—  SPECIALTY CONFERENCE  —

Breast Pathology

Case 2 - Clear Cells of Toker in the Nipple Epidermis

Timothy W. Jacobs
Virginia Mason Medical Center
Seattle, WA





Click on slide thumbnail images for an enlarged view.

If you have any difficulties viewing these slides, email the webmaster.



Clinical History:
A 39 year old woman underwent prophylactic mastectomy. The nipple was unremarkable clinically, as well as on gross pathologic examination. Images are from a routine section of the nipple.


Case 2 - Figure 1
Case 2 - Figure 2
Case 2 - Figure 3

Case 2 - Figure 4
Case 2 - Figure 5
Case 2 - Figure 6

Case 2 - Figure 7
Case 2 - Figure 8
Cytokeratin 7


Pathologic/Microscopic Findings
The nipple was grossly unremarkable, and routine sections were taken. Photomicrographs provided are from two separate fields of a hematoxylin and eosin (H&E) stained slide (figures 1 through 3, and 4 through 7). At low power in both areas, "clear" or pale staining cells can be appreciated at various levels within the epidermis. At intermediate and high power the cells appear slightly larger than surrounding keratinocytes, with pale/clear cytoplasm, but with bland appearing nuclei, with smooth nuclear membranes and small inconspicuous nucleoli (even at high power). Occasional clusters of these pale cells are seen, some with abortive tubule formation.

An immunostain shows the cells in question to be positive for cytokeratin 7 (figure 8). By immunohistochemistry (not shown), these cells were also positive for cytokeratin Cam 5.2, variably positive for estrogen receptor; the cells were negative for HER2, cytokeratin 5/6, S100 protein and gross cystic disease fluid protein (GCDFP). The cells were also negative for cytokeratin 20.

Of note, the mastectomy specimen was negative for in situ or invasive carcinoma.

Differential Diagnosis
This case raises diagnostic issues regarding clear cells in the epidermis of the nipple. In addition, issues pertaining to the judicious use of immunohistochemistry in this context will be considered.

The differential diagnosis of intraepidermal clear cells is theoretically quite wide, and includes epithelial, melanocytic, neuroendocrine and lymphoid lesions (for an excellent comprehensive review, see reference 1) [1]. However in practice, the entities to consider in the nipple skin are fewer and can be conceptually divided into malignant lesions (which have overtly atypical cells) and benign entities (with bland appearing cells) :
  • Malignant entities:
    • Paget's disease of the nipple

    • Malignant melanoma

    • Squamous cell carcinoma in situ
  • Benign entities:
    • Cleared keratinocytes

    • Clear cells of Toker
Of note, malignant lesions are much rarer than benign entities. Malignant melanoma and squamous cell carcinoma primarily involving the nipple are exceedingly rare (essentially confined to individual case reports [2] ), while Paget's disease of the nipple accounts for approximately 1-4% of breast cancer cases [3]. In contrast, benign entities are common. Cleared keratinocytes are a relatively frequent artifact, not unique to the skin of the nipple. Clear cells of Toker are found in up to 10-12% of nipples on routine hematoxylin and eosin stains [4], with a higher percentage when immunohistochemistry is used (specifically cytokeratin 7.) [5, 6, 7].

The differential diagnosis of cells in the nipple skin which are positive for cytokeratin 7 by immunohistochemistry includes Paget's disease of the nipple, clear cells of Toker and Merkel cells [5].

