Abstract for presentation at 11th International Congress of Human Genetics

Making sense of dHPLC results using knowledge of ancestral haplotypes for BRCA1

  • Dr Elizabeth Spriggs, Molecular Diagnostic Laboratory, Health Sciences Centre, Departments of Biochemistry & Medical Genetics and Pediatrics & Child, Canada
  • Dr Patrick Frosk, Molecular Diagnostic Laboratory, Health Sciences Centre, Winnipeg, Canada
  • Ms Susan Burgess, Molecular Diagnostic Laboratory, Health Sciences Centre, Winnipeg, Canada
  • Mr Rick Jobse, Molecular Diagnostic Laboratory, Health Sciences Centre, Winnipeg, Canada

    • Mutations in BRCA1 and BRCA2 account for about 40% of families with an inherited susceptibility to breast and/or ovarian cancer. Individuals, meeting local criteria for a full screen of the BRCA1 and BRCA2 genes, undergo testing using a combination of genomic PTTs and dHPLC analysis. This analysis covers the coding region and intron-exon boundaries for both BRCA1 and BRCA2.
    There are concerns that a polymorphism in an amplicon may mask a true mutation when analyzed by dHPLC for heteroduplex formation. While profiles of patient samples heterozygous for a specific mutation are expected to produce a characteristic pattern, these profiles are not necessarily unique. Hence a pattern produced by a sample heterozygous for a common polymorphism may be very similar to that produced by a patient sample heterozygous for a unique mutation within the same amplicon, leading to the misidentification of a unique mutation as a polymorphism. There are several polymorphisms in BRCA1 and six occur at a frequency much greater than 1%. As a result, over half of the patient samples screened are routinely identified as being heterozygous for at least one polymorphism, leading to further molecular analysis.
    Given that the BRCA1 gene is in a region of strong linkage disequilibrium, the possibility of using ancestral haplotypes was explored as a means to minimize the need for further analysis. Eighty-six patients were tested for six polymorphisms using DNA sequencing or PCR with restriction enzyme digestion: c.548-58delT, c.1067A>G, c.3548A>G, c.4308T>C, c.4837A>G, c.4956G>A. Only five distinct haplotypes were identified in this patient population and two accounted for 89% of the alleles: the “wild-type” haplotype accounted for 56% of the alleles and the haplotype with four polymorphisms [c.548-58delT, c.3548A>G, c.4308T>C, c.4837A>G] co-inherited accounted for 33% of the alleles. The c.1067A>G polymorphism is found in isolation whereas the c.4956G>A polymorphism is only found on the allele with the four polymorphisms. With this knowledge, we have defined an algorithm to determine which patients, if any, require further investigations based on our understanding of the ancestral haplotypes. A similar study is underway for BRCA2.

    Conference Organiser - ICMS Pty Ltd