—  SYMPOSIUM #55  —

New Frontiers in Breast Pathology
Moderator: Dr. Sunil Lakhani

Section 5 - Comparative Molecular Pathology of Myoepithelial Lesions

Jorge S Reis-Filho & Sunil R Lakhani
The Breakthrough Breast Cancer Research Centre
Institute of Cancer Research, London, UK
and
Molecular & Cellular Pathology, Mayne Medical School
University of Queensland
Queensland Institute of Medical Research and Royal Brisbane and Women's Hospital
Brisbane, Australia


Introduction:
Myoepithelial cells are ectordermally-derived, contractile cells which lie in between luminal epithelial cells and the basal membrane of acini and ducts, forming a network of slender processes that invest the overlying luminal cells in a discontinuous net like pattern [1, 2]. These cells are found in breast, salivary, sweat and lachrymal glands [1, 2, 3].

Myoepithelial cells are multifaceted and posses a dual basal epithelial-myoid nature [1, 2, 3]. The basal epithelial features are best exemplified by the presence of tonofilaments, desmosomes and expression of basal cytokeratins (i.e., cytokeratin 5 and 14), whereas the smooth muscle characteristics is best appreciated by ultrastructural analysis in the form of actin microfilaments with dense bodies, and by the expression of smooth muscle actin, smooth muscle myosin heavy chain and calponin [1-3].

The seminal studies by Dardick et al. [4, 5, 6, 7, 8, 9] paved the way for a proper identification of myoepithelial tumours. By recognising the remarkable morphological plasticity of neoplastic myoepithelial cells, which may assume epithelioid, plasmacytoid (hyaline), spindle, clear and even oncocytic cytomorphology, it has become possible to identify tumours composed of cells with myoepithelial differentiation [4, 5, 6, 7, 8, 9]. The morphological plasticity of myoepithelial cells is also reflected at the immunohistochemical level, hence it is not surprising that not all myoid markers are expressed in individual cases [10, 11]. With the advent of novel myoepithelial markers that are neither directly related to the epithelial features nor to the smooth muscle apparatus of myoepithelial cells, including p63 [12], maspin [13], p-cadherin [14, 15], 14-3-3 sigma (14-3-3σ) [16], nerve growth factor receptor (NGFR) [17] and caveolin 1 [14, 18], it has become possible to define the spectrum of lesions with myoepithelial differentiation or myoepithelial cell participation.

In salivary glands, tumours with myoepithelial differentiation and/or participation have been thoroughly characterised [19]. More recently, a new group of benign and malignant soft tissue tumours with myoepithelial differentiation has also been described [10, 20]; some of these lesions bear remarkable histological and immunohistochemical similarities with chondroid syringomas.

For a long time, it was believed that breast neoplasms with myoepithelial differentiation would be remarkably rare and the spectrum of lesions would encompass only adenoid cystic carcinomas, low-grade adenosquamous carcinomas, adenomyoepitheliomas, poorly-differentiated myoepithelial rich carcinomas and malignant myoepitheliomas [21]. However, with comprehensive characterisation of myoepithelial cells at the morphological, ultrastructural, transcriptomic [22] and proteomic [23] levels, the spectrum of breast neoplasms with myoepithelial differentiation has significantly expanded. Currently, apart from the above lesions, it is known that the vast majority of metaplastic breast carcinomas and approx 15% of invasive breast carcinomas show features of myoepithelial differentiation .

The morphological and immunohistochemical features of breast tumours with myoepithelial differentiation and how they compare with their respective salivary gland counterparts have been comprehensively reviewed elsewhere [24, 25]. In this hand-out, the molecular genetic features of breast tumours with myoepithelial differentiation are reviewed and compared with those of salivary gland and soft tissue myoepithelial tumours.

Benign Myoepithelioma
Few examples of benign myoepithelioma of the breast have been reported mostly as case reports or small series. These tumours frequently affect female patients and are show a modal distribution in the 5th-8th decades [24]. The histopathological features and immunohistochemical profiles of myoepitheliomas of the salivary glands and have been comprehensively reviewed [5, 19, 24]. Pure myoepithelial tumours have a wide spectrum of morphological appearances, being composed of varying amounts of spindle cells, basaloid cells, plamacytoid/ hyaline cells and clear cells [5, 19, 24]. Interestingly, the majority of myoepitheliomas of the salivary glands and the soft tissues are composed of cells of two or more types, whereas in the breast, most pure myoepitheliomas consist of cells with spindle cytomorphology [5, 10, 19, 24]. Areas with chrondroid, squamous and sebaceous metaplasia are not uncommon. In salivary glands, it is currently accepted that basal cell adenomas, pleomorphic adenomas and myoepitheliomas are part of a continuum of lesions harbouring varying degrees of myoepithelial cell participation [19, 26, 27]. At the immunohistochemical level, myoepitheliomas of all sites express different combinations of basal Cks, myoid markers, p63, glial fibrillary acidic protein (GFAP) and NGFR [10, 17, 19, 24, 26-28].

