To the Editors:
We have read with great interest the article of Kazakov et al1 on cutaneous adenolipoma. Over the past 8 years, we have compiled a series of 11 cases mostly seen in consultation by one of us (B.Z.) confirming Dr Kazakov's findings and adding some new observations.
Clinical details are summarized in Table 1. Briefly, patients comprised 6 men and 5 women with a mean age of 53 years. Three cases presented as nevus lipomatosus located at the groin and buttock. Seven of the lesions were classified as spindle cell lipoma and were located on the upper arm, thigh (n = 3), head, neck, or wrist, one as a complex apocrine angiomatous hamartoma with a main lipomatous component located at the groin. Physical examination in this latter patient did not reveal other cutaneous abnormalities, and there was no history of trauma. Each of the lesions revealed an eccrine/apocrine glandular and ductal component of varying proportions.
The specimens of nevus lipomatosus (cases 1, 2, and 3) showed ill-defined lesions with a papillomatous epidermis and a deposition of mature fat cells in the superficial dermis accompanied by septae of collagenous fibrous tissue thinning out with depth. Deeply within the subcutaneous tissue were secretory coils and ducts of eccrine glands surrounded by loosely arranged fibrous tissue rich in mucin (Fig. 1).
The 7 spindle cell lipomas (cases 4–10) were quite different from each other: 2 were located solely in the dermis (cases 5 and 7), 3 both in the dermis and subcutis (cases 6, 8, and 10), and 2 were purely subcutaneous (cases 4 and 9), and in contrast to the others, one was encapsulated (case 4). The dermal cases were rather ill-defined and intermingled by fibrous tissue. All these lesions showed mature adipocytes accompanied by a proliferation of bland spindle cells set in a mucinous to collagenous stroma in places with conspicuous mast cells. Case 7 in addition showed floret-like giant cells. The degree of lipomatous differentiation varied and was only moderately present in case 5 with strong positivity of the spindle cell population for CD34, however. In this case, there were slightly dilated coils of eccrine sweat glands being present close to the lesion's margin surrounded by slender spindle cells set in a collagenous stroma rich in mucin. Case 4 showed secretory coils and dilated ducts of eccrine glands close to the periphery but obviously within the lesion. Similar to case 8, the glandular elements revealed altered polarity of luminal cells as described previously.1 The glandular component of cases 6–10 was apocrine in nature with dilated secretory elements and ducts, again surrounded by spindle cell lipomatous elements (Fig. 2). Case 6 presented as a recurrence. Case 10 in addition showed a marked inflammatory component consisting of T-lymphocytes and B-lymphocytes without evidence of kappa or lambda restriction. All these spindle cell lipomas showed a moderate degree of vascularity.
Case 11 was comparatively different from the others and showed an ill-defined lesion with a papillomatous epidermis and a free grenz zone. The reticular and deep dermis revealed prominent fibrous tissue, which was highly vascularized and intermingled by mature adipocytes. In addition to the lipomatous, vascular, and fibrous components, there were large numbers of apocrine secretory elements with prominent dilatation of ducts. In some places, these were intimately associated with the lipogenic and vascular component. The epithelial secretory and ductal cells were compressed and thus partially flattened. Also, curled nerve trunks, compressed sebaceous glands, and a mild inflammatory component consisting of lymphocytes and macrophages were present (Fig. 3).
Neoplasms may show differentiation toward one cell line, such as adipocytes in classic lipoma, sometimes toward 2 cell lines, such as adipocytes and capillaries with fibrin thrombi in angiolipoma, or may be even more complex, involving several cell lines, such as adipocytes, vessels, collagen, epidermis, and sometimes follicular–sebaceous structures in nevus lipomatosus,2 which is then known as a hamartoma. Besides mature adipocytes, spindle cell lipomas show spindle cells and mucin and thus may be regarded as a hamartoma as well.3
Moreover, adipocytic metaplasia in the skin is seen in a variety of benign epithelial proliferations with adnexal differentiation, lesions of hair follicle (such as fibrofolliculoma/trichodiscoma, folliculosebaceous cystic hamartoma, and desmoplastic trichoepithelioma, the latter sometimes being associated with a melanocytic nevus as well), lesions of sweat and sudoriferous glands (such as syringofibroadenoma, eccrine angiomatous hamartoma, and mixed tumor of the skin, eccrine, and apocrine types), and is also commonly seen in congenital and acquired melanocytic nevi.
