Virchows Arch DOI 10.1007/s00428-014-1676-5
ORIGINAL ARTICLE
Hyalinizing clear cell carcinoma with EWSR1-ATF1 fusion gene: report of three cases with molecular analyses Takafumi Nakano & Hidetaka Yamamoto & Toshimitsu Nishijima & Sadafumi Tamiya & Hideki Shiratsuchi & Torahiko Nakashima & Shizuo Komune & Yoshinao Oda
Received: 19 June 2014 / Revised: 2 October 2014 / Accepted: 20 October 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Hyalinizing clear cell carcinoma (HCCC) is a lowgrade salivary gland carcinoma characterized by clear cells and hyalinized stroma. Recently, the EWSR1-ATF1 fusion gene was found in HCCCs. We herein describe three cases of HCCC identified in one male and two females, ranging in age from 27 to 67 years. The tumors were located in the root of tongue, nasopharynx, and soft palate. They were composed of nested or cord-like proliferations of epithelial cells with clear to pale eosinophilic cytoplasm, embedded in hyalinized and focally fibroedematous stroma. Tumor-associated lymphoid proliferation and pseudoepitheliomatous hyperplasia were also observed in each one case. MAML2 fusions specific to mucoepidermoid carcinoma were not detected in any of the three cases. We found EWSR1-ATF1 in two of three HCCCs using reverse transcription polymerase chain reaction (RTPCR) with our original primer sets designed to detect the fusion gene transcripts in formalin-fixed paraffin-embedded (FFPE) tissues. EWSR1 rearrangement was also confirmed by fluorescence in situ hybridization (FISH) on FFPE sections in two cases. There was a good concordance between the two
Electronic supplementary material The online version of this article (doi:10.1007/s00428-014-1676-5) contains supplementary material, which is available to authorized users. T. Nakano : H. Yamamoto (*) : T. Nishijima : S. Tamiya : Y. Oda Department of Anatomic Pathology, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan e-mail: [email protected] T. Nakano : T. Nishijima : H. Shiratsuchi : T. Nakashima : S. Komune Department of Otolaryngology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan S. Tamiya Department of Pathology, Kitakyushu Municipal Medical Center, Kitakyushu, Japan
methods (two positive cases and one negative case by both RT-PCR and FISH). Therefore, RT-PCR and FISH using FFPE tissue may be ancillary tools to confirm the diagnosis of HCCC. Keywords Hyalinizing clear cell carcinoma . EWSR1-ATF1 . Fusion gene . FISH . RT-PCR
Introduction Recently, characteristic fusion genes have been found in some salivary gland carcinomas, including EWSR1-ATF1 in hyalinizing clear cell carcinoma (HCCC), CRTC1- or CRTC3-MAML2 in mucoepidermoid carcinoma (MEC), ETV6-NTRK3 in mammary analogue secretory carcinoma, and MYB-NFIB in adenoid cystic carcinoma [1–5]. HCCC is a rare low-grade salivary gland carcinoma, characterized by clear to pale eosinophilic cells arranged in cords, trabeculae, or nests within a hyalinized stroma. HCCC is also known as clear cell carcinoma, not otherwise specified, in the recent WHO classification [6]. HCCC has a generally indolent clinical course and rarely shows lymph node or distant metastases [7, 8]. Morphologically, HCCC resembles a clear cell variant of MEC, as both tumors are composed of clear to eosinophilic cells, and mucinous differentiation can be seen not only in MEC but also in HCCC [1, 9]. Therefore, the histopathological distinction between HCCC and MEC is occasionally difficult. Recently, Antonescu et al. [1] first reported the presence of EWSR1 rearrangement by fluorescence in situ hybridization (FISH) in 18 of 22 (82 %) cases of HCCC, and the presence of EWSR1-ATF1 fusion transcript in which exon 11 of EWSR1 was fused to exon 3 of ATF1 by reverse transcriptasepolymerase chain reaction (RT-PCR) in both cases examined.
Virchows Arch
Subsequently, Shah et al. [2] confirmed the EWSR1 rearrangement in 13 of 15 (87 %) HCCCs by FISH. Here we report additional three cases of HCCC, two of which were positive for both EWSR1 rearrangement by FISH and EWSR1-ATF1 by RT-PCR using paraffin-embedded tissue and modified primer sets.
Materials and methods The excised specimens were fixed in a solution of 10 % formaldehyde and embedded in paraffin, and stained with hematoxylin-eosin. In all cases, additional stains were also carried out, including periodic acid-Schiff (PAS) with or without prior diastase treatment, mucicarmine and Alcian blue staining. Immunohistochemical staining Immunohistochemical staining was performed using 4-μmthick, formalin-fixed paraffin-embedded (FFPE) tissue, and the primary antibodies used in this study are listed in Supplemental Table 1. A biotin-free, horseradish peroxidase enzyme-labeled polymer method (Envision+ system, Dako, Carpinteria, CA, USA) with hematoxylin counterstaining was used. The immunohistochemistry results were scored as fellows: negative (no staining or 50 %). RT-PCR Total RNA was isolated from FFPE tissue using the miRNeasy FFPE Kit (Qiagen, Valencia, CA, USA), and first-strand cDNA was synthesized using Superscript III Transcriptase (Invitrogen, Carlsbad, CA, USA), according to the manufacturer’s instructions. To detect EWSR1-ATF1 fusion gene transcripts, we performed RT-PCR using our original primers designed to fit the FFPE samples as follows: the forward primer, 5′-caaggattaaatgacagtgtgactc-3′, and the reverse primer, 5′-ctttctgtgaggagcctatg-3′. In addition, RT-PCR for CRTC1-MAML2 and CRTC3-MAML2 was performed as previously reported [3]. Then, the sequence was confirmed by direct sequencing methods with an ABI Prism 310 sequence analyzer (Applied Biosystems, Foster City, CA, USA). In all three cases, the presence of sufficient quality of cDNA was confirmed by positive amplification of the phosphoglycerate kinase (PGK) (data not shown). EWSR1 break-apart FISH FISH was also performed using 4-μm-thick, unstained tissue sections and the EWSR1 break-apart probe set (Vysis LSI EWSR1 dual-color, break-apart rearrangement probe;
Abbott Molecular, Des Plaines, IL, USA), with the Vysis Paraffin Pretreatment IV and Post-hybridization Wash Buffer Kit (Abbott Molecular). The 5′ EWSR1 signal was labeled with SpectrumOrange (orange), and the 3′ EWSR1 signal was labeled with SpectrumGreen (green) [10]. The hybridized slides were reviewed on a fluorescence microscope (BZ9000, Keyence, Osaka, Japan) at ×60 magnification with oil immersion, using a DAPI/Green/Red triple band-pass filter set. The results were scored by evaluating 100 tumor cell nuclei per case. Tumor cells with nuclear overlap and indistinguishable separate nuclei were excluded from analysis. A split signal was defined by 5′ and 3′ signals observed at a distance greater than one-time signal width, and signals separated by less than that were regarded as fused signals. Tumor cells showing split or isolated 5′ signals were considered positive rearrangements. We interpreted a result as positive if more than 20 % of tumor cells showed positive rearrangement [1]. All cases of HCCC had fused 5′ and 3′ (yellow) signals in non-neoplastic cells as an internal control, indicating successful hybridization, and they were thereby eligible for FISH analysis.