Final Diagnosis:
Clear Cells of Toker in the Nipple Epidermis

Case Discussion
The case presented here has classic clinical, morphologic and immunohistochemical features of clear cells of Toker. The intraepidermal clear cells originally described by Dr. Cyril Toker in 1970 and which bear his name are an incidental microscopic finding [4]. (Our patient had no clinical symptoms attributable to her nipple, and as noted, the nipple was grossly unremarkable.) Similar to the cells shown in this case, Toker cells have a bland cytologic appearance, are slightly larger than surrounding keratinocytes, with pale staining ("clear") cytoplasm. The nuclei are similar in size to those of surrounding keratinocytes with inconspicuous or small nucleoli, fine chromatin and smooth nuclear contours. Toker cells may be seen at any level within the epidermis. Of note, they are often found near openings of major lactiferous ducts. They may be present scattered singly or in clusters, with occasional lumen or abortive tubule formation, as in our case. Definitions of what constitutes so-called "Toker cell hyperplasia" vary widely (e.g. clusters of >3 cells, 7 cells, 20 cells or glands), however the relative number of cells and/or degree of clustering do not appear to have clinical significance [6, 7, 8]. Immunohistochemistry results for our case were as described for Toker cells. Toker cells are currently considered to be derived from ductal epithelium. Consequently, their immunophenotype is similar to normal breast ductal epithelium, including positive immunostaining for cytokeratin 7, cytokeratin Cam 5.2 and epithelial membrane antigen (EMA.) [5, 9]. Estrogen and progesterone receptor immunostaining is variable, and usually reflects that seen in the underlying benign lactiferous ducts. Toker cells are negative for HER2 (an important distinguishing feature from Paget cells which frequently are positive.) The cells are usually negative for GCDFP by immunohistochemistry, and mucin histochemistry (not done in our case) is also usually negative. In addition Toker cells are negative for cytokeratin 20. In contrast, Merkel cells are cytokeratin 20 positive, but may also be cytokeratin 7 positive. Merkel cells also differ from Toker cells in that they are restricted to the basal layer, have small round nuclei, scant cytoplasm , do not occur in clusters and are not easily seen on routine H&E stains [5]. (In addition to being cytokeratin 20 negative, the cells in our case were seen on H&E and had none of the morphologic features of Merkel cells.)

In contrast to the bland appearing Toker cells, the cells of Paget's disease of the nipple are pleomorphic and obviously malignant appearing; (this point seems obvious, but it is key to their distinction from Toker cells.) Paget cells are large, usually with round to oval nuclei which have irregular nuclear contours, and prominent nucleoli. Their cytoplasm is eosinophilic or amphophilic, often abundant and may contain mucin. Paget cells are usually present singly or in small aggregates, very rarely forming glands [10, 11]. Paget cells may take up melanin pigment released by epidermal cells or melanocytes, mimicking melanoma [12, 13]. Paget's disease of the nipple is associated with underlying breast carcinoma in ~95% of cases (ductal carcinoma in situ, usually high grade and/or invasive carcinoma.) In addition, unlike Toker cells which are an incidental finding, Paget's disease may present clinically with a scaly nipple lesion, redness, mass or discharge. (Absence of a clinical lesion however does not preclude the diagnosis of Paget's disease which may be clinically occult.) [14, 15]. Paget cells are glandular in origin by immunohistochemistry and ultrastructurally and consequently they are positive for cytokeratin 7, Cam 5.2 and EMA (just like Toker cells!) [5, 16]. Paget cells are usually positive for CEA and may be GCDFP positive. They are negative for high molecular weight cytokeratin and CK5/6, in contrast to pagetoid squamous cell carcinoma which would be CK5/6 positive. Unlike extramammary Paget's disease which may be cytokeratin 20 positive, Paget's disease of the nipple is usually cytokeratin 20 negative (akin to the usual CK7 positive/CK20 negative breast phenotype.) [17, 18]. Paget cells are usually similar morphologically and by immunostaining profile to the underlying carcinoma present [19, 20]. Because associated carcinomas are usually high grade (particularly DCIS), Paget cells are often estrogen and progesterone receptor (ER/PR) negative (95% of cases) and often HER2 positive (79-100% of cases) [21, 22, 23, 24]. Therefore, negative ER/PR immunostaining certainly does not preclude the diagnosis. However, HER2 staining if positive is particularly useful (especially in the differential diagnosis with melanoma, squamous cells carcinoma and Toker cells, which are all negative.) S100 immunostaining is variable in Paget's disease of the nipple (0-26% of cases positive, depending on the study) [25, 26]. Positive S100 staining therefore does not preclude the diagnosis of Paget's or rule in melanoma (a pitfall, particularly in Paget cells which have taken up pigment.) Paget cells will be negative for other melanocytic markers, e.g. HMB45. Although melanoma and squamous cell carcinoma were not serious considerations in our case based on H&E morphology, the cells were also negative by immunohistochemistry for S100 and cytokeratin 5/6 ruling out these entities. (See Table below for summary).

