Abstract

Abstract Homologous recombination (HR) repair defects are of potential therapeutic relevance in a variety of different cancers. Numerous studies have investigated the rate of BRCA1/2 mutations in triple negative breast cancer, and current clinical studies are investigating the efficacy of agents targeting HR deficiency in this breast cancer subtype. A more comprehensive assay for HR defects might expand the number of patients likely to benefit from these therapies, and may expand their utility to other breast cancer subtypes. Recently three DNA-based measures of HR deficiency (HRD) have been developed based on whole genome tumor LOH profiles, telomeric allelic imbalance, or large-scale state transitions. These will be referred herein as HRD-LOH, HRD-TAI and HRD-LST respectively. All 3 scores are highly correlated with defects in BRCA1/2 and other pathway genes in breast or ovarian cancer, and are associated with sensitivity to platinum agents. 213 invasive breast tumor samples and matched normal tissue blocks were obtained from 3 commercial vendors. The samples were selected to contain approximately equal numbers of all subtypes of breast cancer as defined by IHC analysis of ER, PR, and HER2. BRCA1/2 mutation screening and BRCA1 promoter methylation analysis was performed, and genome wide SNP profiles were generated. These data were used to calculate HRD-LOH, HRD-TAI, and HRD-LST scores. Somatic and germline BRCA1/2 mutations were detected in all subtypes of breast cancer at significant levels with the total mutation frequency ranging from 7.8 – 16.4% depending on subtype. In contrast BRCA1 promoter methylation was confined almost exclusively to triple negative tumors (19.7%). Overall BRCA1/2 deficiency ranged from approximately 10% in ER+/Her2- tumors up to approximately 36% in triple negative tumors. HRD-LOH, HRD-TAI, and HRD- LST scores have previously been shown to be highly significantly associated with BRCA1/2 status in both breast and ovarian cancer. In this dataset all 3 scores showed significant association with BRCA1/2 status for the entire dataset, in addition significant association was observed between the scores and BRCA1/2 status in each of the individual tumor subtypes. The 3 scores were found to be highly correlated with one another, but all 3 were still significant in multivariate analysis. This dataset is not of sufficient size to determine which of these scores is best able to identify BRCA1/2 deficient tumors. It is likely that a combination of the 3 scores will prove to be the most robust predictor of HR deficiency. This study has demonstrated significant levels of BRCA1/2 deficiency across all subtypes of breast cancer. All 3 HR deficiency assays showed significant association with BRCA1/2 deficiency regardless of breast cancer subtype. The 3 scores are highly correlated, but also additive and a combination of all 3 is likely to provide the best predictor of HR deficiency. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P6-05-10.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call