Conversely, eccrine/apocrine elements in lipogenic lesions seem to be rare with around 30 reported cases in the literature1,4–7 and have recently also been observed in one spindle cell lipoma.1 This latter observation is confirmed by 7 new cases in our series. In addition, we observed eccrine glandular components in 3 cases of nevus lipomatosus. The glandular changes described so far ranged from a single eccrine secretory coil to numerous areas with dilated eccrine gland structures. Apocrine elements present in 6 of our cases have rarely been encountered in the literature.6
Because of the lack of specific adipocytic immunohistochemical markers and the common finding of entrapped, metaplastic, or even neoplastic8 adipose tissue in various other neoplasms, cytogenetics has become a valuable tool in the investigation of soft tissue tumors suspicious of adipocytic differentiation. Molecular studies using cytogenetic techniques, such as fluorescence in situ hybridization, have confirmed the heterogeneity of lipogenic lesions outlining different chromosomal abnormalities for benign, intermediate, and malignant lesions. Whereas a number of aberrations may be observed in ordinary lipomas (the 12q13-15 region being the most commonly involved), characteristic chromosomal abnormalities have been reported for lesions such as chondroid lipoma [t(11;16)(q13;p13)], lipoblastoma (8q11-13 rearrangement), atypical lipomatous tumor/well-differentiated liposarcoma and dedifferentiated liposarcoma (ring/giant marker chromosome with amplified 12q13-15 material, involving MDM2 and CDK4), and myxoid/round cell liposarcoma [mostly t(12;16)(q13;p11)]. In addition, cytogenetics has supported the concept of spindle cell/pleomorphic lipoma being variations of one clinicopathologic entity.9 Interestingly, angiolipomas have a normal karyotype, which favors a different pathogenesis from that of pure lipomas.10
Cytogenetic investigations in adenolipomas are still pending. Our data and data from the literature confirm that eccrine/apocrine glandular and ductal components may occur in benign lipogenic lesions raising several pathogenetic possibilities. One concept is that these lesions develop from peripheral adipose tissue of eccrine glands,5 another concept is that the glands are entrapped by adipose proliferation.4 Compression and partial occlusion of glands and ducts by adipocytic growth may be an additional component leading to cystic dilatation and glandular epithelial hyperplasia, especially evident and possibly causative in case 11, located at the groin. While supernumerary anogenital mammary-like glands enter the differential diagnosis,7 in our latter case, the glandular component was extensive and intimately associated with an adipocytic, fibrous, and vascular component, suggesting a hamartomatous process—similar or related to eccrine angiomatous hamartoma with lipomatous metaplasia. To the best of our knowledge, this apocrine variant has not been previously described in the literature.
The authors thank the following dermatopathologists, pathologists and clinicians who kindly contributed case materials and clinical data: Dr H. Hudler, Winzendorf, Austria; Dr P. Kind and Dr I. Kiehlmann, Offenbach, Germany; Dr P. Mastan, Vienna, Austria; Dr M.-E. Olszewsky, Fuerth, Germany; Dr E. Rupert, Fieberbrunn, Austria; Dr C.G. Schirren, Darmstadt, Germany.
1. Kazakov DV, Spagnolo DV, Kacerovska D, et al.. Cutaneous adenolipoma: extending the spectrum of changes in the lipomatous and epithelial components. Am J Dermatopathol. 2011;33:56–59.
2. Ansai S-J, Kimura T, Kawana S. A clinicopathologic study of folliculosebaceous cystic hamartoma. Am J Dermatopathol. 2010;8:815–820.
3. Zelger B, Zelger BG, Plörer A, et al.. Dermal spindle cell lipoma: plexiform and nodular variants. Histopathology. 1995;27:533–540.
4. Hitchcock MG, Hurt MA, Santa Cruz DJ. Adenolipoma of the skin: a report of nine cases. J Am Acad Dermatol. 1993;29:82–85.
5. Ait-Ourhrouil M, Grosshans E. Le lipome périsudoral [in French]. Ann Dermatol Venereol. 1997;124:845–848.
6. Antúnez P, Santos-Briz A, Munoz E, et al.. Cutaneous apocrine cystic adenolipoma. Am J Dermatopathol. 2005;27:240–242.
7. Pantanowitz L, Henneberry JM, Otis CN, et al.. Adenolipoma of the external female genitalia.Int J Gynecol Pathol. 2008;27:297–300.
8. Lee J-C, Fletcher CDM. Malignant fat-forming solitary fibrous tumor (so-called “lipomatous hemangiopericytoma”): clinicopathologic analysis of 14 cases. Am J Surg Pathol. 2011;35:1177–1185.
9. Nishio J. Contributions of cytogenetics and molecular cytogenetics to the diagnosis of adipocytic tumors. J Biomed Biotechnol. 2011;2011:524067 Epub 2011 January 11. Review.
10. Sciot R, Akerman M, Dal Cin P, et al.. Cytogenetic analysis of subcutaneous angiolipoma: further evidence supporting its difference from ordinary pure lipomas: a report of the CHAMP study group. Am J Surg Pathol. 1997;21:441–444.