Results Clinical summary Clinicopathologial findings are summarized in Table 1. The three patients were one man and two women, ranging in age from 27 to 67 years (median, 52 years). The tumors were located in the root of tongue, nasopharynx, and soft palate and ranged from 18 to 25 mm in maximum diameter. No lymph node metastasis was detected in any of the cases at diagnosis. During the follow-up (3–75 months, median 18 months), all patients were alive without tumor recurrence, lymph node metastasis, or distant metastasis. Histopathological findings Microscopically, the tumors consisted of a proliferation of clear cells with distinctive cell borders arranged in sheets, nests, cords, and trabecular patterns (Fig. 1a). The tumor cells infiltrated the tumor margins. Hyalinized eosinophilic material was observed within or around the tumor nests (Fig. 1a). Two cases (Cases 2 and 3) showed focal fibroedematous stroma between or around tumor nests (Fig. 1b). In addition, tumor cells occasionally showed pale eosinophilic cytoplasm (Fig. 1b). Neither perineural invasion, vascular invasion, nor tumor necrosis was present in any of our cases. Tumor cells had cytoplasmic glycogen, which was highlighted by PAS after diastase digestion (D-PAS) (Fig. 1c and d). In all three cases, goblet-like cells and small amounts of mucin were
Virchows Arch Table 1 Clinicopathological findings of three cases of hyalinizing clear cell carcinoma Case Age Sex Site
1
52
M
2
27
F
3
67
F
First sign or symptom
Size Treatment (mm)
Root of Painless swelling lesion of 18 tongue root of tongue Nasopharynx Hearing loss and feeling of 25 fullness in the right ear Soft palate Painless swelling lesion of 20 soft palate
Local Lymph node/distant Outcome recurrence metastasis
Surgical resection, adjuvant – radiotherapy (61.4 Gy) Surgical resection – Surgical resection, adjuvant – radiotherapy (32 Gy)
–
NED, 3 months
–
NED, 18 months
–
NED, 75 months
NED no evidence of disease
also focally seen, which was confirmed by D-PAS, Alcian blue, and mucicarmine staining (Fig. 1d–f). In addition, DPAS highlighted basement membrane-like material surrounding the tumor nests (Fig. 1d). In Case 1, overlying squamous epithelium showed pseudoepitheliomatous hyperplasia (PEH) and gradual transition from PEH to HCCC tumor cells with eosinophilic or clear cytoplasm (Fig. 2a and b). These features resembled a clear cell variant of squamous cell carcinoma (SCC). In Case 3, at the periphery of the tumor, lymphoid aggregates were noted with lymphoid follicles having germinal centers (Fig. 3), indicating a tumor-associated lymphoid proliferation (TALP). Immunohistochemical findings The results of immunohistochemical staining are summarized in Table 2. The tumors were positive for pan-cytokeratin (CK) (3/3 cases), low-molecular-weight CK (diffuse 1, focal 2), CK13 (focal and weak 1), CK 14 (2/3), CK 17 (diffuse 2, focal 1), epithelial membrane antigen (EMA) (diffuse 2, focal and weak 1), and p63 (3/3). S100 protein and alpha-smooth muscle actin were negative. In Case 1, overlying squamous epithelium with pseudoepitheliomatous hyperplasia showed CK17 positivity and reduced CK13 expression (Fig. 2c and d). Molecular genetic findings The results of molecular tests are also summarized in Table 2. Using RT-PCR, 136 bp of EWSR1-ATF1 fusion transcripts were detected in two of three cases (66.7 %) (Cases 1 and 2). The direct sequence of EWSR1-ATF1 fusion transcripts revealed that exon 11 of EWSR1 was fused to exon 3 of ATF1 (Fig. 4). Neither CRTC1-MAML2 nor CRTC3-MAML2 fusion transcript was identified in any of the cases. EWSR1 rearrangement was confirmed using FISH in two of three cases (66.7 %) (Cases 1 and 2). Positive rearrangement signals were observed in 79 % (split signals in 60 % and isolated 5′ signals in 19 %) and 69 % (split signals in 57 % and isolated 5′ signals in 12 %) of tumor cells in cases 1 and 2, respectively (Fig. 5). In case 3, a large number (97 %) of tumor
cells displayed fused signals, indicating absence of rearrangement. Altogether, two cases (Cases 1 and 2) were positive for EWSR1 rearrangement by both RT-PCR and FISH.