The molecular genetic features of benign myoepitheliomas of the breast are poorly characterised. Gain of 8q appears to be the most frequent unbalanced chromosomal change [29]. Few myoepitheliomas have been subjected to molecular analysis. Karyotypic information is available only for salivary gland myoepitheliomas: whilst one showed either no changes [30] other case harboured a fairly complex karyotype 46,XY,t(1;12)(q25;q12),del(9)(q22.1q22.3),del(13)(q12q22) [31].With the scant data available in the literature, it seems that breast, salivary gland and soft tissue myoepitheliomas harbour few unbalanced chromosomal changes as defined by CGH, with the recurrent rearrangements mapping to chromosomes 6, 9 and 12.

Pleomorphic Adenoma (PA) (aka, Benign Mixed Tumour)
PA the most frequent tumour in salivary glands and is not uncommon in cutaneous tissues, where it is known as chondroid syringoma. On the other hand, breast PA is exceedingly rare, with <80 cases reported to date [24, 32, 33]. These tumours have been found in patients ranging in age from 19 to 85 years and usually affects female patients (>15F:1M), with only four cases reported in men [24, 32, 33, 34]. Breast PA is more frequently located in the subareolar region, and cases showing intraductal growth are on record. An association with intraductal papillomas has been described. Breast PA is histologically similar to its salivary gland and cutaneous counterparts [24]. Although the vast majority of breast PAs have a benign course, Hayes et al. [32] have recently described three cases of malignant transformation in breast PAs, two in the form of grade-3 invasive ductal carcinoma NOS and areas of high-grade metaplastic carcinoma with chondroid matrix and one in the form of matrix producing metaplastic carcinoma [32].

The dual epithelial-myoepithelial nature of the neoplastic cells of breast PA is easily appreciated at the ultrastructural and immunohistochemical level. Epithelial cells lining lumina of the neoplastic glandular structures are reactive to antibodies against LMW cytokeratins, epithelial membrane antigen (EMA), and CEA [24, 32]. Antibodies against SMA, p63, S-100 protein and CD10 may highlight the outer rim of myoepithelial cells. In addition, myoepithelial markers, SMA, S-100 protein, calponin, CD10 and Ck14 are positive in the spindle cells found in the myxoid and cartilaginous areas [24, 32].

Salivary gland PAs harbour frequent rearrangements involving 8q12 and 12q15, which are seen in 40% and 8% of the tumours, respectively [35, 36]. The target gene of 8q12 rearrangements is the pleomorphic adenoma gene 1 (PLAG1), which encodes a developmentally regulated zinc finger protein, whereas rearrangements of 12q15 affect HMGA2, a gene that encodes one of the high mobility group (HMG) proteins [35, 36]. Even in pleomorphic adenomas with normal karyotypes, cryptic rearrangements of one of these genes may be identified. In approximately 25% of PAs, gross chromosomal abnormalities and translocations involving other genetic loci have been described. Only one case of breast PA has been subjected to interphase FISH to evaluate the presence of rearrangements involving 6p21 and 12q15 [33]. Although the myoepithelial component of that tumour showed immunohistochemical positivity for both HMGA2 and HMGIY, no rearrangements were found [33].

We have recently studied one case of salivary gland pleomorphic adenoma and observed a high level gain of 12q13.3-q21.1, which was shown to be present in both epithelioid, spindle and oncocytic cells, giving support to the common origin of different components of pleomorphic adenomas [37]. Similar high level gains have been found in carcinomas ex-PAs [38] and malignant mixed tumours [39, 40]. It has been suggested that gains of 12q14-q15 contribute to malignant transformation of pleomorphic adenomas [40].

Adenomyoepithelioma
In the breast, perhaps the best characterised lesion with myoepithelial differentiation is adenomyoepithelioma, which was first described by Hamperl in 1970 [1] and the concept expanded by Kiaer et al. [41], Eusebi et al. [42] and Tavassoli et al [43]. The similarities between adenomyoepitheliomas and epithelial-myoepithelial carcinomas of the salivary glands, lung and skin at the histological and immunohistochemical levels are striking. Although epithelial-myoepithelial carcinoma is considered a low-grade malignant neoplasm [19, 44], the clinical behaviour of adenomyoepitheliomas has been a matter of debate [24, 29, 43, 45, 46]; whilst some believe that these lesions are best considered borderline/ low malignant potential tumours, with potential for local recurrences and a rather low proclivity for distant metastasis, others claim that when strict criteria are applied, all adenomyoepitheliomas behave in a benign fashion [46].

