Sry-related HMG-Box gene 10 (SOX10) protein is a

Transcription

1 A Newly Developed Mouse Monoclonal SOX10 Antibody Is a Highly Sensitive and Specific Marker for Malignant Melanoma, Including Spindle Cell and Desmoplastic Melanomas David Tacha, PhD; Weimin Qi, PhD, MD; Seong Ra, MD; Ryan Bremer, PhD; Charlie Yu, MD; Joseph Chu, BS; Laura Hoang, PhD; Bruce Robbins, MD Context. Recent immunohistochemical studies have demonstrated Sry-related HMG-Box gene 10 (SOX10) expression in malignant melanomas, malignant peripheral nerve sheath tumors, a subset of breast carcinomas, and gliomas. SOX10 has shown important clinical utility in its ability to detect desmoplastic and spindle cell melanomas. To date, most publications have employed a research use only goat polyclonal SOX10 antibody for immunohistochemical staining. Objective. To describe the development of a new mouse monoclonal SOX10 antibody (BC34) and evaluate its immunohistochemical staining profile in a wide range of normal and neoplastic tissues, with an emphasis on melanoma. Design. SOX10 antibody was optimized for staining using a polymer detection system and visualization with diaminobenzidine. Results. In normal tissues, SOX10 was expressed in skin melanocytes and eccrine cells, breast myoepithelial and lobular epithelial cells, salivary gland myoepithelial cells, peripheral nerve Schwann cells, and central nervous system glial cells. SOX10 was expressed in 238 of 257 melanomas (92.6%), including 50 of 51 of both spindle cell and desmoplastic melanomas (98%). SOX10 was expressed in 100% of nevi (20 of 20) and schwannomas (28 of 28). In other neoplasms, SOX10 was expressed in 18 of 109 invasive ductal breast carcinomas (16.5%). All other carcinomas were negative for SOX10. SOX10 was identified in 25 of 52 central nervous system neoplasms, primarily in astrocytomas (22 of 41; 53.7%), and in 4 of 99 various sarcomas examined (4.0%). Conclusions. The newly developed mouse monoclonal SOX10 antibody BC34 is highly sensitive and specific for malignant melanoma, including desmoplastic and spindle cell variants, and appears highly suitable for clinical use. (Arch Pathol Lab Med. 2015;139: ; doi: / arpa oa) Sry-related HMG-Box gene 10 (SOX10) protein is a human transcription factor important for neural crest, peripheral nervous system, and melanocytic cell development. 1 4 The SOX10 nuclear protein has been shown to be widely expressed in normal human tissues, including melanocytes, Schwann cells, breast and salivary gland myoepithelial cells, and oligodendroglial cells. 1 3,5,6 It is also expressed in malignant tumors, such as melanoma, malignant peripheral nerve sheath tumors, and a subset of breast Accepted for publication May 19, Published as an Early Online Release December 1, From the Chief Medical Office (Dr Tacha), and the Departments of Research and Development, Biocare Medical, LLC, Concord, California (Drs Qi, Bremer, Yu, and Hoang, and Mr Chu); and the San Diego Pathologists Medical Group, San Diego, California (Drs Ra and Robbins). Dr Tacha is a founder, chief scientific officer, and stockholder of Biocare Medical, LLC, Concord, California. Drs Qi, Bremer, Yu, and Hoang, and Mr Chu are employees of Biocare Medical, LLC. The other authors have no relevant financial interest in the products or companies described in this article. Reprints: David Tacha, PhD, Biocare Medical, 4040 Pike Ln, Concord, CA ( dtacha@biocare.net). carcinomas. Most oligodendrogliomas, but also a large percentage of astrocytomas and glioblastomas, have also been shown to express SOX10. 3,5 SOX10 is also present in benign tumors, such as melanocytic nevi, schwannomas, and neurofibromas. 