Unexpected maspin immunoreactivity in Merkel cell carcinoma

To select the cases, a database of 3,410 consecutive cutaneous tumors was evaluated. From these, 1,196 were malignant tumors, and the other 2,214 being benign tumors or pseudotumors. Only 3 out of 1,196 cutaneous malignant tumors were diagnosed as MCCs (0.25 %).

MCC preponderantly produces early metastases in regional and/or distant lymph nodes. Most of the cases (53 %) are diagnosed in stage III with metastases in more than 4 lymph nodes, and only 13 % of them being identified in stage I [3], such in our cases. About 20–30 % of MCCs are diagnosed with distant metastases [15, 16]. The 5-year survival rate depends on the tumor size, ranging from 66–75 % in tumors smaller than 2 cm to 50–60 % in those larger than 2 cm. Positive margins, absence of postoperative radio-chemotherapy, patient’s age (older than 75 years), relapses, and metastases are also considered important prognostic factors [3, 15‐17]. The 5-year survival rate decreases from 42–52 % in node-positive MCCs to 17–18 % in cases with distant metastases [18]. In every non-metastatic case (stages I and II), wide excision with safety limits and sentinel lymph node biopsy is recommended, followed by radiotherapy [13]. In node-positive MCCs (stage III), treatment of the nodal basin with lymphadenectomy and radiotherapy should be performed [16]. In cases with distant metastases (stage IV), platinum-based chemotherapy and/or immunotherapy should be associated [16]. Oblimersem sodium can be used in bcl-2 positive cases or lorvotuzamab mertansine in CD56-positive MCCs, but the results are not very well known [16]. The newest drug proposed to be used for patients with metastatic MCC is the PI3K pathway inhibitor called Idelalisib that was recently approved by the Food and Drug Administration for application in B-cell lymphoma [19].

Diagnosis of MCC is very difficult and should be based on the clinico-pathological parameters such as tumor location, patient’s age (being more frequent in sun-exposed areas and older people) correlated with the histological neuroendocrine aspect, and absence of the contact of tumor cells with the epidermis. In few cases epidermotropism and additional divergent components such as squamous, follicular, porocarcinoma, sarcomatous, glandular, and neuroblastic were noted [5]. However, the final diagnosis depends on the tumor profile and should take into account a metastasis from tumors with round cells (small cell lung carcinoma, neuroendocrine carcinoma of other organs, neuroblastoma), other cutaneous carcinomas with round cells (sweat gland carcinoma, basal cell carcinoma with neuroendocrine differentiation, small cell squamous cell carcinoma, primary cutaneous small cell carcinoma), lymphomas, and melanomas.

The immunohistochemical characteristics expression of neuroendocrine markers correlated with perinuclear dot-like positivity of keratin 20 is considered specific for MCC [4, 20]. However, unusual immunopositivity of the tumor cells was also reported (Table 2). The monoclonal antibody CM2B4 marker was also introduced in 2009 in clinical practice, which acts against a predicted antigenic epitope on the MCV T-antigen and can be added in the daily diagnosis panel of antibodies [20, 21]. In some of the cases, keratin 20 can be negative, especially in carcinomas unrelated with MCV infection [22]. Unusual positivity was reported for markers such as TTF-1 [23], CD57, PAX-5, TDT (terminal deoxynucleotidyl transferase) [24], and maspin, first reported in this study.

Differentiation of MCC from cutaneous metastases of neuroendocrine carcinoma is difficult and can be based on CEA negativity (that is usually positive in tumors of the gastrointestinal tract and pancreas) and inconstant positivity for keratin 20 [5, 6]. Predominantly, TTF-1 negativity and PAX-5 positivity of MCC is a diagnostic tool in differentiation from a metastatic lung cancer, although inconstant positivity was also observed in MCC [16, 23, 24]. Regarding the primary cutaneous small cell carcinoma, this lesion is characterized by the complete absence of nucleoli, which are well visible in MCC, and keratin 20 negativity [25]. Moreover, the primary cutaneous small cell carcinomas including basal cell carcinoma with neuroendocrine differentiation are negative for keratin 20 [20, 26].

The first description of the Merkel cells was performed in 1875 by Friedrich Sigmund Merkel, who called them “tastzellen” or “touch cell” [27]. Further studies proved, using the electron microscope, that they are located in the basal layer of epidermis and dermis and play a role on slowly adapting mechanoreceptors to sense touch and hair movement [9, 16]. Merkel cells display a monomorphic aspect with scanty cytoplasm and nuclei with fine chromatin; the MCC showing proliferation of similar cells with nuclear pleomorphism.

The role of MCV infection was also recognized in 2008 as a predisposing factor for genesis of MCC. However, due to the fact that MCC can arise in the background of chronic radiodermatitis in patients negative for MCV [28] and mostly occurs in the sun-exposed areas, supposition that ultraviolets can induce activation, proliferation, and malignization of some pluripotent stem cells was evolved [16]. Also, some studies have shown that MCC with divergent differentiation is an aggressive subtype, in whose development MCV is not involved [4, 5].

In this paper, unusual nuclear maspin maspin positivity was noted in both of the cases that occured in the sun exposed areas (case 1 - face and case 3 - upper trunk), without any correlation with the mitotic rate, depth of infiltration, or the quality of the resection margins. The normal epidermis showed a cytoplasmic positivity. However, being about first report in literature about maspin expression in MCC, it is difficult to emit suppositions about its role in this cutaneous tumor, further studies being necessary to confirm its positivity in larger cohorts.

Maspin (mammary serine protease inhibitor) is a member of the serine protease inhibitor family that is knows to play a tumor suppressor role in several malignant epithelial tumors such as colorectal or gastric carcinomas [29‐31]. However, its prognostic role depends on the subcellular localization, the p53-mediated nuclear positivity usually indicating a more aggressive behavior, a higher risk for tumor relapse and lymph node metastases, whereas loss of expression proved to induce a higher risk for distant metastases, at least for gastrointestinal malignant tumors [30, 31].

In tumors of the skin, maspin immunoreactivity was described in 97 % of squamous cell carcinomas and 88 % of basal cell carcinomas, but also in malignant melanomas, more frequent in sun-exposed areas [32, 33]. Few than 25 papers regarding maspin expression in cutaneous tumors have been published to date. However, there are limited data regarding correlation of maspin immunoreactivity with the prognosis. In squamous cell carcinomas, maspin positivity rate was higher in early stages compared with the advanced staged tumors (61 % vs. 39 %) and in non-metastatic tumors compared with cases that displayed lymph node positivity (67 % vs. 33 %). In bout squamous- and basal cell carcinomas it was especially displayed by tumors of the head and neck area (70 % vs. 30 %) [34]. A possible sun-activated maspin-induced DNA damage was also supposed [33]. On the other hand, nuclear expression proved to have a tumor suppressor role in basal cell carcinoma [34] but indicated poorer survival in melanomas [33].

The supposition about role of maspin in carcinogenesis of non-melanoma skin tumors that include MCC should be tested in further studies, on large cohorts.

This report shows that Maspin positivity in Merkel cell carcinoma might be related on sun exposure.

Written informed consent was obtained from the patients for publication of this Case Report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.