Discussion We used FFPE tissues to successfully identify EWSR1-ATF1 in HCCC. Only one previous report shows EWSR1-ATF1 in HCCCs by RT-PCR; in that study, 425 bp of EWSR1-ATF1 fusion transcripts were detected using frozen samples in two cases of HCCC [1]. We modified the primers for FFPE specimens, making it possible to detect a shorter size (136 bp) of EWSR1-ATF1 fusion transcripts in two of three cases. Moreover, EWSR1 rearrangement was also detected in two of three cases by FISH. There was agreement between the results of RT-PCR and those of FISH. Thus, the evidence indicates that our primers are suitable for detecting EWSR1ATF1 using RT-PCR and FFPE tissues. In our series, one of three HCCCs (33 %) was negative for EWSR1 gene rearrangement. Likewise, in the previous studies, a total of 6 of 37 HCCCs (16.2 %) lacked this rearrangement, even though FISH was successful [1, 2]. These results suggest that a minor subset of HCCC might have an as-yet-unknown genetic abnormality that is probably unrelated to the EWSR1 gene. In one case (Case 3) in our series, the tumor was surrounded by an inflammatory infiltration with lymphoid follicles having germinal centers, indicating a convincing example of TALP (Fig. 3). TALP has not been clearly described in HCCC in the literature. The literature notes a chronic inflammatory infiltrate in five cases of HCCC, but they all lacked apparent lymphoid follicles [11–13]. It is not surprising that TALP is present in HCCC, because TALP is a common phenomenon in some types of low-grade salivary gland tumors such as MEC and acinic cell carcinoma [14]. In our Case 1, it was difficult to distinguish between HCCC and SCC with clear cell change at routine histological diagnosis, because there was a morphological transition between the PEH and the HCCC area (Fig. 2a and b). PEH has been previously reported in three cases of HCCC [2]. PEH is a
Virchows Arch
Fig. 1 a HCCC showed a proliferation of clear cells with distinctive cell borders, arranged in sheets, nests, cords, and trabecular patterns, accompanied by characteristic hyalinizing stroma (HE staining) (Case 1). b Tumor cells occasionally have pale eosinophilic cytoplasm and fibroedematous stroma (HE staining) (Case 2). c–f PAS revealed
significant cytoplastic glycogen (c), which was digested by diastase (d). Intracytoplasmic mucin was focally detected by PAS with prior diastase digestion (D-PAS) (d), Alcian blue staining (e), and mucicarmine staining (f). D-PAS highlighted basement membrane-like materials surrounding the tumor nests (d) (Case 3)
benign epithelial change that can be easily confused with SCC. Although it has been reported that the expression profile of CK13 and CK17 is different between normal oral squamous epithelium (CK13+/CK17−) and squamous cell carcinoma (CK13−/CK17+) [15, 16], the expression pattern of CK13 and CK17 is variable in dysplasia and hyperplasia of oral mucosa. In our case, both HCCC cells and overlying squamous epithelium with PEH showed reduced CK13 expression and CK17 overexpression (Fig. 2c and d). In this
context, special attention needs to be paid to interpretation of the immunohistochemical results, because the expression patterns of CK13 and CK17 may be the same among SCC, PEH, and HCCC. One of the main differential diagnoses of HCCC is mucindepleted clear cell variant MEC. Both types of tumor consist of epithelial cells with clear to eosinophilic cytoplasm with various degrees of mucinous differentiation [1, 8]. Indeed, all our cases of HCCC focally had intracytoplasmic mucin,
Virchows Arch
Fig. 2 a, b Overlying hyperplastic squamous epithelium showed irregular, drop-down growth indicating pseudoepitheliomatous hyperplasia (PEH) (a). In connection with PEH, HCCC tumor cells with eosinophilic cytoplasm (arrow) and those with clear cytoplasm (arrowhead) showed gradual transitions, mimicking squamous cell carcinoma with clear cell
change (Case 1). c, d PEH showed decreased expression of CK13 (c) and CK17 protein overexpression (d). More than half of the HCCC tumor cells showed immunoreactivity for CK17 (d), but CK13 was negative (c) (Case 1)
Table 2 Immunohistochemical and molecular results in this study
Fig. 3 Tumor-associated lymphoid proliferation (TALP) was illustrated by lymphoid follicles having germinal centers around the tumor (Case 3)
Case no.
1
2
3
Pan-cytokeratin (AE1/AE3) CK-LMW (CAM5.2) CK13
++ + −
++ + +, weak
++ ++ −
CK14 CK17 EMA p63 S100 protein SMA EWSR1 EWSR1 CRTC1/CRTC3-MAML2
++ ++ +, weak ++ − − Positive Positive −
++ ++ ++ ++ − − Positive Positive −
− + ++ ++ − − − − −
FISH RT-PCR RT-PCR
CK-LMW low-molecular-weight cytokeratin, CK cytokeratin, EMA epithelial membrane antigen, SMA smooth muscle actin, ++ diffuse positive, + focal positive, − negative
Virchows Arch Fig. 4 Detection of EWSR1ATF1 fusion in HCCC. This image of the sequencing of an RT-PCR product shows fusion between EWSR1 exon 11 and ATF1 exon 3 (Case 1)
mimicking goblet cells in MEC. In addition, both HCCC and MEC preferentially occur in the minor salivary glands [1, 7, 8, 17–20]. Although there are confusingly similar histopathological features, characteristic fusion genes (CRTC1/CRTC3MAML2 in MECs and EWSR1-ATF1 in HCCCs) can be helpful for the differential diagnosis [1–3, 19–21]. Only one case of high-grade salivary mucoepidermoid carcinoma with EWSR1-POU5F1 has been reported to date [22]; however, EWSR1 rearrangement in MEC should be further elucidated. According to Antonescu et al. [1], none of 15 cases of HCCC showed MAML2 or POU5F1 rearrangement by FISH. Another important differential diagnosis of HCCC is clear cell variant of myoepithelial carcinoma. Both types of tumor share the histological appearance such as clear to eosinophilic cytoplasm and basement membrane-like materials. The distinctive feature of myoepithelial carcinoma is the expression of myoepithelial markers including S-100 protein and alphasmooth muscle actin; HCCC is generally negative for these markers. Indeed, all our cases of HCCC were negative for these markers, irrespective of EWSR1 rearrangement (Table 2). In summary, we report additional three cases of HCCC, two of which were positive for EWSR1 rearrangement by FISH and EWSR1-ATF1 fusion transcript by RT-PCR using FFPE tissues and modified primer sets. There was good
Fig. 5 Break-apart FISH for EWSR1 rearrangement. The majority of tumor cells show a split (separated) signal pattern of one green and one orange (arrow). The intact EWSR1 is shown by a fused (yellow) signal or closely juxtaposed orange and green signals (arrowhead) (case 1)
concordance between FISH and RT-PCR. One case of HCCC showed PEH, mimicking SCC with clear cell change. Therefore, FISH and RT-PCR using FFPE sections constitute ancillary tools to confirm the diagnosis of HCCC. In addition, we report a unique case of HCCC with convincing TALP. Acknowledgments The authors are grateful to Ms. Tateishi, Ms. Matsumoto, and Ms. Nakamizo for their technical assistance. We also appreciate the technical support from the Research Support Center, Graduate School of Medical Sciences, Kyushu University. The English usage in this article was reviewed by KN International (http://www.kninter. com/). Conflict of interest statement We declare that we have no conflict of interest.