Histologically, adenomyoepitheliomas show distinctive growth patterns, including tubular, papillary or solid; however tumours often show a combination of these. Regardless of the predominant growth pattern, the basic morphological structure of adenomyoepitheliomas (and epithelial-myoepithelial carcinomas of other sites) is a round glandular lumen lined by an inner layer of cuboidal, eosinophilic luminal epithelial cells and an outer layer of spindle, often clear myoepithelial cells lying on a basement membrane [24, 43, 45]. When solid areas are present, they are predominantly composed of myoepithelial cells [24, 43, 45]. When tumours are solely composed of spindle shaped, myoepithelial cells, they are best classified as pure myoepitheliomas. At the ultrastructural and immunohistochemical levels, the inner cell population shows unequivocal features of luminal epithelial cells, including short microvilli at the luminal surface and positivity for luminal Cks, whereas the abluminal population shows definite features of myoepithelial cells, including positivity for basal Cks, myoid markers, p63 and NGFR [12, 17, 24].

In adenomyoepitheliomas and epithelial-myoepithelial carcinomas, the proliferation rates found in myoepithelial cells are greater than those seen in the epithelial compartment [47, 48], suggesting that the epithelial component is terminally differentiated. There is a paucity of data on the molecular genetic profiles of lung and salivary gland epithelial-myoepithelial carcinomas [49, 50]. It has been reported that up to 20% of these lesions are aneuploid [49, 50], however there are no data on the chromosomal aberrations found in these lesions. Breast adenomyoepitheliomas have been analysed by means of comparative genomic hybridisation [29, 51]. Using an arbitrary histological definition for benign, borderline and malignant adenomyoepithelial tumours, Hungermann et al. [29] demonstrated a stepwise increase in the number of unbalanced chromosomal changes from benign to malignant lesions. The most frequent recurrent chromosomal aberrations were gains of 8q (26%), 13q (18%), 6p and 12p (both, 15%). and deletions of 17p (41%), 12q and 14q (both, 15%) [29]. Interestingly, both Hungermann et al. [29] and Jones et al. [51] demonstrated an increased complexity of the genomic changes when comparing recurrent/ metastatic adenomyoepitheliomas with their respective index cases. In addition, a reciprocal chromosomal translocation t(8;16)(p23;q21) has been described in a case of adenomyoepithelioma by cytogenetic and fluorescent in situ hybridisation analysis [52].

Adenoid Cystic Carcinoma (AdCC)
Adenoid cystic carcinoma is a tumour composed of epithelial and myoepithelial cells and has been reported in several anatomical sites, including salivary gland, breast, lung and prostate. In the breast, AdCCs are rare, accounting for <1% of all breast carcinomas [53, 54, 55]. In contrast to similar tumours in other organs, breast AdCC shows a less aggressive clinical behaviour [24, 53, 55, 56].

Histologically, breast AdCC shows three different architectural patterns as the salivary analogue: trabecular-tubular, cribriform, solid [24, 53, 55, 56]. AdCC of any anatomical site is characterised by two types of spaces: true glandular lumina which contain a granular secretion of PASD positive neutral mucosubstances and the myxoid 'cylinders' composed of basement membrane material or basophilic mucin. These spaces are lined by two types of cells: the basaloid type, which lines the spaces containing stroma and basement membrane-like material; and a population of cells with abundant, bright eosinophilic cytoplasm, which surround true glandular lumina containing mucinous material. Recently, a variant of breast AdCCs entirely composed of basal cells has been described and the term basaloid adenoid cystic carcinoma of the breast has been coined to refer to these lesions [24, 53, 55, 56]. Squamous and sebaceous metaplasia in AdCC of different anatomical sites are not uncommon. AdCC of the breast has been described in association with adenomyoepithelioma and with low grade syringomatous (adenosquamous) carcinoma, giving further support to the concept of a spectrum of epithelial–myoepithelial neoplasms [24]. More intriguingly, it has been suggested that microglandular adenosis may be a precursor of breast AdCC and that atypical microglandular adenosis should be considered AdCC in situ.

Immunohistochemical analyses have demonstrated that basaloid cells are positive for vimentin and CK14 and focally for myoepithelial markers, including SMA, calponin, p63 and maspin [24, 53, 54, 55, 56]. Ultrastructurally, basaloid cells show features of myoepithelial differentiation, such as thin cytoplasmic filaments and well developed desmosomes; nevertheless, most frequently they present nondescript features with rare cytoplasmic organelles. The glandular cell type is usually positive for CK7 and shows short microvilli along the luminal surface [24, 53, 54, 55, 56]. Positivity for oestrogen receptor is rarely seen [24, 53, 54, 55, 56, 57].

It has recently been described that AdCC of the salivary glands, breast and lungs consistently show c-kit expression [53, 54, 57, 58], however in most of these studies, heat-induced antigen retrieval was used. Although immunohistochemical positivity for c-kit may help differentiate adenoid cystic carcinomas from their mimics [54], its biological and therapeutic significance remains unclear, as no activating KIT mutations have been found in AdCCs [59, 60]. Furthermore, there are anecdotal cases of c-kit positive but mutation negative metastatic AdCC treated with imatinib mesylate, a small molecule inhibitor of c-kit and other related protein tyrosine kinase, that have rapidly progressed [61], suggesting that c-kit tyrosine kinase inhibitors may have a limited role in the treatment of these lesions [62, 63].