1 3,5 7 SOX10 is highly expressed in desmoplastic melanomas (DMs) and spindle cell melanomas (SCMs). 1 Desmoplastic melanomas and SCMs are usually positive for S100 protein but are often negative or only focally positive with other melanoma markers. Incidence rates of DM have steadily increased during the past 15 years, 8 and DM is prone to misdiagnosis not only clinically, but also histologically. 9,10 There has been an extensive effort to find a suitable marker to differentiate SCMs and DMs from their pathologic mimics. Among melanocyte markers, SOX10 has shown most promise. To date, most publications have used an immunohistochemistry method using a research use only goat polyclonal SOX10. 1,7,11 18 We herein describe the development of a new mouse monoclonal SOX10 hybridoma (BC34) and evaluate its sensitivity and specificity in normal and neoplastic tissues, with an emphasis on melanoma and its subtypes. 530 Arch Pathol Lab Med Vol 139, April 2015 SOX10, and Desmoplastic and Spindle Cell Melanomas Tacha et al

2 Figure 1. Sry-related HMG-Box gene 10 (SOX10) Western blot. MATERIALS AND METHODS Generation of SOX10 Hybridoma Host Immunization. Female BALB/c (6 to about 8 weeks old) mice were immunized intraperitoneally with 100 lg of human SOX10 recombinant protein per mouse in complete Freund adjuvant. Three weeks later, the mice were boosted with another 100 lg of human SOX10 per mouse in incomplete Freund adjuvant 4 more times in 3-week intervals. Mice were bled from the tails and sera were collected and stored at 208C. Hybridoma. Hybridomas producing antibodies to SOX10 were generated by standard techniques from splenocytes of SOX10-immunized BALB/c mice. Splenocytes from SOX10-immunized mice were fused to P3-X63-Ag myeloma cells. Hybridoma supernatants were screened by enzyme-linked immunoassay for antibody reactivity to SOX10. Hybridoma clones were then selected based on their specificity and specificity on formalinfixed, paraffin-embedded sections of targeted tissues. Enzyme-Linked Immunoassay. Host anti-sera immune responses to SOX10 were measured by enzyme-linked immunoassay and were measured in a microtiter plate reader. Anti-SOX10 antibodies (BC34) monoclonal antibody was isotyped using a mouse monoclonal antibody isotyping kit (Invitrogen, Carlsbad, California). Clone BC34 isotype was identified as a mouse immunoglobulin (Ig) G1. The selected hybridoma cells from clone BC34 were cultured in culture medium supplemented with 10% fetal bovine serum. The hybridoma cells were injected into BALB/c mice to produce antibody ascites. The antibody ascites were further purified on a protein A affinity column. The IgG concentration was measured spectrophotometrically. Antibody Cross-Reactivity Tested by Western Blotting. The purified monoclonal antibody SOX10 was characterized by Western blotting. Full-length human SOX10-transfected cell lysates were separated using gel electrophoresis. Protein blots were visualized by incubating SOX10 antibody for 60 minutes at room temperature, followed by incubation with peroxidaseconjugated goat anti-mouse immunoglobulins (Figure 1). Table 1. Sry-Related HMG-Box Gene 10 (SOX10) Expression in Normal Tissue Types (n ¼ 34) a Organ Cerebrum Cerebellum Adrenal Ovary Pancreas Thyroid Parathyroid Testis Bone Spleen Tonsil Thymus Bone marrow Larynx Lung Heart Pituitary Breast Esophagus Stomach Small intestine Colon Liver Salivary gland Kidney Bladder Prostate Uterus Uterine cervix Skeletal muscle Skin Peripheral nerve Placenta Mesothelium a þ, positive;, negative. SOX10 Expression þ (glial cells) þ (glial cells) (bronchial myoepithelial cells) þ (myoepithelial cells, subset of lobular