Somatic mutations in early onset luminal breast cancer

Breast cancer arising in very young patients may be biologically distinct; however, these tumors have been less well studied. We characterized a group of very young patients (≤ 35 years) for BRCA germline mutation and for somatic mutations in luminal (HER2 negative) breast cancer. Thirteen of 79 unselected very young patients were BRCA1/2 germline mutation carriers. Of the non-BRCA tumors, eight with luminal subtype (HER2 negative) were submitted for whole exome sequencing and integrated with 29 luminal samples from the COSMIC database or previous literature for analysis. We identified C to T single nucleotide variants (SNVs) as the most common base-change. A median of six candidate driver genes was mutated by SNVs in each sample and the most frequently mutated genes were PIK3CA, GATA3, TP53 and MAP2K4. Potential cancer drivers affected in the present non-BRCA tumors include GRHL2, PIK3AP1, CACNA1E, SEMA6D, SMURF2, RSBN1 and MTHFD2. Sixteen out of 37 luminal tumors (43%) harbored SNVs in DNA repair genes, such as ATR, BAP1, ERCC6, FANCD2, FANCL, MLH1, MUTYH, PALB2, POLD1, POLE, RAD9A, RAD51 and TP53, and 54% presented pathogenic mutations (frameshift or nonsense) in at least one gene involved in gene transcription. The differential biology of luminal early-age onset breast cancer needs a deeper genomic investigation.


Exome sequencing
50 ng gDNA from tumor and blood were used to prepare exome libraries using the Illumina Nextera Rapid Capture Expanded kit (Illumina, Inc., San Diego, CA -USA/FC-140-1004) according to the following steps: 1. Tagmentation of Genomic DNA: the genomic DNA was tagmented (tags and fragments) using Nextera transposome. The Nextera transposome simultaneously fragments the genomic DNA and adds adapter sequences to the ends, allowing amplification by PCR in subsequent processes. Next the tagmented DNA was purified.
2. First PCR Amplification: the purified tagmented DNA was amplified via a limited-cycle PCR program. Barcodes index 1 (i7) and index 2 (i5) needed for sequencing, as well as common adapters (P5 and P7) required for cluster generation and sequencing were then added. After wards, beads were used to purify the library DNA and remove unwanted products.
3. First Hybridization: the DNA library was mixed with capture probes to targeted regions of interest. Furthermore, the samples were organized in two pools of eight samples each, using different indices.
4. First Capture: the probes hybridized to the targeted regions of interest were captured using streptavidin beads. Two heated wash procedures were used to remove nonspecific binding from the beads. The enriched libraries were then eluted from the beads and prepared for a second round of hybridization. 5. Second Hybridization: the eluted DNA libraries from the first enrichment round were combined with additional capture probes to targeted regions of interest. This second hybridization was required to ensure high specificity of the captured regions.
6. Second Capture: the probes hybridized to the targeted regions of interest were captured using streptavidin beads. Two heated wash procedures were used to remove nonspecific binding from the beads. The enriched libraries were then eluted from the beads and prepared for a second round of hybridization. At the end of this process, captured library DNA was purified using beads.
7. Second PCR Amplification: the captured libraries were amplified via a limited-cycle PCR program. After wards, beads were used to purify the enriched library DNA and remove unwanted products.
The exome capture libraries were quality-controlled using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA), and quantified using the KAPA SYBR FAST qPCR Kits (Kapa Biosystems, Wilmington, MA, USA, part #KK4602) prior to cluster generation on an Illumina cBot. Pooled libraries were loaded on six lanes of one flow cell, using 2x100bp paired-end reads and sequenced on an Illumina HiSeq 1000 platform, targeting 201,121 exons, comprising 62 mega base pairs (Mbp) of the genome and with a median of 95.3% of targeted bases covered at least 30-fold across the sample set.    synonymous means the mutation rates were calculated for all SNVs whose function was labeled as anything except synonymous.