TP53 gene mutations have been demonstrated in salivary gland adenoid cystic carcinomas [64]; however, there are no data for breast lesions.

A comparative analysis between a primary salivary gland AdCC cell line and normal salivary gland tissue by 2-D-DIGE and MALDI-TOF mass spectrometric analysis revealed upregulation of stathmin, maspin, fibrin beta and sialic acid binding immunoglobulin-like lectin 8, and downregulation of enoyl coenzyme A hydratase short chain 1 (ECHS 1), serin proteinase inhibitor B 1 (SERPIN B1), superoxide dismutase 2 (SOD 2), aminolevulinate delta-dehydratase (ALAD) and pro-apolipoprotein [65]. Although these results are interesting, it is not clear whether these genes are truly differentially expressed in AdCC or merely reflect the enrichment of cells with myoepithelial phenotype in the AdCC sample.

The genetic profiles of salivary gland AdCC have been extensively characterised. The most frequent changes are deletions of 6q, 12q and 13q and gains of chromosome 19. In addition, rearrangements involving chromosomes 6q and 9p have also been reported. On the other hand, data on the genetic profile of breast AdCC are scant; the karyotypes of two tumours have been described: 46,XX,inv(9) [66] and 46,XX,t(4;4)(q21;q35),t(5;11)(q13;q21)/46,XX,+1,der(1;16)(q10;p10) [67].

Malignant Myoepithelioma/ Myoepithelial Carcinoma
Myoepithelial carcinoma of the breast and salivary glands is an entity that has received great attention in the literature in the last few years [24]. In a way akin to benign myoepitheliomas, the neoplastic cells that compose this lesion may have a plethora of morphological appearances; however, they often show nuclear atypia and increased mitotic activity. Both breast and salivary glands malignant myoepitheliomas appear to frequently arise from an underlying benign tumour; whilst in the salivary glands, pleomorphic adenomas and benign myoepitheliomas are frequently found in association with myoepithelial carcinomas; there are several examples of breast myoepithelial carcinomas associated with adenomyoepitheliomas.

In the latest World Health Organisation Classification of Tumours [70], malignant myoepithelioma of the breast was defined as an infiltrating tumour composed purely of myoepithelial cells, however it is not clear whether this evidence should be gathered at the histological, immunohistochemical or ultrastructural level. This definition is further complicated by the fact that bona fide myoepithelial carcinomas frequently harbour an abortive myoepithelial phenotype, lacking expression of α-smooth muscle actin and Ck14 in 50% and 47% of the cases, respectively [11]. In addition, aberrant expression of low molecular weight cytokeratins (eg, Ck 7) and other markers not usually seen in normal myoepithelial cells (eg, epithelial membrane antigen) may be observed [11].

Given the plasticity of neoplastic myoepithelial cells [4, 5, 6, 7, 8, 9], myoepithelial carcinomas of all sites may show a multitude of cytomorphological components. However, when arising in the salivary glands, these lesions have a typical low power appearance characterised by multinodularity and the presence of tongue-like processes infiltrating the adjacent tissues [19]. On the other hand, breast myoepithelial carcinomas do not necessarily show the same histological features, lesions with predominantly intraductal or intralobular growth patterns have been reported [24]. Although most of breast myoepithelial carcinomas are reported to be composed of a solid proliferation of spindle cells, with eosinophilic cytoplasm and very atypical nuclei, some may be deceptively bland. In addition, squamous elements and chondroid and osseous metaplasia admixed with the spindle cell proliferation may be found.

The criteria for malignancy in myoepithelial tumours are contentious; however invasion of adjacent tissues should be considered the minimum requirement for myoepithelial carcinoma. Although this is a valid criteria for salivary gland lesions, it is insufficient for soft tissue lesions, as 48.3% of all myoepithelial tumours in one series showed invasive borders and yet were not associated with recurrence or metastasis [10]. Cytologic atypia, although a helpful diagnostic feature, is not present in all cases. High mitotic activity and proliferation index have also been reported to be important characteristics of myoepithelial carcinomas [71]. Nagao et al. [71] have suggested that myoepithelial tumours of the salivary glands with > 7 mitotic figures/ 10 HPF or a Ki-67 proliferation index > 10% would be best classified as myoepithelial carcinomas. However, myoepithelial carcinomas with fewer than 7 mitotic figure/10 HPF are on record [72] and some have argued that further validation of these thresholds is still required before it is used in diagnostic practice [19].