epithelial cells) (rare stromal cells in muscularis propria and mucosa) (rare stromal cells in muscularis propria and mucosa) (rare stromal cells in muscularis propria and mucosa) þ (myoepithelial cells) þ (melanocytes, eccrine myoepithelial cells, subset of eccrine epithelial cells) þ (Schwann cells) Tissue Microarrays. Tissue microarrays (TMAs) were constructed in-house and/or purchased from US Biomax (Rockville, Maryland). SOX10 (BC34) was evaluated in various types of malignant melanomas, including cutaneous melanoma, metastatic melanoma, SCMs and DMs, and melanoma variants (n ¼ 257), in schwannomas (n ¼ 28), and in nevi (compound, intradermal, and junctional; n ¼ 20). To test SOX10 for specificity, a normal TMA (n ¼ 34), and whole-tissue sections and TMAs in various neoplastic tissues (n ¼ 764) were evaluated. Immunohistochemistry. Formalin-fixed, paraffin-embedded whole tissues and tissue TMAs were deparaffinized and hydrated to water. Slides were then exposed to 5 minutes of hydrogen peroxide and washed in water. The TMAs were immersed in an antigen retrieval solution (modified citrate buffer, ph 6.0) and placed in a pressure cooker (Decloaking Chamber, Biocare Medical, Concord, California) at 1258C for 30 seconds. Tissue slides were cooled to 808C and washed in water. SOX10 (BC34) was diluted and optimized at 1:100 in a modified Tris diluent, ph 6.0, and tissues were incubated for 30 minutes and then rinsed in Trisbuffered saline. A goat anti-mouse horseradish peroxidase or alkaline phosphatase polymer detection (MACH 2, Biocare Medical) was applied to tissue sections for 30 minutes at room temperature. Sections were then rinsed in Tris-buffered saline and sections were incubated for either 5 minutes in 3,3 0 -diaminobenzidine or 10 minutes in Fast Red chromogen. Arch Pathol Lab Med Vol 139, April 2015 SOX10, and Desmoplastic and Spindle Cell Melanomas Tacha et al 531

3 Figure 2. Sry-related HMG-Box gene 10 (SOX10) in normal tissues. A, Normal skin. B, Normal breast. C, Normal salivary gland. D, Peripheral nerve. E, Cerebellum. F, Cerebrum (original magnification 310). Scoring Method. Each case was deemed positive if 1% or more of cells were observed staining. Conversely, less than 1% of tumor cells staining determined a case to be negative. Only nuclear staining was considered positive. RESULTS Immunohistochemical staining with the monoclonal SOX10 antibody produced strong nuclear staining in positive tissues, with essentially no background staining. In normal tissues (n ¼ 34), SOX10 was expressed in skin melanocytes, a portion of eccrine glandular epithelial and myoepithelial cells, breast myoepithelial cells as well as a subset of lobular epithelial cells, salivary gland myoepithelial cells, peripheral nerve, Schwann cells, and brain glial cells (Figure 2). SOX10 also stained small numbers of stromal cells in the muscularis propria and lamina propria throughout the digestive tract and rare stromal cells in other various tissues (total population of cells stained was,1%); however, all other cell components in these tissues showed no expression and these tissues were therefore scored as negative (Table 1). SOX10 was expressed in 238 of 257 melanomas (92.6%; Figure 3; Table 2). Notably, 50 of 51 SCMs and DMs (98%) were positive for SOX10 (Figure 4, A through C), and SOX10 also stained sarcomatoid melanoma (Figure 4, D). In benign cases, SOX10 was positive in schwannomas (Figure 5, A) and in nevi (20 of 20; Figure 5, B; Table 2). In other neoplasms (Table 3), SOX10 was expressed in 18 of 109 invasive ductal breast carcinomas (16.5%; Figure 6, A). SOX10 stained myoepithelial cells surrounding breast ductal carcinoma in situ (Figure 6, B). SOX10 staining was identified in 25 of 52 central nervous system neoplasms, 532 Arch Pathol Lab Med Vol 139, April 2015 SOX10, and Desmoplastic and Spindle Cell Melanomas Tacha et al

