Immunohistochemistry for CD117 (KIT) Is Effective in Distinguishing Cutaneous Adnexal Tumors with Apocrine/Eccrine or Sebaceous Differentiation from Other Epithelial Tumors of the Skin Keisuke Goto Department of Diagnostic Pathology, Kainan Hospital, 396 Minamihonden, Maegasu-cho, Yatomi-city, Aichi, 498-8502 Japan
Corresponding author: Keisuke Goto, MD Department of Diagnostic Pathology, Kainan Hospital, 396 Minamihonden, Maegasu-cho, Yatomi-city, Aichi, 498-8502 Japan E-mail: [email protected] Tel: +81-(0)567-65-2511; Fax: +81-(0)567-67-3697 Keywords: c-KIT protein, immunohistochemistry, skin neoplasms
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/cup.12492
This article is protected by copyright. All rights reserved
ACKNOWLEDGEMENTS The author thanks Drs. Yoshimi Oshitani (Department of Dermatology, Kainan Hospital), Toshiaki Sato (Department of Plastic Surgery, Kainan Hospital), Yukio Ishikawa (Pathology Division, Itabashi Medical Laboratory), and Makoto Takeuchi (Department of Dermatology, Tsushima Municipal Hospital) for supplying valuable materials, and thanks Mr. Shinya Hattori and Ms. Mayumi Arakawa for technical assistance.
ABSTRACT Background: The aim of this study was to evaluate the effectiveness of CD117 immunostaining for diagnosing cutaneous adnexal tumors, which has not been previously established. Methods: CD117 immunostaining was performed on representative sections of 87 clinicopathologically confirmed cutaneous apocrine/eccrine tumors, 13 sebaceous tumors, 52 follicular tumors, 55 keratinocytic tumors, 24 metastatic tumors of the skin of known origin, and 1 mammary Paget disease. Staining of 5–50% of cells was considered focally positive, and staining of 51–100% of cells was considered diffusely positive. Staining intensity was semi-quantitatively graded as weak, moderate, or strong.
This article is protected by copyright. All rights reserved
Results: Seventy-seven (95.1%) tumors with apocrine/eccrine differentiation were CD117-positive, but 1 mucinous carcinoma, 1 apocrine carcinoma, and 3 primary cutaneous extramammary Paget diseases were CD117-negative. For most positive cases, staining was diffuse (64/82 cases) and moderate to strong in intensity (74/82). Twelve (92.3%) sebaceous tumors were CD117-positive, showing mainly focal (9/12) and weak (7/12) positivity. Of the 52 follicular tumors, all 4 cases of pilomatricoma and other 5 tumors were positive for CD117. Of the 55 keratinocytic tumors, only 2 cases were CD117-positive. The 24 metastatic tumors and mammary Paget disease were CD117-negative. Conclusions: CD117 may be useful for the differential diagnosis of cutaneous adnexal tumors.
This article is protected by copyright. All rights reserved
INTRODUCTION CD117 (KIT) is a transmembrane tyrosine kinase receptor operating in cell signal transduction and is normally activated upon binding of its ligand, the stem cell factor, in several cell types. Ligand binding results in a phosphorylation cascade, resulting in the activation of various transcription factors that regulate cell differentiation, migration, adhesion, apoptosis, and proliferation (1). KIT-dependent cell types include mast cells, some hematopoietic stem cells, germ cells, melanocytes, and Cajal cells of the gastrointestinal tract (2). Neoplasms of these cells – mastocytosis, acute myeloid leukemia, seminoma, malignant melanoma, and gastrointestinal stromal tumor – are often induced by KIT activation through gain-of-function mutations (1,3–5), and most of these tumors are examples of CD117-positive malignancies (2,6–16). Other normal cells that express CD117 include some Langerhans cells, cerebellar neurons, and epithelial cells in breast, kidney, and skin adnexa (6,7,17). CD117 may also be expressed in some sarcomas (Ewing sarcoma (18–20), angiosarcoma (16), Kaposi sarcoma (21), alveolar soft part sarcoma (22), extraskeletal myxoid chondrosarcoma (22), etc.) and epithelial neoplasms (adenoid cystic carcinoma (23); polymorphous low-grade adenocarcinoma (24); Merkel cell carcinoma (25–28); some small cell carcinomas (27,29); thymic carcinoma (30); chromophobe renal cell carcinoma (31);
This article is protected by copyright. All rights reserved
renal oncocytoma (31); juxtaglomerular cell tumor (32); and a few lung, ovarian, and breast carcinomas (16,22)). Among these mesenchymal and epithelial neoplasms, few of these tumors have any evidence of KIT gain-of-function mutations. Instead, some of these tumors have been reported to have concomitant overexpression of both KIT and stem cell factor, resulting in an autocrine loop (33–36). Previous reports have demonstrated that certain adnexal tumors, including primary cutaneous adenoid cystic carcinoma (37), clear cell hidradenoma (38), and basal cell carcinoma (39), are also positive for CD117. Although CD117 immunohistochemistry is often utilized as a conclusive diagnostic method, especially for mastocytosis, seminoma, gastrointestinal stromal tumor, adenoid cystic carcinoma, and thymic carcinoma (16), the utility of CD117 immunostaining in cutaneous adnexal tumors has not yet been established. The aim of the present study was to evaluate the effectiveness of CD117 immunostaining for diagnosing cutaneous adnexal tumors.
MATERIALS AND METHODS Cases A total of 87 cutaneous adnexal tumors with apocrine/eccrine differentiation, 13 sebaceous
This article is protected by copyright. All rights reserved
tumors, 52 follicular tumors, 55 keratinocytic tumors, 24 metastatic carcinomas of the skin, and 1 mammary Paget disease were retrospectively retrieved from the pathology archive of Kainan Hospital and the author’s consultation file from 2000 to 2014 (only pilomatricomas, basal cell carcinomas, and keratinocytic tumors from 2013 to 2014); all guidelines for experimental investigation with human subjects required by the institution were followed. The 87 apocrine/eccrine tumors included 29 poromas, 10 hidradenomas, 15 mixed tumors of the skin, 6 primary cutaneous extramammary Paget diseases, and 6 other various carcinomas with apocrine/eccrine differentiation, among others (Table 1). For combined apocrine/eccrine adnexal tumors with 2 or more tumor components, the predominant component was named for its representative diagnosis. The 13 sebaceous tumors included 5 sebaceomas and 8 sebaceous carcinomas. The 52 follicular tumors included 4 pilomatricomas, 5 trichoblastomas/trichoepitheliomas, 25 basal cell carcinomas, and 6 keratoacanthomas, among others (Table 1). Although the precise histogenesis of keratoacanthomas has been a matter of debate for a long time (40,41), keratoacanthomas were conventionally included in follicular tumor category since it is not an essential problem in this study. The 55 keratinocytic tumors included 15 seborrheic keratoses, 8 actinic keratoses, 14 Bowen diseases, and 18 squamous cell carcinomas. All 24 metastatic tumors were clinicopathologically
This article is protected by copyright. All rights reserved
identified by primary origin, which included 6 lung (4 adenocarcinomas and 2 squamous cell carcinomas), 3 breast (3 invasive carcinomas with no special type), 1 esophagus (1 squamous cell carcinoma), 4 stomach (3 signet-ring cell carcinomas and 1 poorly differentiated adenocarcinoma), 2 colorectum (2 moderately differentiated tubular adenocarcinomas), 1 common bile duct (1 poorly differentiated adenocarcinoma), 4 pancreas (4 ductal adenocarcinomas), 1 kidney (1 clear cell renal cell carcinoma), 1 prostate (1 adenocarcinoma), and 1 endocervix (1 adenocarcinoma). Metastatic tumors of unknown primary origin were excluded from this study. One mammary Paget disease, namely a breast lobular carcinoma in situ showing marked ductal spreading, was also registered in this study. The study did not include any cases of Merkel cell carcinomas, melanocytic lesions, hematological neoplasms, or mesenchymal tumors. The hematoxylin and eosin-stained slides of all cases, along with clinical features and the relevant immunostained preparations, were reviewed, and diagnoses were confirmed. Difficult cases for which diagnoses could not be clinicopathologically confirmed were removed from this study. Immunohistochemistry For each case, 4 μm-thick sections including the largest tumor area from representative blocks were routinely deparaffinized. Each section was exposed to 3% hydrogen peroxide for
This article is protected by copyright. All rights reserved
5 minutes to block endogenous peroxidase activity and then washed in deionized water for 2 or 3 minutes. The preparations were autoclaved in Bond epitope retrieval solution 2 (EDTA-based pH 9.0 solution) for antigen retrieval. The primary antibody used was a polyclonal anti-CD117 antibody (rabbit polyclonal, clone A4502, dilution 1:200; Dako, Glostrup, Denmark). The slides were incubated for 15 minutes at room temperature with the primary antibody and subsequently labeled with the Bond-III system (Leica Biosystems, Nussloch, Germany). Diaminobenzidine was used as the chromogen and hematoxylin as the counter-stain. The internal positive (eccrine ducts) and negative (epidermal keratinocytes) controls were utilized in almost all cases, but for cases without these controls, normal skin tissues were prepared for use as an appropriate control. Cell membrane staining, not only cytoplasmic staining, in at least 5% of tumor cells in each section was considered positive, and the extent of positive staining was classified as focal (5% to 50% of tumor cells in each section) or diffuse (> 50% to 100%). The intensity of staining was graded as follows: weak (recognized only in high-power view); moderate (recognized easily in low-power view but weaker than the staining in normal eccrine ducts); and strong (same as or stronger than the staining in normal eccrine ducts). The chi-square test was used to evaluate significant differences in CD117 positivity among
This article is protected by copyright. All rights reserved
the tumor groups. Evaluation of Normal Skin Tissue In addition to the skin tumor sections, some sections of non-neoplastic cutaneous and subcutaneous tissue were also obtained for each case. CD117 immunohistochemistry was concurrently or separately performed on these tissues to identify CD117-positive normal skin cells. These normal skin tissues included samples from the head, neck, axilla, body trunk, extremities, groin, and other various sites.