Myoepithelial carcinomas of the breast appear to have an immunohistochemical profile [73] and a clinical behaviour similar to that described for basal-like breast carcinomas: the majority show an aggressive clinical course, high frequency of haematogenous spread and a peculiar proclivity to disseminate to the brain and lungs. There is some evidence that pure spindle cell breast carcinomas and myoepithelial carcinomas are the same entity [73, 74, 75, 76] and some have suggested that calling these lesions monophasic spindle cell metaplastic carcinomas or myoepithelial carcinomas is only a matter of preference [77]. Furthermore, some examples of matrix producing breast carcinomas are morphologically and immunohistochemically indistinguishable from some types of myoepithelial carcinomas found in the salivary glands.

Myoepithelial carcinomas express markers of epithelial and myoid differentiation. Kermix [78], high molecular weight cytokeratins (clone 34 bE12) [78] and p63 [79, 80] are useful markers to determine whether a pure spindle cell lesion shows epithelial differentiation. Recently, we have demonstrated that up to 70% of metaplastic breast carcinomas show EGFR overexpression and that EGFR gene amplification is the underlying genetic mechanism in one third of the cases [81, 82]. Array CGH analysis of these tumours revealed that the SRO in cases with EGFR gene amplification mapped only to SEC61G, LANCL2 and EGFR [82].

Comparative genomic hybridisation has been used to define the molecular genetic profiles of myoepithelial carcinomas of the breast and salivary glands. In both studies, a relative paucity of unbalanced chromosomal changes was seen [29, 72, 83]. However, the regions recurrently affected appear to be distinct: whilst Jones et al. [83] reported recurrent losses of 16q and 17p in 30% of breast myoepithelial carcinomas, Hungermann et al. [72] identified gains of 8q (29%), 1q and chromosome 5 (both, 23%) as the most frequent changes in salivary gland myoepithelial carcinomas. In addition, some of the cases described by Hungermann et al. [29] under the term of breast 'monophasic adenomyoepithelial carcinoma, spindle cell type' could also have been classified as examples of myoepithelial carcinomas. In those tumours, deletions of 17p and 14q and gains of 8q and 7q were the most frequent changes. We have subjected two cases of salivary gland myoepithelial carcinomas to array CGH analysis; both tumours showed recurrent deletions of 5q12.1-qter, 8p23.3-p12, gains of 1q21.1-q25.3, 8p12-qter and 18q11.21-q12.1 and amplification of 12q13.3-q14.1. The latter region has been implicated in the progression of salivary gland PAs and encompasses several oncogene candidates, including cyclin dependent kinase 4 (CDK4), sarcoma amplified sequence (SAS, aka tetraspanin 31 – TSPAN31) and glioma-associated oncogene homolog (GLI).

References:
  1. Hamperl, H., The myothelia (myoepithelial cells). Normal state; regressive changes; hyperplasia; tumors. Curr Top Pathol, 1970. 53: p. 161-220.

  2. Lakhani, S.R. and M.J. O'Hare, The mammary myoepithelial cell--Cinderella or ugly sister? Breast Cancer Res, 2001. 3(1): p. 1-4.

  3. Yaziji, H., A.M. Gown, and N. Sneige, Detection of stromal invasion in breast cancer: the myoepithelial markers. Adv Anat Pathol, 2000. 7(2): p. 100-9.

  4. Dardick, I., Malignant myoepithelioma of parotid salivary gland. Ultrastruct Pathol, 1985. 9(1-2): p. 163-8.

  5. Dardick, I., et al., Salivary gland myoepithelioma variants. Histological, ultrastructural, and immunocytological features. Virchows Arch A Pathol Anat Histopathol, 1989. 416(1): p. 25-42.

  6. Dardick, I., M.J. Thomas, and A.W. van Nostrand, Myoepithelioma--new concepts of histology and classification: a light and electron microscopic study. Ultrastruct Pathol, 1989. 13(2-3): p. 187-224.

  7. Dardick, I. and A.W. van Nostrand, Myoepithelial cells in salivary gland tumors--revisited. Head Neck Surg, 1985. 7(5): p. 395-408.

  8. Dardick, I., et al., Pleomorphic adenoma, I: Ultrastructural organization of "epithelial" regions. Hum Pathol, 1983. 14(9): p. 780-97.

  9. Dardick, I., et al., Pleomorphic adenoma, II: Ultrastructural organization of "stromal" regions. Hum Pathol, 1983. 14(9): p. 798-809.

  10. Hornick, J.L. and C.D. Fletcher, Myoepithelial tumors of soft tissue: a clinicopathologic and immunohistochemical study of 101 cases with evaluation of prognostic parameters. Am J Surg Pathol, 2003. 27(9): p. 1183-96.

  11. Savera, A.T., et al., Myoepithelial carcinoma of the salivary glands: a clinicopathologic study of 25 patients. Am J Surg Pathol, 2000. 24(6): p. 761-74.