4 Figure 3. Sry-related HMG-Box gene 10 (SOX10) in melanoma. A, Melanoma with melanin pigment. B, Cutaneous melanoma. C, Metastatic melanoma in lymph node. D, Epithelioid melanoma (original magnifications 310 [A] and 320 [B through D]). primarily in astrocytomas (22 of 41; 53.7%; Figure 6, C; Table 3); and in limited cases it was expressed in glioblastoma and ependymoma (Table 3). All other carcinomas studied were negative for SOX10. Among sarcomas, SOX10 staining was observed in 2 of 31 leiomyosarcomas (6.5%), in 1 of 22 rhabdomyosarcomas (4.5%), and in 1 of 13 malignant fibrous histiocytomas (7.7%), with all other sarcomas negative (Table 4). In addition, SOX10 was positive in schwannomas (28 of 28; Table 4). Evaluation of neuroendocrine tumors showed SOX10 to be negative in all carcinoid tumors of the digestion tract, including colon, stomach, small intestine, and appendix. In lung and intestinal carcinoid tumors, a few sustentacular cells were positive for SOX10, but neoplastic cells appeared negative (Figure 6, D). COMMENT Previous studies have shown SOX10 to be a highly sensitive and specific nuclear marker for melanoma in both primary and metastatic lesions. 1,11 15 In our study, 115 of 119 primary cutaneous melanomas (96.6%) stained with SOX10. This compares well with the study by Nonaka et al, 7 where SOX10 nuclear expression was found in 76 of 78 melanomas (97%). In a study by Shin et al, 13 SOX10 was expressed in 100% of DMs and was negative in all histologic mimics, such as spindle cell carcinoma, atypical fibroxanthoma, and Table 2. Melanocytic Lesions (n ¼ 277) Melanoma No. of Cases SOX10 þ, No. Positive, % Melanoma (cutaneous) Metastatic melanoma Spindle cell melanoma Desmoplastic melanoma Desmoplastic/spindle cell mixed features Benign nevus (various) Abbreviation: SOX10, Sry-related HMG-Box gene 10. Arch Pathol Lab Med Vol 139, April 2015 SOX10, and Desmoplastic and Spindle Cell Melanomas Tacha et al 533

5 Figure 4. Sry-related HMG-Box gene 10 (SOX10) expression in spindle cell/desmoplastic/sarcomatoid melanomas. A, Spindle cell melanoma. B, Desmoplastic melanoma. C, In situ component of desmoplastic melanoma. D, Sarcomatoid melanoma (original magnifications 310 [A] and 320 [B through D]). Figure 5. Sry-related HMG-Box gene 10 (SOX10) expression in schwannoma and nevus. A, Schwannoma. B, Benign nevus (original magnifications 310 [A] and 320 [B]). sarcomas. Karamchandani et al 16 also showed comparable results, with only 1 of 78 cases of various sarcomas (1.3%) expressing SOX10. In our study, SOX10 was expressed in 50 of 51 DMs and SCMs (98%), and was expressed in 4 of 99 various types of sarcomas (4.0%; Table 2). Nonaka et al 7 showed SOX10 to be expressed in 38 of 77 malignant peripheral nerve sheath tumors (49%). Our study was limited, including only 2 malignant peripheral nerve sheath tumors, both of which were negative. Desmoplastic melanoma and SCM can present diagnostic challenges for the pathologist because of histologic mimics and limitations with immunohistochemical staining (Figure 4). Although S100 usually stains DM, other melanoma markers, such as HMB-45 and Melan-A, have been shown 534 Arch Pathol Lab Med Vol 139, April 2015 SOX10, and Desmoplastic and Spindle Cell Melanomas Tacha et al