RESULTS The results are summarized in Table 1. Eighty-two (94.3%) of the 87 cutaneous tumors with apocrine/eccrine differentiation were positive for CD117. Most the positive cases of apocrine/eccrine tumors exhibited diffuse (> 50%; 64/82) and moderate to strong (74/82) staining (Figure 1). Poromas tended to show focal and weak positivity for CD117, and typically the cuticular tumor cells expressed CD117 more frequently and stronger than the poroid tumor cells (Figure 1C). Only 5 cases, including 1 mucinous carcinoma, 1 apocrine carcinoma, and 3 primary cutaneous extramammary Paget diseases, were negative for CD117. Of the 13 sebaceous tumors, 12 (92.3%) cases were positive for CD117, and the staining
This article is protected by copyright. All rights reserved
tended to be focal (9/12) and weak (7/12) compared to that of apocrine/eccrine tumors (Figure 2A–D). The singular CD117-negative sebaceous tumor was a case of well-differentiated sebaceous carcinoma showing marked sebacytic differentiation; this case was difficult to differentiate from sebaceous adenoma morphologically. In contrast, no CD117 expression was shown in most (82.7%) follicular tumors (Table 1), but exceptionally, all pilomatricomas (4/4) were positive for CD117 (Figure 2E). The other CD117-positive follicular tumors included 1 trichoblastoma and 4 basal cell carcinomas, although most cases (24/25) of basal cell carcinomas showed CD117 expression in the colonizing melanocytes that composed at least 1% of cells in the tumor nests (Figure 2F). The CD117-positive trichoblastoma case showed apparent sebaceous differentiation in the center of tumor nests as well as apparent differentiation toward follicular germinative epithelium and specific follicular stroma. Only 2 (3.6%) cases of the 55 keratinocytic tumors were positive for CD117, but in a few cases of Bowen disease and squamous cell carcinoma, extremely limited (< 5%) and weak staining of the tumor cells was observed. In addition, CD117 expression was observed in the colonizing melanocytes of most cases (14/15) of seborrheic keratoses, as well as basal cell carcinomas; these colonizing melanocytes composed at least 1% of cells in tumor nests. None of the 24 metastatic tumors or the case of mammary Paget disease was
This article is protected by copyright. All rights reserved
positive for CD117. These 24 cases did not have any CD117-positive tumor cells. There were highly significant differences in CD117 expression frequency between apocrine/eccrine tumors and follicular (94.3% vs. 17.3%; P < 0.001), keratinocytic (94.3% vs. 3.6%; P < 0.001), and metastatic tumors (94.3% vs. 0.0%; P < 0.001), as well as between sebaceous tumors and follicular (92.3% vs. 17.3%; P < 0.001), keratinocytic (92.3% vs. 3.6%; P < 0.001), and metastatic tumors (92.3% vs. 0.0%; P < 0.001). In the normal tissues, CD117 was expressed only in melanocytes, mast cells, some Langerhans cells, eccrine units, and apocrine ducts, but not in the apocrine glands (Figure 3A, 3B). The CD117-positive cells of eccrine units included the basal cells of the intradermal ducts, luminal cells of the intradermal ducts, dark cells of the eccrine glands, and clear cells of the eccrine glands, and the stain intensity of basal cells of the intradermal ducts was weaker than that of the other 3 cells. The superficial portion of intradermal eccrine ducts was almost negative for CD117, as were the poroid cells and cuticular cells of the intraepidermal ducts. The apocrine duct cells were positive for CD117, but the apocrine gland cells were all negative. The staining intensity of the eccrine units was relatively weaker than that of the melanocytes and mast cells. In hair follicules including hair bulbs, no apparent CD117-positive cells were seen except for many colonizing melanocytes.
This article is protected by copyright. All rights reserved
DISCUSSION In cutaneous apocrine/eccrine tumors, which could be mistaken for other cutaneous adnexal tumors, keratinocytic tumors, or metastatic tumors, CD117 was expressed in 94.3% of cases, and in most cases, the immunoreactivity was diffuse with moderate to strong intensity. However, only 5 apocrine/eccrine tumors (5.7%) were negative for CD117. The first of these 5 tumors was a mucinous carcinoma, a malignancy which is thought to be apocrine in nature (42) and is well-known to frequently lack expression of characterized primary cutaneous tumor markers, including D2-40, p63, and CK5/6 (43). The second CD117-negative case was an apocrine carcinoma, a male axillary tumor with prominent signet-ring cell or histiocytoid features. Apocrine carcinoma is a primary cutaneous tumor with apocrine gland differentiation (42), so the negative staining result might correspond to the fact that normal apocrine glands do not express CD117. The other 3 cases of CD117-negative apocrine/eccrine tumors were cases of primary extramammary Paget disease, the origin of which has been postulated to be Toker cells rather than apocrine or eccrine structures (44,45). However, because the sample size for each tumor type was low, the proper frequency distribution of CD117 positivity for each tumor type could not be evaluated in this study.
This article is protected by copyright. All rights reserved
Further studies with adequate case numbers are needed to confirm the frequency of CD117 expression in these tumor types. The diagnosis of primary cutaneous extramammary Paget disease is often challenging because
it
is
difficult
to
distinguish,
not
only
morphologically
but
also
immunohistochemically, from certain cases of secondary extramammary Paget disease, such as those of colorectal, prostate, bladder, or endocervical origin (46). The present literature shows that there are few CD117-positive cases among colorectal and prostatic adenocarcinomas (16), in contrast, 3 (50.0%) of the 6 extramammary Paget disease cases in this study expressed CD117. Although the case number is so small, CD117-positivity may be useful in discriminating primary from secondary extramammary Paget disease. CD117 immunostaining could be very useful in differentiating poromas from keratinocytic tumors such as seborrheic keratoses, since all 29 cases of poromas in this study showed CD117 expression and all 15 cases of seborrheic keratoses had no detectable CD117 expression. In contrast, porocarcinomas showed very limited positivity of CD117 in this study (Table 1). Therefore, it might be difficult to utilize CD117 immunostaining for distinguishing porocarcinoma from squamous cell carcinomas or other keratinocytic tumors, especially in the setting of small biopsy specimens.