  12. Reis-Filho, J.S. and F.C. Schmitt, Taking advantage of basic research: p63 is a reliable myoepithelial and stem cell marker. Adv Anat Pathol, 2002. 9(5): p. 280-9.

  13. Reis-Filho, J.S., F. Milanezi, and F.C. Schmitt, Maspin is expressed in the nuclei of breast myoepithelial cells. J Pathol, 2002. 197(2): p. 272-3; author reply 273-4.

  14. Charafe-Jauffret, E., et al., Gene expression profiling of breast cell lines identifies potential new basal markers. Oncogene, 2006. 25(15): p. 2273-84.

  15. Kovacs, A. and R.A. Walker, P-cadherin as a marker in the differential diagnosis of breast lesions. J Clin Pathol, 2003. 56(2): p. 139-41.

  16. Simpson, P.T., et al., Distribution and significance of 14-3-3sigma, a novel myoepithelial marker, in normal, benign, and malignant breast tissue. J Pathol, 2004. 202(3): p. 274-85.

  17. Reis-Filho, J.S., et al., Distribution and significance of nerve growth factor receptor (NGFR/p75NTR) in normal, benign and malignant breast tissue. Mod Pathol, 2006. 19(2): p. 307-19.

  18. Hurlstone, A.F., et al., Analysis of the CAVEOLIN-1 gene at human chromosome 7q31.1 in primary tumours and tumour-derived cell lines. Oncogene, 1999. 18(10): p. 1881-90.

  19. Simpson, R.H.W., Myoepithelial tumours of the salivary gland. Curr Diagn Pathol, 2002. 8: p. 328-337.

  20. Michal, M. and M. Miettinen, Myoepitheliomas of the skin and soft tissues. Report of 12 cases. Virchows Arch, 1999. 434(5): p. 393-400.

  21. Foschini, M.P. and V. Eusebi, Carcinomas of the breast showing myoepithelial cell differentiation. A review of the literature. Virchows Arch, 1998. 432(4): p. 303-10.

  22. Jones, C., et al., Expression profiling of purified normal human luminal and myoepithelial breast cells: identification of novel prognostic markers for breast cancer. Cancer Res, 2004. 64(9): p. 3037-45.

  23. Page, M.J., et al., Proteomic definition of normal human luminal and myoepithelial breast cells purified from reduction mammoplasties. Proc Natl Acad Sci U S A, 1999. 96(22): p. 12589-94.

  24. Pia-Foschini, M., et al., Salivary gland-like tumours of the breast: surgical and molecular pathology. J Clin Pathol, 2003. 56(7): p. 497-506.

  25. Bennett, A.K., S.E. Mills, and M.R. Wick, Salivary-type neoplasms of the breast and lung. Semin Diagn Pathol, 2003. 20(4): p. 279-304.

  26. Prasad, A.R., et al., The myoepithelial immunophenotype in 135 benign and malignant salivary gland tumors other than pleomorphic adenoma. Arch Pathol Lab Med, 1999. 123(9): p. 801-6.

  27. Zarbo, R.J., et al., Salivary gland basal cell and canalicular adenomas: immunohistochemical demonstration of myoepithelial cell participation and morphogenetic considerations. Arch Pathol Lab Med, 2000. 124(3): p. 401-5.

  28. Kilpatrick, S.E., et al., Mixed tumors and myoepitheliomas of soft tissue: a clinicopathologic study of 19 cases with a unifying concept. Am J Surg Pathol, 1997. 21(1): p. 13-22.

  29. Hungermann, D., et al., Adenomyoepithelial tumours and myoepithelial carcinomas of the breast--a spectrum of monophasic and biphasic tumours dominated by immature myoepithelial cells. BMC Cancer, 2005. 5: p. 92.

  30. Martins, C., et al., Benign salivary gland tumors: a cytogenetic study of 21 cases. J Surg Oncol, 1995. 60(4): p. 232-7.

  31. el-Naggar, A.K., et al., Cytogenetic analysis of a primary salivary gland myoepithelioma. Cancer Genet Cytogenet, 1999. 113(1): p. 49-53.

  32. Hayes, M.M., et al., Carcinoma ex-pleomorphic adenoma of the breast. Report of three cases suggesting a relationship to metaplastic carcinoma of matrix-producing type. Virchows Arch, 2005. 446(2): p. 142-9.

  33. Sato, K., et al., Pleomorphic adenoma (benign mixed tumor) of the breast: a case report and review of the literature. Pathol Res Pract, 2005. 201(4): p. 333-9.

  34. Molland, J.G., et al., Pleomorphic adenoma of the parotid and breast in a male patient. Pathology, 2005. 37(3): p. 263-5.

  35. Geurts, J.M., et al., Expression of reciprocal hybrid transcripts of HMGIC and FHIT in a pleomorphic adenoma of the parotid gland. Cancer Res, 1997. 57(1): p. 13-7.

