarcoma (malignant PEComa) mimicking gastrointestinal stromal tumour of the rectum. Histopathology 46:115–117 12. Fadare O, Parkash V, Yilmaz Y, Mariappan MR, Ma L, Hileeto D, Qumsiyeh MB, Hui P (2004) Perivascular epithelioid cell tumor (PEComa) of the uterine cervix associated with intraabdominal “PEComatosis”: A clinicopathological study with comparative genomic hybridization analysis. World J Surg Oncol 19:35 13. Fink D, Marsden DE, Edwards L, Camaris C, Hacker NF (2004) Malignant perivascular epithelioid cell tumor (PEComa) arising in the broad ligament. Int J Gynecol Cancer 14:1036–1039 14. Folpe AL, Goodman ZD, Ishak KG, Paulino AF, Taboada EM, Meehan SA, Weiss SW (2000) Clear cell myomelanocytic tumor of the falciform ligament/ligamentum teres: a novel member of the perivascular epithelioid clear cell family of tumors with a predilection for children and young adults. Am J Surg Pathol 24:1239–1246 15. Folpe AL, McKenney JK, Li Z, Smith SJ, Weiss SW (2002) Clear cell myomelanocytic tumor of the thigh: report of a unique case. Am J Surg Pathol 26:809–812 16. Folpe AL, Mentzel T, Lehr HA, Fisher C, Balzer BL, Weiss SW (2005) Perivascular epithelioid cell neoplasms of soft tissue and gynecologic origin: a clinicopathologic study of 26 cases and review of the literature. Am J Surg Pathol 29:1558–1575 17. Fukunaga M (2004) Perivascular epithelioid cell tumor of the uterus: a case report. Int J Gynecol Pathol 23:287–291
470 18. Fukunaga M (2005) Perivascular epithelioid cell tumor of the uterus: report of four cases. Int J Gynecol Pathol 24:341–346 19. Gan MF, Jin M, Yu CK, Cai JF (2006) Perivascular epithelioid cell tumor (PEComa) of the uterine. Zhonghua Binglixue Zazhi 35:314–315 20. Gao Z, Bhuiya T, Anderson A (2004) Perivascular epithelioid cell tumour (PEComa) of the uterus associated with malignant neoplasm of the female genital tract. J Obstet Gynaecol 24:600–604 21. Greene LA, Mount SL, Schned AR, Cooper K (2003) Recurrent perivascular epithelioid cell tumor of the uterus (PEComa): an immunohistochemical study and review of the literature. Gynecol Oncol 90:677–681 22. Harris GC, McCulloch TA, Perks G, Fisher C (2004) Malignant perivascular epithelioid cell tumour (“PEComa”) of soft tissue: a unique case. Am J Surg Pathol 28:1655–1658 23. Hornick JL, Fletcher CD (2006) PEComa: what do we know so far? Histopathology 48:75–82 24. Iyengar P, Deangelis DD, Greenberg M, Taylor G (2005) Perivascular epithelioid cell tumor of the orbit: a case report and review of the literature. Pediatr Dev Pathol 8:98–104 25. Jeon IS, Lee SM (2005) Multimodal treatment using surgery, radiotherapy, and chemotherapy in a patient with a perivascular epithelioid cell tumor of the uterus. J Pediatr Hematol/Oncol 27:681–684 26. Kim HJ, Lim SJ, Choi H, Park K (2006) Malignant clear-cell myomelanocytic tumor of broad ligament—a case report. Virchows Arch 448:867–870 27. Koutlas IG, Pambuccian SE, Jessurun J, Manivel JC, Gopalakrishnan R (2005) Perivascular epithelioid cell tumor of the oral mucosa. Arch Pathol Lab Med 129:690–693 28. Lehman NL (2003) Malignant PEComa of the skull base. Am J Surg Pathol 28:1230–1232 29. Mai KT, Belanger EC (2006) Perivascular epithelioid cell tumour (PEComa) of the soft tissue. Pathology 38:415–420 30. Mentzel T, Reisshauer S, Rutten A, Hantschke M, Soares de Almeida LM, Kutzner H (2005) Cutaneous clear cell myomelanocytic tumour: a new member of the growing family of perivascular epithelioid cell tumours (PEComas). Clinicopathological and immunohistochemical analysis of seven cases. Histopathology 46:498–504 31. Pan CC, Jong YJ, Chai CY, Huang SH, Chen YJ (2006) Comparative genomic hybridization study of perivascular epithelioid cell tumor: molecular genetic evidence of perivascular
Virchows Arch (2007) 450:463–470
32.
33.
34.
35.
36.
37. 38.
39.
40.
41.
42. 43.
epithelioid cell tumor as a distinctive neoplasm. Hum Pathol 37:606–612 Pan CC, Yang AH, Chiang H (2003) Malignant perivascular epithelioid cell tumor involving the prostate. Arch Pathol Lab Med 127:96–98 Parfitt JR, Bella AJ, Izawa JI, Wehrli BM (2006) Malignant neoplasm of perivascular epithelioid cells of the liver. Arch Pathol Lab Med 130:1219–1222 Parfitt JR, Bella AJ, Wehrli BM, Izawa JI (2006) Primary PEComa of the bladder treated with primary excision and adjuvant interferon-alpha immunotherapy: a case report. BMC Urol 22:20 Pea M, Bonetti F, Zamboni G, Martignoni G, Fiore-Donati L, Doglioni C (1991) Clear cell tumor and angiomyolipoma. Am J Surg Pathol 15:199–202 Pea M, Bonetti F, Zamboni G, Martignoni G, Riva M, Colombari R, Mombello A, Bonzanini M, Scarpa A, Ghimenton C (1991) Melanocyte-marker-HMB-45 is regularly expressed in angiomyolipoma of the kidney. Pathology 23:185–188 Sherr CJ, Roberts JM (1995) Inhibitors of mammalian G1 cyclindependent kinases. Genes Dev 9:1149–1163 Silva EG, Deavers MT, Bodurka DC, Malpica A (2004) Uterine epithelioid leiomyosarcomas with clear cells: reactivity with HMB-45 and the concept of PEComa. Am J Surg Pathol 28:244–249 Soucek T, Pusch O, Wienecke R, DeClue JE, Hengstschlager M (1997) Role of the tuberous sclerosis gene-2 product in cell cycle control. Loss of the tuberous sclerosis gene-2 induces quiescent cells to enter S phase. J Biol Chem 272:29301–29308 Utikal J, Udart M, Leiter U, Peter RU, Krahn G (2005) Additional cyclin D (1) gene copies associated with chromosome 11 aberrations in cutaneous malignant melanoma. Int J Oncol 21:597–605 Vang R, Kempson RL (2002) Perivascular epithelioid cell tumor (‘PEComa’) of the uterus: a subset of HMB-45-positive epithelioid mesenchymal neoplasms with an uncertain relationship to pure smooth muscle tumors. Am J Surg Pathol 26:1–13 Weinberg RA (1995) The retinoblastoma protein and cell cycle control. Cell 81:323–330 Yamamoto H, Oda Y, Yao T, Oiwa T, Kobayashi C, Tamiya S, Kawaguchi K, Hino O, Tsuneyoshi M (2006) Malignant perivascular epithelioid cell tumor of the colon: report of a case with molecular analysis. Pathol Int 56:46–50
