Abstract P3-04-07: Functional BRCA-pathway defects in sporadic breast cancers rarely show a significant reduction in BRCA-pathway proteins

Abstract

Abstract Background: Array comparative genomic hybridization (aCGH) studies have revealed that a subgroup of sporadic breast cancers have a similar genetic landscape as BRCA-mutation carriers, characterized by widespread large-size deletions across the genome. We have previously shown that a functional assay of homologous recombination, using radiation-induced RAD51 foci, correlates closely with the aCGH findings. The prevalence of functional defects in the BRCA-pathway is estimated to be ∼25%, and present in all subtypes of breast cancer regardless of estrogen receptor or HER2 status. Here we sought to investigate the candidate genes and proteins that may drive this BRCA-like phenotype in sporadic breast cancers. Methods: Fresh human breast cancer samples were obtained to query the BRCA-pathway, under an IRB-approved study. As previously, the ability to assemble RAD51 foci was the defining functional assay. A 4-day growth assay assessed whether BRCA-pathway defective tumors were sensitive to crosslinking agents. Nanostring-based gene expression analysis was performed to interrogate the transcript level of key players in the BRCA-pathway. Massively parallel whole exome sequencing analysis is in progress, since tumor and normal germline DNA was obtained for all these samples. Results: Previously we found the incidence of BRCA-pathway defective tumors was 18/71 (25.3%). The current analysis relates to samples obtained in the last 3 years, including 42 patients in the initial series and 18 additional samples, for which DNA, RNA and immunohistochemistry analyses were successfully performed. Known mutation carriers were excluded from the study. RAD51 focus induction revealed a bimodal distribution: no induction (0.75-1.25) or an induction by > 75%. 17 tumors were repair-defective and 43 were proficient, consistent with our previous analysis. 16 of the 17 repair-defective tumors had the lowest IC50 to mitomycin C, further supporting the functional phenotype classification. Since many of the BRCA-pathway proteins are cell cycle regulated, we determined that cell proliferation markers (e.g. Ki67, PCNA) were not different between the repair defective and proficient tumors. Nanostring gene expression analysis revealed no difference in the expression levels of BRCA1, BRCA2 and RAD51. Since BRCA1 protein recruitment was defective in the majority of repair-defective tumors, we also analysed RNF8, RNF168, RAP80, ABRAXAS, MERIT40 or BRCC36 and found no abnormality. The methylation status of the BRCA1 promoter is being tested for all cases, however our results suggest that reduced mRNA and protein levels of BRCA1, BRCA2 or RAD51 are not a major cause of BRCA-pathway defects. Conclusions: The cause of BRCA-pathway defects in sporadic breast cancer cannot be explained by a reduction in mRNA or protein for members of the BRCA-pathway in the vast majority of cases. Methylation of BRCA1 to produce a reduction in mRNA sufficient to cause a repair defective phenotype is likely to be relatively uncommon. Whole exome analysis may reveal a common explanation for this biologically and clinically relevant phenotype. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P3-04-07.