References 1. Antonescu CR, Katabi N, Zhang L et al (2011) EWSR1-ATF1 fusion is a novel and consistent finding in hyalinizing clear-cell carcinoma of salivary gland. Genes Chromosomes Cancer 50:559–570 2. Shah AA, Legallo RD, van Zante A et al (2013) EWSR1 genetic rearrangements in salivary gland tumors: a specific and very common feature of hyalinizing clear cell carcinoma. Am J Surg Pathol 37:571– 578 3. Nakano T, Yamamoto H, Hashimoto K et al (2013) HER2 and EGFR gene copy number alterations are predominant in high-grade salivary mucoepidermoid carcinoma irrespective of MAML2 fusion status. Histopathology 63:378–392 4. Skálová A, Vanecek T, Sima R et al (2010) Mammary analogue secretory carcinoma of salivary glands, containing the ETV6NTRK3 fusion gene: a hitherto undescribed salivary gland tumor entity. Am J Surg Pathol 34:599–608 5. Mitani Y, Li J, Rao PH et al (2010) Comprehensive analysis of the MYB-NFIB gene fusion in salivary adenoid cystic carcinoma: incidence, variability, and clinicopathologic significance. Clin Cancer Res 16:4722–4731 6. Ellis G (2005) Clear cell carcinoma, not otherwise specified. In: Barnes L, Eveson JW, Reichart P, Sidransky D (eds) World health organization classification of tumours. Pathology and genetics of head and neck tumours. IARC Press, Lyon, pp 227–228 7. O’Sullivan-Mejia ED, Massey HD, Faquin WC, Powers CN (2009) Hyalinizing clear cell carcinoma report of eight cases and a review of literature. Head Neck Pathol 3:179–185 8. Milchgrub S, Gnepp DR, Vuitch F, Delgado R, Albores-Saavedra J (1994) Hyalinizing clear cell carcinoma of salivary gland. Am J Surg Pathol 18:74–82 9. Weinreb I (2013) Hyalinizing clear cell carcinoma of salivary gland: a review and update. Head Neck Pathol 7(Suppl 1):S20–S29
Virchows Arch 10. Yamaguchi U, Hasegawa T, Morimoto Y et al (2005) A practical approach to the clinical diagnosis of Ewing’s sarcoma/primitive neuroectodermal tumour and other small round cell tumours sharing EWS rearrangement using new fluorescence in situ hybridisation probes for EWSR1 on formalin fixed, paraffin wax embedded tissue. J Clin Pathol 58:1051–1056 11. Félix A, Rosa JC, Nunes JF, Fonseca I, Cidadão A, Soares J (2002) Hyalinizing clear cell carcinoma of salivary glands a study of extracellular matrix. Oral Oncol 38:364–368 12. Milchgrub S, Vuitch F, Saboorian MH, Hameed A, Wu H, AlboresSaavedra J (2000) Hyalinizing clear-cell carcinoma of salivary glands in fine-needle aspiration. Diagn Cytopathol 23:333–337 13. Uzochukwu NO, Shrier DA, Lapoint RJ (2007) Clear cell carcinoma the base of the tongue: MR imaging findings. AJNR Am J Neuroradiol 28:127–128 14. Auclair PL (1994) Tumor-associated lymphoid proliferation in the parotid gland. A potential diagnostic pitfall. Oral Surg Oral Med Oral Pathol 77:19–26 15. Mikami T, Cheng J, Maruyama S et al (2011) Emergence of keratin 17 vs. loss of keratin 13: their reciprocal immunohistochemical profiles in oral carcinoma in situ. Oral Oncol 47:497–503 16. Kitamura R, Toyoshima T, Tanaka H et al (2012) Association of cytokeratin 17 expression with differentiation in oral squamous cell carcinoma. J Cancer Res Clin Oncol 138:1299–1310
17. Kauzman A, Tabet JC, Stiharu TI (2011) Hyalinizing clear cell carcinoma a case report and review of the literature. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 112:e26–e34 18. Solar AA, Schmidt BL, Jordan RC (2009) Hyalinizing clear cell carcinoma: case series and comprehensive review of the literature. Cancer 115:75–83 19. Nakayama T, Miyabe S, Okabe M et al (2009) Clinicopathological significance of the CRTC3-MAML2 fusion transcript in mucoepidermoid carcinoma. Mod Pathol 22:1575–1581 20. Tirado Y, Williams MD, Hanna EY, Kaye FJ, Batsakis JG, El-Naggar AK (2007) CRTC1/MAML2 fusion transcript in high grade mucoepidermoid carcinomas of salivary and thyroid glands and Warthin’s tumors: implications for histogenesis and biologic behavior. Genes Chromosomes Cancer 46:708–715 21. Schwarz S, Stiegler C, Müller M et al (2011) Salivary gland mucoepidermoid carcinoma is a clinically, morphologically and genetically heterogeneous entity: a clinicopathological study of 40 cases with emphasis on grading, histological variants and presence of the t(11;19) translocation. Histopathology 58:557–570 22. Möller E, Stenman G, Mandahl N et al (2008) POU5F1, encoding a key regulator of stem cell pluripotency, is fused to EWSR1 in hidradenoma of the skin and mucoepidermoid carcinoma of the salivary glands. J Pathol 215:78–86
ORIGINAL ARTICLE
Hyalinizing clear cell carcinoma with EWSR1-ATF1 fusion gene: report of three cases with molecular analyses Takafumi Nakano & Hidetaka Yamamoto & Toshimitsu Nishijima & Sadafumi Tamiya & Hideki Shiratsuchi & Torahiko Nakashima & Shizuo Komune & Yoshinao Oda
Received: 19 June 2014 / Revised: 2 October 2014 / Accepted: 20 October 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Hyalinizing clear cell carcinoma (HCCC) is a lowgrade salivary gland carcinoma characterized by clear cells and hyalinized stroma. Recently, the EWSR1-ATF1 fusion gene was found in HCCCs. We herein describe three cases of HCCC identified in one male and two females, ranging in age from 27 to 67 years. The tumors were located in the root of tongue, nasopharynx, and soft palate. They were composed of nested or cord-like proliferations of epithelial cells with clear to pale eosinophilic cytoplasm, embedded in hyalinized and focally fibroedematous stroma. Tumor-associated lymphoid proliferation and pseudoepitheliomatous hyperplasia were also observed in each one case. MAML2 fusions specific to mucoepidermoid carcinoma were not detected in any of the three cases. We found EWSR1-ATF1 in two of three HCCCs using reverse transcription polymerase chain reaction (RTPCR) with our original primer sets designed to detect the fusion gene transcripts in formalin-fixed paraffin-embedded (FFPE) tissues. EWSR1 rearrangement was also confirmed by fluorescence in situ hybridization (FISH) on FFPE sections in two cases. There was a good concordance between the two