6 Figure 6. Sry-related HMG-Box gene 10 (SOX10) expressed in various neoplastic tissues. A, Breast, invasive ductal carcinoma. B, Myoepithelial cells surrounding breast ductal cell carcinoma in situ. C, Astrocytoma. D, Intestinal carcinoid (sustentacular cells) (original magnifications 320 [A and C] and 310 [B and D]). to be negative in most cases. 14 Other histologic mimics of DM include spindled fibroblasts or histiocytes within prior excision scars. Ramos-Herberth et al 15 demonstrated that SOX10 was less likely than S100 to be expressed by background fibroblasts and histiocytes within scars, and thus SOX10 was superior to S100 in these types of cases. Despite the lack of specificity of S100, pathologists still use S100 in the diagnosis of neural crest derived tumors; however, SOX10 has been shown to be highly specific and to be rarely expressed in non-schwannian and nonmelanocytic tumors in the differential diagnosis of melanoma and peripheral nerve sheath tumors. A recommended approach would include both S100 and SOX10 in the diagnosis of melanoma or peripheral nerve sheath tumors. 16 In our study, SOX10 was negative in the vast majority of nonmelanocytic tumors (Table 3). These findings show concordance with those of other studies. 1,7 SOX10 has been demonstrated in a subset of breast carcinomas, including basal-like or triple-negative carcinomas, and in metaplastic carcinomas. This finding thus supports the concept that these neoplasms may show myoepithelial differentiation. 18 In our study, SOX10 nuclear staining was expressed in normal breast myoepithelial cells, as well as in a subset of breast lobular epithelial cells (Figure 2, D), and was expressed in 16.5% of invasive ductal carcinomas. SOX10 has been shown to stain various types of brain tumors. 5,6 Bannykh et al 5 demonstrated that most oligodendrogliomas, and a large fraction of astrocytomas and glioblastomas, expressed SOX10, correlating with our results. Our findings of SOX10 expression in malignant melanomas (96.6%) and in benign nevi and schwannomas (100%) demonstrated a high concordance with other studies using the well-published SOX10 goat polyclonal antibody. 1,10 15,19 The use of a research use only goat primary antibody may be satisfactory for research purposes; however, it may not be generally accepted in a clinical setting. Polyclonal antibodies are also notorious for lot-to-lot variation and may produce unwanted nonspecific background staining. In the study by Mohamed et al, 1 nonspecific background staining was observed in photomicrographs in both melanoma and in breast cancer. In a study by Zhong et al, 19 SOX10 was strongly positive in the cytoplasm of benign prostate tissue and was weakly positive in the cytoplasm of prostate cancer tissues. In our study, the new monoclonal SOX10 antibody produced clean nuclear staining with essentially no background staining. Arch Pathol Lab Med Vol 139, April 2015 SOX10, and Desmoplastic and Spindle Cell Melanomas Tacha et al 535

7 Table 3. Sry-Related HMG-Box Gene 10 (SOX10) Expression in Various Neoplastic Tissues (n ¼ 637) Neoplasm No. SOX10 þ, No. Positive, % Lung carcinomas Adenocarcinoma Bronchioloalveolar carcinoma Squamous cell carcinoma Adenosquamous cell carcinoma Large cell carcinoma Malignant mesothelioma Colon adenocarcinoma Breast invasive ductal carcinoma Prostate adenocarcinoma Urothelial carcinoma Renal cell carcinoma (clear cell) Liver (hepatocellular carcinoma) Esophageal adenocarcinoma/squamous cell carcinoma Testicular seminoma Ovarian serous/endometrioid adenocarcinoma Adrenal pheochromocytoma Thyroid papillary carcinoma Pancreas adenocarcinoma Islet cell tumor Non-Hodgkin lymphoma (B-cell and T-cell) Hodgkin lymphoma Low-grade neuroendocrine tumor (carcinoid tumor): digestive tract and lung Cervix adenocarcinoma/squamous cell carcinoma Skin carcinomas Basal cell