A Complex Genetic Basis for Breast Cancer Subtypes.
Thomas P Stricker, Christopher D Brown, Vanessa Montoya, Megan McNerney, Ralf Kittler, Subhradip Karmakar, Robert L Grossman, Kevin P White. University of Chicago, IL; University of. Illinois at Chicago
Background: Breast cancer is a heterogenous disease, with subtypes that have different morphologies, gene expression profiles, biological behaviors and treatments. Cancer is also a genetic disease, and both inherited and acquired variants have been implicated in carcinogenesis. However, the genetic contribution to disease heterogeneity is largely unknown. We hypothesize that tumors with similar behavior and morphology will have a shared set of genetic variants, distinct from genetic variants that define other subtypes.
Design: We used next generation sequencing of RNA to identify expressed single nucleotide variants (SNVs) within 11 ER+ and 15 TN breast cancers. Participants in the study span a range of ages, tumor stages, and race. We have designed a bioinformatic pipeline to accurately genotype these samples, as well as quantify gene expression level and isoform usage. Genotypes were annotated based on type (synonymous, non-synonymous or in untranslated regions) and, using previously established databases, whether alleles had previously been seen in the human population (termed common) or are novel alleles (termed rare). Finally, the predicted deleteriousness of each variants was assessed using a comparative evolutionary genomic methods.
Results: To determine whether shared single nucleotide variants (SNVs) classify distinct genetic subtypes of breast cancer, we examined genetic variation in these samples by principal component analysis. Common neutral polymorphisms segregate tumors by genetic ancestry, mixing the phenotypically defined ER+ and TN tumor subtypes. However, rare deleterious variants, which potentially contribute directly to cancer phenotypes, generate two genetically distinct groups, corresponding to ER+ and TN tumors. Interestingly, genotyping of normal tissue for these patients showed that only 2.3% of these rare deleterious alleles that distinguish breast cancer subtypes are somatic mutations. Many of these rare deleterious alleles are in genes in cancer related pathways and processes, such as apoptosis and mitosis.
Conclusions: Thus, ER+ and TN tumors are genetically distinct diseases and these genetic distinctions are driven by rare inherited deleterious alleles. These results suggest the possibility of using genotypes to guide prevention and screening prior to the development of disease.
Tuesday, March 1, 2011 1:00 PM
Poster Session IV # 26, Tuesday Afternoon