Cylindromas are benign skin appendage tumors. They can be seen in conjunction with spiradenomas and trichoepitheliomas. Cases of spiradenocylindromas, demonstrating characteristics of both spiradenoma and cylindroma in the same tumor mass, have also been observed, suggesting similar derivation of both tumors.
They most commonly occur on the head and neck as solitary or multiple tumors. Solitary cylindromas occur sporadically and typically are not inherited. Multiple tumors are observed in an autosomal dominantly inherited manner. When nodules enlarge and coalesce on the scalp, they form the distinctive turban tumor feature.
Malignant cylindromas are very rare. Malignant transformation may develop within solitary cylindromas, or they may complicate the multiple variant (more common).
Cylindromas are appendage tumors previously thought to be of apocrine differentiation. While phenotypic features differ between cylindromas and spiradenomas, recent studies have shown immunohistological and cytomorphological overlap, with both tumors exhibiting apocrine, eccrine, secretory, and ductal features. Therefore, the cellular origin of cylindromas remains unknown. Cylindromas are most likely a very primitive sweat gland tumor differentiating toward either the eccrine or apocrine line.
Brooke-Spiegler syndrome (BSS) has been described as an autosomal dominant disease characterized by the development of multiple skin appendage tumors such as cylindromas, trichoepitheliomas, and spiradenomas, with a variable preponderance of any of the aforementioned subsets. Other lesions reported with BSS include parotid basal cell adenomas, organoid nevi, syringomas, and basal cell carcinomas. Despite variable phenotypic expressions of a predominant tumor in BSS, the gene responsible for multiple cylindromas, CYLD, is localized to band 16q12-q13. The mechanism of genotypic similarity and phenotypic variance is not yet understood.[1, 2]
In 2006, Zhang et al reported a large consanguineous Chinese family with BSS demonstrating intrafamily phenotypic variability. Upon examination, some persons only manifested discrete, small, skin-coloured growths, while the proband manifested an expansion of multiple large growths on the nose and numerous dome-shaped papules on the scalp. Biopsy showed both trichoepitheliomas and cylindromas in the affected persons. By sequence analysis, Zhang et al identified a recurrent mutation 2272C→T (R758X) of the CYLD gene in the affected familial persons that had been previously identified in other ethnic kindreds with familial cylindromatosis.
In 2007, Stegmeier et al noted that the CYLD gene encodes a deubiquitinating enzyme. The enzyme removes Lys-63–linked ubiquitin chains from I-kappaB kinase signaling components. By this mechanism, the enzyme inhibits NF-kappaB pathway activation. They demonstrated that CYLD is also required for the cell's timely entry into mitosis. Consistent with a cell-cycle regulatory function, CYLD localizes to microtubules in interphase and the midbody during telophase. CYLD 's protein levels decrease as cells exit from mitosis. Stegmeier et al identified the protein kinase Plk1 as a potential target of CYLD as a regulator of mitotic entry, and they suggested this because of the physical interaction and similar loss-of-function and over-expression phenotypes.
These findings raise the possibility that, as with other genes that regulate tumorigenesis, CYLD has both tumor-suppressing (apoptosis regulation) and tumor-promoting activities (enhancer of mitotic entry). They suggested that this additional function of CYLD could provide an explanation for the benign nature of most cylindroma lesions.
Massoumi and Paus and explained the manner in which CYLD interferes with tumor necrosis factor-alpha or Toll-like receptor–mediated signaling and with JNK or NF-kappaB-dependent p65/50 signaling to limit inflammation. Additionally, the manner by which CYLD interferes with activation of the proto-oncogene BCL3 and with cyclin D1 expression to limit tumorigenesis was also explained. Finally, the researchers discussed how tumor growth-promoting agents or UV light and inflammatory mediators may activate CYLD.
Researchers have noted a MYB-NFIB gene fusion, which provides a new genetic link between dermal cylindroma and adenoid cystic carcinoma.
Cylindromas are uncommon. The exact incidence is not known.
The exact international incidence of cylindromas is not known.
Most cylindromas remain benign; however, at least 14 reports have described malignant transformation. Multiple cylindromas can cover the entire scalp and cause the disfiguring turban tumor appearance, which necessitates extensive reconstructive surgery.
No racial disparity is reported for cylindromas.
The incidence of cylindroma is more common in females than in males. Female-to-male ratios of 6:1 and 9:1 have been reported.
Solitary cylindromas are lesions that affect middle-aged and elderly persons. Multiple, inherited cylindromas usually begin in early adulthood and increase in size and number throughout life.
Except for BSS, pertinent findings are largely limited to the skin. Histologically similar tumors have been found in the breast, parotid glands, salivary glands, lacrimal gland of the eye, Bartholin glands, the brain, lungs, and kidneys.
The cause of sporadic, solitary cylindromas is largely unknown; however, genetic studies of sporadic cylindromas show loss of heterozygosity at and around the CYLD locus in a substantial number of cases, suggesting that this gene also plays a role in the development of sporadic tumors.
The tyrosine kinase pathway might be a treatment target in tumors with defective CYLD.
Familial cylindromatosis is inherited in an autosomal dominant fashion, and the responsible gene, CYLD, is located on band 16q12-13. Tumors exhibit loss of heterozygosity, implicating the gene as a tumor suppressor gene.
