Allelotypes of primary cutaneous melanoma and benign melanocytic nevi.

A multistep genetic model of tumorigenesis, based on genetic alterations in benign and primary malignant lesions, has been proposed for neoplasms such as colonic carcinoma. However, evidence for a similar genetic progression in melanoma has relied heavily on findings in cultured lesions or metastases. We have investigated every autosomal arm for loss of heterozygosity in 41 primary cutaneous melanomas and 32 benign melanocytic nevi, and have investigated several chromosome arms that show loss in melanoma in 27 Spitz nevi (a nevus with histological similarities to melanoma). Loss of heterozygosity in primary melanoma was identified most frequently on chromosomes 9p (46%) at loci near the p16INK4 gene, 10q (31%), 6q (31%), and 18q (22%); loss of these chromosome arms were related to the progression of the melanoma. Only two benign melanocytic nevi (both of which showed atypical features on histology) demonstrated genetic alterations, including p9 loss in one case. In addition, two Spitz nevi contained interstitial deletions on chromosome 9p. Our findings show that loss of heterozygosity of 9p is not confined to melanoma, but that other uncultured melanocytic lesions can also display loss of this chromosome arm, and that other genetic changes (e.g., loss of 10q, 6q, and 18q) may be important in conveying the malignant phenotype to melanoma.

Prognostic significance of allelic losses in primary melanoma.

Loss of genetic material, including loss of loci on chromosome arms 6q, 9p, and 10q, occurs frequently in cutaneous melanoma but infrequently in benign melanocytic nevi or other melanocytic lesions, suggesting that these genetic alterations are important in the development and progression of melanoma. To examine whether allelic loss is of prognostic importance in melanoma, disease-free survival was related to loss of heterozygosity on 6q, 9p and 10q in 83 individuals with sporadic primary cutaneous melanoma. Loss of chromosome arms 6q and 10q were each significantly associated with a poorer clinical outcome (P=0.013 and P=0.001 respectively). In a subgroup of 41 subjects whose primary tumours were allelotyped, the fractional allelic loss (FAL) at 39 autosomal arms also significantly correlated with disease-free survival (P=0.013), with an increase in FAL associated with a poorer outcome; this association remained significant when controlled for tumour thickness (P=0.035). In addition, a greater proportion of cells were immunopositive for Ki67 antigen, p53 and p21WAF1 protein in the primary melanomas than in the benign melanocytic nevi, however, only p53 over-expression was significantly associated with improved survival (P=0.041).

A gene for monilethrix is closely linked to the type II keratin gene cluster at 12q13.

Monilethrix is an uncommon hereditary disorder of hair and nail which produces hair fragility and a variable alopecia. Many of the dystrophic hairs have a unique beaded morphology. Ultrastructural changes suggest a defect in the microfilament structure of the cortex of the hair shaft,and hence the cysteine-rich trichocyte keratins are candidate genes. Here, in two families with autosomal dominant monilethrix, we have excluded linkage to the type I keratin gene cluster on chromosome 17q, but show that the disorder is closely linked to the type II keratin cluster on 12q, where genes for basic trichocyte keratins are found. The combined maximum lod score for D12S96 was 12.27 at theta=0.0. This is the first mapping of a primary human hair disorder and the first evidence implicating a defect of the word 'hard' keratins of hair and nail disease.

Mapping of monilethrix to the type II keratin gene cluster at chromosome 12q13 in three new families, including one with variable expressivity.

Monilethrix is an autosomal dominant disorder chiefly affecting hair. The degree of hair dystrophy is highly variable, as is the presence of additional features, such as follicular keratoses. In three British families of monilethrix, linkage has recently been reported to the type II keratin gene cluster at chromosome 12q13, and it has been suggested that the disease is due to a defect in the hard keratins of hair and nail. If monilethrix is a keratin disorder, we would predict that some pedigrees might map to the type I keratin gene cluster on 17q where hard keratin genes are also found. We have now studied clinically and by linkage analysis three new and unrelated pedigrees from England, Scotland and Spain, the first of which showed a variant phenotype. In this family the disease was expressed in four of 12 cases only as a follicular-keratosis of the neck, elbows and knees, and without clinical or historical evidence of hair anomalies; non-penetrance in an obligate carrier was also observed. In all three families, we have established linkage to a series of microsatellite markers at the type II locus at 12q13 (Zmax = 6.34 at theta = 0.00 for D12S368) and have excluded linkage from the type I keratin gene cluster on 17q. It remains probable that monilethrix is a disorder of hard keratins, but at present there is no evidence that it is due to defects in type I keratins.