To the Editor:
We read with great interest the findings of the recent study on cutaneous nodular (apocrine) hidradenoma (HA) by MacKinnon et al,1 in which the authors demonstrate the presence of solid proliferations of myoepithelial cells in a conspicuous minority (17%) of otherwise typical HA. According to the authors, this finding “illustrates a biological link between HA and adenomyoepithelioma (AME).” Despite recognizing that “further work is required to establish distinct diagnostic categories in this spectrum,” the authors propose a tumor volume threshold of 25% for the myoepithelial population as a diagnostic criterion to distinguish between HA and cutaneous adenomyoepithelioma (AME).1
We commend the authors for highlighting the possible occurrence of solid foci with myoepithelial differentiation within conventional HA. HA has been classically known to be composed of 5 morphological types of epithelial cells, in variable combination and proportion: clear, polygonal, squamous, oxyphilic, and mucinous cells.2,3 Several authors have emphasized the absence of true myoepithelial differentiation in HA,2,3 and this histologic feature has been regarded as a diagnostic clue in the differential diagnosis with other entities such as aggressive digital papillary adenocarcinoma as well as mixed tumor and myoepithelioma.2–4 Indeed, this concept was recently reiterated by Wiedemeyer et al4 in their study on HA on acral sites, based on evidence of absent immunohistochemical (IHC) staining for S100 and smooth muscle actin. Thanks to their in-depth IHC assessment of cutaneous HA with a large panel of myoepithelial immunomarkers (including CK8/18, calponin, and Glial fibrillary acidic protein), MacKinnon et al1 seem to have convincingly confuted the notion that evidence of myoepithelial differentiation is inconsistent with a diagnosis of HA.
At the same time, however, we not entirely agree with the authors' conclusions regarding the biological relationship between HA and cutaneous AME.1 AME is a benign neoplasm that most frequently occurs in the breast5; AME has been regarded as a myoepithelial lesion, due to the presence of a dominant myoepithelial population. Differently from pure myoepithelial neoplasms such as myoepithelioma and myoepithelial carcinoma, however, AME is characterized by a combined proliferation of myoepithelial cells and glandular cells.5 The elementary structure of stereotypical AME is constituted by small, round-to-oval lumina surrounded by cuboidal glandular cells, which are encompassed by polygonal or spindle-shaped myoepithelial cells.5,6 The myoepithelial component is typically predominant, compressing the glandular lumina, thus resulting in areas virtually devoid of glands. Some AME cases may show a prominent myxochondroid matrix, thus overlapping with pleomorphic adenoma of the breast; in these cases, differential diagnosis based on sole morphology may be somewhat arbitrary.5 Importantly, it should be emphasized that even among breast pathologists, no universal consensus has been reached regarding AME diagnostic criteria, and diagnosing AME on purely morphological grounds often represents a challenging task.5,6
In light of the obvious analogies between the mammary gland and cutaneous apocrine glands, it seems logical that AME may indeed develop also in the skin, and that the frequency of such occurrence may have been underestimated.6 As already stated by other authors,7 several cases of cutaneous AME in the past are likely to have been misclassified as clear cell HA because of morphological similarities between the clear myoepithelial cells in AME and classical clear cells of HA. Indeed, one case of the series of MacKinnon et al1 probably represented (as recognized by the authors) a legitimate cutaneous AME, initially misdiagnosed as HA on routine sections. As a rule, conducting extensive IHC evaluation is paramount to establish a diagnosis of cutaneous AME.1,7
The conclusions of MacKinnon et al1 regarding the biological relationship between HA and AME are drawn solely based on morphological and IHC grounds, with no attempt to integrate such data with investigations of the genetic background behind these neoplasms. We believe that molecular pathology will play an increasingly significant supporting role in the diagnosis of difficult adnexal neoplasms, as it already occurs in breast and salivary gland pathology. To better exemplify this point, it was recently shown that some of the previously diagnosed cases of MAML2 translocation-negative mucoepidermoid carcinoma of the salivary glands actually represented hyalinizing clear cell carcinoma, thanks to the detection of EWSR1 pathogenetic translocations.8 Interestingly, recent research has revealed that most cases of breast AME develop as a result of PIK3CA or AKT1 oncogenic mutations, with HRAS mutations driving the development of an ER-negative aggressive phenotype.9 Furthermore, the presence of HMGA2 aberrations in a small minority of breast AME cases seems to support a biologic link with pleomorphic adenoma, as well as with epithelial–myoepithelial carcinoma of the salivary glands.9,10 On the other hand, approximately half of cutaneous HA cases harbor an oncogenic CRTC1/CRTC3-MAML2 fusion gene.11,12 Importantly, in the skin, this genetic alteration seems to be specific for HA and mucoepidermoid carcinoma; fittingly, so-called “poroid HA” has been consistently shown to lack CRTC1/CRTC3-MAML2 translocations, instead harboring YAP1 aberrations, in line with other neoplasms of the poroma family (hidroacanthoma simplex, classic poroma, dermal duct tumor, and porocarcinoma).13
In sum, we believe that, instead of solely relying on an arbitrary volume threshold for the myoepithelial population as a diagnostic tool to separate between HA and cutaneous AME, any attempt to establish a definite distinction between these 2 neoplasms should not prescind from integration with corresponding genetic data. In light of the stark differences between known oncogenic alterations in breast AME and in cutaneous (and breast) HA, it appears at least possible that legitimate cases of cutaneous AME may show a genetic background more in keeping with breast AME than with cutaneous HA; further research is warranted in this regard. In the meantime, the existence of a continuous spectrum encompassing cutaneous HA and cutaneous AME, albeit solely based on IHC evidence of solid myoepithelial foci in a minority of cutaneous HA cases, remains an interesting hypothesis. We agree with MacKinnon et al1 that in-depth IHC staining for myoepithelial markers may at present be the only feasible strategy to avoid missing a diagnosis of cutaneous AME in suspicious cases.
