Cylindromatosis, Familial

A number sign (#) is used with this entry because of evidence that familial cylindromatosis is caused by heterozygous mutation in the CYLD gene (605018) on chromosome 16q12.

See also Brooke-Spiegler syndrome (BRSS; 605041) and multiple familial trichoepithelioma-1 (MFT1; 601606), which are allelic disorders with overlapping phenotypes.

Description

The disorders classically referred to as familial cylindromatosis, Brooke-Spiegler syndrome, and multiple familial trichoepithelioma were originally described as distinct clinical entities. Patients with BRSS develop multiple skin appendage tumors including cylindromas, trichoepitheliomas, and spiradenomas. Patients with familial cylindromatosis have only cylindromas, and those with MFT1 have only trichoepitheliomas. However, because these disorders show overlapping phenotypic features, and because different manifestations of each have been described within a single family, many consider these disorders to represent a phenotypic spectrum of a single disease entity (Guggenheim and Schnyder, 1961; Welch et al., 1968; Gerretsen et al., 1995; Lee et al., 2005; Bowen et al., 2005; Young et al., 2006; Saggar et al., 2008).

Van Balkom and Hennekam (1994), who preferred the designation 'dermal eccrine cylindromatosis' for familial cylindromatosis, provided a review. 'Eccrine' referred to histologic evidence that the tumors may originate from the eccrine sweat glands.

Blake and Toro (2009) provided a detailed review of the spectrum of disorders associated with CYLD mutations.

Clinical Features

Ancell (1842) and Spiegler (1899) described a familial syndrome characterized by tumors of skin appendages, now known as cylindromas (Lee et al., 2005).

Baden (1962) noted that cylindromatosis can clinically resemble neurofibromatosis (NF1; 162200).

Welch et al. (1968) presented family data supporting the view that 'Ancell-Spiegler' cylindromas and 'Brooke-Fordyce' trichoepitheliomas were manifestations of a single entity.

Harper (1971) provided a dramatic example of cylindromatosis that developed into a 'turban tumor,' covering the scalp.

Vernon et al. (1988) described a solitary, apparently benign lung cylindroma in a 42-year-old woman with multiple cutaneous cylindromas. She died from coronary occlusion which occurred precociously in other members of her family, suggesting no connection between the coronary artery disease and the cylindromas.

Gerretsen et al. (1995) described a large family in which cutaneous cylindromas occurred in members of 5 generations with 30 affected persons (11 male, 19 female). Trichoepitheliomas and milia were also observed. Female-to-female, female-to-male, male-to-female, and male-to-male inheritance was observed. Penetrance reached 100% in adult life.

Poblete Gutierrez et al. (2002) reported a 4-generation German family in which 4 individuals had skin appendage tumors inherited in an autosomal dominant pattern. The eldest affected member, deceased at the time of the report, had turban tumor-like cylindromas of the scalp and nasolabial region confirmed by histologic examination. Small nasolabial tumors from a woman in the third generation showed cylindromas on histology. The youngest family member, in the fourth generation, had cylindromas of the nasolabial region and trichoepitheliomas of the nose and scalp. Poblete Gutierrez et al. (2002) commented on the intrafamilial as well as intraindividual phenotypic variability.

Stoll et al. (2004) reported a mother and daughter with familial cylindromatosis. The dermal eccrine cylindroma arose as small solitary lesions in the mother at age 28 years. Other tumors arose, and she had surgery multiple times. The daughter developed lesions on the frontal part of the scalp at age 23 years. Family history revealed that the mother's sister was also affected.

Mapping

Studying 2 families with cylindromatosis, Biggs et al. (1995) found strong evidence for linkage to chromosome 16q12-q13. There was consistent loss of the wildtype allele near these markers in 19 tumors from 4 individuals, suggesting that the gene was a tumor suppressor gene.

Verhoef et al. (1998) studied a large Dutch family in which affected members had originally been given the diagnosis of tuberous sclerosis. Linkage was excluded from both of 2 chromosomal regions involved in tuberous sclerosis complex on 9q34 (TSC1; 191100) and 16p13 (TSC2; 613254). Reevaluation of the clinical and pathologic data led to a change of the working diagnosis to autosomal dominant cylindromatosis. Subsequent linkage analysis showed a lod score of 3.02 with marker D16S308 at chromosome 16q12-q13.

Biggs et al. (1996) examined polymorphic markers on each chromosome, some of which are close to known tumor suppressor genes, in 25 tumors from 4 individuals with familial cylindromatosis. No loss of heterozygosity (LOH) was detected other than that at loci on chromosome 16q. They suggested that the candidate gene, which they symbolized CYLD1, may be the only tumor suppressor gene implicated in the development of cylindromas. They also demonstrated LOH using markers on chromosome 16q in 8 of 14 (57%) sporadic cylindromas, indicating that the CYLD gene is probably involved in the genesis of both familial and sporadic cylindromas.

Takahashi et al. (2000) evaluated 19 families with this disorder by a combination of genetic linkage analysis and LOH in cylindromas from affected persons. All 15 informative families showed linkage to 16q12-q13, thus providing no evidence for genetic heterogeneity. Recombinant mapping placed the gene in an interval of approximately 1 Mb. There was no evidence of haplotype sharing between families.

Molecular Genetics

Bignell et al. (2000) identified 21 different germline mutations in the CYLD gene in affected members of 21 families with cylindromatosis. Six somatic mutations were identified in 1 patient with sporadic disease and 5 patients with familial disease. All mutations predicted truncation or absence of the encoded protein (see, e.g., 605018.0001-605018.0002).

In affected members of a German family with cylindromas, including 1 patient who also had trichoepitheliomas, suggesting BRSS, Poblete Gutierrez et al. (2002) identified a heterozygous truncating mutation in the CYLD gene (605018.0003). The results indicated that a single CYLD mutation can result in phenotypically different tumor types, indicating that cylindromas and trichoepitheliomas are allelic disorders.

Young et al. (2006) identified a heterozygous mutation in the CYLD gene (605018.0008) in a 73-year-old man with cylindromatosis and turban tumor syndrome and in his 2 children with multiple familial trichoepitheliomas without cylindromas. The findings suggested that the 2 disorders represent phenotypic variation of a single genetic defect.

Saggar et al. (2008) performed genetic analysis of 25 probands with familial skin appendage tumors. In total, 18 mutations in CYLD, including 6 novel mutations, were identified in 25 probands (72%). The mutation frequencies among distinct phenotypes were 85% for BRSS, 100% for FC, and 44% for MFT1. The majority of the mutations resulted in truncated proteins. There were no apparent genotype-phenotype correlations. Saggar et al. (2008) concluded that mutations in the CYLD gene underlie all 3 disorders, but that the reasons for phenotypic variability remain to be explored.

History

Schmidt-Baumler (1931) raised the question of X-linked dominant inheritance. The pedigree of Blandy et al. (1961) showed an affected male who had all daughters affected and all sons unaffected.

Up to the end of 1954, Evans et al. (1966) found 47 reported cases of cylindroma, of which 30 were female. The authors noted that the term cylindroma had also been applied by Billroth (1859) to a type of adenocarcinoma arising in salivary gland tissue.

Hart (1973) suggested that the lesions of hereditary multiple benign cystic epithelioma could be discerned in ancient Parthian coins.