This article is protected by copyright. All rights reserved
CD117 was also expressed in 92.3% of the 13 sebaceous tumors, but the reactivity was relatively focal and with weak intensity. The CD117-positive frequency of sebaceous tumors might decrease in the setting of small biopsy specimens; however, this study included 2 small incisional samples, both of which were evaluated as CD117-positive. Focal and weak positivity of CD117 in skin tumors could suggestive of sebaceous tumors, though it might be not easy to differentiate sebaceous tumors from keratinocytic tumors, metastatic tumors, or apocrine/eccrine tumors based only on CD117 immunostaining. In this study, follicular tumors were basically negative for CD117, but all (4/4) of the pilomatricomas expressed CD117 focally (3/4) and weakly (3/4). These 4 pilomatricomas included many viable matrical tumor cells; CD117 was expressed only in the viable matrical tumor cells and not in the shadow or squamoid cells. Pilomatricomas were positive for CD117; however, histomorphological diagnosis of pilomatricoma is generally not difficult. The other exceptional CD117-positive samples among follicular tumors were 1 trichoblastoma and 4 basal cell carcinomas. The trichoblastoma with CD117 expression clearly showed sebaceous differentiation in the center of tumor nests by morphological examination. Based on the CD117 immunoreactivity of the trichoblastoma, follicular tumors with sebaceous differentiation might exhibit CD117 expression. In comparison, the 4 basal
This article is protected by copyright. All rights reserved
cell carcinomas with CD117 expression did not show apparent sebaceous differentiation. Although the existing literature shows that CD117 expression is frequently noted in basal cell carcinomas (92.4% of 66 cases) (39), the present study revealed that basal cell carcinomas are infrequently positive for CD117 expression (16.0% of 25 cases). However, there was a variable number (> 1% of tumor composing cells) of CD117-positive melanocytes colonizing tumor nests (96.0% of 25 cases) in this study as well. The CD117 positivity of the colonizing melanocytes was usually stronger than that of eccrine units and other cutaneous adnexal tumors, and the melanocytes had multiple long dendritic structures that were useful for identifying these cells (Figure 3C–F). It is important to recognize that the CD117-positive colonizing melanocytes may be easily mistaken for CD117-positive tumor cells of basal cell carcinoma or other tumors. No CD117 expression was seen in almost all (53/55) of the keratinocytic tumors, which included seborrheic keratoses, actinic keratoses, Bowen diseases, and squamous cell carcinomas. However, in this study, seborrheic keratoses were often (93.3% of 15 cases) mixed with some colonizing melanocytes, as well as basal cell carcinomas were often (96.0% of 25 cases), so CD117 immunostaining needs to be carefully evaluated in these cases. Seborrheic keratosis can be morphologically similar to poroma or hidradenoma simplex,
This article is protected by copyright. All rights reserved
actinic keratosis and Bowen disease can be similar to extramammary Paget disease or superficial sebaceous tumors, and squamous cell carcinoma can be similar to sebaceous carcinoma or porocarcinoma. In the differential diagnosis of these tumors, CD117 immunoreactivity may be effective in ruling out the possibility of keratinocytic tumors. All 24 metastatic tumors, some of which could mimic primary cutaneous tumors, were completely negative for CD117 in this study, so CD117-positive findings could exclude the possibility of metastases. However, it should be noted that only a few carcinomas, including some small cell carcinomas, salivary gland carcinomas, renal carcinomas, lung carcinomas, breast carcinomas, and ovarian carcinomas, have been reported to express CD117 in varying degrees (16,22). In
conclusion,
it
was
demonstrated
that
CD117
expression,
assessed
by
immunohistochemistry, is specific for cutaneous adnexal tumors with apocrine/eccrine and sebaceous differentiation, as well as pilomatricomas. CD117 could be a useful marker for the differential diagnosis of cutaneous adnexal tumors, particularly for distinguishing apocrine/eccrine or sebaceous tumors from follicular, keratinocytic, or metastatic tumors. However, this study could not evaluate the proper frequency distribution of CD117 positivity for each tumor type because of the limited sample size. Future studies with large sample sizes
This article is protected by copyright. All rights reserved
are necessary to validate these findings.
REFERENCES 1. Lennartsson J, Ronnstrand L. Stem cell factor receptor/c-Kit: from basic science to clinical implications. Physiol Rev 2012; 92: 1619. 2. Jordan JH, Walchshofer S, Jurecka W, et al. Immunohistochemical properties of bone marrow mast cells in systemic mastocytosis: evidence for expression of CD2, CD117/Kit, and bcl-x(L). Hum Pathol 2001; 32: 545. 3. Kitamura Y, Hirota S. Kit as a human oncogenic tyrosine kinase. Cell Mol Life Sci 2004; 61: 2924. 4. Beadling C, Jacobson-Dunlop E, Hodi FS, et al. KIT gene mutations and copy number in melanoma subtypes. Clin Cancer Res 2008; 14: 6821. 5. Vita M, Tisserand JC, Chauvot de Beauchêne I, et al. Characterization of S628N: A Novel KIT Mutation Found in a Metastatic Melanoma. JAMA Dermatol 2014; DOI: 10.1001/jamadermatol.2014.1437. 6. Arber DA, Tamayo R, Weiss LM. Paraffin section detection of the c-kit gene product (CD117) in human tissues: value in the diagnosis of mast cell disorders. Hum Pathol
This article is protected by copyright. All rights reserved
1998; 29: 498. 7. Graadt van Roggen JF, van Velthuysen ML, Hogendoorn PC. The histopathological differential of diagnosis of gastrointestinal stromal tumours. J Clin Pathol 2001; 54: 96. 8. Wells SJ, Bray RA, Stempora LL, Farhi DC. CD117/CD34 expression in leukemic blasts. Am J Clin Pathol 1996; 106: 192. 9. Tian Q, Frierson HF Jr, Krystal GW, Moskaluk CA. Activating c-kit gene mutations in human germ cell tumors. Am J Pathol 1999; 154: 1643. 10. Matsuda R, Takahashi T, Nakamura S, et al. Expression of the c-kit protein in human solid tumors and in corresponding fetal and adult normal tissues. Am J Pathol 1993; 142: 339. 11. Natali PG, Nicotra MR, Winkler AB, Cavaliere R, Bigotti A, Ullrich A. Progression of human cutaneous melanoma is associated with loss of expression of c-kit proto-oncogene receptor. Int J Cancer 1992; 52: 197. 12. Takahashi H, Saitoh K, Kishi H, Parsons PG. Immunohistochemical localization of stem cell factor (SCF) with comparison of its receptor c-Kit proto-oncogene product (c-KIT) in melanocytic tumours. Virchows Arch 1995; 427: 283.
This article is protected by copyright. All rights reserved
13. Ohashi A, Funasaka Y, Ueda M, Ichihashi M. c-KIT receptor expression in cutaneous malignant melanoma and benign melanotic naevi. Melanoma Res 1996; 6: 25. 14. Montone KT, van Belle P, Elenitsas R, Elder DE. Proto-oncogene c-kit expression in malignant melanoma: protein loss with tumor progression. Mod Pathol 1997; 10: 939. 15. Isabel ZY, Fitzpatrick JE. Expression of c-kit (CD117) in Spitz nevus and malignant melanoma. J Cutan Pathol 2006; 33: 33. 16. Miettinen M, Lasota J. KIT (CD117): A review on expression in normal and neoplastic tissues, and mutations and their clinicopathologic correlation. Appl Immunohistochem Mol Morphol 2005; 13: 205. 17. Miliaras D, Karasavvidou F, Papanikolaou A, Sioutopoulou D. KIT expression in fetal, normal adult, and neoplastic renal tissues. J Clin Pathol 2004; 57: 463. 18. Scotlandi K, Manara MC, Strammiello R, et al. C-kit receptor expression in Ewing’s sarcoma: lack of prognostic value but therapeutic targeting opportunities in appropriate conditions. J Clin Oncol 2003; 21: 1952. 19. González I, Andreu EJ, Panizo A, et al. Imatinib inhibits proliferation of Ewing tumors cells mediated by the stem cell factor/KIT receptor pathway, and sensitizes
This article is protected by copyright. All rights reserved
cells to vincristine and doxorubicin-induced apoptosis. Clin Cancer Res 2004; 10: 751. 20. Ahmed A, Gilbert-Barness E, Lacson A. Expression of c-kit in Ewing family of tumors: a comparison of different immunohistochemical protocols. Pediatr Dev Pathol 2004; 7: 342. 21. Kandemir NO, Gun BD, Bahadir B, et al. c-Kit (CD117) expression in classic Kaposi’s sarcoma. Clin Exp Dermatol 2010; 35: 525. 22. Espinosa I, Lee CH, Kim MK, et al. A novel monoclonal antibody against DOG1 is a sensitive and specific marker for gastrointestinal stromal tumors. Am J Surg Pathol 2008; 32: 210. 23. Holst VA, Marshall CE, Moskaluk CA, Frierson HF Jr. KIT protein expression and analysis of c-kit gene mutation in adenoid cystic carcinoma. Mod Pathol 1999; 12: 956. 24. Penner CR, Folpe AL, Budnick SD. C-kit expression distinguishes salivary gland adenoid cystic carcinoma from polymorphous low-grade adenocarcinoma. Mod Pathol 2002; 15: 687. 25. Su LD, Fullen DR, Lowe L, Uherova P, Schnitzer B, Valdez R. CD117 (KIT