Electronic supplementary material The online version of this article (doi:10.1007/s00428-014-1676-5) contains supplementary material, which is available to authorized users. T. Nakano : H. Yamamoto (*) : T. Nishijima : S. Tamiya : Y. Oda Department of Anatomic Pathology, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan e-mail: [email protected] T. Nakano : T. Nishijima : H. Shiratsuchi : T. Nakashima : S. Komune Department of Otolaryngology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan S. Tamiya Department of Pathology, Kitakyushu Municipal Medical Center, Kitakyushu, Japan
methods (two positive cases and one negative case by both RT-PCR and FISH). Therefore, RT-PCR and FISH using FFPE tissue may be ancillary tools to confirm the diagnosis of HCCC. Keywords Hyalinizing clear cell carcinoma . EWSR1-ATF1 . Fusion gene . FISH . RT-PCR
Introduction Recently, characteristic fusion genes have been found in some salivary gland carcinomas, including EWSR1-ATF1 in hyalinizing clear cell carcinoma (HCCC), CRTC1- or CRTC3-MAML2 in mucoepidermoid carcinoma (MEC), ETV6-NTRK3 in mammary analogue secretory carcinoma, and MYB-NFIB in adenoid cystic carcinoma [1–5]. HCCC is a rare low-grade salivary gland carcinoma, characterized by clear to pale eosinophilic cells arranged in cords, trabeculae, or nests within a hyalinized stroma. HCCC is also known as clear cell carcinoma, not otherwise specified, in the recent WHO classification [6]. HCCC has a generally indolent clinical course and rarely shows lymph node or distant metastases [7, 8]. Morphologically, HCCC resembles a clear cell variant of MEC, as both tumors are composed of clear to eosinophilic cells, and mucinous differentiation can be seen not only in MEC but also in HCCC [1, 9]. Therefore, the histopathological distinction between HCCC and MEC is occasionally difficult. Recently, Antonescu et al. [1] first reported the presence of EWSR1 rearrangement by fluorescence in situ hybridization (FISH) in 18 of 22 (82 %) cases of HCCC, and the presence of EWSR1-ATF1 fusion transcript in which exon 11 of EWSR1 was fused to exon 3 of ATF1 by reverse transcriptasepolymerase chain reaction (RT-PCR) in both cases examined.
Virchows Arch
Subsequently, Shah et al. [2] confirmed the EWSR1 rearrangement in 13 of 15 (87 %) HCCCs by FISH. Here we report additional three cases of HCCC, two of which were positive for both EWSR1 rearrangement by FISH and EWSR1-ATF1 by RT-PCR using paraffin-embedded tissue and modified primer sets.
Materials and methods The excised specimens were fixed in a solution of 10 % formaldehyde and embedded in paraffin, and stained with hematoxylin-eosin. In all cases, additional stains were also carried out, including periodic acid-Schiff (PAS) with or without prior diastase treatment, mucicarmine and Alcian blue staining. Immunohistochemical staining Immunohistochemical staining was performed using 4-μmthick, formalin-fixed paraffin-embedded (FFPE) tissue, and the primary antibodies used in this study are listed in Supplemental Table 1. A biotin-free, horseradish peroxidase enzyme-labeled polymer method (Envision+ system, Dako, Carpinteria, CA, USA) with hematoxylin counterstaining was used. The immunohistochemistry results were scored as fellows: negative (no staining or 50 %). RT-PCR Total RNA was isolated from FFPE tissue using the miRNeasy FFPE Kit (Qiagen, Valencia, CA, USA), and first-strand cDNA was synthesized using Superscript III Transcriptase (Invitrogen, Carlsbad, CA, USA), according to the manufacturer’s instructions. To detect EWSR1-ATF1 fusion gene transcripts, we performed RT-PCR using our original primers designed to fit the FFPE samples as follows: the forward primer, 5′-caaggattaaatgacagtgtgactc-3′, and the reverse primer, 5′-ctttctgtgaggagcctatg-3′. In addition, RT-PCR for CRTC1-MAML2 and CRTC3-MAML2 was performed as previously reported [3]. Then, the sequence was confirmed by direct sequencing methods with an ABI Prism 310 sequence analyzer (Applied Biosystems, Foster City, CA, USA). In all three cases, the presence of sufficient quality of cDNA was confirmed by positive amplification of the phosphoglycerate kinase (PGK) (data not shown). EWSR1 break-apart FISH FISH was also performed using 4-μm-thick, unstained tissue sections and the EWSR1 break-apart probe set (Vysis LSI EWSR1 dual-color, break-apart rearrangement probe;
Abbott Molecular, Des Plaines, IL, USA), with the Vysis Paraffin Pretreatment IV and Post-hybridization Wash Buffer Kit (Abbott Molecular). The 5′ EWSR1 signal was labeled with SpectrumOrange (orange), and the 3′ EWSR1 signal was labeled with SpectrumGreen (green) [10]. The hybridized slides were reviewed on a fluorescence microscope (BZ9000, Keyence, Osaka, Japan) at ×60 magnification with oil immersion, using a DAPI/Green/Red triple band-pass filter set. The results were scored by evaluating 100 tumor cell nuclei per case. Tumor cells with nuclear overlap and indistinguishable separate nuclei were excluded from analysis. A split signal was defined by 5′ and 3′ signals observed at a distance greater than one-time signal width, and signals separated by less than that were regarded as fused signals. Tumor cells showing split or isolated 5′ signals were considered positive rearrangements. We interpreted a result as positive if more than 20 % of tumor cells showed positive rearrangement [1]. All cases of HCCC had fused 5′ and 3′ (yellow) signals in non-neoplastic cells as an internal control, indicating successful hybridization, and they were thereby eligible for FISH analysis.
