Author(s): Mark Landers, Rhonda Meredith, Jerry Lee, Yipeng Wang, Byung-In Lee, Joseph Monforte; AltheaDx, San Diego, CA
Sub-category: Triple Negative Breast Cancer
Category: Breast Cancer – Triple-negative/Cytotoxics/Local Therapy
Meeting: 2012 ASCO Annual Meeting
Abstract No: 1058
Citation: J Clin Oncol 30, 2012 (suppl; abstr 1058)
Background: TNBC is an aggressive subtype of breast cancer accounting for 10-15% of all cases. TNBC tumors (ER-/PR-/HER-) are more common in patients with BRCA mutations. BRCA mutations leading to homologous DNA repair deficient tumors enhance the efficacy of chemotherapy and PARP inhibitors. BRCA mutations have been identified in 20% of patients without family history. Identification of germline and somatic BRCA mutations in unselected patients could increase the number of patients who benefit from these therapies. Determining BRCA mutational status from FFPE and fresh frozen specimens may enable clinical studies in these patient populations.We describe the development of an NGS BRCA mutational assay compatible with FFPE and fresh frozen samples using tumor/adjacent normal matched tissues. Methods: Matched samples were purchased from Cureline. gDNA was isolated by lysis/column purification (Qiagen) and enriched for BRCA exons/flanking regions (Halogenomics Selector). Fragment libraries were constructed (Ion Torrent frag express) and prepared for sequencing by emPCR (Ion Torrent Template Xpress). Libraries were sequenced for 65 cycles (Ion Torrent PGM) yielding 2-4M reads/sample. Variants were called from tMAP aligned reads by GATK and VarScan. Overlapped exonic variants were filtered by p-value (<0.0001) from VarScan. Results: In the first patient set normalized average depth of BRCA exon coverage was 64X and 72X per 150K reads in FFPE and fresh frozen tissues respectively covering 95-100% of target. hg19 alignment rates varied between 97-99% across all samples. Similar numbers of variants were called in both FFPE (12) vs. fresh frozen (13) with a corresponding mean duplicate removed depth of coverage of 23.3X and 42.4X at the called positions. 10/13 calls in fresh frozen overlapped with those in FFPE. A tumor specific somatic frameshift insertion in BRCA2 was detected in both FFPE and fresh frozen tissues. Conclusions: Results indicate that NGS mediated BRCA mutational analysis demonstrates equivalent utility in both FFPE and fresh frozen tumor samples although more sequencing reads are required to produce equivalent depth of coverage starting from FFPE samples.