This article is protected by copyright. All rights reserved
receptor) expression in Merkel cell carcinoma. Am J Dermatopathol 2002; 24: 289. 26. Feinmesser M, Halpern M, Kaganovsky E, et al. c-kit expression in primary and metastatic merkel cell carcinoma. Am J Dermatopathol 2004; 26: 458. 27. Yang DT, Holden JA, Florell SR. CD117, CK20, TTF-1, and DNA topoisomerase II-alpha antigen expression in small cell tumors. J Cutan Pathol 2004; 31: 254. 28. Swick BL, Srikantha R, Messingham KN. Specific analysis of KIT and PDGFR-alpha expression and mutational status in Merkel cell carcinoma. J Cutan Pathol 2013; 40: 623. 29. Pan CX, Yang XJ, Lopex-Beltran A, et al. c-kit expression in small cell carcinoma of the urinary bladder: prognostic and therapeutic implications. Mod Pathol 2005; 18: 320. 30. Pan CC, Chen PCH, Chiang H. KIT (CD117) is frequently overexpressed in thymic carcinomas but is absent in thymomas. J Pathol 2004; 202: 375. 31. Wang HY, Mills SE. KIT and RCC are useful in distinguishing chromophobe renal cell carcinoma from the granular variant of clear cell renal cell carcinoma. Am J Surg Pathol 2005; 29: 640. 32. Kim HJ, Kim CH, Choi YJ, Ayala AG, Amirikachi M, Ro JY. Juxtaglomerular cell
This article is protected by copyright. All rights reserved
tumor of kidney with CD34 and CD117 immunoreactivity: report of 5 cases. Arch Pathol Lab Med 2006; 130: 707. 33. Hibi K, Takahashi T, Sekido Y, et al. Coexpression of the stem cell factor and the c-kit genes in small-cell lung cancer. Oncogene 1991; 6: 2291. 34. Krystal GW, Hines SJ, Organ CP. Autocrine growth of small cell lung cancer mediated by coexpression of c-kit and stem cell factor. Cancer Res 1996; 56: 370. 35. Hines SJ, Organ C, Kornstein MJ, Krystal GW. Coexpression of the c-kit and stem cell factor genes in breast carcinomas. Cell Growth Differ 1995; 6: 769. 36. Inoue M, Kyo S, Fujita M, et al. Coexpression of c_ABL, c-KIT, and platelet-derived growth factor receptor-beta in ovarian serous carcinoma and normal ovarian surface epithelium. Cancer 2003; 98: 758. 37. Ramakrishnan R, Chaudhry IH, Ramadial P, et al. Primary cutaneous adenoid cystic carcinoma: a clinicopathologic and immunohistochemical study of 27 cases. Am J Surg Pathol 2013; 37: 1603. 38. Volmat KE, Cummings TJ, Wang WH, et al. Clear cell hidradenoma. Arch Pathol Lab Med 2005; 129: e113. 39. Terada T. Expression of NCAM (CD56), chromogranin A, synaptophysin, c-KIT
This article is protected by copyright. All rights reserved
(CD117) and PDGFRA in normal non-neoplastic skin and basal cell carcinoma: an immunohistochemical study of 66 consecutive cases. Med Oncol 2013; 30: 444. 40. Ito Y, Kurokawa I, Nishimura K, et al. Keratin and filaggrin wxpression in keratoacanthoma. J Eur Acad Dermatol Venereol 2008; 22: 353. 41. Misago N, Takai T, Toda S, et al. The changes in the expression levels of follicular markers in keratoacanthoma depend on the stage: keratoacanthoma is a follicular neoplasm exhibiting infundibular/isthmic differentiation without expression of CK15. J Cutan Pathol 2014; 41: 437. 42. Requena L, Kutzner H, Hurt MA, et al. Malignant tumours with apocrine and eccrine differentiation. In: LeBoit PE, Burg G, Weedon D, et al, eds. Pathology and Genetics Skin Tumours. Lyon: IARC Press, 2006; 125. 43. Levy G, Finkelstein A, McNiff JM. Immunohistochemical techniques to compare primary vs. metastatic mucinous carcinoma of the skin. J Cutan Pathol 2010; 37: 411. 44. Marucci G, Betts CM, Golouh R, Peterse JL, Foschini MP, Eusebi V. Toker cells are probably precursors of Paget cell carcinoma: a morphological and ultrastructural description. Virchows Arch 2002; 441: 117. 45. Willman JH, Golitz LE, Fitzpatrick JE. Vulvar clear cells of Toker: precursors of
This article is protected by copyright. All rights reserved
extramammary Paget’s disease. Am J Dermatopathol 2005; 27: 185. 46. Perrotto J, Abbott JJ, Ceilley RI, Ahmed I. The role of immunohistochemistry in discriminating primary from secondary extramammary Paget disease. Am J Dermatopathol 2010; 32: 137.
This article is protected by copyright. All rights reserved
Table 1. Comparative CD117 immunohistochemistry of non-Merkel cell epithelial tumors of the skin Tumor type
Positivity,
Extent
Intensity
F18,
W8,
(94%)
D64
S47
Syringofibroadenomatous hyperplasia
1/1 (100%)
F1, D0
W1, M0, S0
Hidrocystoma
5/5 (100%)
F0, D5
W0, M1, S4
Hidrocystadenoma
2/2 (100%)
F1, D1
W0, M1, S1
Hidroacanthoma simplex
2/2 (100%)
F0, D2
W0, M0, S2
Poroma
29/29
F7,
W3, M8, S18
(100%)
D22
10/10
F2, D8
W0, M7, S3
n (%) Tumors
with
apocrine/eccrine 82/87
differentiation
Hidradenoma
M27,
(100%) Hidradenoma papilliferum
1/1 (100%)
F0, D1
W0, M0, S1
Spiradenoma
6/6 (100%)
F0, D6
W0, M0, S6
Cylindroma
1/1 (100%)
F0, D1
W0, M1, S0
This article is protected by copyright. All rights reserved
Syringoma
1/1 (100%)
F0, D1
W1, M0, S0
Mixed tumor of the skin
15/15
F4,
W0, M5, S10
(100%)
D11
Porocarcinoma
2/2 (100%)
F2, D0
W2, M0, S0
Ductal carcinoma
1/1 (100%)
F0, D1
W0, M1, S0
Cribriform carcinoma
1/1 (100%)
F0, D1
W0, M1, S0
Adenoid cystic carcinoma
1/1 (100%)
F0, D1
W0, M0, S1
Endocrine mucin-producing sweat gland 1/1 (100%)
F0, D1
W0, M1, S0
carcinoma Mucinous carcinoma
0/1 (0%)
Apocrine carcinoma
0/1 (0%)
Extramammary Paget disease
3/6 (50%)
F1, D2
W1, M1, S1
12/13
F9, D3
W7, M4, S1
Tumors with sebaceous differentiation
(92%) Sebaceoma
5/5 (100%)
F4, D1
W3, M1, S1
Sebaceous carcinoma
7/8 (86%)
F5, D2
W4, M3, S0
9/52 (17%)
F7, D2
W6, M3, S0
Tumors with follicular differentiation
This article is protected by copyright. All rights reserved
Trichofolliculoma Trichilemmal
0/2 (0%) cyst/Proliferating 0/5 (0%)
trichilemmal cyst Desmoplastic trichilemmoma
0/1 (0%)
Inverted follicular keratosis
0/4 (0%)
Pilomatricoma
4/4 (100%)
F3, D1
W3, M1, S0
Trichoblastoma/Trichoepithelioma
1/5 (20%)
F1, D0
W0, M1, S0
Basal cell carcinoma
4/25 (16%)
F3, D1
W3, M1, S0
Keratoacanthoma
0/6 (0%) F2, D0
W2, M0, S0
Tumors with keratinocytic differentiation
2/55 (4%)
Seborrheic keratosis
0/15 (0%)
Actinic keratosis
1/8 (13%)
F1, D0
W1, M0, S0
Bowen disease
1/14 (7%)
F1, D0
W1, M0, S0
Squamous cell carcinoma
0/18 (0%)
Metastatic carcinomas
0/24 (0%)
Mammary Paget disease
0/1 (0%)
D, diffuse (> 50%); F, focal (5–50%); M, moderate; S, strong; W, weak
This article is protected by copyright. All rights reserved
Figure legends
Figure 1. Most apocrine/eccrine tumors showed CD117-positivity to various degrees. A poroma with positive CD117 expression (A); in high-power view, the tumor cells and melanocytes in the basal cell layer of the existing epidermis were positive for CD117 (B). CD117-positivity of the cuticular tumor cells tended to be more frequent than that of the poroid tumor cells (C). Other CD117-positive apocrine/eccrine tumors included cylindroma (D), hidradenoma papilliferum (E), mixed tumor of the skin (F), cribriform carcinoma (G),
This article is protected by copyright. All rights reserved
and primary extramammary Paget disease (H). Images were taken with magnification of ×4 (A, E, F), ×10 (H), ×20 (C, D, G), or ×40 (B).
Figure 2. CD117 expression in sebaceous and follicular tumors. CD117 was typically expressed focally and weakly in sebaceomas (A. hematoxylin and eosin stain, B. CD117 immunostain) and sebaceous carcinoma (C. hematoxylin and eosin stain, D. CD117 immunostain). Pilomatricomas frequently showed CD117-positivity (E). The tumor cells in basal cell carcinomas were usually negative for CD117 (F). Images were taken using magnification of ×10 (E), ×20 (A, B), or ×40 (C, D, F).
This article is protected by copyright. All rights reserved
Figure 3. CD117 expression in the normal skin tissues and in the colonizing melanocytes of cutaneous adnexal tumors. CD117 was expressed in the eccrine units but not in the apocrine glands (A. hematoxylin and eosin stain, B. CD117 immunostain). Basal cell carcinomas contained numerous colonizing melanocytes (C. hematoxylin and eosin stain, D & E. melanosome [mouse anti-human monoclonal, clone HMB45, dilution 1:100; Dako, Glostrup, Denmark] immunostain with Giemsa as the counter-stain, F. CD117 immunostain with Giemsa as the counter-stain). Images were taken using magnification of ×10 (C, D), ×20 (A, B), or ×40 (E, F).