With regard to "bland" appearing cells, rare reports of Paget's disease of the nipple have been attributed to lobular carcinoma in situ (LCIS) (as opposed to the usual DCIS, with pleomorphic cells). A recent interesting case report describes a patient with bland clear intraepidermal cells in both nipples, with LCIS involving nearby lactiferous ducts in a pagetoid fashion [27]. By immunohistochemistry the intraepidermal cells were positive for cytokeratin 7 and estrogen receptor and negative for HER2 and e-cadherin. (Toker cells in contrast appear to be positive for e-cadherin; personal experience.) This was considered to be case of Paget's disease due to LCIS. In practice, the distinction between LCIS involving the nipple skin versus Toker cells may be difficult as both have quite bland appearing cells. In addition, e-cadherin immunostaining may be difficult to interpret on these single cells in a background of strongly positive keratinocytes. Pragmatically, the distinction may be moot as the management implications for either are not that of classic Paget's disease of the nipple with malignant appearing cells and underlying DCIS and/or invasive carcinoma.

Lastly, "atypical" Toker cells have also been described. Recently, 12.5% Toker cells in one study were considered "atypical" and reported to have slightly larger nuclei and more cytoplasm (but preserved nuclear-cytoplasmic ratio), slightly irregular nuclear membranes, and with more prominent eosinophilic nucleoli [6]. (Of note, our case did not have these morphologic features.) These authors found that immunostains for CD138 and p53 were useful adjuncts for distinguishing Toker cells from Paget cells. All Toker cells were p53 negative and 18/19 were CD138 negative, while amongst Paget's cases, 6/10 were p53 positive and 7/10 were CD138 positive. Of note, all Toker cells in this study were HER2 negative and 9/10 Paget's cases were HER2 positive. The clinical significance of so called "atypical" Toker cells remains to be determined.

Differential Diagnosis of Clear Cells in Nipple Epidermis

Paget's disease Toker cells Squamous cell carcinoma in situ Melanoma
Cytokeratin 7 Positive Positive Negative Negative
Cam 5.2 Positive Positive Negative Negative
EMA Positive Positive Positive or Negative Negative
CK5/6 Negative Negative Positive Negative
S100 Positive or Negative Positive or Negative Negative Positive
HMB45 Negative Negative Negative Positive
ER/PR Usually negative, (may be positive) Variably positive or negative Negative Negative
HER2 Often Positive, (may be negative) Negative Negative Negative
Mucin May be positive Negative Negative Negative
Morphology Malignant Bland Malignant Malignant

Conclusions
Clear cells of Toker are an incidental and benign microscopic finding, without specific clinical significance. However, their importance lies in the recognition of these mimickers by pathologists, to avoid misdiagnosis of malignant entities, particularly Paget's disease of the nipple. A key diagnostic feature of Toker cells is their bland cytologic appearance as opposed to the overtly atypical cells found in malignant lesions. Judicious use of immunohistochemistry as an adjunct necessitates familiarity with immunostaining profiles common to benign and malignant entities (e..g. Toker and Paget's both are cytokeratin 7 positive) as well as knowledge of useful contrasting features (e.g. frequent HER2 positivity of Paget's, while Toker cells, melanoma and squamous carcinoma are negative.) Above all, a diagnostic approach which integrates clinical and imaging findings, H&E morphology and immunophenotype is most prudent in this challenging area.

References
  1. Kohler S, Rouse RV, Smoller BR. The differential diagnosis of pagetoid cells in the epidermis. Mod Pathol. 1998;11(1):79-92.

  2. Venkataseshan VS, Budd DC, Un Kim D, Hutter RV. Intraepidermal squamous carcinoma (Bowen's disease) of the nipple. Hum Pathol. 1994 Dec;25(12):1371-4.

  3. Chen CY, Sun LM, Anderson BO. Paget disease of the breast: Changing patterns of incidence, clinical presentation, and treatment in the U.S. Cancer. 2006 Oct 1;107(7):1448-58.

  4. Toker C. Clear cells of the nipple epidermis. Cancer. 1970 Mar;25(3):601-10.

  5. Lundquist K, Kohler S, Rouse RV. Intraepidermal cytokeratin 7 expression is not restricted to Paget cells but is also seen in Toker cells and Merkel cells. Am J Surg Pathol. 1999 Feb;23(2):212-9.

  6. Di Tommaso L, Franchi G, Destro A, Broglia F, Minuti F, Rahal D, et al. Toker cells of the breast. Morphological and immunohistochemical characterization of 40 cases. Hum Pathol. 2008 Sep;39(9):1295-300.