Results Clinical summary Clinicopathologial findings are summarized in Table 1. The three patients were one man and two women, ranging in age from 27 to 67 years (median, 52 years). The tumors were located in the root of tongue, nasopharynx, and soft palate and ranged from 18 to 25 mm in maximum diameter. No lymph node metastasis was detected in any of the cases at diagnosis. During the follow-up (3–75 months, median 18 months), all patients were alive without tumor recurrence, lymph node metastasis, or distant metastasis. Histopathological findings Microscopically, the tumors consisted of a proliferation of clear cells with distinctive cell borders arranged in sheets, nests, cords, and trabecular patterns (Fig. 1a). The tumor cells infiltrated the tumor margins. Hyalinized eosinophilic material was observed within or around the tumor nests (Fig. 1a). Two cases (Cases 2 and 3) showed focal fibroedematous stroma between or around tumor nests (Fig. 1b). In addition, tumor cells occasionally showed pale eosinophilic cytoplasm (Fig. 1b). Neither perineural invasion, vascular invasion, nor tumor necrosis was present in any of our cases. Tumor cells had cytoplasmic glycogen, which was highlighted by PAS after diastase digestion (D-PAS) (Fig. 1c and d). In all three cases, goblet-like cells and small amounts of mucin were
Virchows Arch Table 1 Clinicopathological findings of three cases of hyalinizing clear cell carcinoma Case Age Sex Site
1
52
M
2
27
F
3
67
F
First sign or symptom
Size Treatment (mm)
Root of Painless swelling lesion of 18 tongue root of tongue Nasopharynx Hearing loss and feeling of 25 fullness in the right ear Soft palate Painless swelling lesion of 20 soft palate
Local Lymph node/distant Outcome recurrence metastasis
Surgical resection, adjuvant – radiotherapy (61.4 Gy) Surgical resection – Surgical resection, adjuvant – radiotherapy (32 Gy)
–
NED, 3 months
–
NED, 18 months
–
NED, 75 months
NED no evidence of disease
also focally seen, which was confirmed by D-PAS, Alcian blue, and mucicarmine staining (Fig. 1d–f). In addition, DPAS highlighted basement membrane-like material surrounding the tumor nests (Fig. 1d). In Case 1, overlying squamous epithelium showed pseudoepitheliomatous hyperplasia (PEH) and gradual transition from PEH to HCCC tumor cells with eosinophilic or clear cytoplasm (Fig. 2a and b). These features resembled a clear cell variant of squamous cell carcinoma (SCC). In Case 3, at the periphery of the tumor, lymphoid aggregates were noted with lymphoid follicles having germinal centers (Fig. 3), indicating a tumor-associated lymphoid proliferation (TALP). Immunohistochemical findings The results of immunohistochemical staining are summarized in Table 2. The tumors were positive for pan-cytokeratin (CK) (3/3 cases), low-molecular-weight CK (diffuse 1, focal 2), CK13 (focal and weak 1), CK 14 (2/3), CK 17 (diffuse 2, focal 1), epithelial membrane antigen (EMA) (diffuse 2, focal and weak 1), and p63 (3/3). S100 protein and alpha-smooth muscle actin were negative. In Case 1, overlying squamous epithelium with pseudoepitheliomatous hyperplasia showed CK17 positivity and reduced CK13 expression (Fig. 2c and d). Molecular genetic findings The results of molecular tests are also summarized in Table 2. Using RT-PCR, 136 bp of EWSR1-ATF1 fusion transcripts were detected in two of three cases (66.7 %) (Cases 1 and 2). The direct sequence of EWSR1-ATF1 fusion transcripts revealed that exon 11 of EWSR1 was fused to exon 3 of ATF1 (Fig. 4). Neither CRTC1-MAML2 nor CRTC3-MAML2 fusion transcript was identified in any of the cases. EWSR1 rearrangement was confirmed using FISH in two of three cases (66.7 %) (Cases 1 and 2). Positive rearrangement signals were observed in 79 % (split signals in 60 % and isolated 5′ signals in 19 %) and 69 % (split signals in 57 % and isolated 5′ signals in 12 %) of tumor cells in cases 1 and 2, respectively (Fig. 5). In case 3, a large number (97 %) of tumor
cells displayed fused signals, indicating absence of rearrangement. Altogether, two cases (Cases 1 and 2) were positive for EWSR1 rearrangement by both RT-PCR and FISH.
