Program I will combine data on the changing clinical and dermoscopic classification over time, with genetic information (germline and somatic DNA collected from saliva and naevi).
To study the associations between clinical and dermoscopic phenotype, and genotype.
To determine the contribution of novel candidate pigmentation and naevogenic genes to naevus count, morphology and dermoscopic patterns.
To provide information to allow the development of risk prediction models for melanoma based on pigmentation and naevogenic phenotype.
To perform whole exome sequencing of up to 30 melanoma patients and 30 controls to search for genetic polymorphisms associated with contrasting naevus phenotype traits.
Objectives: Using data from the CRE Core Study, as well as additional data from an ongoing case-control study of naevi (Brisbane Naevus Morphology Study (BNMS), NHMRC Project Grant APP1062935), Program I will:
Document the natural history of naevi over the three years of the Core Study.
Characterise the phenotypic and genotypic factors associated with naevus growth and evolution.
Assess the association between genetic factors and the number, size, distribution, profile, colour; and overall dermoscopic pattern of naevi.
Proposed research: To undertake detailed clinical and genetic studies, using established study procedures.
Data on melanoma risk factors and dermatologic characteristics (naevus number, size, distribution, profile, colour and overall dermoscopic pattern), and DNA suitable for genetic association analysis, will be available from 1200 participants (as part of BNMS), including a high-risk melanoma group (n=600; personal or family history of melanoma), and a matched general population control group (n=600). Significant naevi (>5 mm) will be classified on their predominant dermoscopic pattern and will be grouped by age decade (20-69 years) and body site. The signature naevus pattern (pattern seen in >40% of naevi) will be calculated for each individual. For each naevus, we will identify its dermoscopic sub-class and assess which sub-classes are most highly associated with melanoma.
Pigmentation and naevogenic gene sequencing and high density SNP genotyping is routinely performed on genomic DNA prepared from saliva samples as part of the BNMS. SNP screening in genes or gene regions identified through recent GWAS and whole exome sequencing studies, to assess their contribution to pigmentation and overall clinical and dermoscopic naevus signature patterns, will be performed in our collections. We will also perform high density SNP genotyping which currently consists of a total of 500,000 SNPs (240K exome marker and 240K tagging SNPs) suitable for a European ancestry population. In total we will generate SNP genotypes from approx. 1000 cases and 600 controls.
Exome sequencing of extreme naevus phenotypes. We propose to perform whole exome sequencing of up to 30 melanoma patients and 30 matched controls with extremes of naevus phenotypes to discover genetic polymorphisms that may underlie such differences. The output from these experiments will be a database of exon sequences suitable for bioinformatic analysis.
Statistical testing of SNP associations and gene-gene interactions. A sample of 600 melanoma cases and 600 controls gives 80% power (critical p=0.05) to replicate an existing locus with genotypic relative risk of 1.4. Power to detect quantitative trait loci is better: we obtain 80% power to replicate a locus explaining only 0.6% of the variance of naevus count (critical p=0.05), and 3.25% of variance of a locus at a genome-wide significant level (critical p=5x10-8). The CRE team has experience in analysis of whole genome sequence data, and will lead the testing of individual variants as well as burden tests in our known risk loci. The primary phenotypes tested for SNP association will be melanoma status, naevus counts (total, regional) and dominant dermoscopic naevus type. We will also test for novel associations with other melanoma risk phenotypes such as skin, hair and eye color. For traits that are categorical or ordinal in nature, we will fit log-linear and ordinal logistic models, including interaction terms testing gene by environment and gene by gene (epistatic) interactions.
Expected outcomes: The total number of naevi is already recognized as the most important risk factor for melanoma after pigmentation phenotype, and is thought to reflect one of the two fundamental biological pathways to melanoma. However, the understanding of this relationship is still superficial. Combining clinical and dermoscopic classification of naevi with genetic information will provide new avenues for better early prediction and diagnosis of melanoma and translatable individual risk estimates.
Significance and innovation: This Program will improve our understanding of both the progression of naevi towards melanoma, as well as the genetic basis of the disease. Such information will have direct application to clinical practice by improving the potential of melanoma early detection and diagnosis.