A skin biopsy may be performed. Light microscopy with ordinary hematoxylin and eosin (H&E) staining is sufficient for diagnosis of cylindroma.
Cylindroma is a dermal tumor without attachment to the epidermis. The lesion is composed of numerous oval and polygonal nests molded into a jigsawlike pattern. Masses of epithelial cells are surrounded and penetrated by a hyaline sheath closely resembling a basement membrane. This sheath separates the tumor from the dermal mesenchyme, yet does not interfere with tumor growth and proliferation.
Malignant cylindromas demonstrate islands of cells displaying marked anaplasia and pleomorphism of nuclei. Mitoses are increased and are abnormal. Besides invasion into surrounding tissue, loss of the delicate hyaline sheath occurs.
Tumor islands are composed of 2 cell types. Peripheral cells are small and highly basophilic; palisading is suggested. Larger, more pale-staining cells are seen centrally. Small tubal lumina are sometimes found with careful observation.
A lack of lymphoid tissue is a histological feature that differentiates cylindromas from spiradenomas. Spiradenomas show a unique prominent presence of lymphocytes. Cylindromas, on the other hand, demonstrate a large number of prominent dendritic cells that most likely represent Langerhans cells that permeate the tumor. S-100 protein–, HLA-DR–, and CD1a-positive cells can be seen in cylindromas and represent the existence of Langerhans cells.
Hyaline bands, which surround tumor islands, are mostly composed of type IV collagen. This is equivalent to the subepidermal lamina densa. Fragments of anchoring fibrils, identical to type VII collagen, are also seen. These anchoring fibrils can also be seen embedded in basement membrane on electron microscopy (EM).
The hyaline membrane is highly enlarged compared with the dermoepidermal junction, as observed by EM. EM demonstrates a basement membrane of 2.7-4.3 µm, compared with the dermoepidermal junction average basement membrane thickness of 60-90 nm. EM also demonstrates that cylindromas lack clear distinction of lamina densa. Numerous abnormal inclusions of fibrillar material are noted and appear similar to lamina lucida.
Although hemidesmosomes (HDs) are seen with EM, they are irregularly spaced and can only be seen in high power, compared with normal basal keratinocytes, in which HDs are seen at low power. HDs noted in the cylindroma cells have greater size variation and half the normal number of basal keratinocytes.
Immunohistochemical studies on cylindromas demonstrate a variation in cell matrix proteins compared with normal epidermal basal keratinocytes. Alpha-6-beta-4 integrin expression in tumor cells is weaker. In normal basal keratinocytes, laminin-5 is understood as necessary for HD and basement membrane formation. Further studies have shown an improper processing of laminin-5 in cylindromas. This may explain the lower percentage of HDs in cylindromas. Cylindromas also demonstrate low expression of alpha-6-beta-4 integrins. These changes may be the cause the structural abnormalities seen in the basement membranes of cylindromas.
Immunohistochemical analysis has demonstrated myoepithelial, apocrine, eccrine, ductal, and secretory features in both cylindromas and spiradenomas. Alpha-SMA, indicating myoepithelial differentiation, has been shown to be expressed in the basaloid cells of both tumors. Both tumors also demonstrate S-100 protein expression, a marker designating eccrine differentiation and apocrine markers, human milk fat globulin, and lysozyme. In addition, expression of keratin polypeptides 10 and 14, specific for ductal epithelium, and keratins 7, 8, and 18, specific for secretory cells, have been observed in both cylindromas and spiradenomas.
Nerve growth factor, S-100 protein, CD44, and CD34 are other markers that have been found to be expressed in or surrounding eccrine coils and are not expressed in the eccrine duct or apocrine gland. These markers have all been found to varying degrees in cylindromas. Other immunohistochemical studies linking cylindromas to eccrine differentiation include positive expression of cytokeratins 19 and 1/10/11. IKH-4 is a monoclonal antibody specific for the eccrine gland and will not stain apocrine glands. This marker can also be used to differentiate eccrine from apocrine tumors and is positive in eccrine cylindromas and spiradenomas.
Immunohistochemical studies linking cylindromas to apocrine differentiation include the expression of alpha-1-antichymotrypsin, alpha smooth muscle actin (1A4), and cytokeratins 8 and 18.
Staining for cytokeratin 15, a marker specific for hair follicle stem cells, has also been shown in some cylindromas and spiradenomas. The expression of follicular, apocrine, and eccrine features in cylindromas suggests that the tumor may be derived from epithelial stem cells of immature differentiation.
Markers expressed in cylindromas and both eccrine and apocrine glands include epithelial membrane antigen, carcinoembryonic antigen, mucinlike carcinoma-associated antigen (B12), laminin, collagen IV, fibronectin, and CD34(QBEND/10).
Follow-up care of patients with multiple cylindromas is recommended because of the tendency for new lesions to develop. Follow-up care is also recommended because of the risk of malignant degeneration.
A cylindroma occasionally erodes through the skull, causing hemorrhage and meningitis.
Anal and rectal spiradenocylindroma-like basaloid carcinomas have been reported with CYLD gene mutations defined.
Scattered reports exist of malignant transformation of isolated cylindromas, so this must be a consideration for regular follow-up for patients with cylindromas.
The prognosis is not good with malignancy because visceral metastasis frequently follows.