This article is protected by copyright. All rights reserved
Corresponding author: Keisuke Goto, MD Department of Diagnostic Pathology, Kainan Hospital, 396 Minamihonden, Maegasu-cho, Yatomi-city, Aichi, 498-8502 Japan E-mail: [email protected] Tel: +81-(0)567-65-2511; Fax: +81-(0)567-67-3697 Keywords: c-KIT protein, immunohistochemistry, skin neoplasms
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/cup.12492
This article is protected by copyright. All rights reserved
ACKNOWLEDGEMENTS The author thanks Drs. Yoshimi Oshitani (Department of Dermatology, Kainan Hospital), Toshiaki Sato (Department of Plastic Surgery, Kainan Hospital), Yukio Ishikawa (Pathology Division, Itabashi Medical Laboratory), and Makoto Takeuchi (Department of Dermatology, Tsushima Municipal Hospital) for supplying valuable materials, and thanks Mr. Shinya Hattori and Ms. Mayumi Arakawa for technical assistance.
ABSTRACT Background: The aim of this study was to evaluate the effectiveness of CD117 immunostaining for diagnosing cutaneous adnexal tumors, which has not been previously established. Methods: CD117 immunostaining was performed on representative sections of 87 clinicopathologically confirmed cutaneous apocrine/eccrine tumors, 13 sebaceous tumors, 52 follicular tumors, 55 keratinocytic tumors, 24 metastatic tumors of the skin of known origin, and 1 mammary Paget disease. Staining of 5–50% of cells was considered focally positive, and staining of 51–100% of cells was considered diffusely positive. Staining intensity was semi-quantitatively graded as weak, moderate, or strong.
This article is protected by copyright. All rights reserved
Results: Seventy-seven (95.1%) tumors with apocrine/eccrine differentiation were CD117-positive, but 1 mucinous carcinoma, 1 apocrine carcinoma, and 3 primary cutaneous extramammary Paget diseases were CD117-negative. For most positive cases, staining was diffuse (64/82 cases) and moderate to strong in intensity (74/82). Twelve (92.3%) sebaceous tumors were CD117-positive, showing mainly focal (9/12) and weak (7/12) positivity. Of the 52 follicular tumors, all 4 cases of pilomatricoma and other 5 tumors were positive for CD117. Of the 55 keratinocytic tumors, only 2 cases were CD117-positive. The 24 metastatic tumors and mammary Paget disease were CD117-negative. Conclusions: CD117 may be useful for the differential diagnosis of cutaneous adnexal tumors.
This article is protected by copyright. All rights reserved
INTRODUCTION CD117 (KIT) is a transmembrane tyrosine kinase receptor operating in cell signal transduction and is normally activated upon binding of its ligand, the stem cell factor, in several cell types. Ligand binding results in a phosphorylation cascade, resulting in the activation of various transcription factors that regulate cell differentiation, migration, adhesion, apoptosis, and proliferation (1). KIT-dependent cell types include mast cells, some hematopoietic stem cells, germ cells, melanocytes, and Cajal cells of the gastrointestinal tract (2). Neoplasms of these cells – mastocytosis, acute myeloid leukemia, seminoma, malignant melanoma, and gastrointestinal stromal tumor – are often induced by KIT activation through gain-of-function mutations (1,3–5), and most of these tumors are examples of CD117-positive malignancies (2,6–16). Other normal cells that express CD117 include some Langerhans cells, cerebellar neurons, and epithelial cells in breast, kidney, and skin adnexa (6,7,17). CD117 may also be expressed in some sarcomas (Ewing sarcoma (18–20), angiosarcoma (16), Kaposi sarcoma (21), alveolar soft part sarcoma (22), extraskeletal myxoid chondrosarcoma (22), etc.) and epithelial neoplasms (adenoid cystic carcinoma (23); polymorphous low-grade adenocarcinoma (24); Merkel cell carcinoma (25–28); some small cell carcinomas (27,29); thymic carcinoma (30); chromophobe renal cell carcinoma (31);
This article is protected by copyright. All rights reserved
renal oncocytoma (31); juxtaglomerular cell tumor (32); and a few lung, ovarian, and breast carcinomas (16,22)). Among these mesenchymal and epithelial neoplasms, few of these tumors have any evidence of KIT gain-of-function mutations. Instead, some of these tumors have been reported to have concomitant overexpression of both KIT and stem cell factor, resulting in an autocrine loop (33–36). Previous reports have demonstrated that certain adnexal tumors, including primary cutaneous adenoid cystic carcinoma (37), clear cell hidradenoma (38), and basal cell carcinoma (39), are also positive for CD117. Although CD117 immunohistochemistry is often utilized as a conclusive diagnostic method, especially for mastocytosis, seminoma, gastrointestinal stromal tumor, adenoid cystic carcinoma, and thymic carcinoma (16), the utility of CD117 immunostaining in cutaneous adnexal tumors has not yet been established. The aim of the present study was to evaluate the effectiveness of CD117 immunostaining for diagnosing cutaneous adnexal tumors.
MATERIALS AND METHODS Cases A total of 87 cutaneous adnexal tumors with apocrine/eccrine differentiation, 13 sebaceous
This article is protected by copyright. All rights reserved
tumors, 52 follicular tumors, 55 keratinocytic tumors, 24 metastatic carcinomas of the skin, and 1 mammary Paget disease were retrospectively retrieved from the pathology archive of Kainan Hospital and the author’s consultation file from 2000 to 2014 (only pilomatricomas, basal cell carcinomas, and keratinocytic tumors from 2013 to 2014); all guidelines for experimental investigation with human subjects required by the institution were followed. The 87 apocrine/eccrine tumors included 29 poromas, 10 hidradenomas, 15 mixed tumors of the skin, 6 primary cutaneous extramammary Paget diseases, and 6 other various carcinomas with apocrine/eccrine differentiation, among others (Table 1). For combined apocrine/eccrine adnexal tumors with 2 or more tumor components, the predominant component was named for its representative diagnosis. The 13 sebaceous tumors included 5 sebaceomas and 8 sebaceous carcinomas. The 52 follicular tumors included 4 pilomatricomas, 5 trichoblastomas/trichoepitheliomas, 25 basal cell carcinomas, and 6 keratoacanthomas, among others (Table 1). Although the precise histogenesis of keratoacanthomas has been a matter of debate for a long time (40,41), keratoacanthomas were conventionally included in follicular tumor category since it is not an essential problem in this study. The 55 keratinocytic tumors included 15 seborrheic keratoses, 8 actinic keratoses, 14 Bowen diseases, and 18 squamous cell carcinomas. All 24 metastatic tumors were clinicopathologically
This article is protected by copyright. All rights reserved
identified by primary origin, which included 6 lung (4 adenocarcinomas and 2 squamous cell carcinomas), 3 breast (3 invasive carcinomas with no special type), 1 esophagus (1 squamous cell carcinoma), 4 stomach (3 signet-ring cell carcinomas and 1 poorly differentiated adenocarcinoma), 2 colorectum (2 moderately differentiated tubular adenocarcinomas), 1 common bile duct (1 poorly differentiated adenocarcinoma), 4 pancreas (4 ductal adenocarcinomas), 1 kidney (1 clear cell renal cell carcinoma), 1 prostate (1 adenocarcinoma), and 1 endocervix (1 adenocarcinoma). Metastatic tumors of unknown primary origin were excluded from this study. One mammary Paget disease, namely a breast lobular carcinoma in situ showing marked ductal spreading, was also registered in this study. The study did not include any cases of Merkel cell carcinomas, melanocytic lesions, hematological neoplasms, or mesenchymal tumors. The hematoxylin and eosin-stained slides of all cases, along with clinical features and the relevant immunostained preparations, were reviewed, and diagnoses were confirmed. Difficult cases for which diagnoses could not be clinicopathologically confirmed were removed from this study. Immunohistochemistry For each case, 4 μm-thick sections including the largest tumor area from representative blocks were routinely deparaffinized. Each section was exposed to 3% hydrogen peroxide for
This article is protected by copyright. All rights reserved
5 minutes to block endogenous peroxidase activity and then washed in deionized water for 2 or 3 minutes. The preparations were autoclaved in Bond epitope retrieval solution 2 (EDTA-based pH 9.0 solution) for antigen retrieval. The primary antibody used was a polyclonal anti-CD117 antibody (rabbit polyclonal, clone A4502, dilution 1:200; Dako, Glostrup, Denmark). The slides were incubated for 15 minutes at room temperature with the primary antibody and subsequently labeled with the Bond-III system (Leica Biosystems, Nussloch, Germany). Diaminobenzidine was used as the chromogen and hematoxylin as the counter-stain. The internal positive (eccrine ducts) and negative (epidermal keratinocytes) controls were utilized in almost all cases, but for cases without these controls, normal skin tissues were prepared for use as an appropriate control. Cell membrane staining, not only cytoplasmic staining, in at least 5% of tumor cells in each section was considered positive, and the extent of positive staining was classified as focal (5% to 50% of tumor cells in each section) or diffuse (> 50% to 100%). The intensity of staining was graded as follows: weak (recognized only in high-power view); moderate (recognized easily in low-power view but weaker than the staining in normal eccrine ducts); and strong (same as or stronger than the staining in normal eccrine ducts). The chi-square test was used to evaluate significant differences in CD117 positivity among
This article is protected by copyright. All rights reserved
the tumor groups. Evaluation of Normal Skin Tissue In addition to the skin tumor sections, some sections of non-neoplastic cutaneous and subcutaneous tissue were also obtained for each case. CD117 immunohistochemistry was concurrently or separately performed on these tissues to identify CD117-positive normal skin cells. These normal skin tissues included samples from the head, neck, axilla, body trunk, extremities, groin, and other various sites.
