PD03-06: Basal-Like Breast Tumors Are Associated with Frequent Micro Genomic Aberrations in Important Cancer Genes.

Abstract

Abstract Background: The basal-like subtype has been shown to have increased genomic instability relative to other breast cancer subtypes and this may reflect processes involved in the pathogenesis of this disease. Basal-like tumor specific regions of aberration have previously been identified by large-scale genomic analyses in terms of cytobands and specific genes such as RAD17, RAD50 and chr 10p. There is, however, some evidence that these subtype associated genomic events extend to smaller scale genomic aberrations as well. A previous study (Saal et al, 2008) described one such event in the Basal-like SUM149 breast cancer cell, which contains a small amplification in exon 2 of the PTEN gene. We suspect that primary breast tumors may harbor similar events, and that these events may have previously been undetectable using the current generation of whole genome copy number arrays. In this study, we sought to identify small regions of genomic change that might contribute to the tumor development of primary breast tumor by high density, gene centric CGH arrays on both cell lines and primary tumors and then validate the presence of these regions using mRNA-seq. Methods: A custom Agilent tiling array for CGH (244,000 probes at a 200bp tiling resolution) was created to identify small regions of genomic change, and was focused on previously identified basal-specific regions and genes. One important application of this array is to refine common regions of DNA copy number variations in basal-like breast tumors and discover small-scale genomic changes unable to be detected by lower-resolution genome wide arrays. Tumor genomic DNA from 100 patients and 10 breast cancer cell lines was labeled and hybridized to these arrays. Log intensity ratio plots were generated for each sample for each gene (128 different genes tested) and analyzed for copy number changes. mRNA-seq was also performed on a subset of these samples (14 tumors, 4 cell lines). Results: We confirmed the presence of the PTEN micro-amplification in SUM149 and by using mRNA-seq, also showed that this resulted in loss of expression of all exons downstream of this event. In addition to large-scale amplifications and deletions, our data also showed that primary tumor breast cancer genomes frequently contain areas of small-scale copy number gains and losses, termed microevents. The genomic landscape of the basal-like subtype also exhibited many of these microevents, as basal-like tumors have the highest incidence of these events. Conversely, Luminal A tumors, the subtype with the most stable genome, exhibited the lowest numbers of microevents and lowest numbers of gross copy number aberrations. Our results also indicated that RB1 and PTEN, two genes known to be involved in basal-like breast tumor formation, showed a particularly high incidence of these microevents. These microevents may alter expression of the involved gene as well, as suggested by data from microarray and mRNA-seq studies. Conclusion: Using a high probe density, gene-centric aCGH microarray, we present evidence of small-scale genomic aberrations that may contribute to gene inactivation, and thus, genomic instability and tumor formation through a mechanism not detected using conventional copy number analyses. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr PD03-06.