Discussion We used FFPE tissues to successfully identify EWSR1-ATF1 in HCCC. Only one previous report shows EWSR1-ATF1 in HCCCs by RT-PCR; in that study, 425 bp of EWSR1-ATF1 fusion transcripts were detected using frozen samples in two cases of HCCC [1]. We modified the primers for FFPE specimens, making it possible to detect a shorter size (136 bp) of EWSR1-ATF1 fusion transcripts in two of three cases. Moreover, EWSR1 rearrangement was also detected in two of three cases by FISH. There was agreement between the results of RT-PCR and those of FISH. Thus, the evidence indicates that our primers are suitable for detecting EWSR1ATF1 using RT-PCR and FFPE tissues. In our series, one of three HCCCs (33 %) was negative for EWSR1 gene rearrangement. Likewise, in the previous studies, a total of 6 of 37 HCCCs (16.2 %) lacked this rearrangement, even though FISH was successful [1, 2]. These results suggest that a minor subset of HCCC might have an as-yet-unknown genetic abnormality that is probably unrelated to the EWSR1 gene. In one case (Case 3) in our series, the tumor was surrounded by an inflammatory infiltration with lymphoid follicles having germinal centers, indicating a convincing example of TALP (Fig. 3). TALP has not been clearly described in HCCC in the literature. The literature notes a chronic inflammatory infiltrate in five cases of HCCC, but they all lacked apparent lymphoid follicles [11–13]. It is not surprising that TALP is present in HCCC, because TALP is a common phenomenon in some types of low-grade salivary gland tumors such as MEC and acinic cell carcinoma [14]. In our Case 1, it was difficult to distinguish between HCCC and SCC with clear cell change at routine histological diagnosis, because there was a morphological transition between the PEH and the HCCC area (Fig. 2a and b). PEH has been previously reported in three cases of HCCC [2]. PEH is a
Virchows Arch
Fig. 1 a HCCC showed a proliferation of clear cells with distinctive cell borders, arranged in sheets, nests, cords, and trabecular patterns, accompanied by characteristic hyalinizing stroma (HE staining) (Case 1). b Tumor cells occasionally have pale eosinophilic cytoplasm and fibroedematous stroma (HE staining) (Case 2). c–f PAS revealed
significant cytoplastic glycogen (c), which was digested by diastase (d). Intracytoplasmic mucin was focally detected by PAS with prior diastase digestion (D-PAS) (d), Alcian blue staining (e), and mucicarmine staining (f). D-PAS highlighted basement membrane-like materials surrounding the tumor nests (d) (Case 3)
benign epithelial change that can be easily confused with SCC. Although it has been reported that the expression profile of CK13 and CK17 is different between normal oral squamous epithelium (CK13+/CK17−) and squamous cell carcinoma (CK13−/CK17+) [15, 16], the expression pattern of CK13 and CK17 is variable in dysplasia and hyperplasia of oral mucosa. In our case, both HCCC cells and overlying squamous epithelium with PEH showed reduced CK13 expression and CK17 overexpression (Fig. 2c and d). In this
context, special attention needs to be paid to interpretation of the immunohistochemical results, because the expression patterns of CK13 and CK17 may be the same among SCC, PEH, and HCCC. One of the main differential diagnoses of HCCC is mucindepleted clear cell variant MEC. Both types of tumor consist of epithelial cells with clear to eosinophilic cytoplasm with various degrees of mucinous differentiation [1, 8]. Indeed, all our cases of HCCC focally had intracytoplasmic mucin,
Virchows Arch
Fig. 2 a, b Overlying hyperplastic squamous epithelium showed irregular, drop-down growth indicating pseudoepitheliomatous hyperplasia (PEH) (a). In connection with PEH, HCCC tumor cells with eosinophilic cytoplasm (arrow) and those with clear cytoplasm (arrowhead) showed gradual transitions, mimicking squamous cell carcinoma with clear cell
change (Case 1). c, d PEH showed decreased expression of CK13 (c) and CK17 protein overexpression (d). More than half of the HCCC tumor cells showed immunoreactivity for CK17 (d), but CK13 was negative (c) (Case 1)
Table 2 Immunohistochemical and molecular results in this study
Fig. 3 Tumor-associated lymphoid proliferation (TALP) was illustrated by lymphoid follicles having germinal centers around the tumor (Case 3)
Case no.
1
2
3
Pan-cytokeratin (AE1/AE3) CK-LMW (CAM5.2) CK13
++ + −
++ + +, weak
++ ++ −
CK14 CK17 EMA p63 S100 protein SMA EWSR1 EWSR1 CRTC1/CRTC3-MAML2
++ ++ +, weak ++ − − Positive Positive −
++ ++ ++ ++ − − Positive Positive −
− + ++ ++ − − − − −
FISH RT-PCR RT-PCR
CK-LMW low-molecular-weight cytokeratin, CK cytokeratin, EMA epithelial membrane antigen, SMA smooth muscle actin, ++ diffuse positive, + focal positive, − negative
Virchows Arch Fig. 4 Detection of EWSR1ATF1 fusion in HCCC. This image of the sequencing of an RT-PCR product shows fusion between EWSR1 exon 11 and ATF1 exon 3 (Case 1)
mimicking goblet cells in MEC. In addition, both HCCC and MEC preferentially occur in the minor salivary glands [1, 7, 8, 17–20]. Although there are confusingly similar histopathological features, characteristic fusion genes (CRTC1/CRTC3MAML2 in MECs and EWSR1-ATF1 in HCCCs) can be helpful for the differential diagnosis [1–3, 19–21]. Only one case of high-grade salivary mucoepidermoid carcinoma with EWSR1-POU5F1 has been reported to date [22]; however, EWSR1 rearrangement in MEC should be further elucidated. According to Antonescu et al. [1], none of 15 cases of HCCC showed MAML2 or POU5F1 rearrangement by FISH. Another important differential diagnosis of HCCC is clear cell variant of myoepithelial carcinoma. Both types of tumor share the histological appearance such as clear to eosinophilic cytoplasm and basement membrane-like materials. The distinctive feature of myoepithelial carcinoma is the expression of myoepithelial markers including S-100 protein and alphasmooth muscle actin; HCCC is generally negative for these markers. Indeed, all our cases of HCCC were negative for these markers, irrespective of EWSR1 rearrangement (Table 2). In summary, we report additional three cases of HCCC, two of which were positive for EWSR1 rearrangement by FISH and EWSR1-ATF1 fusion transcript by RT-PCR using FFPE tissues and modified primer sets. There was good
Fig. 5 Break-apart FISH for EWSR1 rearrangement. The majority of tumor cells show a split (separated) signal pattern of one green and one orange (arrow). The intact EWSR1 is shown by a fused (yellow) signal or closely juxtaposed orange and green signals (arrowhead) (case 1)
concordance between FISH and RT-PCR. One case of HCCC showed PEH, mimicking SCC with clear cell change. Therefore, FISH and RT-PCR using FFPE sections constitute ancillary tools to confirm the diagnosis of HCCC. In addition, we report a unique case of HCCC with convincing TALP. Acknowledgments The authors are grateful to Ms. Tateishi, Ms. Matsumoto, and Ms. Nakamizo for their technical assistance. We also appreciate the technical support from the Research Support Center, Graduate School of Medical Sciences, Kyushu University. The English usage in this article was reviewed by KN International (http://www.kninter. com/). Conflict of interest statement We declare that we have no conflict of interest.
