Emerging paradigms in cancer genetics: some important findings from high-density single nucleotide polymorphism array studies.

High-density single nucleotide polymorphism (SNP) mapping arrays have identified chromosomal features whose importance to cancer predisposition and progression is not yet clearly defined. Of interest is that the genomes of normal somatic cells (reflecting the combined parental germ-line contributions) often contain long homozygous stretches. These chromosomal segments may be explained by the common ancestry of the individual's parents and thus may also be called autozygous. Several studies link consanguinity to higher rates of cancer, suggesting that autozygosity (a genomic consequence of consanguinity) may be a factor in cancer predisposition. SNP array analysis has also identified chromosomal regions of somatic uniparental disomy (UPD) in cancer genomes. These are chromosomal segments characterized by loss of heterozygosity (LOH) and a normal copy number (two) but which are not autozygous in the germ-line or normal somatic cell genome. In this review, we will also discuss a model [cancer gene activity model (CGAM)] that may explain how autozygosity influences cancer predisposition. CGAM can also explain how the occurrence of certain chromosomal aberrations (copy number gain, LOH, and somatic UPDs) during carcinogenesis may be dependent on the germ-line genotypes of important cancer-related genes (oncogenes and tumor suppressors) found in those chromosomal regions.