carcinoma Squamous cell carcinoma CNS neoplasms Astrocytoma Glioblastoma Ependymoma Medulloblastoma Abbreviation: CNS, central nervous system. Table 4. Sry-Related HMG-Box Gene 10 (SOX10) Expression in Soft Tissue Tumors (n ¼ 127) Soft Tissue Tumors No. SOX10 þ, No. Positive, % Schwannoma (neurilemmoma) Leiomyosarcoma Rhabdomyosarcoma Fibrosarcoma Dermatofibrosarcoma protuberans Liposarcoma Angiosarcoma Neurofibrosarcoma Malignant fibrous histiocytoma CONCLUSION In summary, we have shown that the newly developed mouse monoclonal SOX10 antibody (BC34) is highly sensitive and specific for malignant melanoma, including desmoplastic and spindle cell variants, and appears highly suitable for clinical use. References 1. Mohamed A, Gonzalez RS, Lawson D, et al. SOX10 expression in malignant melanoma, carcinoma, and normal tissues. Appl Immunohistochem Mol Morphol. 2013;21(6): Pusch C, Hustert E, Pfeifer D, et al. The SOX10/Sox10 gene from human and mouse: sequence, expression, and transactivation by the encoded HMG domain transcription factor. Hum Genet. 1998;103(2): Mollaaghababa R, Pavan WJ. The importance of having your SOX on: role of SOX10 in the development of neural crest-derived melanocytes and glia. Oncogene. 2003;22(20): Bondurand N, Kobetz A, Pingault V, et al. Expression of the SOX10 gene during human development. FEBS Lett. 1998;432(3): Bannykh SI, Stolt CC, Kim J, et al. Oligodendroglial-specific transcriptional factor SOX10 is ubiquitously expressed in human gliomas. J Neurooncol. 2006; 76(2): Britsch S, Goerich DE, Riethmacher D, et al. The transcription factor Sox10 is a key regulator of peripheral glial development. Genes Dev. 2001;15(1): Nonaka D, Chiriboga L, Rubin BP. Sox10: a pan-schwannian and melanocytic marker. Am J Surg Pathol. 2008;32(9): Feng Z, Wu X, Chen V, et al. Incidence and survival of desmoplastic melanoma in the United States, J Cutan Pathol. 2011;38(8): Conley J, Lattes R, Orr W. Desmoplastic malignant melanoma (a rare variant of spindle cell melanoma). Cancer. 1971;28(4): Busam KJ. Desmoplastic melanoma. Clin Lab Med. 2011;31(2): George E, McClain SE, Slingluff CL, et al. Subclassification of desmoplastic melanoma: pure and mixed variants have significantly different capacities for lymph node metastasis. J Cutan Pathol. 2009;36(4): Palla B, Su A, Binder S, Dry S. SOX10 expression distinguishes desmoplastic melanoma from its histologic mimics. Am J Dermatopathol. 2013;35(5): Shin J, Vincent JG, Cuda JD, et al. Sox10 is expressed in primary melanocytic neoplasms of various histologies, but not in fibrohistiocytic proliferations and histiocytoses. J Am Acad Dermatol. 2012;67(4): Palla B, Su A, Binder S, Dry S. SOX10 expression distinguishes desmoplastic melanoma from its histologic mimics. Am J Dermatopathol. 2013;35(5): Ramos-Herberth FI, Karamchandani J, Kim J, et al. SOX10 immunostaining distinguishes desmoplastic melanoma from excision scar. J Cutan Pathol. 2010; 37(9): Karamchandani JR, Nielsen TO, van de Rijn M, et al. SOX10 and S100 in the diagnosis of soft-tissue neoplasms. Appl Immunohistochem Mol Morphol. 2012;20(5): Tsuta K, Raso MG, Kalhor N, et al. Sox10-positive sustentacular cells in neuroendocrine carcinoma of the lung. Histopathology. 2011;58(2): Cimino-Mathews A, Subhawong AP, Elwood H, et al. Neural crest transcription factor Sox10 is preferentially expressed in triple-negative and metaplastic breast carcinomas. Hum Pathol. 2013;44(6): Zhong WD, Qin GQ, Dai QS, et al. SOXs in human prostate cancer: implication as progression and prognosis factors. BMC Cancer. 2012;12: Arch Pathol Lab Med Vol 139, April 2015 SOX10, and Desmoplastic and Spindle Cell Melanomas Tacha et al