References 1. Antonescu CR, Katabi N, Zhang L et al (2011) EWSR1-ATF1 fusion is a novel and consistent finding in hyalinizing clear-cell carcinoma of salivary gland. Genes Chromosomes Cancer 50:559–570 2. Shah AA, Legallo RD, van Zante A et al (2013) EWSR1 genetic rearrangements in salivary gland tumors: a specific and very common feature of hyalinizing clear cell carcinoma. Am J Surg Pathol 37:571– 578 3. Nakano T, Yamamoto H, Hashimoto K et al (2013) HER2 and EGFR gene copy number alterations are predominant in high-grade salivary mucoepidermoid carcinoma irrespective of MAML2 fusion status. Histopathology 63:378–392 4. Skálová A, Vanecek T, Sima R et al (2010) Mammary analogue secretory carcinoma of salivary glands, containing the ETV6NTRK3 fusion gene: a hitherto undescribed salivary gland tumor entity. Am J Surg Pathol 34:599–608 5. Mitani Y, Li J, Rao PH et al (2010) Comprehensive analysis of the MYB-NFIB gene fusion in salivary adenoid cystic carcinoma: incidence, variability, and clinicopathologic significance. Clin Cancer Res 16:4722–4731 6. Ellis G (2005) Clear cell carcinoma, not otherwise specified. In: Barnes L, Eveson JW, Reichart P, Sidransky D (eds) World health organization classification of tumours. Pathology and genetics of head and neck tumours. IARC Press, Lyon, pp 227–228 7. O’Sullivan-Mejia ED, Massey HD, Faquin WC, Powers CN (2009) Hyalinizing clear cell carcinoma report of eight cases and a review of literature. Head Neck Pathol 3:179–185 8. Milchgrub S, Gnepp DR, Vuitch F, Delgado R, Albores-Saavedra J (1994) Hyalinizing clear cell carcinoma of salivary gland. Am J Surg Pathol 18:74–82 9. Weinreb I (2013) Hyalinizing clear cell carcinoma of salivary gland: a review and update. Head Neck Pathol 7(Suppl 1):S20–S29
Virchows Arch 10. Yamaguchi U, Hasegawa T, Morimoto Y et al (2005) A practical approach to the clinical diagnosis of Ewing’s sarcoma/primitive neuroectodermal tumour and other small round cell tumours sharing EWS rearrangement using new fluorescence in situ hybridisation probes for EWSR1 on formalin fixed, paraffin wax embedded tissue. J Clin Pathol 58:1051–1056 11. Félix A, Rosa JC, Nunes JF, Fonseca I, Cidadão A, Soares J (2002) Hyalinizing clear cell carcinoma of salivary glands a study of extracellular matrix. Oral Oncol 38:364–368 12. Milchgrub S, Vuitch F, Saboorian MH, Hameed A, Wu H, AlboresSaavedra J (2000) Hyalinizing clear-cell carcinoma of salivary glands in fine-needle aspiration. Diagn Cytopathol 23:333–337 13. Uzochukwu NO, Shrier DA, Lapoint RJ (2007) Clear cell carcinoma the base of the tongue: MR imaging findings. AJNR Am J Neuroradiol 28:127–128 14. Auclair PL (1994) Tumor-associated lymphoid proliferation in the parotid gland. A potential diagnostic pitfall. Oral Surg Oral Med Oral Pathol 77:19–26 15. Mikami T, Cheng J, Maruyama S et al (2011) Emergence of keratin 17 vs. loss of keratin 13: their reciprocal immunohistochemical profiles in oral carcinoma in situ. Oral Oncol 47:497–503 16. Kitamura R, Toyoshima T, Tanaka H et al (2012) Association of cytokeratin 17 expression with differentiation in oral squamous cell carcinoma. J Cancer Res Clin Oncol 138:1299–1310
17. Kauzman A, Tabet JC, Stiharu TI (2011) Hyalinizing clear cell carcinoma a case report and review of the literature. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 112:e26–e34 18. Solar AA, Schmidt BL, Jordan RC (2009) Hyalinizing clear cell carcinoma: case series and comprehensive review of the literature. Cancer 115:75–83 19. Nakayama T, Miyabe S, Okabe M et al (2009) Clinicopathological significance of the CRTC3-MAML2 fusion transcript in mucoepidermoid carcinoma. Mod Pathol 22:1575–1581 20. Tirado Y, Williams MD, Hanna EY, Kaye FJ, Batsakis JG, El-Naggar AK (2007) CRTC1/MAML2 fusion transcript in high grade mucoepidermoid carcinomas of salivary and thyroid glands and Warthin’s tumors: implications for histogenesis and biologic behavior. Genes Chromosomes Cancer 46:708–715 21. Schwarz S, Stiegler C, Müller M et al (2011) Salivary gland mucoepidermoid carcinoma is a clinically, morphologically and genetically heterogeneous entity: a clinicopathological study of 40 cases with emphasis on grading, histological variants and presence of the t(11;19) translocation. Histopathology 58:557–570 22. Möller E, Stenman G, Mandahl N et al (2008) POU5F1, encoding a key regulator of stem cell pluripotency, is fused to EWSR1 in hidradenoma of the skin and mucoepidermoid carcinoma of the salivary glands. J Pathol 215:78–86