RESULTS The results are summarized in Table 1. Eighty-two (94.3%) of the 87 cutaneous tumors with apocrine/eccrine differentiation were positive for CD117. Most the positive cases of apocrine/eccrine tumors exhibited diffuse (> 50%; 64/82) and moderate to strong (74/82) staining (Figure 1). Poromas tended to show focal and weak positivity for CD117, and typically the cuticular tumor cells expressed CD117 more frequently and stronger than the poroid tumor cells (Figure 1C). Only 5 cases, including 1 mucinous carcinoma, 1 apocrine carcinoma, and 3 primary cutaneous extramammary Paget diseases, were negative for CD117. Of the 13 sebaceous tumors, 12 (92.3%) cases were positive for CD117, and the staining
This article is protected by copyright. All rights reserved
tended to be focal (9/12) and weak (7/12) compared to that of apocrine/eccrine tumors (Figure 2A–D). The singular CD117-negative sebaceous tumor was a case of well-differentiated sebaceous carcinoma showing marked sebacytic differentiation; this case was difficult to differentiate from sebaceous adenoma morphologically. In contrast, no CD117 expression was shown in most (82.7%) follicular tumors (Table 1), but exceptionally, all pilomatricomas (4/4) were positive for CD117 (Figure 2E). The other CD117-positive follicular tumors included 1 trichoblastoma and 4 basal cell carcinomas, although most cases (24/25) of basal cell carcinomas showed CD117 expression in the colonizing melanocytes that composed at least 1% of cells in the tumor nests (Figure 2F). The CD117-positive trichoblastoma case showed apparent sebaceous differentiation in the center of tumor nests as well as apparent differentiation toward follicular germinative epithelium and specific follicular stroma. Only 2 (3.6%) cases of the 55 keratinocytic tumors were positive for CD117, but in a few cases of Bowen disease and squamous cell carcinoma, extremely limited (< 5%) and weak staining of the tumor cells was observed. In addition, CD117 expression was observed in the colonizing melanocytes of most cases (14/15) of seborrheic keratoses, as well as basal cell carcinomas; these colonizing melanocytes composed at least 1% of cells in tumor nests. None of the 24 metastatic tumors or the case of mammary Paget disease was
This article is protected by copyright. All rights reserved
positive for CD117. These 24 cases did not have any CD117-positive tumor cells. There were highly significant differences in CD117 expression frequency between apocrine/eccrine tumors and follicular (94.3% vs. 17.3%; P < 0.001), keratinocytic (94.3% vs. 3.6%; P < 0.001), and metastatic tumors (94.3% vs. 0.0%; P < 0.001), as well as between sebaceous tumors and follicular (92.3% vs. 17.3%; P < 0.001), keratinocytic (92.3% vs. 3.6%; P < 0.001), and metastatic tumors (92.3% vs. 0.0%; P < 0.001). In the normal tissues, CD117 was expressed only in melanocytes, mast cells, some Langerhans cells, eccrine units, and apocrine ducts, but not in the apocrine glands (Figure 3A, 3B). The CD117-positive cells of eccrine units included the basal cells of the intradermal ducts, luminal cells of the intradermal ducts, dark cells of the eccrine glands, and clear cells of the eccrine glands, and the stain intensity of basal cells of the intradermal ducts was weaker than that of the other 3 cells. The superficial portion of intradermal eccrine ducts was almost negative for CD117, as were the poroid cells and cuticular cells of the intraepidermal ducts. The apocrine duct cells were positive for CD117, but the apocrine gland cells were all negative. The staining intensity of the eccrine units was relatively weaker than that of the melanocytes and mast cells. In hair follicules including hair bulbs, no apparent CD117-positive cells were seen except for many colonizing melanocytes.
This article is protected by copyright. All rights reserved
DISCUSSION In cutaneous apocrine/eccrine tumors, which could be mistaken for other cutaneous adnexal tumors, keratinocytic tumors, or metastatic tumors, CD117 was expressed in 94.3% of cases, and in most cases, the immunoreactivity was diffuse with moderate to strong intensity. However, only 5 apocrine/eccrine tumors (5.7%) were negative for CD117. The first of these 5 tumors was a mucinous carcinoma, a malignancy which is thought to be apocrine in nature (42) and is well-known to frequently lack expression of characterized primary cutaneous tumor markers, including D2-40, p63, and CK5/6 (43). The second CD117-negative case was an apocrine carcinoma, a male axillary tumor with prominent signet-ring cell or histiocytoid features. Apocrine carcinoma is a primary cutaneous tumor with apocrine gland differentiation (42), so the negative staining result might correspond to the fact that normal apocrine glands do not express CD117. The other 3 cases of CD117-negative apocrine/eccrine tumors were cases of primary extramammary Paget disease, the origin of which has been postulated to be Toker cells rather than apocrine or eccrine structures (44,45). However, because the sample size for each tumor type was low, the proper frequency distribution of CD117 positivity for each tumor type could not be evaluated in this study.
This article is protected by copyright. All rights reserved
Further studies with adequate case numbers are needed to confirm the frequency of CD117 expression in these tumor types. The diagnosis of primary cutaneous extramammary Paget disease is often challenging because
it
is
difficult
to
distinguish,
not
only
morphologically
but
also
immunohistochemically, from certain cases of secondary extramammary Paget disease, such as those of colorectal, prostate, bladder, or endocervical origin (46). The present literature shows that there are few CD117-positive cases among colorectal and prostatic adenocarcinomas (16), in contrast, 3 (50.0%) of the 6 extramammary Paget disease cases in this study expressed CD117. Although the case number is so small, CD117-positivity may be useful in discriminating primary from secondary extramammary Paget disease. CD117 immunostaining could be very useful in differentiating poromas from keratinocytic tumors such as seborrheic keratoses, since all 29 cases of poromas in this study showed CD117 expression and all 15 cases of seborrheic keratoses had no detectable CD117 expression. In contrast, porocarcinomas showed very limited positivity of CD117 in this study (Table 1). Therefore, it might be difficult to utilize CD117 immunostaining for distinguishing porocarcinoma from squamous cell carcinomas or other keratinocytic tumors, especially in the setting of small biopsy specimens.