  7. Nofech-Mozes S, Hanna W. Toker cells revisited. Breast J. 2009 Jul-Aug;15(4):394-8.

  8. van der Putte SC, Toonstra J, Hennipman A. Mammary Paget's disease confined to the areola and associated with multifocal Toker cell hyperplasia. Am J Dermatopathol. 1995 Oct;17(5):487-93.

  9. Yao DX, Hoda SA, Chiu A, Ying L, Rosen PP. Intraepidermal cytokeratin 7 immunoreactive cells in the non-neoplastic nipple may represent interepithelial extension of lactiferous duct cells. Histopathology. 2002 Mar;40(3):230-6.

  10. Shousha S. Glandular Paget's disease of the nipple. Histopathology. 2007 May;50(6):812-4.

  11. Barnes PJ, Dumont RJ, Higgins HG. Acinar pattern of mammary Paget's disease: a case report. Breast J. 2007 Sep-Oct;13(5):520-6.

  12. Culberson JD, Horn RC, Jr. Paget's disease of the nipple; review of twenty-five cases with special reference to melanin pigmentation of Paget cells. AMA Arch Surg. 1956 Feb;72(2):224-31.

  13. Azzopardi JG, Eusebi V. Melanocyte colonization and pigmentation of breast carcinoma. Histopathology. 1977 Jan;1(1):21-30.

  14. Paget J. On the disease of the mammary areola preceding cancer of the mammary gland. St Bartholomew Hospital Reports. 1874;10:87-90.

  15. Gullett AE, Prieto VG, Lopez A, Klein K, Arun B, Bedrosian I, et al. Type of Mammary Paget Disease Is Associated with Disease Free Survival (Meeting Abstract). Mod Pathol. 2010;23(Suppl 1):Abstract 507.

  16. Smith KJ, Tuur S, Corvette D, Lupton GP, Skelton HG. Cytokeratin 7 staining in mammary and extramammary Paget's disease. Mod Pathol. 1997;10(11):1069-74.

  17. Chaudary MA, Millis RR, Lane EB, Miller NA. Paget's disease of the nipple: a ten year review including clinical, pathological, and immunohistochemical findings. Breast Cancer Res Treat. 1986;8(2):139-46.

  18. Ordonez NG, Awalt H, Mackay B. Mammary and extramammary Paget's disease. An immunocytochemical and ultrastructural study. Cancer. 1987 Mar 15;59(6):1173-83.

  19. Wood WS, Hegedus C. Mammary Paget's disease and intraductal carcinoma. Histologic, histochemical, and immunocytochemical comparison. Am J Dermatopathol. 1988 Jun;10(3):183-8.

  20. Cohen C, Guarner J, DeRose PB. Mammary Paget's disease and associated carcinoma. An immunohistochemical study. Arch Pathol Lab Med. 1993 Mar;117(3):291-4.

  21. Meissner K, Riviere A, Haupt G, Loning T. Study of neu-protein expression in mammary Paget's disease with and without underlying breast carcinoma and in extramammary Paget's disease. Am J Pathol. 1990;137(6):1305-9.

  22. Wolber RA, Dupuis BA, Wick MR. Expression of c-erbB-2 oncoprotein in mammary and extramammary Paget's disease. Am J Clin Pathol. 1991;96(2):243-7.

  23. Anderson JM, Ariga R, Govil H, Bloom KJ, Francescatti D, Reddy VB, et al. Assessment of Her-2/Neu status by immunohistochemistry and fluorescence in situ hybridization in mammary Paget disease and underlying carcinoma. Appl Immunohistochem Mol Morphol. 2003 Jun;11(2):120-4.

  24. Bianco MK, Vasef MA. HER-2 gene amplification in Paget disease of the nipple and extramammary site: a chromogenic in situ hybridization study. Diagn Mol Pathol. 2006 Sep;15(3):131-5.

  25. Gillett CE, Bobrow LG, Millis RR. S100 protein in human mammary tissue--immunoreactivity in breast carcinoma, including Paget's disease of the nipple, and value as a marker of myoepithelial cells. J Pathol. 1990 Jan;160(1):19-24.

  26. Ramachandra S, Gillett CE, Millis RR. A comparative immunohistochemical study of mammary and extramammary Paget's disease and superficial spreading melanoma, with particular emphasis on melanocytic markers. Virchows Arch. 1996 Dec;429(6):371-6.

  27. Sahoo S, Green I, Rosen PP. Bilateral paget disease of the nipple associated with lobular carcinoma in situ. Arch Pathol Lab Med. 2002 Jan;126(1):90-2.