  36. Voz, M.L., et al., PLAG1, the main translocation target in pleomorphic adenoma of the salivary glands, is a positive regulator of IGF-II. Cancer Res, 2000. 60(1): p. 106-13.

  37. Di Palma, S., et al., Oncocytic change in pleomorphic adenoma: molecular evidence in support of an origin in neoplastic cells. J Clin Pathol, 2006.

  38. Roijer, E., et al., Translocation, deletion/amplification, and expression of HMGIC and MDM2 in a carcinoma ex pleomorphic adenoma. Am J Pathol, 2002. 160(2): p. 433-40.

  39. Tsang, Y.T., et al., Amplification of MGC2177, PLAG1, PSMC6P, and LYN in a malignant mixed tumor of salivary gland detected by cDNA microarray with tyramide signal amplification. Cancer Genet Cytogenet, 2004. 152(2): p. 124-8.

  40. Rao, P.H., et al., Nonsyntenic amplification of MYC with CDK4 and MDM2 in a malignant mixed tumor of salivary gland. Cancer Genet Cytogenet, 1998. 105(2): p. 160-3.

  41. Kiaer, H., et al., Adenomyoepithelial adenosis and low-grade malignant adenomyoepithelioma of the breast. Virchows Arch A Pathol Anat Histopathol, 1984. 405(1): p. 55-67.

  42. Eusebi, V., et al., Adenomyoepithelioma of the breast with a distinctive type of apocrine adenosis. Histopathology, 1987. 11(3): p. 305-15.

  43. Tavassoli, F.A., Myoepithelial lesions of the breast. Myoepitheliosis, adenomyoepithelioma, and myoepithelial carcinoma. Am J Surg Pathol, 1991. 15(6): p. 554-68.

  44. Fulford, L.G., et al., Epithelial-myoepithelial carcinomas of the bronchus. Am J Surg Pathol, 2001. 25(12): p. 1508-14.

  45. Rosen, P.P., Adenomyoepithelioma of the breast. Hum Pathol, 1987. 18(12): p. 1232-7.

  46. McLaren, B.K., et al., Adenomyoepithelioma: clinical, histologic, and immunohistologic evaluation of a series of related lesions. Am J Surg Pathol, 2005. 29(10): p. 1294-9.

  47. Koyama, M., et al., Immunohistochemical assessment of proliferative activity in mammary adenomyoepithelioma. Histopathology, 1997. 31(2): p. 134-9.

  48. Fonseca, I. and J. Soares, Proliferating cell nuclear antigen immunohistochemistry in epithelial-myoepithelial carcinoma of the salivary glands. Arch Pathol Lab Med, 1993. 117(10): p. 993-5.

  49. Cho, K.J., et al., Epithelial-myoepithelial carcinoma of salivary glands. A clinicopathologic, DNA flow cytometric, and immunohistochemical study of Ki-67 and HER-2/neu oncogene. Am J Clin Pathol, 1995. 103(4): p. 432-7.

  50. Fonseca, I. and J. Soares, Epithelial-myoepithelial carcinoma of the salivary glands. A study of 22 cases. Virchows Arch A Pathol Anat Histopathol, 1993. 422(5): p. 389-96.

  51. Jones, C., R. Tooze, and S.R. Lakhani, Malignant adenomyoepithelioma of the breast metastasizing to the liver. Virchows Arch, 2003. 442(5): p. 504-6.

  52. Gatalica, Z., et al., Gene expression profile of an adenomyoepithelioma of the breast with a reciprocal translocation involving chromosomes 8 and 16. Cancer Genet Cytogenet, 2005. 156(1): p. 14-22.

  53. Azoulay, S., et al., KIT is highly expressed in adenoid cystic carcinoma of the breast, a basal-like carcinoma associated with a favorable outcome. Mod Pathol, 2005. 18(12): p. 1623-31.

  54. Mastropasqua, M.G., et al., Immunoreactivity for c-kit and p63 as an adjunct in the diagnosis of adenoid cystic carcinoma of the breast. Mod Pathol, 2005. 18(10): p. 1277-82.

  55. Acs, G., et al., Microglandular adenosis with transition into adenoid cystic carcinoma of the breast. Am J Surg Pathol, 2003. 27(8): p. 1052-60.

  56. Page, D.L., Adenoid cystic carcinoma of breast, a special histopathologic type with excellent prognosis. Breast Cancer Res Treat, 2005. 93(3): p. 189-90.

  57. Crisi, G.M., et al., Expression of c-kit in adenoid cystic carcinoma of the breast. Am J Clin Pathol, 2005. 124(5): p. 733-9.

  58. Aslan, D.L., et al., c-kit expression in adenoid cystic carcinoma does not have an impact on local or distant tumor recurrence. Head Neck, 2005. 27(12): p. 1028-34.

  59. Holst, V.A., et al., KIT protein expression and analysis of c-kit gene mutation in adenoid cystic carcinoma. Mod Pathol, 1999. 12(10): p. 956-60.