This article is protected by copyright. All rights reserved
CD117 was also expressed in 92.3% of the 13 sebaceous tumors, but the reactivity was relatively focal and with weak intensity. The CD117-positive frequency of sebaceous tumors might decrease in the setting of small biopsy specimens; however, this study included 2 small incisional samples, both of which were evaluated as CD117-positive. Focal and weak positivity of CD117 in skin tumors could suggestive of sebaceous tumors, though it might be not easy to differentiate sebaceous tumors from keratinocytic tumors, metastatic tumors, or apocrine/eccrine tumors based only on CD117 immunostaining. In this study, follicular tumors were basically negative for CD117, but all (4/4) of the pilomatricomas expressed CD117 focally (3/4) and weakly (3/4). These 4 pilomatricomas included many viable matrical tumor cells; CD117 was expressed only in the viable matrical tumor cells and not in the shadow or squamoid cells. Pilomatricomas were positive for CD117; however, histomorphological diagnosis of pilomatricoma is generally not difficult. The other exceptional CD117-positive samples among follicular tumors were 1 trichoblastoma and 4 basal cell carcinomas. The trichoblastoma with CD117 expression clearly showed sebaceous differentiation in the center of tumor nests by morphological examination. Based on the CD117 immunoreactivity of the trichoblastoma, follicular tumors with sebaceous differentiation might exhibit CD117 expression. In comparison, the 4 basal
This article is protected by copyright. All rights reserved
cell carcinomas with CD117 expression did not show apparent sebaceous differentiation. Although the existing literature shows that CD117 expression is frequently noted in basal cell carcinomas (92.4% of 66 cases) (39), the present study revealed that basal cell carcinomas are infrequently positive for CD117 expression (16.0% of 25 cases). However, there was a variable number (> 1% of tumor composing cells) of CD117-positive melanocytes colonizing tumor nests (96.0% of 25 cases) in this study as well. The CD117 positivity of the colonizing melanocytes was usually stronger than that of eccrine units and other cutaneous adnexal tumors, and the melanocytes had multiple long dendritic structures that were useful for identifying these cells (Figure 3C–F). It is important to recognize that the CD117-positive colonizing melanocytes may be easily mistaken for CD117-positive tumor cells of basal cell carcinoma or other tumors. No CD117 expression was seen in almost all (53/55) of the keratinocytic tumors, which included seborrheic keratoses, actinic keratoses, Bowen diseases, and squamous cell carcinomas. However, in this study, seborrheic keratoses were often (93.3% of 15 cases) mixed with some colonizing melanocytes, as well as basal cell carcinomas were often (96.0% of 25 cases), so CD117 immunostaining needs to be carefully evaluated in these cases. Seborrheic keratosis can be morphologically similar to poroma or hidradenoma simplex,
This article is protected by copyright. All rights reserved
actinic keratosis and Bowen disease can be similar to extramammary Paget disease or superficial sebaceous tumors, and squamous cell carcinoma can be similar to sebaceous carcinoma or porocarcinoma. In the differential diagnosis of these tumors, CD117 immunoreactivity may be effective in ruling out the possibility of keratinocytic tumors. All 24 metastatic tumors, some of which could mimic primary cutaneous tumors, were completely negative for CD117 in this study, so CD117-positive findings could exclude the possibility of metastases. However, it should be noted that only a few carcinomas, including some small cell carcinomas, salivary gland carcinomas, renal carcinomas, lung carcinomas, breast carcinomas, and ovarian carcinomas, have been reported to express CD117 in varying degrees (16,22). In
conclusion,
it
was
demonstrated
that
CD117
expression,
assessed
by
immunohistochemistry, is specific for cutaneous adnexal tumors with apocrine/eccrine and sebaceous differentiation, as well as pilomatricomas. CD117 could be a useful marker for the differential diagnosis of cutaneous adnexal tumors, particularly for distinguishing apocrine/eccrine or sebaceous tumors from follicular, keratinocytic, or metastatic tumors. However, this study could not evaluate the proper frequency distribution of CD117 positivity for each tumor type because of the limited sample size. Future studies with large sample sizes
This article is protected by copyright. All rights reserved
are necessary to validate these findings.
REFERENCES 1. Lennartsson J, Ronnstrand L. Stem cell factor receptor/c-Kit: from basic science to clinical implications. Physiol Rev 2012; 92: 1619. 2. Jordan JH, Walchshofer S, Jurecka W, et al. Immunohistochemical properties of bone marrow mast cells in systemic mastocytosis: evidence for expression of CD2, CD117/Kit, and bcl-x(L). Hum Pathol 2001; 32: 545. 3. Kitamura Y, Hirota S. Kit as a human oncogenic tyrosine kinase. Cell Mol Life Sci 2004; 61: 2924. 4. Beadling C, Jacobson-Dunlop E, Hodi FS, et al. KIT gene mutations and copy number in melanoma subtypes. Clin Cancer Res 2008; 14: 6821. 5. Vita M, Tisserand JC, Chauvot de Beauchêne I, et al. Characterization of S628N: A Novel KIT Mutation Found in a Metastatic Melanoma. JAMA Dermatol 2014; DOI: 10.1001/jamadermatol.2014.1437. 6. Arber DA, Tamayo R, Weiss LM. Paraffin section detection of the c-kit gene product (CD117) in human tissues: value in the diagnosis of mast cell disorders. Hum Pathol
This article is protected by copyright. All rights reserved
1998; 29: 498. 7. Graadt van Roggen JF, van Velthuysen ML, Hogendoorn PC. The histopathological differential of diagnosis of gastrointestinal stromal tumours. J Clin Pathol 2001; 54: 96. 8. Wells SJ, Bray RA, Stempora LL, Farhi DC. CD117/CD34 expression in leukemic blasts. Am J Clin Pathol 1996; 106: 192. 9. Tian Q, Frierson HF Jr, Krystal GW, Moskaluk CA. Activating c-kit gene mutations in human germ cell tumors. Am J Pathol 1999; 154: 1643. 10. Matsuda R, Takahashi T, Nakamura S, et al. Expression of the c-kit protein in human solid tumors and in corresponding fetal and adult normal tissues. Am J Pathol 1993; 142: 339. 11. Natali PG, Nicotra MR, Winkler AB, Cavaliere R, Bigotti A, Ullrich A. Progression of human cutaneous melanoma is associated with loss of expression of c-kit proto-oncogene receptor. Int J Cancer 1992; 52: 197. 12. Takahashi H, Saitoh K, Kishi H, Parsons PG. Immunohistochemical localization of stem cell factor (SCF) with comparison of its receptor c-Kit proto-oncogene product (c-KIT) in melanocytic tumours. Virchows Arch 1995; 427: 283.
This article is protected by copyright. All rights reserved
13. Ohashi A, Funasaka Y, Ueda M, Ichihashi M. c-KIT receptor expression in cutaneous malignant melanoma and benign melanotic naevi. Melanoma Res 1996; 6: 25. 14. Montone KT, van Belle P, Elenitsas R, Elder DE. Proto-oncogene c-kit expression in malignant melanoma: protein loss with tumor progression. Mod Pathol 1997; 10: 939. 15. Isabel ZY, Fitzpatrick JE. Expression of c-kit (CD117) in Spitz nevus and malignant melanoma. J Cutan Pathol 2006; 33: 33. 16. Miettinen M, Lasota J. KIT (CD117): A review on expression in normal and neoplastic tissues, and mutations and their clinicopathologic correlation. Appl Immunohistochem Mol Morphol 2005; 13: 205. 17. Miliaras D, Karasavvidou F, Papanikolaou A, Sioutopoulou D. KIT expression in fetal, normal adult, and neoplastic renal tissues. J Clin Pathol 2004; 57: 463. 18. Scotlandi K, Manara MC, Strammiello R, et al. C-kit receptor expression in Ewing’s sarcoma: lack of prognostic value but therapeutic targeting opportunities in appropriate conditions. J Clin Oncol 2003; 21: 1952. 19. González I, Andreu EJ, Panizo A, et al. Imatinib inhibits proliferation of Ewing tumors cells mediated by the stem cell factor/KIT receptor pathway, and sensitizes
This article is protected by copyright. All rights reserved
cells to vincristine and doxorubicin-induced apoptosis. Clin Cancer Res 2004; 10: 751. 20. Ahmed A, Gilbert-Barness E, Lacson A. Expression of c-kit in Ewing family of tumors: a comparison of different immunohistochemical protocols. Pediatr Dev Pathol 2004; 7: 342. 21. Kandemir NO, Gun BD, Bahadir B, et al. c-Kit (CD117) expression in classic Kaposi’s sarcoma. Clin Exp Dermatol 2010; 35: 525. 22. Espinosa I, Lee CH, Kim MK, et al. A novel monoclonal antibody against DOG1 is a sensitive and specific marker for gastrointestinal stromal tumors. Am J Surg Pathol 2008; 32: 210. 23. Holst VA, Marshall CE, Moskaluk CA, Frierson HF Jr. KIT protein expression and analysis of c-kit gene mutation in adenoid cystic carcinoma. Mod Pathol 1999; 12: 956. 24. Penner CR, Folpe AL, Budnick SD. C-kit expression distinguishes salivary gland adenoid cystic carcinoma from polymorphous low-grade adenocarcinoma. Mod Pathol 2002; 15: 687. 25. Su LD, Fullen DR, Lowe L, Uherova P, Schnitzer B, Valdez R. CD117 (KIT
