TRPS1 Expression in Endocrine Mucin-Producing Sweat Gland Carcinoma: Diagnostic Utility and Pitfalls

Abstract

To the Editor: Endocrine mucin-producing sweat gland carcinoma (EMPSGC) is a low-grade adenocarcinoma originating from sweat glands, exhibiting neuroendocrine differentiation. It is often considered the cutaneous analogue of solid papillary carcinoma of the breast.1,2 EMPSGC is an uncommon adnexal carcinoma, most frequently found in the eyelids and periorbital regions, with a predilection for older women. Histopathologically, EMPSGC typically manifests as relatively well-demarcated dermal nodules consisting of basaloid cells, exhibiting various growth patterns, such as cystic, solid, papillary, or cribriform. The tumor cells are commonly associated with mucin, extracellular and intracellular. Infiltrating glands with conspicuous pools of extracellular mucin can also be observed, mimicking primary cutaneous mucinous adenocarcinoma. Because of this feature, some consider EMPSGC a precursor lesion of primary cutaneous mucinous adenocarcinoma.3 Immunophenotypically, tumor cells of EMPSGC express CK7, GCDFP15, GATA-3, ER, PR, INSM1, synaptophysin, and chromogranin, while lacking CK20 immunoreactivity. This unique immunoprofile is usually helpful in distinguishing EMPSGCs from their morphologic mimics such as basal cell carcinomas (BCCs). However, in a superficially shaved biopsy with small lesional cells, there is potential for overlooking and misdiagnosing EMPSGCs as BCCs, especially considering BCCs' high prevalence in routine dermatopathology practice. Recent research has spotlighted TRPS1 as a diagnostic marker for carcinomas of mammary origin.4 However, subsequent studies have shown that TRPS1 expression is not confined to mammary neoplasms; it is also present in various cutaneous neoplasms, including extramammary Paget disease,5 squamous cell carcinoma in situ,5 cutaneous adnexal neoplasms,6 and even non-neoplastic reactive fibroblasts/myofibroblasts.7 Given the morphologic and immunophenotypic parallels between EMPSGC and solid papillary carcinoma of the breast, it is plausible to expect TRPS1 expression in EMPSGCs. Currently, there are limited data regarding the utility of TRPS1 in distinguishing EMPSGCs from their mimics such as BCCs and Merkel cell carcinomas (MCCs). To delve deeper into this, we conducted a pilot study to evaluate TRPS1 expression in a series of EMPSGC and its morphologic mimics. After Institutional Review Board approval (IRB#: 2022-0662), and with a waiver of informed consent, we gathered cases of EMPSGC (n = 10), BCC (n = 25), and MCC (n = 25) from our departmental pathology archives spanning years 2021–2023. All hematoxylin and eosin (H&E)-stained slides and relevant immunohistochemical studies, if applicable, were scrutinized to affirm the original diagnoses before retrieving the corresponding formalin-fixed, paraffin-embedded tissue blocks. Paraffin sections, freshly cut to 4–5 μm thickness from each formalin-fixed, paraffin-embedded tissue block, underwent immunohistochemical analysis using a monoclonal anti-TRPS1 rabbit antihuman antibody (Abcam, Waltham, MA, EPR16171, 1:2000) and a Leica Bond Max autostainer system (Leica Biosystems, GmbH, Nussloch, Germany) with standardized automated protocols. Positive immunoreactivity was defined as nuclear expression of TRPS1 in more than 5% of tumor cells. Intensity of TRPS1 expression was categorized into four levels: none (0), weak (1+), moderate (2+), and strong (3+), with innate eccrine glands serving as the internal control for strong (3+) intensity. In addition, the extent of TRPS1 expression was further stratified into four groups: none (0%–5%), focal (6%–25%), patchy (26%–75%), and diffuse (>75%). The immunohistochemical results of all tested cases were reviewed by two board-certified pathologists. TRPS1 expression was observed in all cases of EMPSGC (100%; 10/10), consistently demonstrating strong (3+) and diffuse expression (Fig. 1 and Table 1). In contrast, most BCCs (92%; 23/25) were negative for TRPS1 (Fig. 2 and Table 1). When present, TRPS1 expression in BCCs was only focal and weak. All MCCs (100%; 25/25) consistently lacked TRPS1 expression (Fig. 2 and Table 1). Although distinguishing MCCs from EMPSGCs is usually straightforward based on cytomorphology alone, the differentiation between BCCs and EMPSGCs can occasionally be challenging, particularly when the lesion arises in the eyelid or periorbital region, which is a common site for EMPSGCs. In addition, BCCs can, albeit rarely, exhibit neuroendocrine differentiation.8 Therefore, our findings are intriguing, as the presence of diffuse and strong TRPS1 expression, in conjunction with positivity for neuroendocrine markers, is likely to favor a diagnosis of EMPSGC over BCC. According to a recently published report,6 and our internal experiences with various cutaneous adnexal neoplasms, cutaneous carcinomas of eccrine origin can diffusely express TRPS1. However, no primary cutaneous adnexal carcinoma known to date, besides EMPSGC, strongly and diffusely expresses TRPS1 while simultaneously exhibiting neuroendocrine differentiation. Thus, this immunoprofile (ie, TRPS1+/neuroendocrine markers+) may be valuable in confirming the diagnosis of EMPSGC within the appropriate histopathologic and clinical context. One should, however, be aware that cutaneous metastasis of breast carcinoma can also exhibit strong and diffuse TRPS1 expression. Although metastatic breast carcinoma to the areas where EMPSGCs frequently arise (eg, eyelid, periorbital region) would be unusual, one should definitively rule out this possibility through documented clinical history or imaging studies. Furthermore, as with any other immunohistochemical studies, it is unwise to rely solely on one marker. If necessary, a panel of IHCs should be performed to avoid potential diagnostic pitfalls.FIGURE 1.: Representative two examples of endocrine mucin-producing sweat gland carcinoma (A, H&E, original magnification ×10; B, TRPS1, original magnification ×20; C, H&E, original magnification ×4; D, INSM1, original magnification ×10; E, TRPS1, original magnification ×10). TABLE 1. - Summary of TRPS1 Expression of EMPSGC, BCC, and MCC EMPSGC (n = 10) BCC (n = 25) MCC (n = 25) Extent of expression None 0 23 25 Focal 0 2 0 Patchy 0 0 0 Diffuse 10 0 0 Intensity of expression None 0 23 25 Weak 0 2 0 Moderate 0 0 0 Strong 10 0 0 FIGURE 2.: Representative examples of BCC (A, H&E, original magnification ×10; B, TRPS1, original magnification ×10) and MCC (C, H&E, original magnification ×10; D, TRPS1, original magnification ×10). Note that stromal/mesenchymal cells around the tumor nests of BCC are positive for TRPS1, whereas the tumor cells themselves are devoid of TRPS1 expression (B).In summary, we have demonstrated strong and diffuse TRPS1 expression in EMPSGCs, a finding, to our knowledge, not previously reported. Based on our pilot study, TRPS1 may serve as an additional tool in the diagnosis of EMPSGC. In cases where there is immunoreactivity with neuroendocrine markers, such as INSM1, synaptophysin, and chromogranin, concurrent strong and diffuse TRPS1 immunoreactivity is likely to favor a diagnosis of EMPSGC. However, it is important to note that our pilot study is limited because of the small cohort size. Therefore, further studies with a larger cohort are necessary to fully validate this finding and more accurately assess the utility of TRPS1 in this setting.

Topics & Concepts