  60. Jeng, Y.M., C.Y. Lin, and H.C. Hsu, Expression of the c-kit protein is associated with certain subtypes of salivary gland carcinoma. Cancer Lett, 2000. 154(1): p. 107-11.

  61. Lin, C.H., et al., Unexpected rapid progression of metastatic adenoid cystic carcinoma during treatment with imatinib mesylate. Head Neck, 2005. 27(12): p. 1022-7.

  62. Ochel, H.J., et al., Effects of imatinib mesylate on adenoid cystic carcinomas. Anticancer Res, 2005. 25(5): p. 3659-64.

  63. Hotte, S.J., et al., Imatinib mesylate in patients with adenoid cystic cancers of the salivary glands expressing c-kit: a Princess Margaret Hospital phase II consortium study. J Clin Oncol, 2005. 23(3): p. 585-90.

  64. Yamamoto, Y., et al., DNA analysis at p53 locus in adenoid cystic carcinoma: comparison of molecular study and p53 immunostaining. Pathol Int, 1998. 48(4): p. 273-80.

  65. Nakashima, D., et al., Protein expression profiling identifies maspin and stathmin as potential biomarkers of adenoid cystic carcinoma of the salivary glands. Int J Cancer, 2006. 118(3): p. 704-13.

  66. Mark, H.F., I. Hanna, and D.R. Gnepp, Cytogenetic analysis of salivary gland type tumors. Oral Surg Oral Med Oral Pathol Oral Radiol Endod, 1996. 82(2): p. 187-92.

  67. Pandis, N., et al., Chromosome analysis of 97 primary breast carcinomas: identification of eight karyotypic subgroups. Genes Chromosomes Cancer, 1995. 12(3): p. 173-85.

  68. Rutherford, S., et al., Chromosome 6 deletion and candidate tumor suppressor genes in adenoid cystic carcinoma. Cancer Lett, 2006. 236(2): p. 309-17.

  69. Rutherford, S., et al., Mapping of candidate tumor suppressor genes on chromosome 12 in adenoid cystic carcinoma. Lab Invest, 2005. 85(9): p. 1076-85.

  70. Tavassoli, F.A. and P. Deville, Pathology and genetics of tumours of the breast and female genital organs. 2003, Lyon: IARC Press.

  71. Nagao, T., et al., Salivary gland malignant myoepithelioma: a clinicopathologic and immunohistochemical study of ten cases. Cancer, 1998. 83(7): p. 1292-9.

  72. Hungermann, D., et al., Relative paucity of gross genetic alterations in myoepitheliomas and myoepithelial carcinomas of salivary glands. J Pathol, 2002. 198(4): p. 487-94.

  73. Reis-Filho, J.S., et al., Metaplastic breast carcinomas are basal-like tumours. Histopathology, 2006. in press.

  74. Leibl, S. and F. Moinfar, Metaplastic breast carcinomas are negative for Her-2 but frequently express EGFR (Her-1): potential relevance to adjuvant treatment with EGFR tyrosine kinase inhibitors? J Clin Pathol, 2005. 58(7): p. 700-4.

  75. Dunne, B., et al., An immunohistochemical study of metaplastic spindle cell carcinoma, phyllodes tumor and fibromatosis of the breast. Hum Pathol, 2003. 34(10): p. 1009-15.

  76. Reis-Filho, J.S., et al., Novel and classic myoepithelial/stem cell markers in metaplastic carcinomas of the breast. Appl Immunohistochem Mol Morphol, 2003. 11(1): p. 1-8.

  77. Rosai, J., Rosai and Ackerman's surgical pathology. Vol. 2. 2004, Edinburgh: Mosby.

  78. Adem, C., et al., Wide spectrum screening keratin as a marker of metaplastic spindle cell carcinoma of the breast: an immunohistochemical study of 24 patients. Histopathology, 2002. 40(6): p. 556-62.

  79. Reis-Filho, J.S. and F.C. Schmitt, p63 expression in sarcomatoid/metaplastic carcinomas of the breast. Histopathology, 2003. 42(1): p. 94-5.

  80. Koker, M.M. and C.G. Kleer, p63 expression in breast cancer: a highly sensitive and specific marker of metaplastic carcinoma. Am J Surg Pathol, 2004. 28(11): p. 1506-12.

  81. Reis-Filho, J.S., et al., Metaplastic breast carcinomas exhibit EGFR, but not HER2, gene amplification and overexpression: immunohistochemical and chromogenic in situ hybridization analysis. Breast Cancer Res, 2005. 7(6): p. R1028-35.

  82. Reis-Filho, J., et al., EGFR amplification and lack of activating mutations in metaplastic breast carcinomas. J Pathol, 2006.

  83. Jones, C., et al., Comparative genomic hybridization analysis of myoepithelial carcinoma of the breast. Lab Invest, 2000. 80(6): p. 831-6.