This article is protected by copyright. All rights reserved
receptor) expression in Merkel cell carcinoma. Am J Dermatopathol 2002; 24: 289. 26. Feinmesser M, Halpern M, Kaganovsky E, et al. c-kit expression in primary and metastatic merkel cell carcinoma. Am J Dermatopathol 2004; 26: 458. 27. Yang DT, Holden JA, Florell SR. CD117, CK20, TTF-1, and DNA topoisomerase II-alpha antigen expression in small cell tumors. J Cutan Pathol 2004; 31: 254. 28. Swick BL, Srikantha R, Messingham KN. Specific analysis of KIT and PDGFR-alpha expression and mutational status in Merkel cell carcinoma. J Cutan Pathol 2013; 40: 623. 29. Pan CX, Yang XJ, Lopex-Beltran A, et al. c-kit expression in small cell carcinoma of the urinary bladder: prognostic and therapeutic implications. Mod Pathol 2005; 18: 320. 30. Pan CC, Chen PCH, Chiang H. KIT (CD117) is frequently overexpressed in thymic carcinomas but is absent in thymomas. J Pathol 2004; 202: 375. 31. Wang HY, Mills SE. KIT and RCC are useful in distinguishing chromophobe renal cell carcinoma from the granular variant of clear cell renal cell carcinoma. Am J Surg Pathol 2005; 29: 640. 32. Kim HJ, Kim CH, Choi YJ, Ayala AG, Amirikachi M, Ro JY. Juxtaglomerular cell
This article is protected by copyright. All rights reserved
tumor of kidney with CD34 and CD117 immunoreactivity: report of 5 cases. Arch Pathol Lab Med 2006; 130: 707. 33. Hibi K, Takahashi T, Sekido Y, et al. Coexpression of the stem cell factor and the c-kit genes in small-cell lung cancer. Oncogene 1991; 6: 2291. 34. Krystal GW, Hines SJ, Organ CP. Autocrine growth of small cell lung cancer mediated by coexpression of c-kit and stem cell factor. Cancer Res 1996; 56: 370. 35. Hines SJ, Organ C, Kornstein MJ, Krystal GW. Coexpression of the c-kit and stem cell factor genes in breast carcinomas. Cell Growth Differ 1995; 6: 769. 36. Inoue M, Kyo S, Fujita M, et al. Coexpression of c_ABL, c-KIT, and platelet-derived growth factor receptor-beta in ovarian serous carcinoma and normal ovarian surface epithelium. Cancer 2003; 98: 758. 37. Ramakrishnan R, Chaudhry IH, Ramadial P, et al. Primary cutaneous adenoid cystic carcinoma: a clinicopathologic and immunohistochemical study of 27 cases. Am J Surg Pathol 2013; 37: 1603. 38. Volmat KE, Cummings TJ, Wang WH, et al. Clear cell hidradenoma. Arch Pathol Lab Med 2005; 129: e113. 39. Terada T. Expression of NCAM (CD56), chromogranin A, synaptophysin, c-KIT
This article is protected by copyright. All rights reserved
(CD117) and PDGFRA in normal non-neoplastic skin and basal cell carcinoma: an immunohistochemical study of 66 consecutive cases. Med Oncol 2013; 30: 444. 40. Ito Y, Kurokawa I, Nishimura K, et al. Keratin and filaggrin wxpression in keratoacanthoma. J Eur Acad Dermatol Venereol 2008; 22: 353. 41. Misago N, Takai T, Toda S, et al. The changes in the expression levels of follicular markers in keratoacanthoma depend on the stage: keratoacanthoma is a follicular neoplasm exhibiting infundibular/isthmic differentiation without expression of CK15. J Cutan Pathol 2014; 41: 437. 42. Requena L, Kutzner H, Hurt MA, et al. Malignant tumours with apocrine and eccrine differentiation. In: LeBoit PE, Burg G, Weedon D, et al, eds. Pathology and Genetics Skin Tumours. Lyon: IARC Press, 2006; 125. 43. Levy G, Finkelstein A, McNiff JM. Immunohistochemical techniques to compare primary vs. metastatic mucinous carcinoma of the skin. J Cutan Pathol 2010; 37: 411. 44. Marucci G, Betts CM, Golouh R, Peterse JL, Foschini MP, Eusebi V. Toker cells are probably precursors of Paget cell carcinoma: a morphological and ultrastructural description. Virchows Arch 2002; 441: 117. 45. Willman JH, Golitz LE, Fitzpatrick JE. Vulvar clear cells of Toker: precursors of
This article is protected by copyright. All rights reserved
extramammary Paget’s disease. Am J Dermatopathol 2005; 27: 185. 46. Perrotto J, Abbott JJ, Ceilley RI, Ahmed I. The role of immunohistochemistry in discriminating primary from secondary extramammary Paget disease. Am J Dermatopathol 2010; 32: 137.
This article is protected by copyright. All rights reserved
Table 1. Comparative CD117 immunohistochemistry of non-Merkel cell epithelial tumors of the skin Tumor type
Positivity,
Extent
Intensity
F18,
W8,
(94%)
D64
S47
Syringofibroadenomatous hyperplasia
1/1 (100%)
F1, D0
W1, M0, S0
Hidrocystoma
5/5 (100%)
F0, D5
W0, M1, S4
Hidrocystadenoma
2/2 (100%)
F1, D1
W0, M1, S1
Hidroacanthoma simplex
2/2 (100%)
F0, D2
W0, M0, S2
Poroma
29/29
F7,
W3, M8, S18
(100%)
D22
10/10
F2, D8
W0, M7, S3
n (%) Tumors
with
apocrine/eccrine 82/87
differentiation
Hidradenoma
M27,
(100%) Hidradenoma papilliferum
1/1 (100%)
F0, D1
W0, M0, S1
Spiradenoma
6/6 (100%)
F0, D6
W0, M0, S6
Cylindroma
1/1 (100%)
F0, D1
W0, M1, S0
This article is protected by copyright. All rights reserved
Syringoma
1/1 (100%)
F0, D1
W1, M0, S0
Mixed tumor of the skin
15/15
F4,
W0, M5, S10
(100%)
D11
Porocarcinoma
2/2 (100%)
F2, D0
W2, M0, S0
Ductal carcinoma
1/1 (100%)
F0, D1
W0, M1, S0
Cribriform carcinoma
1/1 (100%)
F0, D1
W0, M1, S0
Adenoid cystic carcinoma
1/1 (100%)
F0, D1
W0, M0, S1
Endocrine mucin-producing sweat gland 1/1 (100%)
F0, D1
W0, M1, S0
carcinoma Mucinous carcinoma
0/1 (0%)
Apocrine carcinoma
0/1 (0%)
Extramammary Paget disease
3/6 (50%)
F1, D2
W1, M1, S1
12/13
F9, D3
W7, M4, S1
Tumors with sebaceous differentiation
(92%) Sebaceoma
5/5 (100%)
F4, D1
W3, M1, S1
Sebaceous carcinoma
7/8 (86%)
F5, D2
W4, M3, S0
9/52 (17%)
F7, D2
W6, M3, S0
Tumors with follicular differentiation
This article is protected by copyright. All rights reserved
Trichofolliculoma Trichilemmal
0/2 (0%) cyst/Proliferating 0/5 (0%)
trichilemmal cyst Desmoplastic trichilemmoma
0/1 (0%)
Inverted follicular keratosis
0/4 (0%)
Pilomatricoma
4/4 (100%)
F3, D1
W3, M1, S0
Trichoblastoma/Trichoepithelioma
1/5 (20%)
F1, D0
W0, M1, S0
Basal cell carcinoma
4/25 (16%)
F3, D1
W3, M1, S0
Keratoacanthoma
0/6 (0%) F2, D0
W2, M0, S0
Tumors with keratinocytic differentiation
2/55 (4%)
Seborrheic keratosis
0/15 (0%)
Actinic keratosis
1/8 (13%)
F1, D0
W1, M0, S0
Bowen disease
1/14 (7%)
F1, D0
W1, M0, S0
Squamous cell carcinoma
0/18 (0%)
Metastatic carcinomas
0/24 (0%)
Mammary Paget disease
0/1 (0%)
D, diffuse (> 50%); F, focal (5–50%); M, moderate; S, strong; W, weak
This article is protected by copyright. All rights reserved
Figure legends
Figure 1. Most apocrine/eccrine tumors showed CD117-positivity to various degrees. A poroma with positive CD117 expression (A); in high-power view, the tumor cells and melanocytes in the basal cell layer of the existing epidermis were positive for CD117 (B). CD117-positivity of the cuticular tumor cells tended to be more frequent than that of the poroid tumor cells (C). Other CD117-positive apocrine/eccrine tumors included cylindroma (D), hidradenoma papilliferum (E), mixed tumor of the skin (F), cribriform carcinoma (G),
This article is protected by copyright. All rights reserved
and primary extramammary Paget disease (H). Images were taken with magnification of ×4 (A, E, F), ×10 (H), ×20 (C, D, G), or ×40 (B).
Figure 2. CD117 expression in sebaceous and follicular tumors. CD117 was typically expressed focally and weakly in sebaceomas (A. hematoxylin and eosin stain, B. CD117 immunostain) and sebaceous carcinoma (C. hematoxylin and eosin stain, D. CD117 immunostain). Pilomatricomas frequently showed CD117-positivity (E). The tumor cells in basal cell carcinomas were usually negative for CD117 (F). Images were taken using magnification of ×10 (E), ×20 (A, B), or ×40 (C, D, F).
This article is protected by copyright. All rights reserved
Figure 3. CD117 expression in the normal skin tissues and in the colonizing melanocytes of cutaneous adnexal tumors. CD117 was expressed in the eccrine units but not in the apocrine glands (A. hematoxylin and eosin stain, B. CD117 immunostain). Basal cell carcinomas contained numerous colonizing melanocytes (C. hematoxylin and eosin stain, D & E. melanosome [mouse anti-human monoclonal, clone HMB45, dilution 1:100; Dako, Glostrup, Denmark] immunostain with Giemsa as the counter-stain, F. CD117 immunostain with Giemsa as the counter-stain). Images were taken using magnification of ×10 (C, D), ×20 (A, B), or ×40 (E, F).
This article is protected by copyright. All rights reserved