1. MacKinnon WF, Moss P, Pasternak S, et al. Apocrine hidradenoma and adenomyoepithelioma: entities on a biological continuum of adnexal neoplasia. Am J Dermatopathol. 2020. [Epub ahead of print].
2. Hidradenoma. In: Kazakov DV, Michal M, Kacerovska D, et al, eds. Cutaneous Adnexal Tumours. Philadelphia, PA: Lippincott Williams & Wilkins; 2012:9–23.
3. Apocrine hidradenoma. In: Requena L, Sangüeza O, eds. Cutaneous Adnexal Neoplasms. Cham, Switzerland: Lippincott Williams & Wilkins; 2017:81–102.
4. Wiedemeyer K, Gill P, Schneider M, et al. Clinicopathologic characterization of hidradenoma on acral sites: a diagnostic pitfall with digital papillary adenocarcinoma. Am J Surg Pathol. 2019. [Epub ahead of print].
5. Hayes MM. Adenomyoepithelioma of the breast: a review stressing its propensity for malignant transformation. J Clin Pathol. 2011;64:477–484.
6. González I, Gonzalez ML, Kroh JM, et al. Cutaneous adenomyoepithelioma: report of a case and review of the literature. Am J Dermatopathol. 2016;38:549–552.
7. Tran TAN, Bridge JA, Deharvengt SJ, et al. Primary cutaneous adenomyoepithelioma ex spiradenoma with malignant histologic features, epithelial-myoepithelial carcinoma type: a first case report with molecular studies. Int J Surg Pathol. 2019. [Epub ahead of print].
8. Hsieh MS, Wang H, Lee YH, et al. Reevaluation of MAML2 fusion-negative mucoepidermoid carcinoma: a subgroup being actually hyalinizing clear cell carcinoma of the salivary gland with EWSR1 translocation. Hum Pathol. 2017;61:9–18.
9. Lubin D, Toorens E, Zhang PJ, et al. Adenomyoepitheliomas of the breast frequently harbor recurrent hotspot mutations in PIK3-AKT pathway-related genes and a subset show genetic similarity to salivary gland epithelial-myoepithelial carcinoma. Am J Surg Pathol. 2019;43:1005–1013.
10. El Hallani S, Udager AM, Bell D, et al. Epithelial-myoepithelial carcinoma: frequent morphologic and molecular evidence of preexisting pleomorphic adenoma, common HRAS mutations in PLAG1-intact and HMGA2-intact cases, and occasional TP53, FBXW7, and SMARCB1 alterations in high-grade cases. Am J Surg Pathol. 2018;42:18–27.
11. Winnes M, Mölne L, Suurküla M, et al. Frequent fusion of the CRTC1 and MAML2 genes in clear cell variants of cutaneous hidradenomas. Genes Chromosomes Cancer. 2007;46:559–563.
12. Kuma Y, Yamada Y, Yamamoto H, et al. A novel fusion gene CRTC3-MAML2 in hidradenoma: histopathological significance. Hum Pathol. 2017;70:55–61.
13. Sekine S, Kiyono T, Ryo E, et al. Recurrent YAP1-MAML2 and YAP1-NUTM1 fusions in poroma and porocarcinoma. J Clin Invest. 2019;130:3827–3832.